As part of my dissertation research, I spent the spring monitoring the development of wood frog eggs. Some of the eggs I brought to the lab, removed from the jelly coat and reared in incubators. I left the rest of the eggs in the ponds and checked on them daily until they hatched. Although not part of my research, I thought it would be interesting to take photos comparing the wild eggs and algae symbionts to the lab-reared eggs without. The relationship between this alga species and only a handful of amphibians is one of the most unique stories in biology. To the best of my ability, I’ve chronicled the Green Egg symbiosis story below.

Green Eggs

Folks have been fascinated by the coincidence of algae and amphibian embryos for a long time. One of the first notes was published in 1888. Orr was studying amphibian skeletal and nervous system development. He needed embryos and had collected egg masses of a few local species. He was surprised to see that the salamander eggs he procured seemed “to present a remarkable case of symbiosis.” He noticed that there were algae cells situated inside of the egg membrane.

This is surprising because that membrane, called the vitelline membrane, is impermeable to everything but ions and small molecules, which certainly excludes entire algae cells. So the only other option is that the algae entered the embryo before the vitelline membrane had formed. But this would be no less surprising because, like basically all animals, the vitelline membrane is formed in the ovaries, and in amphibians, extra layers of jelly are added in the oviducts, long before the egg is exposed to pond water.

In the figure above (from Gilbert 1942) depicting the layers surrounding a salamander egg, it is clear to see that there are many layers acting as barriers between the pond water and the egg.

Which came first–the algae or the egg?

So, Orr was left with the plaguing question: Which came first—the algae or the egg? Perplexed, he wrote, “I have not discovered how the Algae enter the membrane, nor what physiological effect they have on the respiration of the embryo, but it seems probable that in the latter respect they may have an important influence.”

It turns out that Orr was right, algae do have an important influence on the embryos, but it was not demonstrated until the 1940s.  The phenomenon continued to perplex scientists for decades. By the 1940s, algae had been reported in the eggs of multiple salamander species and also wood frogs across the continent from California to New England and Virginia. Gilbert (1942) was the first to explore how the algae came to enter the egg and also studied the importance of the algae-egg relationship.

As to the entrance of the algae into the egg, he first assumed that the algae were present in the ovitract of the female salamander. He washed the oviducts of salamanders and tried to culture the solution, but no algae developed. If algae were not present in the ovitract, then they enter from the pond after the eggs are deposited. To confirm, Gilbert took a recently inseminated female and allowed her to lay eggs in an aquarium with pond water then removed her and allowed her to lay another clutch in an aquarium with algae-free tap water. Algal cells were present in the eggs laid in pond water within hours, but no algae grew within the tap water eggs after a few days. At the end of this experiment, Gilbert put the eggs from the tap water aquarium in a natural pond. Within a few days, algae had penetrated the jelly membrane.

By closely studying the eggs, he found that the algae transformed as they penetrated the egg. When free-swimming in the pond, the algal cells are long and oval with flagella used for locomotion. As the algae penetrate the egg layers, they grow in size, become spherical in shape, and lose their flagella. It is these large, non-mobile cell-types that rest on the inner membrane and create the “verdant blanket” (as Gilbert describes it; 1942, p. 220).

To demonstrate symbiosis, he compared hatching rates in egg masses kept in light and dark to prevent photosynthesis, finding that light and algae increased hatching success. Although algalogistis (yep, that’s a thing) hadn’t officially recognized the alga species, folks who worked with amphibian eggs called it Oophila amblystoma (“Oo” and “phila” come from the Greek words “egg” and “love,” and “amblystoma” is the genus name of the salamander species the algae cohabitates with. So, the name just means “an algae that likes salamander eggs”).

Just a few years ago, researchers took a closer look at the Oophila algae. They wondered if this was really one species or a clade of algal species, and if it the latter, did different algal species associate with specific amphibian species? To answer those questions, they sequenced the genome of algae from four species across the continent.

