In The Forever War by Joe Haldeman, the protagonist loses a leg in battle, and is sent to a hospital on the resort planet Heaven. There, to his surprise, the doctors, who have made good use of the two centuries his near-lightspeed maneuvers have bought them, graft a metal leg skeleton to his stump and grow him a new leg: nerves, blood vessels, muscles and all.

More than twenty-five years after The Forever War was published, it looks like growing new limbs and organs may be possible long before humans fly between the stars. The field of tissue engineering has grown from science fiction speculation to real world successes, including lab grown organs and commercially available living tissue for skin grafts.

Notes on the Nature of Cells

Living tissue can be broken into separate cells which can be persuaded to grow and divide in petri dishes incubated at body temperature with the right nutrients. However, even when they will grow in a layer one cell thick over the bottom of a petri dish, the cells will not grow back together to form the tissue they came from, whether it was an artery or a bone or even something relatively simple like skin.

The secret of getting cells to build tissues turns out to lie in the extracellular matrix -- a "living web" of proteins, polysaccharides and proteoglycans that cells both stick to and secrete as they grow into tissues like skin and cartilage. The exact components of the extracellular matrix differ between tissues. For instance, the extracellular matrix of bone also includes hydroxyapatite (a mineral containing calcium, phosphate, and bound water) for strength. If an extracellular matrix is provided for cells in cell culture they will often grow into it, absorbing it and secreting proteins and polysaccharides to "remodel" it for their own use, multiplying to fill it and produce living tissue.

However, because extracellular matrix is so complex, it is difficult to reproduce. It can be approximated with mixtures of proteins like collagen or elastin which are normally found in the extracellular matrix, or fibrin, which causes blood clotting. However, if these proteins come from human tissue (for example, the pooled blood products used until recently to make surgical fibrin "glue"), they may transfer disease from one of the donors to the patient. If they come from animals, patients may develop an immune response to the foreign proteins that remain in the newly grown tissue.

Luckily cells will often be happy with a synthetic extracellular matrix of something easier to obtain and purify. Among other things, they will accept several types of biodegradable plastics. Once they have an extracellular matrix substitute, many types of cells will happily grow on it (and into it, and through it, if it's properly porous), secreting extracellular matrix proteins and breaking down the plastic as they go, until they produce a piece of tissue the same size and shape as the plastic "scaffold" but composed of living cells and newly secreted extracellular matrix. In several cases it has been possible to layer two types of tissue to produce a simple organ, like a blood vessel (a tube of smooth muscle cell tissue lined with a layer of endothelial cells) or a urinary bladder (a "balloon" of smooth muscle tissue lined with urothelial cells).

Notes on the Nature of Plastics

Plastics are long chains, or polymers, of repeating chemical units (monomers). They are usually named "poly" (meaning "many") and their monomer name. For example, polystyrene, which is used to make Styrofoam, is a polymer of styrene.

Biodegradable plastics can be molded into solid objects, like plates and pins, or spun out in very fine fibers, which can be braided together to make thread, or woven (or just matted together like felt) to make a fabric-like mesh.

For tissue scaffolds, it is important that the plastic be porous, to allow for cell attachment and growth, and for nutrient diffusion through the scaffold as the cells are growing into it. There are several ways of generating pores. Salt leaching entails mixing the plastic with grains of a salt, shaping it into the desired form and letting it cool, then soaking it in water to dissolve the salt, leaving pores behind. This method can only be used on thin films, or structures made of thin films, because salt trapped too deeply within the scaffold won't dissolve properly. The thin films can be layered on top of each other to produce thicker scaffolds. Another method is foaming, where carbon dioxide or another gas is bubbled into the molten plastic before it is allowed to cool. This can be used for scaffolds of any shape. Finally, some labs with access to 3D printing systems favor "drawing" the scaffold in layers of molten plastic, where each layer is drawn with appropriate holes, and allowed to harden before the next layer is added.

Growing Tissues: The Major Players

Polylactic acid, sometimes called polylactide, was one of the first biodegradable plastics. It is a polymer of lactic acid, as its name suggests, and is often referred to as PLA.

PLA is degraded by hydrolysis (the breaking of a chemical bond by adding water to it) of the backbone esters of the polymer. The esters are broken at random, so that the PLA chains in the material get shorter and shorter until monomers of lactic acid start to come loose and the plastic essentially dissolves. This process is called "bulk degradation" and is similar to the way a slice of bread will get softer and mushier all through as it sits in a bowl of milk.

