REVIEW doi:10.1006/mthe.1999.0003, available on IDEAL at [Link]

com

Tissue Engineering
Robert Langer
Massachusetts Institute of Technology, Cambridge, Massachusetts

INTRODUCTION sized and manufactured into the desired shape and di-
mensions.
Tissue engineering is an interdisciplinary field that applies (iii) The cells must be uniformly seeded onto or into the
the principles of engineering and the life sciences to the material and grown in a bioreactor.
development of biological substitutes that restore, main- (iv) The engineered structure is placed into the appro-
tain, or improve tissue function. The potential impact of priate in vivo site. Depending on the site and the structure,
tissue engineering from both a therapeutic and an eco- vascularization may be necessary.
nomic standpoint is enormous. The total U.S. health care
costs for patients suffering from tissue loss or organ failure
exceed $400 billion annually. Approximately 8 million TISSUE ENGINEERING: CURRENT STATUS
surgical procedures are performed annually in the United Scientists have attempted to engineer tissues or organs in
States to treat these disorders and 40 –90 million hospital nearly every part of the body. For example, in brain dis-
days are required (1). It has been estimated that the total eases where dopamine production is lost, e.g., Parkin-
market for tissue-engineered products in United States is son’s, immortalized PC12 cells have been encapsulated in
$80 billion annually (2). semipermeable polymer membranes and shown to release
One of the earliest demonstrations that engineering a dopamine for more than 6 months in animal models of
tissue may be possible was by Bisceglie in the 1930s, in this disease (8).
which he encased mouse tumor cells in a polymer mem- Several approaches have been used to aid in nerve re-
brane and inserted them into a pig’s abdominal cavity (3). generation as well. For example, Schwann cells derived
These studies showed that cells could survive and not be from sciatic nerves have been placed in Matrigel and
destroyed by the immune system. This approach was an seeded in polymer membranes (9). In another case, elec-
early example of cell encapsulation, which allows nutri- trically conducting polymers such as polypyrole have
ents and wastes to diffuse through membranes, yet pre- been used to stimulate nerve outgrowth. When PC12 or
vents immune cells or large molecules such as antibodies Schwann cells are placed on such polymers and a small
(which might destroy the encapsulated cells) from enter- voltage (100 mV) is applied, neurite outgrowth is en-
ing. Therefore, the cells can survive. This approach was hanced (10).
later used in the 1970s by Chick and co-workers to create Another tissue under investigation is the cornea. In
semipermeable membranes in which islet cells were en- these studies, corneal epithelial cells were placed on poly-
capsulated to aid in glucose control for diabetes in animal vinyl alcohol hydrogels and transplanted into rabbit cor-
models (4). neas. The gels adhered to the cornea and the cells prolif-
In the late 1970s and 1980s, cells on sheets of collagen, erated for up to 2 weeks (11).
or collagen– glycosaminoglycan composites, were used in Critical issues in creating an implanted liver include
tissue regeneration in an attempt to create new skin (5, 6). enhancing cell survival, maintaining differentiated func-
These sheets were two-dimensional systems, and the ma- tion, developing a significant cell mass, and achieving
terials composing them were naturally occurring, i.e., iso- vascularization. To these ends, several different ap-
lated from animal or fish sources. proaches have been followed. In one case, hepatocytes
The next critical step was creating three-dimensional placed on polymer scaffolds and implanted into animals
structures that enabled large numbers of cells to be were shown to produce albumin and other liver function
housed, required for creating tissues in three dimensions. markers. They also cleared products of bilirubin and urea
In addition, scaffolds composed of chemically produced metabolism, similar to a normal liver (12). In a related
synthetic polymers were introduced and provided signif- approach, hepatocytes were sandwiched between two hy-
icant advantages in tailoring such properties as strength, drated collagen layers. In addition to maintaining their
degradation rates, and incorporation of cell-recognition morphology, the hepatocytes secreted functional markers
properties. Liver cells were used in early studies of this at physiological levels for at least 6 weeks (13).
approach (7), but more than 20 tissue types have been Tissue-engineered livers have been studied not only as
studied subsequently, some of which are discussed below. implants but also as extracorporeal units. In one example,
The strategy of tissue engineering generally involves the porcine hepatocytes are placed in hollow fiber mem-
following steps: branes and connected to a patient. This approach, which
is now in advanced clinical trials, has been used as a
(i) An appropriate cell source must be identified, iso- “bridge to transplant” on nearly 100 patients waiting for
lated, and produced in sufficient numbers. a new liver (14).
(ii) An appropriate biocompatible material that can be There have been several attempts to create an artificial
used as a cell substrate (open system) or cell-encapsula- pancreas. The most common approach has been encap-
tion material (closed system) must be isolated or synthe- sulating pancreatic islet cells inside a semipermeable

