Patterning is a process of subdivision: a group similar cells being partitioned into spatially and/or temporally distinct subsets. Morphogenesis is a process of shape change: cells collectively transforming themselves into a new tissue-scale forms. Both of these happen repeatedly in a developing animal, intertwining to produce ever more structural complexity as a creature goes from fertilized egg to mature adult.
How a sheet of cells is subdivided into unique geometric arrangements
SUMMARY: In multicellular organisms, it is common to find tissues with characteristic patterns (e.g. stripes or spots). For most of these, the local rules that generate those patterns are a mystery. A particularly challenging example is the insect wing. For many species, the “veins” on the wings have predictable geometric features, yet there is also unique pattern of veins on every individual wing. This project used a large dataset of wings and a simple computational model to provide insight into how veins are patterned.
Our goal was to better understand the stained-glass like patterns that make insect wings so captivating. See, for instance, this dragonfly wing:
Insect wings have “veins”. These are the dark lines in the stained glass. Unlike the similarly-named tubes that carry our blood, these are thickened struts that strengthen the wing and serve many other functions.
To a first approximation, wing veins come in two broadly distinguishable types: primary veins have the same overall pattern on left and right wings of an insect, and across individuals of that species. Secondary veins are different—they have a unique pattern on every single wing. They are like fingerprints.
We know a lot about wing development in the fruit fly Drosophila melanogaster, and very little about other species. But, incidentally, fruit flies only have primary veins. As a consequence, we know a lot about the arrangement and development of primary veins, but secondary veins are largely a mystery.
These fingerprint-like patterns are found in many distantly related groups of insects. Here, for instance are a lacewing and a grasshopper:
We collected 100s of images of wings, focusing on dragonflies and damselflies (order: Odonata). Some were original micrographs and others were published wing tracings from the natural history literature. This is just one of many examples of how detailed, descriptive natural history work forms the foundation of “question-driven” biological research.
We developed automated software tools to computationally measure all sorts of quantitative features of wing veins, including their angles, lengths, and connectivities. Across the entire dataset, we have 100,000+ vein segments.
We found that a useful way to characterize a wing is by the distribution of sizes and shapes of the polygons made by the veins. In dragonflies, e.g., every wing is different, but their polygons have the same overall distributions.
We also learned that different groups of insects have characteristic distributions of shapes, and there are clear allometric scaling relationships in vein features among closely related species.
How might these patterns of veins emerge as a wing develops? We propose a geometric model for how the veins are patterned, based on evenly spaced signaling centers that secrete an inhibitory cue:
The model is intentionally made very simple, but it nonetheless can recapitulate patterns that look remarkably life-like:
Our model can also capture many of the basic wing features of some distantly related insect species. It is a falsifiable hypothesis for how development proceeds in wings that have secondary veins. I hope future researchers will put its assumptions to the test.
Building a body, from simple to complex
SUMMARY: The two-spotted field cricket, Gryllus bimaculatus, is an increasingly popular model species. We used a combination of brightfield timelapse microscopy, confocal microscopy, 3D reconstructions, and measurements of egg dimensions to comprehensively describe the morphogenesis and anatomy of the growing cricket embryo.
Extensive research into Drosophila melanogaster embryogenesis has been the basis for most of our understanding of insect developmental processes. However, Drosophila development is likely quite different from that of the ancestral arthropod. By studying the development of other model species, we can uncover generalizable principles of insect development and reconstruct patterns of evolution.
The cricket has a sophisticated functional genetic toolkit, including RNAi and CRISPR-mediated genome modification. However, the existing cricket embryonic staging system was fragmentary and it was based on morphological landmarks that are not easily visible on a live, undissected egg. To address this problem, I described a set of “egg stages.” These stages serve as a guide for identifying the developmental progress of a cricket embryo from fertilization to hatching, based solely on the external appearance of the egg.
I illustrated the visible features of each stage, which were time-matched to close-up micrographs of the embryo after it had been removed from the egg (see some example images here). This resource will facilitate further studies on insect development and serve as a useful point of reference for other studies of wild-type and experimentally manipulated insect development.
I also recorded some time-lapses of cricket embryogenesis:
The above GIF shows katatrepsis, the period of development when many insects do a backwards somersault in the egg. The GIF is an excerpt from this time-lapse of the first eight days of cricket development.
A longer version of this time lapse with detailed annotations has been published as two separate supplemental movies in Donoughe and Extavour (Developmental Biology 2016). Click on the screenshots below to access the movies:
You can download the movies directly here (first 3 days, next 5 days). You can also see the movies and other supplementary information at the article’s page.
Setting aside special cells
SUMMARY: Primordial germ cells are the cells that turn into sperm or egg. Thus, they are the only lineage of cells in a body that give rise to the next generation. For the first time, we uncovered the molecular basis of germ cell formation in an insect other than the fruit fly, discovering that the process is strikingly similar to that of mice.
Two modes of germ cell specification are known in animals. In fruit fly, zebrafish, nematode, and frog, germ cells are specified by maternally supplied germ-line determinants. In mice, by contrast, embryonic cell-to-cell signaling induces cells to become germ cells. Molecular evidence for the latter – inductive germ-line specification – had previously been demonstrated only for the mouse.
In this research, we provided functional evidence for inductive germ cell specification in an invertebrate, by showing that Bone Morphogenetic Protein (BMP) signaling is required for the establishment of embryonic germ cells in a cricket. BMP pathway knockdown caused reduction or loss of germ cells, and elevated levels of BMP signaling caused supernumerary and ectopic germ cells to form. Therefore, like in mice, BMP signaling is both necessary and sufficient to induce germ cell specification in crickets. Thus, BMP-based germ cell induction may be a shared ancestral mechanism in animals.
After this work was published, Lochab and Extavour (Developmental Biology 2017) reviewed the role of BMP signaling in germ cell biology across animals, while Nakamura and Extavour (Development 2016) showed that the cricket ortholog of the gene blimp-1 acts downstream of BMP signaling, much like it does in mice.