The octopus can regenerate an entire arm. This is no simple task. As this close-up of the octopus’s suckers demonstrates, the octopus arm is quite a complex structure. 
Octopus bimaculoides. Specimen: Schadenfreude. Photograph by Z. Yan Wang, Ragsdale Lab, UChicago. Many thanks to Yan and Cliff for this beautiful picture.

The idea of regeneration and de novo cell production captures the interest of many in the public. I am not entirely clear on the reasons that people find regeneration so alluring. But I venture to say that regeneration may be an attractive idea because it offers 1) the hope of reversing ill health and 2) a biological approach to immortality, typically the exclusive domain of religion. Moreover, there appears to be a sense that regeneration can be self-directed, giving us not only access to health and immortality but control over that access. As I previously wrote, the human brain does not regenerate. But is that true of all brains? of all neurons? all nervous systems? That is our topic for the day.

Let’s start by understanding a bit about how regeneration works in various animals where it works really well.  We’ll start with flatworms. Flatworms are bilaterian (symmetrical left to right), invertebrate animals which include such charming members as the tapeworm.

[In case you’re wondering, indented text is tangential to the main point of the post. Read if you are interested. Feel free to skip if you’re not. Increasing levels of indentation simply reflect deeper and deeper entry into the besides-the-point rabbit hole.]

Bilaterian animals are animals that are symmetrical left to right, at least on the outside (internal structures such as the heart may not be symmetrically placed), and at least ancesterally (see flounder below). Having a left and a right implies that there is also a dorsum (back to the sky) and ventrum (“stomach” or toward the ground), Bilaterians also have a front (or head region, in the direction of movement) and a caudal (or tail) end. The vast majority of multicellular animal species are bilateria with sponges and starfish among the radially symmetric exceptions. Bilaterians evolved after radially symmetric animals. Unlike their ancestors, bilaterians develop from 3 germ levels, known as endoderm, mesoderm, and…. drum roll, please…. the lovely ectoderm from which the nervous system derives.

This shows the head of a flounder (thanks to a Poissonerie on Rue Jeanne d’Arc in Paris 75013). Flounder anatomy is odd in a couple of ways. First these flatfish swim forward with one side up (left in this case) and the other side down. Dorsal and ventral then represent the right and left in the orientation that the animal swims in. Second, both eyes are on one side, the left side in this case. According to Wikipedia, the eye on the bottom migrates to the top during development. The fish’s predators always come from above as the fish hide on the sea floor; ergo no need for a bottom eye. Of course these animals are bilaterians despite their secondarily evolved assymetry.

You may wonder why I am writing about flatworms, more formally known as planarians. Well let’s start with the fact that these little animals have a brain!!! In fact, flatworms appear to possess the earliest evolved brain. A brain is defined here as a collection of neurons in the head region of an animal. The inaugural brain, belonging to  flatworms, is the eponymous excuse for Oné Pagán’s delightful book, The First Brain (Oxford University Press, 2014) which covers the brain and so many more planarian-related topics. But this post is not a full book review of The First Brain. Instead what brings us to today’s post is inspiration from something that I learned of in the book, namely that flatworms are champion regenerators.

How good are planarians at regenerating? Well pretty darn good. An entire planarian with a complete nervous system including the brain and the planarian version of eyes (ocelli are actually better termed as eye spots, essentially light-sensitive organs that do not support the formation of a pixelated image) can emerge from tiny bits of an adult worm. Cut a flatworm in half, in quarters, eighths and even down to twentieths; and what emerges is a complete flatworm. [10,000 cells regenerate into a 200,000 cell flatworm.]  It gets even better. As Dr Pagán explains in clear and entertaining prose, flatworms’ astounding ability to regenerate can occur repeatedly, giving rise to the statement by John Dalyell, a Scottish naturalist, “[the flatworm] may almost be called immortal under the edge of the knife. Innumerable sections of the body all become complete and perfect animals.”

Work in Peter Reddien’s laboratory at MIT demonstrates that a single cell, of a very particular type, can lead to the regeneration of an entire worm. Let me explain. Worm regeneration depends on neoblasts, which are cells that divide and thus have the ability to make daughter cells which in turn can make more daughter cells and so on. In this way, neoblasts proliferate. Furthermore, a neoblast is a type of stem cell in that its daughters can be all manner of different types of cells (e.g. muscle, neuron, skin, gasterointestitinal and so on).