12 MOLECULAR THERAPY Vol. 1, No. 1, January 2000

Copyright © The American Society of Gene Therapy
 REVIEW

membrane such as alginate or polyacrylonitrile–polyvinyl cutaneously into mice. After 6 weeks, all experimental
chloride (PAN–PVC) (15). This has shown some success in specimens resembled the normal calf tendons from which
animals and a few attempts have been made in humans. the cells had been isolated, as judged by gross examina-
One critical issue is the biocompatibility of the membrane tion. Histologic evaluation demonstrated organized colla-
used, as fibrous encapsulation of the membrane can pre- gen fibrils with polymer remnants. Ten weeks after im-
vent reproducible insulin outflux. A second issue is islet plantation, histologic evaluation showed parallel linear
sourcing on a sufficiently large scale. organization of collagen bundles throughout the speci-
Urinary structures have been created by several meth- mens, centrally and peripherally. Mechanical analysis of
ods. In one of these, urinary epithelial cells have been neotendon constructs after 8 weeks shows that they have
placed on lactic– glycolic acid copolymer tubes to create tensile strength comparable to and mechanical character-
new ureters as a potential treatment for hypospadius (16). istics similar to those of the normal tendon (24).
In another, autologous chondrocytes have been delivered Major resection of the small bowel often leads to a state
endoscopically to treat urinary reflux in children and of malabsorption and malnutrition known as “short
urinary stress incontinence in women (2). Both of these bowel syndrome.” Parenteral nutrition improves survival,
studies are in advanced clinical trials in humans. One of but prolonged hyperalimentation can often lead to infec-
the first attempts to create an actual whole urinary organ tious complications, loss of vascular access, and progres-
involved constructing polymer composites in which blad- sive liver disease. To examine whether tissue engineering
der cells are placed, an approach that has created new could produce a new intestine, cell–polymer (polyglycolic
bladders in dogs (17). acid) constructs were created using enterocytes isolated
Much effort has focused on creating tissue-engineered from crypt cell-enriched fractions of intestine which were
blood vessels. In one recent study, the bioreactor condi- allowed to attach to the polymer scaffolds in vitro. Mor-
tions under which the tissue-engineered blood vessels are phological analysis by phase-contrast microscopy of in
grown were shown to be essential to this process. By vitro constructs indicated that the cell populations were
placing vascular smooth muscle cells and, subsequently, composed almost exclusively of round-, oval-, and colum-
endothelial cells on a polymer tube and growing them nal-shaped cells as well as distinctive goblet cells. Rare
under pulsatile conditions (similar to what happens in small round cells characteristic of lymphocytes were also
the body when a heart beats), the cells make significantly observed. These cell–polymer constructs were subse-
more collagen and are physically stronger than blood quently implanted in animals and, after 14 days, entero-
vessels produced in static tissue culture. In addition, the cytes were observed organizing into a stratified epithe-
pulsatile-engineered blood vessels possess pharmacologi- lium overlying the polymer fibers and granulation tissue.
cal properties similar to normal vessels. When sutured Successful engraftment occurred 86% of the time (25).
into pigs, the vessels were shown to be patent for up to a More than one million operations annually involve
month, the duration of the study (18). bone repair. Conventionally, bone “ingrowth” is acceler-
Several studies have focused on cartilage, which has ated through the use of either autogenous bone grafts or
been produced by placing chondrocytes on lactic glycolic allogenic bone. The first can be a successful procedure, but
acid copolymer scaffolds and transplanting them into is often material-limited and can cause donor site mor-
several different animal models (19, 20). The reactor con- bidity and contour irregularities. The second can also be
ditions under which the cells are grown in vitro are critical successful, but cell-mediated immune responses to trans-
to successful treatment. By mixing the cells appropriately plantation alloantigens and pathogens can be problem-
during seeding, as well as during cultivation, the cells can atic. To address this problem via tissue engineering, syn-
be made to produce desirable amounts of glycosamino- thetic biodegradable polymers have been used as
glycans and collagen, molecules that are important to the templates onto which cells (osteoblasts or osteocytes) are
eventual mechanical strength of the cartilage tissue (21). seeded prior to implantation. Specimens at 6 weeks are
Creating heart components is the aim of several differ- primarily composed of cartilage, with islands of osteoid
ent studies. Tissue-engineered heart valves have been cre- tissue seen peripherally associated with blood vessel inva-
ated in which endothelial cells and fibroblasts were co- sion. At 10 weeks, these implants show increasing
seeded on polymer scaffolds that formed in the shape of a amounts of new bone growth. Examination at 20 weeks
valve. When these engineered valves are placed into reveals gross bone formation in all specimens based on
sheep they appear to be functional, as judged by ultra- osteoblast and osteocyte activity. Histological evaluation
sound imaging, for up to 6 months (the duration of the with hematoxylin and eosin stains of such specimens
study) (22). Heart muscle cells have also been placed on revealed that they were composed of trabeculated bone
polymer scaffolds and grown under different reactor con- enveloping islands of residual cartilage. Representative of
ditions with various growth media. The resultant engi- endochrondral bone formation, these basophilic areas of
neered structures beat like a normal heart and demon- residual cartilage contained hypertrophied cells. It is also
strate biochemical and pharmacological characteristics significant that specimens demonstrating organized bone
similar to normal heart muscle (23). formation at 20 weeks contained a hypocellular bone
To create a tissue-engineered tendon, tenocytes were marrow. In addition, viewing these specimens under po-
isolated by enzymatic digestion of tendons and grown on larized light revealed the presence of lamellar bone (26).
fibrous polymer pieces in vitro and then transplanted sub- Finally, tissue engineering has implications for gene