It turned out that the Oophila are closely related, close enough to be considered a single clade or species, but within the group there are subclades that tend to associate with each host. However, the researchers also found evidence that different subclades of the algae occasionally defect and switch to other host species.

Friends or just roommates?

Since Gilbert’s era, we’ve learned a lot about the economy of the biological transactions between the algae and the embryo. Using microelectrodes inserted within the egg membrane, Bachmann et al. (1986) measured oxygen concentration within the egg in contrast to the outside water in both light and dark. In the light, the algae worked overtime, producing so much oxygen through photosynthesis that the oxygen production exceeded that needed by the embryo and the environment inside of the egg became super-oxygenated, even when the surrounding water was anoxic (oxygen-poor).

This extra photosynthetic boost is more than just a little helpful. Most amphibian embryos get oxygen by diffusion across the egg membrane from surrounding water or air. Rinder and Friet (1994) wondered if diffusion of oxygen from the surrounding water alone could support amphibian species that had evolved with algal symbionts. They measured oxygen gradients across the cell membrane and injected dye in egg masses to see how much water would be able to flow past the eggs. In wood frogs, they found that the egg mass was loose enough and contained enough channels that water diffusion alone could support respiration. However, spotted salamander egg masses are dense and water does not penetrate to the interior eggs. In their case, the innermost eggs would suffocate without their personal algal oxygen factories. (Note: Hutchinson and Hammen (1958) inferred this relationship many decades before through experimentation rather than directly measuring O² within the eggs.)

This plant-animal relationship is a multifaceted exchange. The reciprocity extends past simple O² and CO² cycling. The algae also benefit from the nitrogen waste produced by the embryo, which limits the ammonia buildup in the egg which is toxic to embryos (Goff and Stein 1978).

Friends or more-than-just-friends?

This symbiotic relationship is certainly interesting, but not altogether surprising. After all, lots of animals host symbiotic microbes, even humans house a panoply in our guts and all over our skin (Gilbert et al. 2018). These types of ectosymbionts, commensal organisms that live in and on other organims’ bodies but outside of the tissue are common.

But the green egg story gets even more fascinating. In 2011, Kerney et al. used a technique that binds only to algae DNA and causes it to glow under fluorescent lights (fluorescent in situ hybridization (FISH)). In doing so, they could see algae cells inside(!) of the salamander tissue and cells. This makes this salamander-algae relationship the only example of endosymbiosis in a vertebrate ever seen! Previously, this type of animal-plant union was thought to only be possible in simpler organisms like coral and mollusks.

Friends or frenemies?

So, not only do these free-swimming algae manage to invade the egg membrane, but they even make it all the way inside of the embryonic cells. To some extent, this makes sense for the algae. Why would you want to live out in the pond water with variable temperatures and predators when you could live inside a nice cozy egg with abundant CO2 and nutrients? But once inside the egg, why would algae invade inside an animal cell where there is less access to light for photosynthesis?

A team of researchers (Burns et al. 2017), including Kerney, wanted to figure out the riddle. They isolated and sequenced the genes expressed by salamander cells with and without algae cells from inside and outside of the animal cell. Inside the animal cells, algae seemed to be under more stress, to the point that they switch from photosynthesis to fermentative energy production, something that would normally only happen in highly unfavorable environments. On the other hand, salamander cells with algae were unfazed and even seemed to suppress their immune response which would ordinarily attack foreign cells, perhaps to keep the algae within. For algae, then, there appears to be a risk-benefit balance that drives them inside of the egg where conditions are favorable, but risk being captured inside of the embryonic animal cells.

Better than Seuss

After over a century of investigation, the story of the Green Eggs has gotten more and more interesting with finer and finer resolution. The saga goes to show how the naturalist’s intrigue can inspire a cascade of illuminating research.

Understanding this complex relationship can also help us understand threats to amphibian populations. Herbicides can kill the symbiotic algae and result in lower hatching success and reduced developmental rates. Olivier and Moon (2009) tested the effect of the herbicide atrazine on spotted salamander egg masses. They found that even low concentrations of the chemical killed the algae. In addition to the direct effects of the chemical, the loss of the symbiont had negative repercussions.