Two key features define stem cells. They can proliferate and they are undifferentiated, at least somewhat. An undifferentiated cell is one that has not pledged to any tissue sorority, so to speak. In vertebrate terms, a perfect stem cell would divide into cells that each could grow up and join the digestive tract, the blood, the bones, muscles or nervous system. This perfect stem cell’s lineage is multipotent, even omnipotent. In reality, pluripotency, the ability to differentiate into several but not all possible cell types, is more typical of stem cells in adult vertebrates.

Radiation kills dividing cells including stem cells (thus radiation is used to kill rapidly dividing cancerous cells). Reddien and his colleagues found that after irradiating worms, a single neoblast survived in 22% of the worms blasted with radiation. These surviving neoblasts happily did what they do so well which is to proliferate, thereby producing an island of identical daughters, called a clone. These clone-producing neoblasts were called cNeoblasts. One cNeoblast produced hundreds of descendants within a couple of weeks. But the question is what happens to all those descendants; which sorority or sororities do they join? To address this question, Reddien and colleagues blasted flatworms with an even higher dose of radiation, a dose which allowed no neoblasts to survive. Even with all the neoblasts off line, these irradiated flatworms took 6 weeks to die, but after 6 weeks of slow degeneration and failure to feed, every worm was dead. With that background, Reddien et al asked whether the introduction of one cNeoblast could rescue the doomed worms. Remarkably, the answer is yes. After the transfer of a single cNeoblast into a lethally irradiated worm, some worms lived beyond 7 weeks. They regained parts such as ocelli and the pharynx, which allowed them to start to feed again by about 8 weeks. There were limitations to the regeneration. First, only 7 of 130 (~5%) transplants lived past 6 weeks. Second, the worms were small and asexual, different from the norm. Nonetheless, this experiment shows that one cNeoblast can rescue a flatworm from certain death.

What is super-duper-cool about planarian regeneration is that it extends to the brain. A full brain can develop from a piece of the adult worm that does not contain any brain cells. Moreover, if the brain is taken out and put back in, in reverse orientation or placed close to the tail instead of the front or even placed into a different flatworm species, the animal recovers normal appearance and function. It is entertaining to think about whether the regenerated brain encodes the same experiences, aka learning, as does the original brain. Would the memories of the original worm be recapitulated in the re-made worm? I want to say “no” but in truth, I don’t know. What is the power of planarian epigenetics?

Returning to our original set of neuro-centric questions, let’s re-assess.

• Which organisms have a central nervous system that can regenerate? Planarians and the the land snail, Limax
• Which organisms have a central nervous system that cannot regenerate? Vertebrates

The situation is a bit different when we turn our attention to the  peripheral nervous system. Most animals have some ability to repair peripheral tissues including peripheral nerves. That list includes Hydra, planarians, octopi, lizards, and the rest of vertebrates, including us. That’s right. We can regrow injured nerves. Not always perfectly but often and well enough. Then it appears that, with respect to regeneration, what separates us from lizards is our inability to regenerate non-neural peripheral tissues whereas what separates us from planarians and Limax  is our inability to regenerate the central nervous system or to make new peripheral neurons (as opposed to simply repairing existing axons). You may have heard a recent story of spinal cord repair after injury. This is not regeneration of lost cells. In fact, spinal cord injury wreaks its devastation not by killing neurons but my interrupting axons. Injured (e.g. crushed or severed) axons regrow in the periphery but not in the central nervous system. The reason for this difference is that the peripheral environment is supportive and the central one quite hostile to axonal re-growth. Thus, to support the re-growth of damaged axons, researchers from University College – London took cells from the nose that help olfactory nerve axons make it through the cribiform plate and into the brain. They placed these cells into the spinal cord and successfully provided a conduit for axons to cross the site of injury in a patient with a very clean cut injury (this patient was injured by a knife and the axons were severed rather than crushed). Bottom line, only animals descended from ancestors that evolved very early in evolution have the ability to regenerate neurons so as to make a new nerve, nerve net, ganglion, spinal cord, or brain.