MOLECULAR THERAPY Vol. 1, No. 1, January 2000 13

Copyright © The American Society of Gene Therapy
 REVIEW

therapy. One critical issue for gene therapy is having a tant challenge. For example, it has been shown that hepa-
large enough number of cells expressing desired gene tocytes can produce different levels of particular proteins
products. In one study the human growth hormone depending on the adhesion of the cells to the material on
(hGH) gene was transfected into rat hepatocytes to see if which the cells are grown (31). To develop tissue-engi-
it was possible to demonstrate in vivo expression of hGH neered organs, it is important to understand how to grow
following transplantation. To accomplish this, the opti- the cells of that organ under conditions that maximize
mal conditions for in vitro gene transfer were first defined the ability of these cells to perform their physiological
in rat hepatocytes. Then, to study these cells in vivo, a roles. Understanding remodeling is also important for
portacaval shunt was performed on rats by ligation of the creating tissues.
portal vein and creation of an end side to side anastomo- Another major area of study is the creation of materials
sis with the vena cava. Seven days later, polymer sponges for tissue engineering. Of particular value is the creation
were implanted into the subcutaneous tissue or mesen- of synthetic materials that have appropriate strength, deg-
tery. The sponges were vascularized over 5 days, and then radation times, microstructure, permeability, and the
the transfected hepatocytes were injected into the ability to contain cell recognition properties that can be
sponges (27). used to regulate cell growth and adhesion. One way that
Histological examination of tissue sections from the this is being attempted is to synthesize polymers in such
cell–polymer construct was performed at different time a way that desired regulatory molecules such as specific
points following hepatocyte implantation. Organized amino acid sequences that can regulate cell behavior, for
plates of viable hepatocytes were seen as late as 15 days example, R-G-D, can be grafted onto them (32).
postimplantation filling the spaces of the prevascularized Still another challenge is the ability to induce vascular-
polymer device. In animals containing implants with the ization of a tissue. One approach being explored involves
transfected cells, serum hGH levels, which were undetect- the use of controlled release of growth factors such as
able prior to cell injection, were found to be within the epidermal growth factor (33). By placing growth factors in
human physiological range on day 2 postimplantation, polymer microcapsules or scaffolds, the growth factor can
but fell thereafter. Although this early study demonstrates be steadily released for weeks, enhancing local vascular-
that tissue-engineering approaches may be useful in as- ization.
sisting in gene therapy, clearly more research is needed In addition, the design of novel polymer structures
(29). through manufacturing techniques such as three-dimen-
sional printing, which can build in complex vascular
FUTURE CHALLENGES structures, is being explored (34). Another challenge in-
volves cyropreservation so that tissues or organs can be
Several significant challenges still face tissue engineering. kept frozen until needed. This is particularly important
