Education Section

Tissue engineering in surgery

J.P. Garner
Biomedical Sciences, Dstl Porton Down, Salisbury, SP4 0JQ
Correspondence to: J.P. Garner, Biomedical Sciences, Dstl Porton Down, Salisbury, SP4 0JQ    

Introduction Tissue engineering Arterial conduits

Breast tissue Current problems References

The demands for repair and renewal of worn out or injured human tissue continue to increase and it is now apparent that this demand cannot be met from human donors. A partial solution may be found in living related and trans-species transplantation but these approaches invoke the problems of disease transfer and ethical dilemmas. Tissue engineering is a new technology that seeks to meet these increasing demands by utilising novel cell culture methods in vitro to provide tissue replacements in vivo. This article reviews the current state of tissue engineering and its potential for use in surgery

INTRODUCTION
There is a continuing impetus to improve the quality of life of patients who survive lifethreatening operations or illnesses and are left with considerable morbidity, such as diabetes following pancreatic resection or short bowel syndrome after multiple resections for Crohn’s disease. Allied with this are the advancements in surgical science which seek to treat other diseases which have previously had a dismal outcome such as fulminant hepatic failure. Abdominal organ transplantation continues to be a successful treatment for some but it represents only a partial solution to these problems; it is an approach fundamentally limited by the dearth of donor organs. This is addressed to some degree by the increasing use of live-related donors for partial organ donation, but the deficit of organs to potential recipients is still huge. Tissue engineering represents a potential answer to this problem. This article seeks to explain the scientific basis of tissue engineering and describe the current status of research in some areas specifically relevant to the general surgeon. It is a burgeoning field with numerous articles being published, attempting to bring the promise of tissue engineered surgical solutions closer to a clinical reality. Unfortunately, there are still significant hurdles to overcome in many areas before tissue engineering becomes the routine approach to treating functional tissue deficits.

TISSUE ENGINEERING
The first accepted example of tissue engineering dates back 70 years to when Bisceglie (1933) wrapped mouse tumour cells in a polymer membrane and successfully transplanted them into the chick embryo peritoneum.¹ One definition of tissue engineering is ‘the science of generating tissue, using laboratory molecular biological techniques and the principles of material engineering, to treat a functional or anatomical defect in vivo’. There are three main ways in which this is achieved.

• The re-introduction into the body of individual or clusters of cells to reproduce a specific - often enzymatic - function. It has been used successfully in pancreatic islet transplantation.² A non-enzymatic example is the transplantation of myoblast cells into the area of a myocardial infarction to limit the area of fibrotic scarring and restore contractile function³

• Implantation of an acellular biomaterial scaffold onto or into which native cells from adjacent uninjured tissue repopulate. It acts as a template for regeneration - an example is the use of a synthetic polymer tube as a guide to peripheral nerve regeneration⁴

• The implantation of pre-formed biomaterial constructs. These are generally cellular and tend to be more representative of the native organ with a greater and more rapid functionality than the other two tissue engineering methods. An example would be the implantation of a bioactive glass construct to reconstruct the middle ear ossicular chain.⁵

The cells used in tissue engineering may be autologous, allogeneic or xenogeneic. Autologous cells are derived from the specific individual into which they will be later reimplanted. Such autologous cells require harvest or donation in advance of need, subsequently coupled with in vitro culture. This imparts both technical and time constraints but does avoid the issues of immunogenicity/rejection and transmission of disease. Allogeneic cells are derived from an individual of the same species other than the recipient. Common examples of allogeneic donations are cadaveric organ transplantation and pooled human blood transfusion. Xenogeneic cells are derived from a species different to that of the recipient. Both allogeneic and xenogeneic cells raise important and potentially insurmountable issues of disease transmission, host rejection and ethical disquiet but retain the potential advantage of being cultured and constructed in advance of need. This would enable the generation of an ‘off-the-shelf’ replacement tissue.

Whatever the source, the cells used may be either adult or embryonic. Embryonic cells are, by definition, pluripotent and may be channelled into whichever cell line is required but this procedure is not technically straightforward and there is a limited supply of allogeneic embryonic cells available. Adult cells may be pluripotent or ‘committed’ to following a predetermined cellular heritage; harvest of adult cells may leave anatomical or functional defects at the donor site. Other than small cell clusters, implanted cells require a scaffold to act as a three-dimensional template.

