BIOL 606 Home

Evolution of Insect Wings

Presentation by: Grant McIntyre

Lecture 5, February 9th, 1998 - BIOLOGY 606
Rapporteur: Gavin F. Hanke
Focal Paper: Averof, M. and S.M. Cohen. 1997. Nature, 385: 627-630.


The origin of insect wings is a very puzzling issue. Most work on the evolution of wings has focused on the dipteran genus Drosophila , with conclusions extrapolated to other insect taxa.

Two hypotheses dominate arguments on the origin of insect wings. The first and perhaps least supported, is that wings evolved from thoracic paranota that were enlarged over time. Enlarged paranota may have acted like parachutes (Packard 1898) such that insects could land with good attitude and be better prepared to avoid predators. These expanded plates later developed articulations that were required for active, flapping flight. Mueller (1987) first proposed this paranotal origin for insect wings. The second, more supported hypothesis, suggests that gills have become elaborated into fully functional control surfaces in an aquatic, ancestral stage. These control surfaces were modified (increased rigidity and reduced weight) to become functional in air. Oken (1811) first suggested that external gill lamellae of insects were converted over time to functional wings. Stoneflies show a functionally intermediate condition where wings were used as sails to cross the water surface and this behavior appears as transitional stage between the use control surfaces (rigid gills) in water and flapping flight in air. Relatively recent work by Wigglesworth and Kukalová-Peck during the 1970s, provided fossil evidence to suggest that wings evolved from modified exites of insect coxae.

Insect wings are composed of two opposed layers of cuticle and epidermis, supported by veins that are filled with haemolymph, neurons and tracheae. This basic wing structure is similar to the structure of gills of immature insects. The similarity of these two structures suggests homology, with further support from the presence of chemo- and stretch receptors in both leg and wing tissues.

Researchers believe that wing articulations are homologous between taxa even though the structure of wings themselves varies among taxa. Muscles attach wings to the thorax and the coxa of the leg in the same thoracic segment. Similar muscles and tracheae are associated with gills of larval insects, and these abdominal gills are in a similar position (dorsolateral to the body) as wing primordia developing on the thorax.

Unfortunately insect fossils are relatively rare, because of their lack of 'hard' parts, and so there is little preserved of the early evolution of winged insects. Grant stated that the interpretation of insect fossil record is also subject to biases of each researcher, limiting consensus on the identity of structures and on evolutionary patterns.

Averof and Cohen (1997) provided a valuable test case for using genetic information to support hypothesized evolutionary transitions. They used imunoflourescence techniques to determine regions where gene products (evidence for gene expression) were present in developing limbs of Artemia and crayfish, and these results were compared with similar observations on Drosophila . The distal-less (Dll ) gene determines the proximal-distal axis of both wings and legs and so is of little use in supporting wing origin hypotheses. The genes nubbin (pdm ) and apterous (ap ) are expressed in the epipod of Artemia and crayfish, showing distinct separation of leg and epipod tissues. These results paralleled information presented on gene expression in Drosophila legs and wings. The research presented shows that there are several shared gene sequences that influence development of both wings and legs. Researchers now must examine developing gills for pdm and ap expression as additional support for gill-wing origins. Recent evidence suggests that pdm also contributes to orienting the proximal-distal axis of developing wings. Strong evidence supporting the gill-wing hypothesis stems from observations that developing Drosophila wing tissue buds off the dorsal portion of a early leg primordium, where gills are attached in crustaceans (Crayfish), and this presumptive wing tissue migrates into a dorsolateral position.

Currently, most researchers support the gill-origin hypothesis for insect wings. Grant noted that several problems and questions persist, even though the evidence presented by Averof and Cohen (1997) provides reasonable evidence supporting a genetic link between legs and wings. Questions that remained include: 1) Does similar genetic pattern represent common ancestry, convergence, or a constraint because of limited genetic material available? and 2) Can genetic, morphological and fossil evidence ever be integrated? Additional studies are needed on development of legs, gills and wings of 'primitive' extant insect taxa. Fossil material also needs re-examination to clarify past interpretations of thoracic structures.