One of these is identifying reliable cell sources. The recent for more complex and larger tissues.
development of human embryonic stem cells potentially In summary, tissue engineering has enormous poten-
provides one important approach to addressing this prob- tial, yet many challenges remain. Nonetheless, several
lem (28). Some of the critical issues in this area involve tissue-engineered skins have received FDA approval and
understanding how to control the differentiation of stem many other tissues as discussed above are in advanced
cell populations into the desired cell types, for example, stages of clinical trials. It is hoped that as the years
liver, cartilage, or others. There are also a variety of tech- progress, tissue engineering will be able to be used not
nical challenges such as creating pure cells and processes only to create new tissue structures but also as vehicles in
to create large populations of such cells to make desired which genes can be placed to deliver large numbers of
tissues. cells to create desired gene products, thereby aiding gene
Another major challenge is the creation of universal therapy as well.
donor cells that would not be rejected by the body. One
interesting approach involves using molecules that could REFERENCES
mask the histocompatibility proteins on the surface of
1
Langer, R., and Vacanti, J. P. (1993). Tissue engineering. Science 260: 920 –926.
cells that normally cause the donor cells to be recognized 2
Vacanti, J. P., and Langer, R. (1999). Tissue engineering: The design and fabrication of
as foreign. This type of approach is being studied using living replacement devices for surgical reconstruction and transplantation. Lancet 354:
porcine cells for transplantation to patients with Parkin- 32–34.
3
Bisceglie V. (1933). Uber die antineoplastische immunitat; heterologe Einpflanzung von
son’s disease (27). Tumoren in Huhner-embryonen. Ztschr. Krebsforsch. 40: 122–140.
Another challenge is developing the appropriate biore- 4
Chick, W. L., Like, A. A., and Lauris, V. (1975). Beta cell culture on synthetic capillaries:
actors to affect desired mass transport or impart mechan- An artificial endocrine pancreas. Science 187: 847– 848.
5
Bell, E., Ehrlich, P., Buttle, D. J., and Nakatsuji, T. (1981). Living tissue formed in vitro and
ical characteristics to the cells. One approach has in- accepted as skin-equivalent of full thickness. Science 221: 1052–1054.
6
volved designing bioreactors that have appropriate Burke, J. F., Yannas, I. V., Quimby, W. C., Bondoc, C. C., and Jung, W. K. (1981).
Successful use of a physiologically acceptable artificial skin in the treatment of extensive
mixing to achieve uniform cell seeding. Mixing is also burn injury. Ann. Surg. 194: 413– 448.
important to create engineered tissues because it aids in 7
Vacanti, J. P., Morse, M. A., Saltzman, W. M., Domb, A. J., Perez-Atayde, A., and Langer,
the transport of nutrients and wastes during tissue culti- R. (1988). Selective cell transplantation using bioabsorbable artificial polymers as matrices.
J. Pediatr. Surg. 23: 3–9.
vation and growth (30). 8
Langer, R., and Vacanti, J. P. (1993). Tissue engineering. Science 260: 920 –926.
Regulation of cell behavior represents another impor- 9
Guenard, V., Kleitman, N., Morrisset, T. K., Bunge, R. P., and Aebischer, P. (1992).

