Fig. 1: Life cycle of a typical sea urchin, Arbacia punctulata.
The transformation from a single celled egg to a complex animal is one of the most amazing processes in biology. It is all the more remarkable because it can often proceed normally despite major perturbations. It is easy to think of development as a program in which each step depends crucially on the previous steps, but there are many examples in biology that violate this assumption. This project focuses on one particularly striking example — larval cloning in echinoderms — in which development appears to proceed from a very different starting point than the "normal" process which starts from a fertilized egg.
Larval cloning has many interesting connections to cell and developmental biology, physiology, ecology, and evolution.
This project is definitely doable by a single dedicated hobbyist, however it would be much easier to do with a group of people. It should make excellent, and varied class projects at the high school or college level.
Sea urchins have a fairly typical animal life cycle (Fig. 1) and developmental process (reviewed in  and  (for development), and ). Like most animals, they start as a single celled fertilized egg which divides numerous times to form a sphere of cells during embryonic development. The embryonic tissues then undergo a series of morphogenetic movements and cell differentiation processes to form a free-living larva which looks quite different from the adult, and lives in a different habitat. Then the larva undergoes metamorphosis, transforming itself into the juvenile urchin. The juvenile urchin grows into a sexually mature adult, which spawns eggs and sperm.
The larva has a complex body with a gut, skeleton, muscles, and a nervous system. It looks nothing like the adult sea urchin, but rather like a tiny swimming Sputnik, with its ciliated arms pointing forward (Fig. 1). The larvae feed on planktonic cells (algae), whereas the adult scrapes food off the bottom.
The juvenile urchin develops inside the larva, turned inside out. At metamorphosis, it flips itself inside out, shedding or breaking down much of the larval body, and forming a new mouth, and reorganizing everything else.
Except for flipping itself inside out at metamorphosis (which is actually surprisingly common), the major features of the urchin life cycle are typical of many animal taxa. However, over the last two decades it has become clear that the larvae of some sea urchins and many other echinoderms have alternative developmental pathways in which the larvae reproduce asexually (larval cloning) in various ways  . In sea urchins, buds form on the larva which resemble earlier embryonic stages (blastulae and gastrulae). The buds come off and develop into new, smaller, larvae. Larvae may also be able to clone by splitting themselves into pieces  .
Some of the interesting questions raised by echinoderm larval cloning include:
- What are the developmental mechanisms that allow the urchin to reorganize its axes and internal structure, and undergo morphogenesis, from multiple starting points?
- What are the cell/developmental mechanisms that allow the urchin to — apparently — step back to a much earlier stage in development? Do these mechanisms involve stem cells or dedifferentiation of differentiated cells?
- What are the selective advantages/disadvantages of larval cloning?
- What are the ecological consequences of larval cloning?
- How has it evolved within and among the echinoderms? And how is it related to budding, fission, twinning, or regeneration in other organisms?
A key requirement for studying these questions is understanding what triggers larval cloning. Only a few studies have investigated this in a handful of taxa   . In particular, in the sand dollar1 Dendraster excentricus, larval cloning is induced by pulses of high food (possibly a cue that conditions are particularly good for larval growth), or fish mucus (a possible cue for the presence of predators; small size after cloning appears to protect against fish predation ). In the purple urchin, Strongylocentrotus purpuratus budding is induced by culture at low pH (pH 7.2 - 7.9 vs 8.1), possibly as a stress response, however these buds fail to develop into new larvae .
Interesting things to test
These are just a few of many possibilities that would be technically feasible in a high school classroom or for a dedicated hobbyist.
Food pulses and fish mucus: McDonald and Vaughn  found that pulses of high food induced budding in the sand dollar Dendraster excentricus. Vaughn et al  also found that fish mucus induced cloning in Dendraster. A food pulse may provide a cue to the larvae that conditions favor increasing larval numbers at smaller size, so that larvae reproduce asexually instead of devoting resources to growth and juvenile development. Fish mucus may provide a clue that fish predators on plankton are abundant, so that smaller, more numerous larvae will have an advantage over larger larvae. Do Dendraster and Arbacia – distantly related species – respond in the same way to the same cues?
pH: Chan et al  suggested that low pH induced the formation of buds in Strongylocentrotus. Daily and annual pH changes are experienced by natural populations . Chan et al 2012 used a pH range that mimicked natural variation and changes in pH expected from to long term ocean acidification (over the next three centuries) due to CO2 emissions. The buds failed to survive, however, it is possible that transient low pH (e.g. over a day-night cycle) could induce the formation of buds that could undergo normal larval development.