Scaffolds may be natural or synthetic, degradable or not. Examples of commonly used scaffolds are: an acellular small intestinal submucosal tube used in the generation of artificial arteries which is a natural non-degradable scaffold, or the polyglactin (Vicryl^TM) scaffolds on which neocartilage is generated - the Vicryl^TM hydrolyses as the cell population increases. The combination of scaffold and cells is known as a construct.

In general, the majority of tissue engineering concepts are still in their infancy. Clinical utility and long-term follow-up is available for only a few products, notably cultured skin substitutes and cartilage renewal systems which are both outside the scope of this article.^6,7 Some organ-specific research is more advanced than others, a state of affairs dictated in part by the relative complexity of the organs concerned and also the strength of the clinical need. Several areas are described below - some in considerable technical detail - in order to illustrate the extent of progress, to date, the volume of work being undertaken and the nature of the problems still to be surmounted before tissue engineering can provide solutions to the problems posed by everyday surgical practice.

Figure 1: A sheet of small intestinal submucosal scaffold. Animal gut is enzymatically treated to render an acellular wafer thin tissue that can act as a scaffold in a wide range of TE applications including arterial conduits, bladder reconstruction and the regeneration of small intestine itself.

Figure 2: A collagen sponge scaffold. Collagen is extracted from pig skins, homogenised and enzymatically treated to produce a collagen solution free of antigenic material. It can be molded and freeze-dried before reinforcement over the base with polyglycolic acid felt to produce a robust scaffold. Reproduced from the Int J of Artif Organs 2001; 24:50-4 with permission from Wichtig Editore.

Figure 3: Collagen sponge scaffold being tested in a small intestine model. A 5cm length of jejunum has been resected and a silicone stent interposed to maintain a patent tube. A sheet of collagen sponge (arrow) is waiting to be wrapped around the tube to mimic the serosal layers. After 4 weeks the silicone tube is removed endoscopically and the neointestine exposed to gut content. Reproduced from the Int J of Artif Organs 2001; 24:50-4 with permission from Wichtig Editore.

Figure 4: The microscopic findings of regenerated small intestine after 16 weeks (Masson-Tricrom staining; 100x magnification). Intestinal mucosa with a villous pattern has covered the surface of the regenerated submucosal tissue. A thin muscular layer is also evident (arrow). Reproduced from the Int J of Artif Organs 2001; 24:50-4 with permission from Wichtig Editore.

Figure 5: An interposition graft of SIS scaffold into an experimental oesophageal defect. Regeneration of all layers of the oesophageal wall occurs. Such therapy may become useful in the treatment of oesophageal atresia or Barratts oesophagus.

ARTERIAL CONDUITS
Given the large numbers of both coronary and peripheral artery bypass grafting that take place each year, the development of a successful, reliable, artificial small calibre arterial conduit has been termed the ‘Holy Grail’ for cardiovascular surgeons.⁸ Whilst Dacron^TM and PTFE are successful large calibre grafts at diameters below about 6 mm, their patency rates are poor.⁹ Before embarking on a project to engineer the ideal small calibre arterial graft it is useful to consider the properties it should possess. It should be elastic and able to withstand repetitive dynamic loading ie pulsatile flow. It must achieve burst pressures significantly greater than systolic pressure and present a non-thrombogenic surface that retains patency. In addition to these absolute requirements, the ideal conduit would also retain vasoactivity in response to circulating chemical modulators, and should integrate into the host system rather than represent a persistent foreign body. Finally, and almost uniquely when considering tissue engineering implants, it must be able to function immediately upon implantation.

Faced with such a formidable task, what advances has tissue engineering made in constructing the perfect artificial artery? The four general approaches of endothelialisation of synthetic grafts, collagen scaffolds, cell self-assembly and seeded biosynthetic scaffolds are outlined below.

