Deletion of a single gene allows axons to grow through a complete spinal cord injury, and to do so in abundance. The discovery, called “spectacular” by another spinal cord injury researcher, appears to overcome a major hurdle in spinal cord repair, and sets the stage for tackling other problems in promoting functional recovery after spinal cord injury.
The gene, called PTEN (phosphatase and tensin homolog), is an inhibitor of another gene, called mTOR (mammalian target of rapamycin). mTOR is active in the embryonic development of the CNS, but is gradually turned off by PTEN.
“mTOR controls protein translation, and is required for most forms of cellular growth,” said Zhigang He, PhD, associate professor of neurology at Harvard Medical School, who led the study published online Aug. 8 before the print edition of Nature Neuroscience. “Our finding is very simple. When mTOR is functional, it promotes growth. PTEN inhibits it.”
In the study, Dr. He turned his attention to spinal cord injury in adult mice. He showed that sprouting of axons above a brainstem lesion was robust in the early post-natal period, but dropped sharply over the next two months. That drop-off correlated with reduced mTOR expression in the injured corticospinal neurons, suggesting a correlation between the mTOR downregulation and the decrease in sprouting.
Based on his prior studies and on work demonstrating that PTEN inhibits mTOR, Dr. He created mice genetically deficient in PTEN. Neurons in these mice retained the ability to sprout in response to injury, in a manner “characteristic of young neurons,” he said.
In order to test whether the sprouting neurons can grow past the injury site, Dr. He induced either partial or complete lesions at T8 in these mice. In mice with active PTEN, axons in either type of injury retract from the injury site. “Nothing can regenerate,” he said.
But in mice without PTEN, “we observed very extensive sprouting, and massive growth toward the lesion site” in both injury types, Dr. He said. There were also significant numbers of axons that grew past the lesion site, either going through it, or for some axons in the partial lesion, going around it on the intact side.
In an independent set of experiments, lesions were performed and analyzed in a double-blind manner. The treatment each mouse had received before surgery was correctly predicted about 95 percent of the time, indicating the result was “consistent and robust enough” to overcome variability in surgical technique common in spinal cord injury models.
As is typical for neuronal regrowth, the axons regrew slowly, extending up to three millimeters after 12 weeks. They projected bilaterally past the injury, unlike the unilateral paths of uninjured axons.
Regrown axons also appeared able to form new synapses, as shown by presence of standard presynaptic markers in regenerated axons, and characteristic ultrastructural features under electron microscopy.
Encouragingly, axonal regrowth later in life did not depend on deleting PTEN in the embryo. Even when PTEN was deleted four weeks after birth, regeneration occurred once the spinal cord was injured.
The current study did not look at functional recovery in the mice, Dr. He said, because the T8 lesion — which is experimentally easy to induce — produces variable functional impairments that are difficult to study carefully. He is now actively pursuing functional studies in mice with higher spinal cord injuries.
Whether PTEN suppression by itself can be useful in treatment of axonal regrowth in humans is unknown. “It's still pretty early stages,” he said. His lab is currently investigating PTEN inhibitors in animal models.
Perhaps the most important direct result of the discovery is that it opens up the field of research in many directions that have previously been closed. “Before our study, there was extremely limited success with regeneration, and there was some concern about whether it was even possible. Our data suggest it is possible,” he said.
Because it has been so difficult to get axons to regenerate at all, investigating aspects of post-lesion regrowth and reconnection has been highly limited. “But now we have that possibility, we can explore in more detail the other problems that need to be overcome.” These include determining how best to reduce the activity of lesion-site inhibitors of axonal growth, including injured myelin. “Now we have the possibility to use our mice, to increase mTOR, and remove the inhibitory molecules, to see which one gives the best regeneration,” he said.
Commenting on the study, Ben Barres, MD, PhD, a professor of neurology and neurological sciences at Stanford University, who studies axonal regeneration and neuron-glia interactions in the CNS, called Dr. He's study a landmark. “To see this work, to see CNS neurons growing through CNS glia so robustly, there is nothing that has been done that is comparable. It is just amazing,” he said.
While it is still only a minority of axons that are regenerating, he added, this is “many more axons that have ever been demonstrated before.”
“Whether they grow back to the right target and form functional connections is now the next question,” Dr. Barres said. “And are they reforming specific synapses, to get restoration of function? The reason so little is known about that is that it has never before been possible to get axons to regrow so robustly to their targets.”
CNS plasticity may be a key to restoring function after regrowth, just as it is in embryonic development, he added.
“Will that be recapitulated in the adult brain?” Dr. Barres asked. “That is far from clear. But I think there is every reason to be hopeful that this will lead to new treatments. It's not going to happen tomorrow, there's a lot of work to do, but there is really new hope here. It just may be possible.”
ARTICLE IN BRIEF
Investigators showed that sprouting of axons above a brainstem lesion in mice with spinal cord injury was robust in the early post-natal period, but dropped sharply over the next two months. That drop-off correlated with reduced mTOR expression in the injured corticospinal neurons, suggesting a correlation between the mTOR downregulation and the decrease in sprouting.