Interestingly, recent reports  suggest that brief exposure to low pH (much lower than used in Chan et al's experiments) stimulates differentiated mammalian cells to become pluripotent stem cells (but see , which came out the same day this project page was started). Is it possible that something similar is happening in echinoderms, which are fairly closely related to vertebrates?
Membrane potential: Membrane potential is one of the most important regulators of cell properties. Notable examples from development include events of fertilization  and metamorphosis , however some recent studies suggest that membrane potential is an under-appreciated, but important, regulator of many developmental processes, including regeneration .
External potassium concentrations strongly influence membrane potential . Long term exposure to high or low potassium concentrations can have less predictable effects, but it would be fairly simple to do short term manipulations of membrane potential by briefly increasing or reducing potassium concentration in the seawater. One could transiently depolarize cells by adding a few milliliters of 0.5 M KCl to the media, and then transferring the larvae back to regular seawater after a few minutes. Alternatively, one could make artificial seawater (see ) with low potassium to transiently hyperpolarize the cell membrane.
Low Calcium: Calcium affects many, many cellular and physiological processes; however cell adhesion is one of the major processes dependent on calcium. While the mechanisms of larval cloning have not been studied in detail, they almost certainly involve cell and matrix rearrangements, and hence are likely to depend on cell adhesion. Therefore low-calcium media may influence the frequency/success of larval cloning. Calcium concentrations could be reduced by making one's own calcium-free artificial seawater  or by using calcium chelators. Larvae will fall apart if kept in calcium-free media for more than a few minutes, but they could be cultured in reduced calcium media.
Others: A large number of other possible experiments might be practical and interesting based on cell biological, developmental, or ecological considerations. For example, are there other factors such as salinity, light, or temperature that correlate well with algal blooms? If so, they might provide signals that larvae could use to decide whether to clone or continue allocating resources to larval growth and development of the juvenile.
A few major molecular pathways control which parts of the embryo become gut (endoderm) versus skin (ectoderm) versus muscle (mesoderm), and which parts become anterior versus posterior, or oral versus aboral . Certain affordable reagents (e.g. NiCl2, LiCl, and others) disrupt different signaling pathways, resulting in distinct, major defects in animal embryos. Certain common pollutants also disrupt some of these pathways . Note that one study on the Wnt/β-catenin pathway has been presented at a meeting .
For discussion of methods of obtaining and culturing urchin embryos and larvae see this article.
In principle, if one observes many larvae frequently enough, one could see larvae in the process of budding off new larvae. In principle, this would be the best way to see if a treatment causes larval cloning. However, budding is rare and unpredictable, and may be too quick to for direct observation of cloning to be a practical method for detecting triggers of cloning when in a classroom setting.
A less ideal, but easier, way to test whether a treatment affects the frequency of larval cloning is to count the number of larvae before and after exposing them to a treatment (or control) condition . Without cloning, the number of larvae should only decrease (due to larval death or loss during transfers). Therefore, if the number of larvae increases, this can be attributed to cloning. However, one must be very careful 1) to count correctly (over counts will appear as cloning), and 2) avoid transferring larvae into or out of dishes accidentally (they can get moved from one dish to another if they get stuck in the transfer pipette, or splashed from one dish to another). Furthermore, larvae have to be healthy enough that they do not die in culture. If larval numbers decline one could not attribute differences in larval numbers to cloning rather than differences in survival.
Assuming one uses the counting approach, rather than directly observing budding, one must consider a number of factors in experimental design. For example: how many larvae should one use, and how many dishes? Increasing the number of larvae per dish, and the number of dishes, makes the experiment more difficult, but increases the statistical power. Cloning is rare, so one needs many larvae per dish to detect it. One also needs multiple dishes per treatment to avoid pseudoreplication. If one only has one dish per treatment, and sees lots of cloning in one treatment but none in another, it could be because of the treatment, or it could be because of some unknown difference between the dishes. Therefore one must use a few dishes per treatment to determine if any difference between treated and control larvae is due to the treatment or due to dish-to-dish variation. (This issue comes up again during statistical analysis.)
A reasonable start might be (but see ) to use 3 dishes per treatment and transfer 10 larvae into each dish at 2 to 5 mL seawater per larva. Immediately apply the treatment, and then culture for 2 – 3 days. Finally, count the larvae, transferring them out of the dish as one counts, to avoid double counting. This does not give much time for the larvae to clone, but should be both sufficient , and short enough that conditions in the culture will remain adequate to minimize larval death.