How Can They Do It When We Can’t?

How can they do it when we can’t?

Everyone knows that many salamanders can regenerate legs and tails while mammals never can (Figs.1, 2). If you tell a layperson that you work on regeneration, they will inevitably ask, “How can they do it when we can’t?” For most of my career, I was obliged to answer, “We don’t know”, and reflect on the fact that the key questions of science really are those simple ones asked by lay people.

The great news of the last few years is that now we do know. At least, now we have some good ideas about how it works, although there are still many gaps to fill. The two key concepts are cell potency and positional information. Cell potency relates to the range of cell types that the cells of a regenerating structure are able to become. Positional information relates to a system of signals that vary with position within the regenerating structure and control what pathway of differentiation those cells adopt.

Let’s start with the cells and consider those cells making up the limb of a vertebrate animal. The skeletal structures and connective tissues of limbs arise in embryonic development from a limb bud—a structure consisting of a loose mass of cells wrapped up in an outer layer. The loose cells are multipotent, meaning that they can become any one of a series of connective tissue cell types: bone, cartilage, tendon, ligament, or dermis (the inner layer of the skin). The other principal structures of a limb: the muscles, nerves and blood vessels, are formed from separate cell populations that grow into the bud during limb development.

When an adult salamander limb is amputated, the connective tissue cells near the cut surface dedifferentiate and become a uniform population of multipotent cells very similar to those of the embryonic limb bud. This fact has only been discovered recently, and it relied on a very powerful modern technology: single-cell RNA sequencing, which enables individual cells to be analysed to find which genes are active and to what extent. This is not the case for the cells of frog tadpole legs, which show very limited regeneration. However, it has been shown that grafting of embryonic limb bud cells into the amputation sites of frogs can improve their regenerative ability. So, to achieve successful regeneration, one requirement is a mechanism to de-differentiate connective tissue cells into multipotent cells similar, or identical to, those of the embryonic limb bud.

But this is only half the story. To regenerate a complex structure like a limb, these cells need to re-differentiate into the various functional cell types arranged in a complex three-dimensional pattern. This is the role of positional information.

Positional information is a concept invented by Lewis Wolpert (1929–2021), a public intellectual well known both in developmental biology and in wider social circles. In the late 1960s, he proposed that developing or regenerating systems must have some kind of embedded information that varies with position, such that multipotent cells can look at the positional information in their locality and use it to decide into what cell type to differentiate. Wolpert postulated the existence of positional signals—for example, gradients of diffusible molecules, which set up the system; and positional values, which were a long-term stable representation of the positional information. This latter concept of positional value is particularly important in regeneration because the information is needed not just for embryonic development, but also to be available to control regeneration after maybe years of the animal’s life.

The identity of many positional signals was discovered between 1980 and 2000 through intensive studies of embryonic development in frogs, mice, fruit flies, and nematode worms. They are composed of small proteins secreted by cells, which are often called “growth factors” or “cytokines” in other contexts. They include molecules discovered by biochemists, like fibroblast growth factors, transforming growth factors, and bone morphogenetic proteins, and others discovered by geneticists with weird names like Wnt and Hedgehog. Most of these factors are active during limb development and again during regeneration. However, they do not persist during adult life. What does persist is the system of positional values, which were only discovered recently.

The key elements of positional value are the genes that define specific body regions during embryonic development. Such genes all encode transcription factors, which are proteins that control the expression of other genes. For example, in the limb, expression of a gene called shox defines the middle part of the limb and of another gene called hox-a13 defines the hand or foot. These genes are not expressed in the mature limb, but they are still “open for business” in the appropriate regions. For example, hox-a13 is permanently repressed in the upper arm by a specific modification of chromosomal proteins, while in the hand it is not repressed and so it can be re-expressed when exposed to the stimuli of regeneration. In other examples, there is very low-level expression of the relevant positional information genes in the adult, which can be upregulated in regeneration. There are also some cell adhesion molecules, regulated by the same genes, that keep the cell populations in appropriate groups as regeneration proceeds.

Although mammals use the same genes to specify positional information in embryonic development, these genes cannot be reactivated in the adult. Mammals have neither the appropriate multipotent cells nor the appropriate positional information to enable successful regeneration. This is why “they can do it and we can’t.” Whether these key elements can be rebuilt using the modern techniques of regenerative medicine remains a question for the future.

_{Featured image by Nomad Sabor on Unsplash.}

OUPblog - Academic insights for the thinking world.