We tend to think of bone as dead structural material, like the girders of a skyscraper, which we call its skeleton. In reality bone is an active, living organ. Bones become longer and thicker as the body grows, and strengthen or weaken according to the load they bear. Bone marrow produces the red and white blood cells that nourish and protect our tissues, and platelets that stop the bleeding at the site of an injury. Bone has remarkable regenerative capacities: under optimal conditions broken bones heal with no scar, new living material fixing the defect as though it had never been. Ordinarily, growth and repair are the only reasons that new bone appears after birth.
The body makes bone through two distinct mechanisms: intramembranous and endochondral formation. In the former, bone cells, or osteoblasts, sit on top of bone and secrete more bony substance, like bricklayers stacking bricks to build a wall. Intramembranous bone formation thickens the top of the skull and widens the shaft of long bones.
Endochondral bone formation is a radically different process, reminiscent of the lost-wax technique that artists use to cast sculptures. The body forms a cartilage model that is then infiltrated, resorbed, and replaced by true bone. This is how bones grow longer, and how fractures heal.
Endochondral bone formation is also the embryo's predominant method of constructing its skeleton in the first place. During the first two months after conception, primitive mesenchymal tissue condenses into pre-cartilage representations of most of the bones in the body, which gradually become cartilage and finally bone. This embryonic skeleton emerges from its undifferentiated ancestor cells in a regular sequence, the way parts of the image in an instant photograph slowly seep into view: first the back structures, then the front, head before tail, trunk before thighs and upper arms, then lower legs and forearms, ankles and wrists, fingers and, very last, toes.
When Kaplan and Zasloff decided to launch an FOP research program at Penn, the first challenge was to characterize the nature and natural history of the disease. The method and distribution of new-bone formation in FOP had never been accurately documented. A century or more of medical literature was filled with ambiguities, contradictions, and errors. Earlier investigators generally assumed that the bone formed through the intramembranous process. Perhaps they could more easily imagine pockets of renegade bone cells running amok. Kaplan's group looked at tissue samples of active FOP lesions, biopsies that had been taken before the right diagnosis was made, and determined conclusively that the disease used the endochondral pathway.
Doctors had always known that heterotopic bone did not appear randomly over the body. The neck and upper back were usually affected first. Some joints were nearly always involved, while others, such as those in the hands and feet, showed symptoms much later and to a lesser degree. Zasloff had begun studying the pattern while at the NIH, and Kaplan's group at Penn, with a steadily growing patient population, systematically surveyed forty-four people joint by joint. Most people might have a hard time accurately dating their personal histories of bumps, cuts, sprains, and breaks, but for people with FOP, injuries tend to mark memorable stages of life. "Like the last time you walk up a stair," says Andy Sando, a man in northern Michigan who lost leg mobility after a fall. "It was April 29, 1985. That's the exact date. I knew after I took that last step: I'm never going to do that again."
The researchers discovered that the age at which ossification began varied from one person to the next, but the sequence of joint involvement was almost always the same: first the neck and spine, then the shoulders, the hips and elbows, the knees and wrists, the ankles, and finally the jaw. Back to front, head to tail, trunk to appendages, proximal to distal—the pattern was hauntingly familiar, reminiscent of the sequence of endochondral bone formation in the embryo. The embryo models its skeleton by condensing undifferentiated mesenchyme cells into cartilage and then bone. In some mysterious and profoundly disturbing way the FOP body was recruiting existing connective tissue and transforming it into bone, bone that often retained the shape of the muscles or ligaments that it had once been.
"These people aren't just forming little bones here and there," Kaplan says. "They are forming a whole extra skeleton. It doesn't necessarily look like the first one, but that's what it is. It's extraordinary that a differentiated tissue like a muscle or a tendon can turn into another differentiated tissue, like a bone. It's like crossing all the lanes of a busy highway. Developmental biology is not known to work that way. You never see the brain turn into the pancreas. You never see the kidney turn into the heart. You never even see the stomach turn into the small intestine. But here you see what seems to be perfectly normal muscle turn into perfectly normal bone. Normal bone. It looks normal. It looks normal on x-ray and under the microscope. It behaves like normal bone—if it bears weight, it gets denser, and if it doesn't bear weight, it becomes osteoporotic. If you break it, it heals, just like a normal fracture. It even contains marrow. It's normal in every way except one: it shouldn't be there."