Lactic acid is a product of glucose (blood sugar) metabolism and is normally found in the body. It is broken down by living tissues into carbon dioxide and water. However, lactic acid is about as acidic as lemon juice, and when PLA is in the last stages of breaking down the lactic acid it gives off can damage some tissues. Though a dash of vinegar makes for a pleasant tang in a salad dressing, very few people will drink vinegar straight out of the bottle; in the same way, a little lactic acid doesn't hurt anything, but a lot of it can. Some tissues, such as bone, may be particularly vulnerable to it, because bone doesn't have many blood vessels to feed the cells and carry off their waste products -- blood vessels which could also help carry away excess lactic acid produced as the PLA breaks down.

Absorbable sutures

PLA has been used for absorbable surgical sutures (stitches) for more than twenty years. The plastic is drawn as fine fibers while it is molten, and the fibers are braided to form thread that is suitably strong to use for sewing tissues together after surgery or trauma. The thread is gradually absorbed by the body, so it isn't necessary to remove the stitches.

Orthopedic implants

Solid PLA is presently used to make plates, pins, and screws for fastening damaged bones back together so they can heal properly. This lets orthopedic surgeons avoid using metal screws and pins, which must either be left in place (sometimes causing trouble later) or removed (requiring a second surgery) after the bones heal. By contrast PLA implants hold the bones in place long enough to let them heal, then degrade and go away. However, possibly because of the buildup of lactic acid near the bone tissue, about 5% of patients have trouble with these implants. This can range from a red tender spot that gradually goes away to an open sore that drains bits of the plastic and takes a long time to heal.

Growing new bone

Porous PLA has been used as a scaffold for bone growth. Bone cells are taken from an experimental animal and persuaded to grow and divide. Then nutrient fluid containing the bone cells is poured over the PLA scaffold, a process called seeding. The cells stick to the outside of the scaffold and grow onto the surface and into the pores. The pores also let the nutrient fluid penetrate the scaffold to nourish the cells inside. The seeded scaffold is then implanted in a laboratory animal (often the one from which the cells were taken in the first place). In the animal, the bone cells grow into and through the scaffold, eventually breaking down the PLA and replacing it with bone.

One possible application of this would be to temporarily replace a damaged bone with plastic, at the same time using the plastic to grow a new piece of the patient's own bone right where it is needed. The implant must be very rigid to take the place of bone. PLA has been strengthened by adding calcium triphosphate or hydroxyapatite -- minerals very similar to those found in real bone. These minerals may also help buffer (neutralize) the lactic acid released as PLA breaks down.

Polyglycolic acid, sometimes called polyglycolide, is one of the most commonly used types of plastic for scaffolds. It is a polymer of glycolic acid; its name is commonly abbreviated PGA. Structurally PGA is much like PLA, except that it lacks a methyl group (a carbon with three hydrogens attached), which makes it more easily wettable, and makes its backbone more accessible to the water that breaks it. This characteristic means it breaks down faster than PLA.

PGA is degraded the same way that PLA is, by bulk hydrolysis -- that is, breaking random esters along its backbone by adding water to them. Glycolic acid is somewhat less acidic than lactic acid, but may still have a pH lowering effect in the last stages of breaking down.

Artificial Skin

Skin consists of two major layers of cells: a bottom layer of fibroblasts which produce the connective tissue called the dermis, and an upper layer of keratinocytes that produce the epidermis. When skin has been killed to its full depth (ranging from 1.5 to 4 millimeters), it won't grow in from the edges of the wound more than an inch or so. This often makes it necessary to supply skin to lay over the top of the damaged area. For patients with serious burns, a large amount of skin may be needed. The hope is that lab-grown skin can fill this need.

Normal (non-cancerous) cells will divide only a limited number of times in cell culture, probably for the same reasons that people and animals age and eventually die. The foreskins left over from infant circumcisions, which would normally be thrown away as medical waste, are a rich source of young fibroblasts, which have many divisions left before they "run down." Suspensions of these fibroblasts are poured over a nonwoven PGA mesh and the cells soak in and stick to the PGA fibers. They grow and divide many times, until they fill the mesh, and secrete extracellular matrix, so that when the PGA is broken down what is left is an artificial dermis, the bottom layer of the skin. This tissue (sold as Dermagraft) is used to help heal foot ulcers in diabetics; though it isn't complete skin, it supplies a dermis that keratinocytes (the epidermis-producing cells) from the wound edges can spread over. Another type of artificial skin, Appligraf, is made with a layer of fibroblasts for the dermis (but grown on collagen, an extracellular matrix protein) covered with a layer of keratinocytes which make an epidermis.