References:

Bachmann, M. D., Carlton, R. G., Burkholder, J. M., and Wetzel, R. G. (1986). Symbiosis between salamander eggs and green algae: microelectrode measurements inside eggs demonstrate effect of photosynthesis on oxygen concentration. Can. J. Zool. 64, 1586–1588.

Burns, J. A., Zhang, H., Hill, E., Kim, E., and Kerney, R. (2017). Transcriptome analysis illuminates the nature of the intracellular interaction in a vertebrate-algal symbiosis. Elife 6. doi:10.7554/eLife.22054.

Gilbert, J. A., Blaser, M. J., Caporaso, J. G., Jansson, J. K., Lynch, S. V., and Knight, R. (2018). Current understanding of the human microbiome. Nat. Med. 24, 392–400.

Gilbert, P. W. (1942). Observations on the Eggs of Ambystoma Maculatum with Especial Reference to the Green Algae Found Within the Egg Envelopes. Ecology 23, 215–227.

Goff, L. J., and Stein, J. R. (1978). Ammonia: basis for algal symbiosis in salamander egg masses. Life Sci. 22, 1463–1468.

Hutchison, V. H., and Hammen, C. S. (1958). Oxygen utilization in the symbiosis of embryos of the salamander, Ambystoma maculatum and the alga, Oophila amblystomatis. Biol. Bull. 115, 483–489.

Kim, E., Lin, Y., Kerney, R., Blumenberg, L., and Bishop, C. (2014). Phylogenetic Analysis of Algal Symbionts Associated with Four North American Amphibian Egg Masses. PLoS One 9, e108915.

Olivier, H. M., and Moon, B. R. (2010). The effects of atrazine on spotted salamander embryos and their symbiotic alga. Ecotoxicology 19, 654–661.

Orr, H. (1888). Memoirs: Note on the Development of Amphibians, chiefly concerning the Central Nervous System; with Additional Observations on the Hypophysis, Mouth, and the Appendages and Skeleton of the Head. J. Cell Sci. s2-29, 295–324.

Pinder, A., and Friet, S. (1994). Oxygen transport in egg masses of the amphibians Rana sylvatica and Ambystoma maculatum: convection, diffusion and oxygen production by algae. J. Exp. Biol. 197, 17–30.

First, here is the happy mother:

I’m so excited! My Robertus tincs are finally calling!!! I just moved the female in with the confirmed male today, so now we just have to hope for true love! #dartfrog #dendrobates #tinctorius #robertus #frog #poisondartfrog #vivarium #amphibian #proudparent #frogfamily

A post shared by Andis A. (@azandis) on Jan 6, 2018 at 1:42pm PST

Day 1:

Here are the first images of the viable eggs, approximately 36-hours after oviposition and Gosner stage 12. At this stage, the embryo has gone through many cell divisions, but if you look closely, you can still make out the individual cells. For the first period of embryonic development, the zygote (fertilized, single-cell eggs) divides into a mass of multiple cells. Around the developmental period in these photos, the cell are beginning to migrate in order to form specific structures in the embryo. The dark indent on the surface, called the blastopore, is where the outer cells are closing in on themselves, like a deflating basketball that is being inverted. As the cells migrate inward, they create cavities that will form the body cavities of the fully formed animal. The blastopore will eventually become the anus, and the inwardly migrating cells will form the lining of the digestive track.

I also thought it might also be interesting to take a look at some of the non-viable eggs that were beginning to decompose. 

Day 3:

Approximately 72 hours, Gosner stage 12. The eggs are still developing. I took a scale shot to give a sense of their size. They measure 3.034 mm (L) and 3.180 mm (R) in diameter, which I estimate makes them about 14.6 (L) and 16.8 (R) cubic mm in volume. It’s worth remembering that, at this point, the embryo is still the same size as when it was a single-cell zygote after initial fertilization, even though it is now comprised of thousands of cells. The cells are dividing, but not growing. In the high magnification photos you can see that the blastopore indent is deforming and elongating. Inside the embryo, cells are beginning to form the notochord which will eventually become the nervous system. Although it doesn’t look like much, most of the structural components of the major life systems are already in place.

