MAKING BONE
Perhaps because they are the last of our remains to dissipate to dust, we think of bones as inanimate things. But they are not. Like hearts and livers, bones are continually built up and broken down in a cycle of construction and destruction. And though they seem so separate from the rest of our bodies, they originate from the same embryonic tissues that make the flesh that covers them. In a very real sense, bone is flesh transformed.
The intimate relationship between bones and flesh can be seen in the origin of the cells that make them. Most bone cells – osteoblasts – are derived from mesoderm, the same embryonic tissue that also gives rise to connective tissue and muscle. The relationship can also be seen in the way that bones form. Buried within each bone are the remains of the cells that made it.
Our various bones are made in two quite different ways. Flat bones, such as those of the cranium, start out in the embryo as a layer of osteoblasts that secrete a protein matrix. Calcium phosphate spicules form upon this matrix and encase the cells. As the bone grows, layers of osteoblasts are added and each is, in turn, entombed by its own secretions. Long bones, such as femurs, do things a bit differently. They start out as the condensations of cells that are visible in an embryo’s developing limbs. These cells, which are also derived from mesoderm, are called chondrocytes and they produce cartilage. The cartilage is a template for the future bone, one that only later becomes invaded by osteoblasts. When the template first appears, it is bone in form but not in substance.
One of the molecules that controls these condensations is bone morphogenetic protein (BMP). It is convenient to speak of it as one molecule, but it is really a family of them. Like so many families of signalling molecules, the BMPs crop up in the most unexpected places in the embryo. It is a BMP that, long before the bones are formed, instructs some the embryo’s cells to become belly rather than back. In older embryos, however, BMPs appear in the condensations of cells that will become future bones. In children and adults, they appear around fractured bones. The remarkable thing about BMPs is their ability to induce bone almost anywhere. If one injects BMPs underneath the skin of a rat, nodules of bone will form that are quite detached from the skeleton, but that look very much like normal bone, even to the extent of having marrow.
To make bone it is not enough that undifferentiated cells condense in the right places and quantities. The cells have to be turned into osteoblasts and chondrocytes. To return to a metaphor that I used earlier, they have to calculate their fates. The gene that calculates the fates of osteoblast happens to be the one responsible for ‘Arnold-head’. This gene encodes a transcription factor called CBFA1. It may be thought that CBFA1 is not very important, since mutations in it result only in a few missing bones. However, Arnold’s descendants are heterozygous for the mutation: only one of their two CBFA1 genes carries the mutant copy. Mice heterozygous for a mutation in the same gene also have soft heads and lack clavicles. But mice that are homozygous for the mutation are literally boneless. Instead of skeletons they have only bands of cartilage threading through their bodies, and their brains are protected by little more than skin. They are completely flexible and they are also dead. Boneless mice die within minutes of being born, asphyxiated for want of a ribcage to support their lungs.
By one of those quirks of genetic history, South Africa is also home to a mutation that has the opposite effect of Arnold’s: one that causes not a deficiency of bone, but rather an excess. Far from having holes in their skulls, the victims of this second mutation have crania that are unusually massive. The mutation’s effects are not obvious at birth. The thick skulls and coarse features that characterise this syndrome only come with age. Unlike the boneless mutation, the extra-bone mutation is often lethal. Its victims usually die in middle age from seizures as the excess bone crushes some vital nerve. Again, unlike the boneless mutation, the thick-skull mutation is recessive and so is expressed in only a handful of people – inbred villagers descended from the original Dutchmen who founded the Cape Colony in the seventeenth century.
The mutation that causes this disorder disables a quite different sort of gene from CBFA1. The protein itself is called sclerostin, after the syndrome sclerosteosis. It is thought to be an inhibitor of BMPs – perhaps it binds to them and so disables them. This is how many BMP inhibitors work. In the early embryo, organiser molecules such as noggin restrict the action of BMP in just this way. Indeed, noggin mutations are responsible for yet another bone-overgrowth syndrome that affects only finger-bones and causes them to fuse together with age, rendering them immobile.
Surplus-bone disorders illustrate the need that our bodies have to keep BMPs under control. Yet fused fingers and even thick skulls are relatively mild manifestations of the ability of BMPs to produce bone in inconvenient places. Another disease shows the extent of what can go wrong when osteoblasts proliferate throughout the body and make bone wherever they please. The disorder is known as fibrodysplasia ossificans progressiva or FOP. It is rare: estimates put the number of people afflicted with it worldwide at about 2500, but only a few hundred are actually known to specialists in the disease. Its most famous victim was an American man by the name of Harry Raymond Eastlack. In 1935, Harry, then a five-year-old, broke his leg while playing with his sister. The fracture set badly and left him with a bowed left femur. Shortly afterwards, he also developed a stiff hip and knee. The stiffness was not, however, caused by the original break, but rather by bony deposits that had grown on his adductor and quadriceps muscles.
As Harry grew older, the bony deposits spread throughout his body. They appeared in his buttocks, chest and neck and also his back. By 1946 his left leg and hip had completely seized up; his torso had become permanently bent at a thirty-degree angle; bony bridges had formed between his vertebrae, and the muscles of his back had turned to sheets of bone. Attempts were made to surgically excise the bone, but it grew back – harder and more pervasive than before. At the age of twenty-three, he was placed in an institution for the chronically disabled. By the time of his death in 1973, his jaws had seized up and he could no longer speak.
Harry Eastlack requested that his skeleton be kept for scientific study, and today it stands in Philadelphia’s Mutter Museum. Bound in extra sheets, struts and pinnacles of bone that ramify across the limbs and ribcage, the skeleton is, in effect, that of a forty-year-old man encased in another skeleton, but one that is inchoate and out of control. The cause of the disease is understood in general terms. The bodies of FOP patients do not respond to tissue trauma in the normal way. Bruises and sprains, instead of being repaired with the appropriate tissue, are repaired with osteoblasts and the new tissue turns to bone. This has all the hallmarks of an error in BMP production or control, but the mutation itself has not yet been identified. The search may well be a long one. FOP patients rarely have children, so the causal gene cannot be mapped by searching through long pedigrees of afflicted families.