Lab-grown cartilage

Cartilage comes in several types -- the elastic cartilage found in ears, the fibrous cartilage found where tendons and ligaments are fixed to bone, and the slippery hyaline cartilage that makes the gliding surfaces of the joints. Because joint cartilage is very important for the smooth working of the joint, and because it does not heal very well when it is injured, there is a great deal of interest in growing hyaline cartilage for transplants. Chondrocytes (cartilage cells) can be harvested from an experimental animal (often a calf, because cattle have large joints that supply a lot of cartilage, and young cells can divide many times) and grown in cell culture. The chondrocytes are seeded into a PGA mesh or a PGA foam scaffold in the shape of the desired part. The chondrocytes grow into the PGA and secrete proteoglycans and collagen into the extracellular matrix between them, until in a week or two the scaffold has become cartilage of the same size and shape. In a dramatic demonstration of the potential of this technique, researchers have affixed cartilage grown in the shape of a human ear to the back of a nude mouse, a mouse specially bred to have a deficient immune system so that it would not reject the foreign tissue.

Copolymers are plastics whose polymers are mixtures of different monomers. PLA-PGA copolymers are often used for biodegradable scaffolds. The backbones of these copolymers have lactic acid alternating with glycolic acid (or sometimes "blocks" of several lactic acids, alternating with several glycolic acids -- these are called block copolymers).

PLA-PGA has been used to grow short lengths of intestinal tissue within the abdominal cavity in rats. The new tissue can be successfully spliced into the animals' gut. In humans who lose intestinal tissue to birth defects or trauma, digestion problems make it hard to get sufficient nutrients from normal food. Such "grow your own" intestinal tissue could supply sufficient nutrient uptake to allow normal development and metabolism.

Strengthening Tissues

Exercise for arteries

Some tissues, like those that make the blood vessels, must hold up under pressure. Just as skeletal muscle changes in response to stress (which is why working out makes you stronger), so the cells and extracellular matrix in blood vessels change in response to the force of the flowing blood. Blood vessels are made of layers of tissue -- smooth muscle on the outside and endothelial cells lining the inside. In working arteries, the smooth muscle cells on the outside of the blood vessels secrete more collagen into the extracellular matrix, making it stronger and less stretchy. Blood vessel tissue can be grown on PGA mesh tubes in the lab, but it is weak and flabby compared to blood vessels in a living body.

The goal of growing blood vessels is to make them available for transplant someday. Weak blood vessels won't work for this -- the sutures sewing the new blood vessel to the old may rip out if the new blood vessel is too flimsy. In addition, a weak blood vessel is vulnerable to aneurysm -- a ballooning out of the blood vessel wall from the pressure of the blood. As the wall of the blood vessel stretches it gets weaker, making it stretch more under the same pressure, until the blood vessel tears. If the blood vessel is big enough, the patient can bleed to death inside before the wound can clot.

Arteries normally grow in a living creature, with a beating heart to supply the regular stress that blood vessels need to grow strong. To mimic these conditions in culture, a tubular scaffold of PGA mesh was seeded with smooth muscle cells, and incubated with nutrient fluid. During incubation, nutrient fluid was also pumped through it at 165 beats per minute, as a fetal heart would pump blood through a developing blood vessel. After 8 weeks of growth under these conditions, the inside of the tube was seeded with endothelial cells, which were allowed to grow over the smooth muscle cells for 3 days. The new blood vessels were stronger than blood vessels grown on the same scaffolds without stretching, and stitches were less likely to rip out of them. The new blood vessels were successfully implanted in pigs, showing they would hold up to the stress of a real heartbeat and a living body.

Building body in heart valves

The heart pumps blood by contracting chambers inside it, forcing the blood out of one chamber and into the next. To keep the blood from flowing backwards, there are valves between the chambers which allow blood to pass in only one direction. Children are sometimes born with defective heart valves, or the valves can start to fail as people age. Valves from pig hearts (treated to remove as much of the foreign proteins as possible), or mechanical (non-degradable) heart valves can be implanted as replacements. However, these substitutes tend to break down over time. Furthermore, because they can't grow with a patient, children with defective heart valves must have their replacements "upgraded" every few years, which requires open heart surgery. Obviously, replacing the bad valve with a new valve grown from the patient's own cells would be ideal. However, heart valves, like arteries, need to be strong enough to resist the hydraulic force of the blood in the heart, and like arteries, they need exercise to grow strong.

To grow a heart valve, researchers started with a model of the heart valve constructed of PGA mesh coated with poly4hydroxybutyrate (P4HB).