Day 5:

The embryos have elongated, measuring are 4.48 mm and 4.54 mm in length, almost 50% longer than yesterday, but the total mass should be about the same.

Gosner stage 18. The embryos developed a lot in just the last 48 hours. The embryos blew right through the end of gastrulation and into neurulation before I got these photos. So unfortunately, we missed the development of the neural plate, neural folding, and embryo elongation. As an overview, in the last photos the embryos were at the end of gastrulation. The cells had differentiated into cell layers, dividing the embryo into the ecotoderm (outer cell layer, eventually forming the skin, teeth, and most of the nervous system), mesoderm (middle cell layer, eventually forming most of the skeleton, internal organs, and muscles), and the endoderm (inner cell layer, eventually forming the lining of internal body cavities like the digestive track and lungs). So, at the end of gastrulation, the future fate of all cells were set, but the embryos still looked just like a ball of cells. Neurulation follows gastrulation, and as the name suggests, is characterized by the development of the proto neural system. These neural cells begin to migrate outward and develop ridges that will fold in on themselves during creation of the neural tube (which will eventually form the brain and spinal cord). (Since I didn’t capture a photo of the tinctorius eggs at this stage, I’ve include an example of a wood frog embryo at this stage as an example.)

As neurulation continues, the embryo starts to elongate and look more like a tadpole. By the later stages of neurulation (as in the photos today) most of the organ systems are fully established.

You can also see the very large yolk. Unlike bird, mammal, and reptile embryos, amphibian yolks form internally as part of the digestive track. Dendrobates eggs have much larger yolks than other frogs due to their particular larval ecology. In the wild, tinctorius eggs are deposited terrestrially, but the larvae are aquatic after hatching and require water immediately. So, these frogs have evolved a parental care strategy wherein the male attends the eggs, and upon hatching, transports them to a larger pool. The rearing pool can be hundreds of meters away from the egg site, and may take multiple days of transport. The large yolk ensures that the tadpoles will be hardy enough for the trip. Being hardy also helps the tadpoles once they are in the rearing pool since tinctorius larvae are cannibalistic.

You’ll also notice that the membrane around the embryo has expanded and is more pronounced. This called the vitelline membrane and it is analogous to the membrane surrounding the yolk in a chicken egg, the one that keeps a sunny-side-up egg from becoming a scrambled egg. Since tincortius eggs are laid out of the water, this membrane (along with the jelly layer) helps prevent drying.

Day 6:

Gosner stage 19. We have gill buds! Those little nubbin-wings on either side are the sprouts of tiny gills. Up to this point, the cells of the embryo have been able to diffuse oxygen without special structures. But as the cells proliferate and some cells are buried deep within the developing embryo, simple diffusion can’t cut it. Over the next two or three days, the gills and circulatory system will take form and start pumping oxygenated fluid throughout.

Day 7:

Gosner stage 20. The gills are elongating and the simple heart, which is basically just a tube at this point, is pumping blood cells through the limited circulatory system.

Day 8:

Gosner stage 21. The gills have formed almost to their external extent.

Tinctorius eggs are laid in oxygen poor conditions compared to many other frogs. As such, they have developed extremely long and filamentous gills compared to the “average” anuran. The greater the surface area of the gills, the more time there is for Co2 and O2 to diffuse across the lining of the gills to the jelly and, after hatching, to the water. If you look closely in the video, you will see the rhythmic expansion and contraction of the embryo. That’s the heart pumping. In the last portion of the video, you should be able to see individual blood cells flowing through the gills in time with the heartbeat.

Also notable at this point is that the nervous system is developing quickly, both the brain itself and the sensory organs. The proto-eye, while not externally visible, is developing as an outgrowth from the brain.

Stay tuned! The embryos are developing quickly and could hatch any time in the next few days!

Day 9:

The gills continue to branch and elongate. By this point, I think the heart has developed into two chambers, a huge architectural achievement from its origin as a single tube.