14 MOLECULAR THERAPY Vol. 1, No. 1, January 2000

Copyright © The American Society of Gene Therapy
 REVIEW
22
Syngeneic Schwann-cells derived from adult nerves seeded in semipermeable guidance Shinoka, T., Ma, P. X., Shum-Tim, D., Breuer, C. K., Cusick, R. A., Zund. G,, Langer, R.,
channels enhance peripheral-nerve regeneration. J. Neurosci. 12: 3310 –3320. Vacanti, J. P., and Mayer, J. E. (1996). Tissue-engineered heart valves: Autologous valve
10
Schmidt, C. E., Shastri, V. R., Vacanti, J. P., and Langer, R. (1997). Stimulation of neurite leaflet replacement study in a lamb model. Suppl. II Circ. 94: 164 –168.
23
outgrowth using an electronically conducting polymer. Proc. Natl. Acad. Sci. USA 94: Carrier, R., Papadaki, M., Rupnick, M., Schoem, F. J., Bursac, N., Langer, R., Freed, L. E.,
8948 – 8953. and Vunjak-Novakovic, G. (1999). Cardiac tissue engineering: Cell seeding, cultivation
11
Kobayashi, H., Ikada, Y., Moritera, T., Ogura, T., and Honda, Y. (1991). Collagen- parameters, and tissue construct characterization. Biotech. Bioeng. 64: 580 –589.
24
immobilized hydrogel as a material for lamellar keratoplasty. J. Appl. Biomat. 2: 261–267. Langer, R., Vacanti, J. P., Vacanti, C. A., Atala, A., Freed, L., and Vunjak-Novakovic, G.
12
Cima, L. G., Vacanti, J., Vacanti, C., Ingber, D., Mooney, D., and Langer, R. (1991). (1995). Tissue engineering: Biomedical applications. Tissue Eng. 1: 151–161.
25
Tissue engineering by cell transplantation using degradable polymer substrate. J. Biomech. Organ, G. M., Mooney, D. J., Hansen, L. K., Schloo, B., and Vanati, J. P. (1993).
Eng. 113: 143–151. Design and transplantation of enterocyte–polymer constructs: A small animal model for
13
Yarmush, M. L., Dunn, J. C. Y., and Tompkins, G. (1992). Assessment of artificial liver neointestinal replacement in short bowel syndrome. Am. Coll. Surg. Surg. Forum
support technology. Cell Transplant. 1: 323–341. 44: 432.
14 26
Watababe, F. D., Mullon, C. J. P., Hewitt, W. R., Arkadopoulos, N., Kahaku, E., Eguchi, Vacanti, C. A., Kim, W., Upton, J., Vacanti, M. P., Mooney, D., Schloo, B., and Vacanti,
S., Khalili, T., Arnaout, W., Shackleton, C. R., Rozga, J., Solomon, B., and Demetriou, A. A. J. P. (1993). Tissue engineering growth of bone and cartilage. Transplant. Proc. 25: 1019.
27
(1997). Clinical experience with a bioartificial liver in the treatment of severe liver failure: A Fontaine, M. J., Hansen, L. K., Thompson, S., Uyama, S., Ingber, D. E., Mulligan, R.,
phase I clinical trial. Ann. Surg. 225: 484 – 494. Langer, R., and Vacanti, J. P. (1993). Transplantation of genetically altered hepatocytes
15
Sullivan, S. J., Maki, T., Borland, K. M., Mahoney, M. D., Solomon, B. A., Muller, T. E., using cell–polymer constructs leads to sustained human growth hormone secretion in vivo.