Let’s finish up with the incredibly interesting question of why some animals regenerate and others don’t. Not all planarians have regenerative capabilities, meaning that highly related (evolutionarily) species can differ in their regenerative capacity. Conversely, non-planarians as diverse as sponges, a radially symmetric animal without a nervous system; Hydra, a radially symmetric animal that possesses a nerve net but no brain; and the lizard, a terrestrial vertebrate with a complex nervous system, all have the capacity to regenerate body parts. As stated in the introduction to a recent virtual symposium, we still do not understand the phylogenetic distribution of regeneration.

The cells that make up a sponge (which is an animal believe it or not: they feed on other organisms rather than relying on sunlight as do plants) can re-form into a whole sponge, even after the individual cells have been dissociated by being sifted through a fine mesh. Hydra (also an animal and also not a bilaterian) can also reorganize their component cells from complete dissociation. They can even reform from 300 cells, even though mature hydra can have more than 150,000 cells. That may not appear any more spectacular than what the sponge does except for the fact that hydra have neurons and sponges do not. Granted the hydra’s neurons would be analogous to peripheral neurons rather than central neurons.

I should say that a distinction has been drawn between regeneration that involves making new cells (epimorphosis) and that which only involves the re-arrangement of existing cells (morphallaxis). Interestingly this distinction was made by Thomas Hunt Morgan, of fruit fly genetics fame, who worked initially on regeneration. He then left behind the field of regeneration to work on inheritance in fruit flies, winning the Nobel Prize in Physiology or Medicine in 1933 for his discovery that genes are collected onto physical structures that we call chromosomes. This post primarily concerns epimorphosis.

What are the clues that can inform our understanding of how regeneration evolved, possibly/probably multiple times? Let’s consider three. First, some animals appear to couple regeneration with autotomy, meaning the loss of a limb or a piece of the body as a strategy for survival. In other words, given the choice between dying or losing a body part (e.g. tail) and living, some animals will lose the body part and survive and in some cases, go on to regenerate the lost body part. Lizards, octopi, and earthworms can all regenerate, impressively growing a new limb or body part that is virtually indistinguishable from the original. One seriously weird example of this is the sea cucumber, which “under threat of predation eject all their viscera and subsequently regrow them” [Quote from Smith and Olds 2011].

I am struck by how bad we humans are at regeneration. Of course, we have some regenerative capacity. We can regenerate skin, bones, or liver But we don’t remake an amputated limb and we don’t re-generate brain regions lost to stroke or trauma. And we’re not even perfect at remaking those tissues that we do regenerate. As an example, I recently scraped off a medium (about 1/3 inch or 8 mm in diameter) size piece of skin off of my thumb knuckle. It is now completely healed. And the outlines of the injured area are totally obvious. In other words, regeneration made the skin close up but did not return the skin to its original state.

The visceral autotomy of sea cucumbers recapitulates, to a certain degree, sea cucumber development in which the viscera are resorbed and then re-made during normal metamorphosis. This brings us to a second potential clue which is that regeneration involves a return to developmental processes. This is such an accepted idea that journal sections and departments typically pair development and repair. In other words, if an injury requires repair, what better method to use than the gold standard that got you there to begin with, viz development?

Finally, regeneration resembles asexual reproduction and related mechanisms that combat aging or senescence. For example, flatworm regeneration serves to replace organs during adulthood. In essence, flatworms replace old and tired cells, effectively turning back the hands of time and making the animal forever young. Whether regeneration renews the same body (anti-senescence) or creates a new body (asexual reproduction), the effect is much the same: a step toward continued existence, perchance immortality.

In sum, regeneration may have evolved as a strategy to evade predators; a mechanism for adult repair; and/or a way to perpetuate an individual organism. Still, which evolutionary considerations tip the balance toward regeneration in some species and preclude regeneration in others? In the end, I don’t understand the evolutionary considerations that lead to the distribution of regeneration that is observed today. I do suspect that animals with advanced regenerative capabilities suffered some evolutionary trade-off. I would further speculate that whatever that trade-off is, we mammals are better for having taken the non-regenerative side of the evolutionary bargain that we took.