The Missing Chapter
237
especially hard, because people are endowed with lesser fracture-healing
abilities than most other animals, for whom nonunions are a non-
problem. Based on what I'd learned about the importance of nerves in
bone healing, I figured we could produce nonunions in rats by cutting
the nerves to the broken leg, particularly if we removed whole segments
of the nerves so they couldn't grow back. I assigned this part of the
project to Dr. Bruce Baker, a young orthopedic surgeon who was then
finishing his residency with an extra year on a fellowship in my lab.
After Bruce had worked out the complicated surgical procedure, we
anesthetized a series of rats, removed the nerve supply to one leg of each
animal, and broke the fibula, or smaller bone of the calf, in a standard
way. Then every day we reanesthetized a few of the rats and took out the
fracture area to mount it for the microscope. At the same time, Bruce
checked the cut nerves to make sure there was no regrowth. Successful
denervation was confirmed by the microscope and by complete paralysis
of the affected leg.
The results were encouraging yet puzzling. The nerves didn't regrow,
and the broken bones took twice the normal six or seven days to heal,
but heal they did, even though theoretically they shouldn't have knit at
all without nerves.
It was well known that the severed end of a nerve would die after a
couple of days, but, since we'd cut the nerves at the same time as we'd
broken the bones, maybe the cut ends had exerted a subdued healing
effect while they remained alive. In another series of animals, we cut the
nerves first. Three days later, after making sure the legs were fully de-
nervated, we operated again to make the fractures. We felt sure the
delay would give us true nonunions. To our surprise, however, the bones
healed faster than they had in our first series, although they still took a
few days longer than normal.
Here was a first-class enigma. The only thing we could think of doing
was to cut the nerves even earlier, six days before the fractures. When
we got that series of slides back, we found that these animals, whose
legs were still completely without nerves, healed the breaks just as fast
and just as well as the normal control animals. Then we took a more
detailed microscopic look at the specimens Bruce had taken from around
the nerve cut. We found that the Schwann cell sheaths were growing
across the gap during the six-day delay. As soon as the perineural sleeve
was mended, the bones began to heal normally, indicating that at least
the healing, or output, signal was being carried by the sheath rather
than the nerve itself. The cells that biologists had considered merely
insulation turned out to be the real wires.