Endothelialisation of synthetic grafts: Early attempts to solve the ‘Holy Grail’ problem were centred on overcoming the inherent thrombogenicity of the luminal surface of small diameter synthetic grafts by seeding them with endothelial cells (ECs). This is the simplest approach, utilising current graft technology and availability, of tackling the thrombogenic nature of its lumina. In order to avoid the issue of immunogenicity, this approach requires an autologous cell source (usually from a previously excised vein or from the microvessels in a liposuction aspirate) which is then cultured in vitro before being seeded onto the graft. This technique has two inherent flaws. Firstly, it is actually very difficult to get ECs to adhere to the luminal surface of the graft, although novel adherence promoters such as a recombinant fibronectin compound have helped.¹⁰ Secondly, it does not confront the issues of vasoactivity and biointegration, but as larger calibre synthetic grafts seem to work well enough without confronting these issues it is open to debate whether this is a major problem.

Collagen scaffolds: Two approaches have been used. Weinberg and Bell (1986) have mimicked the advential and medial layers of native artery using culture derived collagen / fibroblast gel sheets and a similar collagen / smooth muscle cell (SMC) sheet wrapped over each other to form a tubular construct.¹¹ These were then lined by an EC monolayer. This artificial artery represented a major advance in conduit technology and presented theoretical advantages by promoting intercellular signalling via the collagen matrix.¹² Unfortunately, the tensile strength of such constructs was woefully inadequate for physiological use and required external support from a Dacron^TM framework, which introduced the problems of a lack of remodelling and vasoactivity outlined above.¹¹ The second collagen-based approach employs an acellular scaffold for native in vivo growth. Small intestine may be treated by detergent and enzymatic means to leave an acellular small intestine submucosa (SIS) tube. This may be further treated with cross-linked collagen and heparinised before implantation into animal hosts.¹³ The SIS graft represents a robust conduit with good compliance when compared with a standard vein graft. Animal studies have yielded reasonable medium-term patency rates.^14-16 Unfortunately, whereas these animal studies have shown a smooth EC neo-intima when examined at necropsy, it is widely recognised that the human endothelial cells are significantly more reluctant to populate a thrombogenic surface. This is an area of active current research.¹⁷

Cell self-assembly models: L’Heureaux et al. (1995) have grown flat sheets of SMCs and fibroblasts and then rolled them up around a central tubular support. This is removed some eight weeks later after maturation of the construct and subsequently followed by EC seeding. This produced a model of a trilaminar artery with an experimental burst pressure of 2000 mmHg. Gratifyingly, this construct method also preserved many cellular functions, often lost in standard culture environments such as von Willebrand factor expression by endothelial cells and good platelet adhesion inhibition in vitro. Unfortunately, this success did not transfer to the in vivo model with occlusion rates of 50% occuring at one week.¹⁸ It is also not clear whether this arterial conduit would function vasoactively. However, if autologous cells are used, this completely cell-cultured method would obviate the problem of immunogenicity.

Seeded biosynthetic scaffold constructs: By using a degradable scaffold, SMCs can be seeded onto an appropriately sized and shaped template. As the culture progresses the scaffold is gradually replaced by a tubular model of SMCs; these can then be given an EC lining. An important feature of this type of modelling is the use of bioreactor culture vessels.¹⁹ These impart dynamic mechanical stresses into the culture formulation which has been shown to markedly improve cellular orientation into a manner much more closely resembling a native artery.²⁰ This results in higher tensile strength and burst pressure.¹⁷ Niklason et al. (1999) used a polyglycolic tubular mesh and imparted a dynamic pulsatile radial distension force during eight weeks of culture; the polyglycolic mesh had by then been replaced by a SMC layer. After EC seeding and further culture, this bilaminar artificial artery had a burst pressure of 2000mmHg and exhibited serotonin and endothelin-1 vasoactivity. Patency was maintained for up to four weeks after porcine carotid implantation.²¹

Whilst remarkable advances have been made in the pursuit of the artificial small calibre vascular conduit, there are significant hurdles still to be overcome before a tissue engineering vessel is available for clinical use. It has been suggested that it will be at least another decade before a marketable artery will be produced.¹⁷