P4HB is a polymer of 4-hydroxy butyric acid, a type of fatty acid that is a breakdown product of the fats in food. The scaffold was seeded with heart muscle cells and the inside of the scaffold was seeded with endothelial cells in the usual way. A special bioreactor pumped the nutrient fluid through the seeded scaffold, in the same way a living heart pumps blood through a heart valve. The force of the pumping was gradually increased over the next 2 weeks as the cells were growing and secreting extracellular matrix. Endothelial cells orient themselves in the direction that fluid flows over their surfaces, and fibroblasts secrete more extracellular matrix in response to stretching of the tissues where they grow -- so pumping nutrient fluid through the newly forming tissue made it stronger outside and smoother inside, more like a normal heart valve. The new valves were stronger, held stitches better, and had more collagen and glycosaminoglycans (components of the extracellular matrix that make it stronger) than similar valves grown without pumping. The new heart valves were implanted in the lambs from which the endothelial and heart muscle cells were originally taken, and functioned normally for up to 5 months (at which point the last of the sheep were killed, so that the heart valves could be removed and examined). They even grew with the sheep, the way a regular heart valve would, an important ability if tissue engineered heart valves are ever to be used in children.

Growing a simple organ: the urinary bladder

Defects in the bladder are fairly common, resulting from birth defects, cancer, disease or trauma. It is hard to repair the urinary tract with anything but urinary tract tissue, because urine is quite corrosive in the body, and urinary tract tissue is lined with special cells, urothelial cells, which are not damaged by it. A bladder is made of a layer of smooth muscle tissue lined with urothelial tissue to protect it. Researchers at Harvard made bladder-shaped scaffolds of PGA coated with PLA-PGA, seeded the insides with urothelial cells and the outsides with smooth muscle cells, and let the cells grow into the scaffolds for a week. They then replaced the urinary bladders of several beagles with the lab grown bladders. Blood vessels and nerves grew into the bladders, the plastic degraded and disappeared, and the tissue engineered organs functioned normally for up to 11 months (the length of the study).

Other Uses: Drug Delivery

Often drugs are needed in only one part of the body -- antibiotics at the site of infection, for instance, or muscle relaxants at one cramping muscle. But giving drugs by mouth or injection means giving enough to saturate the whole body. Putting the drug in a pellet of biodegradable plastic makes it possible to implant it right where it is needed.

Polyanhydrides are a class of biodegradable plastic that was designed for drug delivery. They are polymers of diacids -- molecules with two carboxyl groups. They are made of combinations of several different diacids -- the amounts and natures of their monomers affect the behavior of the polymer, so the length of time it lasts can be tailored for different purposes. Unlike the other biodegradable plastics, polyanhydrides degrade by surface erosion, so they remain solid, but get smaller and smaller, like a hard candy melting away as it is sucked on. This was so that they would deliver the embedded drug steadily over a long period of time, rather than releasing a little bit for a long time, then the rest all at once as the mushy remains of the polymer finally dissolves.

Drug delivery to the brain

Getting drugs into the brain is difficult. The brain is richly supplied with blood vessels, but the blood-brain barrier closely regulates what molecules may pass into the brain. Many drugs are turned away at the door (or rather, can't pass the special endothelial cells that line blood vessels in the brain). This makes chemotherapy for brain tumors very difficult. Enter the polyanhydrides, and a drug called carmustine (1,3-bis(2-chloroethyl)-1-nitrosourea, called BCNU).

BCNU is used for chemotherapy of a brain cancer called glioblastoma multiforme. BCNU is so toxic to the liver and kidneys that it tends to make patients very sick when it is administered to the whole body. Furthermore, the drug breaks down in a few minutes, so that it has to be taken many times.

However, when BCNU is mixed with polyanhydrides to make what are called "Gliadel wafers" the drug becomes easier on the patient, and harder on the cancer. Brain surgery is necessary anyway, to remove the tumor; when the surgeon is done, a BCNU-containing polyanhydride implant can be fitted into the space the tumor was occupying. The drug is delivered straight to the location where the tumor is most likely to recur, and the rest of the body is spared its toxic effects. The polymer slowly releases the drug over about a month, so that repeated doses aren't necessary. Gliadel wafers were recently approved for medical use.