Monaco, A. P., and Chick, W. L. (1991). Biohybrid artificial pancreas: Long-term implanta- Transplant. Proc. 25: 1002–1004.
28
tion studies in diabetic, pancreatectomized dogs. Science 252: 718 –721. Thomson, J. A., Itskovitz-Eldor, J., Shapiro, S. S., et al. (1998). Embryonic stem cell lines
16
Atala, A., Vacanti, J. P., Peters, C. A., Mandell, J., Retik, A. B., and Freeman, M. R. derived from human blastocysts. Science 282: 1145–1147.
29
(1992). Formation of urothelial structure in vivo from disassociated cells attached to biode- Faustman, D., and Coe, C. (1991). Prevention of xenograft rejection by masking donor
gradable polymer scaffolds in vitro. J. Urol. 148: 658. HLA class I antigens. Science 252: 1700 –1702.
17 30
Oberpenning, F., Meng, J., Yoo, J. J., and Atala, A. (1999). De novo reconstitution of a Vunjak-Novakovic, G., Obradovic, B., Bursac, P., Martin, I., Langer, R., and Vunjak-
functional mammalian urinary bladder by tissue engineering. Nature Biotechnol. 17: 149 – Novakovic, G. (1998). Chondrogenesis in a cell–polymer– bioreactor system. Exp. Cell Res.
155. 14: 193–202.
18 31
Niklason, L. E., Gao, J., Abbott, W. M., Hirschi, K. K., Houser, S., Marini, R., and Langer, Mooney, D., Hansen, L., Vacanti, J., and Langer, R. (1992). Experimental approach to
R. (1999). Functional arteries grown in vitro. Science 284: 489 – 493. elucidate the mechanism of ultrasound-enhanced polymer erosion and release of incorpo-
19
Vacanti, C. A., Langer, R., Scholl, B., and Vacanti, J. P. (1991). Synthetic polymers rated substances. Macromolecules 25: 123–128
32
seeded with chondrocytes provide a template for new cartilage formation. Plast. Reconstr. Barrera, D. A., Zylstra, E., Lansbury, P. T., and Langer, R. (1993). Synthesis and RGD
Surg. 88: 733–759. peptide modification of a new biodegradable copolymer (polyactic acid-co-lysine). J. Am.
20
Freed, L. E., Grande, D. A., Emmanual, J., Kwan, M., Marquis, J. C., Lingbin, Z., Chem. Soc. 115: 11010 –11011.
33
Dunkelman, N., and Langer, R. (1994). Joint resurfacing using allograft chondrocytes and Mooney, D., Kaufmann, P. M., Sano, K., et al. (1996). Localized delivery of epidermal
synthetic biodegradable polymer scaffolds. J. Biomed. Mat. Res. 28: 891– 899. growth factors improves the survival of transplanted hepatocytes. Biotechnol. Bioeng. 50:
21
Vunjak-Novakovic, G., Freed, L., Biron, R. J., and Langer, R. (1996). Effects of mixing on 422– 429.
34
the composition and morphology of tissue-engineered cartilage. Am. Inst. Chem. Eng. J. 42: Griffith, L. G., Wu, B., Cima, M. J., Powers, M. J., Chaignaud, B., and Vacanti, J. P.
850 – 860. (1997). In vitro organogenesis of liver tissue. Ann. N.Y. Acad. Sci. 831: 382–397.

MOLECULAR THERAPY Vol. 1, No. 1, January 2000 15

Copyright © The American Society of Gene Therapy