Researchers have developed a four-dimensional model of spinal-cord injury in mice, which shows how nearly half a million cells in the spinal cord respond over time to injuries of varying severity. The model, known as a cell atlas, could help researchers to resolve outstanding questions and develop new treatments for people with spinal-cord injury (SCI).
“If you know what every single cell on the spinal cord is doing in response to injury, you could use that knowledge to develop tailor-made and mechanism-based therapies,” says Mark Anderson, a neurobiologist at the Swiss Federal Institute of Technology in Geneva, Switzerland, who worked on the atlas. “Things don’t need to be a shot in the dark.”
Anderson and his colleagues used machine-learning algorithms to build the atlas by mapping data from RNA sequencing and other cell-biology techniques. They described the work in a Nature paper published today1 and have made the entire atlas available through an online platform.
The atlas is a valuable resource for testing hypotheses about SCI, says Binhai Zheng, who studies spinal-cord regeneration at the University of California, San Diego. “There are a lot of hidden treasures.”
Injury insights
The researchers examined sections of the spinal cord, sampled from 52 injured and uninjured mice at 1, 4, 7, 14, 30 and 60 days after injury. Their analysis involved 18 experimental SCI conditions, including different types of injury and levels of severity. They used RNA-sequencing tools to explore how 482,825 cells responded to injury over time.
The spinal cord, like the brain, is made of delicate tissue that is isolated from the body’s immune system by physical barriers that restrict the entry of immune cells. But when the spinal cord is damaged, immune cells from the body infiltrate the injury site and activate inflammatory responses. This keeps the injury free of infection, but can also impair healing and make lesions worse. The researchers showed that the influx peaks between 7 and 14 days after injury.
They also found that injury immediately impairs the function of cells that form the blood–spinal cord barrier and the arachnoid barrier — a protective membrane that covers the spinal cord.
Genes associated with dysfunction in these barriers were increasingly switched on in the first four days after injury, but their expression began to decrease on day seven.
The researchers also compared the cellular responses to injury in young and old mice. When SCI occurs, specialized cells called astrocytes form a thin border around the lesion in the spinal cord and seal it to protect adjacent tissues. These protective barriers play a crucial part in wound repair and recovery.
The study found that astrocytes lost their ability to respond to injury and form protective borders around lesions in old animals, but not young ones.
“Looking at the histological images, you could see with your naked eye that these barriers form very robustly in young animals but are completely dysfunctional in old mice,” says Anderson.
As a result, older mice had larger lesions with more extensive neuronal loss and greater invasion of immune cells. Their ability to recover from SCI was also reduced, leading to functional impairments and paralysis.
Gene therapy
Using insights from the atlas, the researchers designed a gene therapy to promote wound repair after SCI in older mice. They used a virus to deliver genes programmed to express three growth factors — EGF, FGF2 and VEGF — to spinal-cord cells. These proteins can boost the growth of astrocytes and cells that form blood–spinal cord barrier.
When injected into the lower thoracic spinal cord in old mice two days before SCI, the treatment increased the number of border-forming astrocytes, reduced the infiltration of harmful immune cells and helped to restore the integrity of the blood–spinal cord barrier. As a result, treated mice had smaller and more contained spinal-cord lesions and recovered their ability to walk just as well as young mice that experienced similar injuries.
The researchers say the gene-therapy component of their study provides proof of principle, but caution that more work is needed before this approach could potentially benefit people with similar injuries.
One key challenge will be controlling the duration of the gene therapy’s effects. “We don’t necessarily want to be stimulating this kind of astrocyte proliferative response chronically,” says Timothy O’Shea, a medical engineer at Boston University in Massachusetts.
Determining the best time to administer such treatments will also be essential. “Those things are still the kind of caveats that I think would need to be worked out from a therapeutic standpoint,” says O’Shea.