A caveat: it is worth mentioning at this point that my descriptions of developmental timing of internal organs are based on “average” anuran development. Not a lot of research has been conducted on dendrobatids, so most of these descriptions reflect ranid development. This is most clear in my attempts to assign Gosner stages to the embryos, as the ranid timing doesn’t sync entirely. Where research has been carried out, I try to give the specific description. 

Day 10:

The gills are still growing, but should be reaching about their maximum extent. Soon, the gills will begin to atrophy and the animal will switch to internal respiration.

The eyes are just beginning to appear.

The embryonic eye forms as an extension of the developing nervous system that grows outward from the neural tube. As it reaches the surface of the embryo, the optic stalk differentiates into the component cell types of the eye. At this stage, the tip of the optic stalk at the skin surface is forming into a cup-like shape. In the coming stages, that cup deepens and widens. At the same time, the lens begins to form across the “rim” of the cup from the ectoderm. As the eye progresses, the lens becomes more “lens shaped” and the skin covering the eye becomes the cornea. If you are interested in learning more about the development of the anuran eye, I would suggest checking out Thomas Reh’s lab webpage.

Day 11:

Today the eyes are well-along in development and obvious.

I outlined the formation of the eye yesterday, but I thought it would be good to also include a diagram of the formation.

To give you a sense of where the embryos’ eyes are heading, below is a close photo of a fully formed larval wood frog eye.

Day 12:

One of the embryos hatched today! The vitelline membrane is ruptured; however, the hatchling remains in the jelly and has not fully liberated itself. In the photo above, the animal on the right is the hatchling.

You’ll notice that the gills have begun to atrophy and shrink. As the heart becomes more efficient and the circulatory system grows, the distribution of oxygen to the body systems requires less surface area for gas exchange with the water. Also around this time, the primitive red blood cells of the early embryo transition into more efficient larval type red blood cells. Interestingly, there are a total of four generations of blood cells from embryo to adult. The cells change in shape, size, and even blood-type over the course of development. Even the organs producing blood cells shifts from kidneys, to the liver, and on to the spleen and bone marrow in adults.

Day 13:

Both of the embryos have hatched out of the vitelline membrane, but remain in the egg jelly. They are very active and respond to movement and light. The gills have diminished quite a bit as the respiratory system switches to internal respiration. The eyes, while still covered by the outer skin layer, look much like a completed larval eye.

Day 14:

One of the hatchlings was deceased when I checked on it this morning. The tissue still looked pretty good; so I fixed it in a formalin solution to preserve its form.

The cornea is almost entirely transparent and the gills have almost complete atrophied. From this point on, the larva will assume the classic tadpole/pollywog body plan that is most familiar of anuran larvae.

References:

Hill, R. W., Wyse, G. A., and Anderson, M. (2016). Animal Physiology. 4 edition. Oxford University Press.

McDiarmid, R. W., and Altig, R. (2000). Tadpoles: the biology of anuran larvae. University of Chicago Press.

Rojas, B. (2014). Strange parental decisions: Fathers of the dyeing poison frog deposit their tadpoles in pools occupied by large cannibals. Behav. Ecol. Sociobiol. 68, 551–559.

Vences, M., Kosuch, J., Lötters, S., Widmer, A., Jungfer, K. H., Köhler, J., et al. (2000). Phylogeny and classification of poison frogs (Amphibia: dendrobatidae), based on mitochondrial 16S and 12S ribosomal RNA gene sequences. Mol. Phylogenet. Evol. 15, 34–40.

Vitt, L., Vitt, L., and Caldwell, J. (2013). Herpetology: an introductory biology of amphibians and reptiles. 4th ed. Academic Press.

Wells, K. D. (2010). The Ecology and Behavior of Amphibians. University of Chicago Press.

Weygoldt, P. (1987). Evolution of parental care in dart poison frogs (Amphibia: Anura: Dendrobatidae). J. Zoolog. Syst. Evol. Res. 25, 51–67.