Drug delivery to bone

Bone infections (osteomyelitis) are hard to treat with injectable antibiotics, because bone tissue doesn't have very many blood vessels, which makes it difficult to get antibiotics to diffuse from the blood into the damaged bone. However, beads of PLA containing antibiotics can be embedded directly in the section of bone where the infection is hiding. The antibiotics are released over several weeks as the PLA is broken down (apparently in this application the "burst" of antibiotics released at the end of the treatment is not a problem). The lactic acid the PLA gives off may be less of a problem for this use because most bacteria that attack the human body don't like acidic conditions any more than most body tissues do.

Where do we go from here?

Promising as this new technology is, it is still limited. Most of the organs that have been grown are composed of only a few layers of cells. Part of the reason for this is that living cells need to obtain nutrients and oxygen, and get rid of their wastes. In the body this is usually done by a network of very fine blood vessels called capillaries, which thread through the tissue. Presently capillaries can't be grown in lab-grown organs, so engineered tissue must be thin enough for nutrients to work their way in from the surrounding fluid, and wastes to work their way out again.

The process of blood vessel growth during development (called "angiogenesis") is being studied with great interest, in part because some cancers stimulate angiogenesis and grow much faster as a result of the extra nutrients the new blood vessels bring. When the tissue signals that cause angiogenesis are understood, it may be possible to layer them into a biodegradable scaffold for one of the larger organs, recruiting blood vessels to nourish the growing organ.

For that matter, as we learn more about how cells organize into tissues and organs, perhaps someday we could regrow whole limbs. One group has already grown the bone, cartilage and tendon for a finger, using biodegradable scaffolds implanted in a mouse. It still would need muscles, nerves, blood vessels and skin, but it's a start. Or what about limbs that humans never had? Anybody want a prehensile tail? It would be a design challenge, to be sure, but tissue engineers of the future could crib from spider monkeys and other creatures that already have tails.

Imagine the fashion possibilities of skin alone. Bio-artists might "paint" in the full range of human skin tones to produce the ultimate in wearable art -- art that really becomes a part of you. How about bas relief sculpture in cartilage, that can be implanted under the skin? Or organs that no mammal ever had -- organs that contain cells that make bacterial enzymes to detoxify poisons, maybe, or organs that make the biological "antifreeze" some arctic fish use, antifreeze that might make a skier or a mountaineer less vulnerable to frostbite.

For now, researchers are trying to grow other organs: kidneys, livers, and corneas, for instance. There are even plans to try to grow a heart in the laboratory (but the scope and expense have been compared to the Apollo project, so be prepared for a long wait). Maybe one day I can come back, like Cordelia Vorkosigan in Lois McMaster Bujold's Memory, from seeing a friend's replacement heart "beating merrily away in its little vat." I'd like that.

Catherine Faber presently lives in Portland, Oregon, putting food on the table by soldering computer parts. She has written a fantasy short story, numerous filk songs (songs about science fiction and fantasy subjects), and a Ph.D. dissertation in Molecular Biology. She would like to be a science writer when she grows up.

References

An, Y. H. et al. (2000) "Pre-clinical In Vivo Evaluation of Orthopaedic Bioabsorbable Devices." Biomaterials 21: 2635-2652.

Behravesh, E. et al. (1999) "Synthetic Biodegradable Polymers for Orthopaedic Applications." Clinical Orthopaedics and Related Research 367S: S118-S125.

Ferber, D. (1999) "Lab-Grown Organs Begin To Take Shape." Science 284: 422-425.

Ferber, D. (1999) "From the Lab to the Clinic." Science 284: 423.

Hoerstrup, S.P. et al. (2000) "Functional Living Trileaflet Heart Valves Grown In Vitro." Circulation Nov 7 2000: III44-III49.

Langer, R. (2000) "Biomaterials in Drug Delivery and Tissue Engineering: One Laboratory's Experience." Accounts of Chemical Research 33(2): 94-101.

Naughton, G. et al. (1997) "A Metabolically Active Human Dermal Replacement for the Treatment of Diabetic Foot Ulcers." Artificial Organs 21(11): 1203-1210.

Naughton, G. (1999) "The Advanced Tissue Sciences Story." Scientific American April 1999: 84-85.

Niklason, L.E. et al. (1999) "Functional Arteries Grown in Vitro." Science 284: 489-493.

Parenteau, N. (1999) "Skin: The First Tissue-Engineered Products." Scientific American April, 1999: 83-84.

Shay, J.W. & Wright, W.E. (2000) "The Use of Telomerized Cells for Tissue Engineering." Nature Biotechnology 18:22-23.

Temenof, J. S. & Mikos, A. G. (2000) "Review: Tissue Engineering for Regeneration of Articular Cartilage." Biomaterials 21: 431-440.