Fast-regenerating mice offer clues for stroke, spinal cord and optic nerve injury

axon regeneration CNS
The CAST mouse from Thailand–genetically distinct from most lab mice–may have the right ingredients for nerve regeneration. (Courtesy Jackson Laboratory)

Second in a two-part series on nerve regeneration. Read part 1.

The search for therapies to spur regeneration after spinal cord injury, stroke and other central nervous system injuries hasn’t been all that successful to date. Getting nerve fibers (axons) to regenerate in mammals, typically lab mice, has often involved manipulating oncogenes or tumor suppressor genes to encourage growth, a move that could greatly increase a person’s risk of cancer.

A study published online last week by Neuron, led by the F.M. Kirby Neurobiology Center at Boston Children’s Hospital, took a completely different tactic.

Seeing little success at first, the researchers wondered whether they were working with the wrong mice. Laboratory mice are pretty much a monoculture—a strain called C57BL/6 tends to predominate. To get more genetic diversity, the team decided to compare regenerative abilities across multiple strains of mice. They turned to the Collaborative Cross, a project launched a decade ago that picked eight “founder” strains of mice and began systematically breeding them, yielding hundreds of progeny strains whose genomes are well delineated.

Michael Costigan, PhD, Clifford Woolf, MD, PhD, and colleagues took these founder strains as a guide and tested their ability to regenerate axons in models of spinal cord injury, optic nerve injury and stroke. Unexpectedly, one semi-wild strain called CAST proved to be a “hyper-regenerator” in all three models. Interestingly, C57BL/6 was the worst nerve regenerator.

The team then did gene expression studies, comparing which genes were turned on and off in injured neurons across the different strains. They hoped to find differences that might point to regeneration-promoting factors.

Regeneration CAST mice
Of the nine mouse strains tested, nerve fibers (axons) from the CAST mice showed the most regeneration in the lab (bottom right). Axons from C57 mice, commonly used in the lab, showed the least regeneration (upper left).

“Gene expression across the whole genome is a completely agnostic test in looking for what genes are important in explaining a trait,” explains Costigan. “People have tried whole genome screens in the past but have gotten so much data that it’s been difficult to analyze. The question we asked was very specific, designed to identify only a handful of the most important genes.”

They were especially interested in CAST and how it diverged from its fellow mice. Typically, mice require a pre-injury to activate a growth program that enables a small amount of central nervous system regeneration. Not only did CAST mice have a much bigger response to such priming, they didn’t always need it: in the ischemic stroke model, the injury from the stroke itself seemed to provide enough impetus for axon growth. “They have a completely different growth program that acts on top of the program activated by pre-injury,” says Costigan.

That growth program centers on the protein activin—the #1 hit on two independent, comparative whole-genome gene-expression studies. When activin was added to the C57BL/6 mice, they became more like the CAST mice: axon growth increased dramatically.

“Activin achieved more regeneration than has ever been seen before without genetic manipulation of tumor suppressors,” says Woolf.

The missing link?

Interestingly, activin is the same protein that allows lizards to regrow their tails and fish to regrow fins. Why isn’t it more evident in mice and other mammals? One theory holds that the growth capacity of neurons is repressed in mammals to keep brain connections stable, allowing us to maintain our thoughts and remain ourselves over our entire lifetime.

The CAST mice could perhaps be seen as a missing link to animals like lizards or newts that regenerate nerves and limbs readily: originally from Thailand, they diverged from other mice relatively early in evolution. (According to Costigan, most current lab strains of mice are closely related genetically, descending from pets kept by Victorian “mouse fanciers,” whereas CAST mice are separated from the lab animals’ common European ancestor by more than 1 million years of evolution.)

“We found a normal mammal that can regenerate its axons without us having to knock tumor suppressor genes out,” Costigan says. “That allows us to look at the mechanisms responsible and to use that knowledge to find therapies down the line.”

Woolf believes that activin could be engineered into a biologic therapeutic agent, or that small drugs that act like activin could be identified by screening. Activin-based interventions may be particularly promising in stroke, where an injury wasn’t required to prime strong axon growth.

Clearly, further investigation is needed, and strong regeneration will likely require a cocktail of factors. “Evolution has placed many fail safes on active neuronal regeneration in the adult central nervous system,” says Costigan. “But the fact that it can be done, as in lizards or newts, means that we can recapitulate it, and activin is an essential component of this cocktail.”