Discussants: Mark Steinhilber and Corey Davis

Discussion opened with the question of whether genes are useful in establishing primitive vs. derived character states, since there are similar HOX gene sequences in vertebrates and insects. Genes shared between crustaceans and insects were believed to be important in the evolution of arthropod structures, but group members argued that these shared sequences are plesiomorphic within the Arthropoda and are of little use in phylogenetic analyses. The group considered gene expression that occurs later in ontogeny to indicate a more derived condition relative to gene sequences that control basic metabolic processes and anatomy. Students questioned whether gene expression in a specific location represents a derived condition.

Students agreed that additional studies are needed for a complete understanding of the role of genetic sequences, gene expression and genetic control on the ontogenetic development of organisms, and to properly understand evolutionary transitions. Caution was suggested in the interpretation of previous research, in that researchers may have focused on specific structures (gills and/or legs) and may have missed the fact that genes may alter many structures, perhaps totally unrelated to the structure(s) under study. The fact that nubbin (pdm) influences the proximal-distal axis of wings and the development of the nervous system, provides evidence that genes may be expressed in different structures at different times. All present agreed that anomalous results must be explained and not glossed over as in the paper by Averof and Cohen (1997).

The pdm and distal-less (Dll ) genes were not considered useful in determining the transition from gills to wings. Students agreed that researchers need to look for evidence of gene expression in wings that is not expressed in thoracic armor, and that researchers should look for expression of similar genes in wings and gills, not just wings and legs. To date, pdm is not expressed in the notum, but is in developing wings, supporting a limb origin for wings rather than from thoracic armor. Students and instructors agreed that the budding of presumptive wing tissue from the dorsal part of the limb bud provides strong evidence that wings evolved from legs at least in Drosophila and that more research is needed on other insects to verify this observation.

The importance of suppresser proteins and/or their genes in regulation of ontogenetic processes was stressed and that future research should focus on how suppresser proteins alter wing development. It was noted that the epipods of Artemia show arrested development, and it was suggested that the same factors used in limiting Artemia epipod growth may have been important in the development, and evolution in other arthropod taxa.

Students asked whether there actually is a structure specific gene in insects. HOX genes have generalized expression during development, and gene expression changes with time and region of the body, so therefore, would be poor choice if looking for structure specificity. Perhaps co-factors, that were mentioned in discussion, may combine with HOX genes to create more specific developmental response in tissues. It was suggested that study of these co-factors may be more valuable in determining the development of structures.

One fundamental question discussed was whether insect wings evolved in water, or in a terrestrial environment. If wings evolved from gills, then this implies that wings had an aquatic origin. There does not appear to be any structures preserved in the fossil material presented that could be considered transitional between wings and legs. The question arose as to whether early uniramians possessed exopodites, and if not, are wings independent of gills?

The most primitive of the extant winged insects have aquatic larvae, and the most 'primitive' of what are considered to be derived winged taxa have aquatic larvae. The aquatic life history phase is considered by Dr. Ball to be a secondary colonization of the aquatic habitat resulting from competition in terrestrial environments. In this scenario, wings evolved in terrestrial environments and are retained in taxa with aquatic larvae. Questions that surfaced following this hypothesis of a terrestrial origin of wings included: 1) Where do larval gills develop? 2) Are mayfly gills new structures or are they derived wings? 3) Did gills develop before wings?

There was a cautionary note forwarded, that we must be sure that the same enzymes and genes are being used in all taxa studied. Perhaps enzymes in one taxon are replaced by a functional analog in another. A final graphic example was presented, that Dll gene expression appears in tube feet of echinoderms, in insect appendages and in vertebrate limbs. Genes that code for external structures are believed to be more subject to convergence than are genes expressed in the internal environment only.

BIOL 606 Home