BREAST TISSUE
An increasing trend towards skin-sparing mastectomy, which maintains the residual skin envelope, has allowed reconstructive options to flourish which potentially lessen the cosmetic impact of autologous tissue transfer.²² Nevertheless, there are often still considerable problems associated with methods of reconstruction such as abdominal wall weakness and herniation after a transabdominal myorecutaneous flap or shoulder weakness and prominent scarring after a latissmus dorsi myorecutaneous reconstruction flap. Given that the problem is one of volume restitution, a cultured adipocyte construct would seem to be an ideal solution. Initial work sought to unravel the cellular and extracellular signalling mechanisms involved in the differentiation of the mesenchymal stem cell via many intermediaries into the adult adipocyte. Amongst a legion of cell-to-cell and intracellular messengers it appears that insulin and insulin-like growth factor-1 (IGF-1) have pre-eminence.²³ Coupling this discovery with the use of poly(lactic-co-glycolic-acid)-polyethylene glycol (PLAG-PEG) microspheres for insulin and IGF-1 delivery, Yuksel et al. (2000) demonstrated significant increases in the weight of rat adipofascial flaps after insulin/IGF-1 delivery sphere implantation, compared with sham spheres.²⁴ This was due to an increase in the number of adipocytes rather than hypertrophy of existing fat cells and suggests a cellular proliferation and differentiation that was absent from the shamtreated side. Subsequent use of similar spheres in a non-adipose environment demonstrated fat generation de novo.²⁵ This work was extended to the insulin/IGF-1 microsphere seeding of a PGLA hemispherical scaffold before implantation into rats.²⁶ Histologically, after 12 weeks there was no residual scaffold, only a fibroelastic/adipose breast shape, albeit 30% smaller than the original scaffold. Work continues to identify methods of increasing yield and rate of growth on these breast scaffolds which ultimately could be used as an aesthetic replacement for the human breast although no evidence regarding the potential longevity of such implants has accrued.

SMALL INTESTINE
Short bowel syndrome remains a difficult clinical problem to manage whether it is a result of congenital atresia or multiple resections for Crohn’s disease or infarction. Permanent total parenteral nutrition has been used effectively in some cases to ameliorate the metabolic and nutritional consequences of short bowel syndrome but is itself beset with problems of hepatic failure, sepsis and access difficulties.²⁷ Delaying small bowel transit by segment inversion is only partially successful and small bowel transplantation is still in its infancy.²⁸ There are still problems of donor organ supply, immunosuppression and rejection - the five-year graft survival rate is only 40-50%.²⁹ This clinical need makes the small intestine one of the most exciting areas of tissue engineering research - not least because small bowel has been used as a scaffold for engineering a variety of other tissues since the 1960s.³⁰

Two differing approaches have been used. Tavakk et al. (2001) have grown neointestinal cyst structures in vivo after implantation of polyglycolic acid scaffolds seeded with partially denatured small bowel tissue derived from an allogenic source.³¹ The cysts are opened after maturation and anastomosed to the in situ intestine. Examination of this neointestinal tissue has shown it to regenerate the normal structural features of the small intestine including mucosa, submucosa, a thin muscularis layer and villus formation.³² At a cellular level, there is normal orientation and distribution of new enterocytes and cellular transport function is intact.³³ In addition, after 20 weeks exposure to the normal intestinal environment in vivo, the neointestine is populated with a normal range of immune cells including B, T and natural killer cells as well as macrophages.³⁴

The second approach has two subtypes, both using a collagen scaffold. Chen and Badylak (2001) extended the techniques that had utilised an SIS tube for blood vessels, dura mater and urinary bladder replication and investigated its potential as a scaffold for small intestinal regeneration.^{13, 35-37} After creation of gut defects by excision of 50% circumference of a small intestinal segment in 23 experimental dogs, SIS patches of approximately 7x3 cm were sutured in place. A further four dogs underwent a small bowel segmental resection with an interposition graft of approximately 6 cm of tubular SIS anastomosed at either end. In the patch group, 20 out of 23 dogs survived, the remainder succumbing to peritonitis from patch leakage. Macroscopically, there was excellent integration of the patches into native bowel with shrinkage of patch size to only 20% of original size at one year. Histological sections at six months demonstrated virtually normal small intestinal wall. The tube grafts were less successful with three out of four dogs suffering anastomotic breakdown and the fourth showing a clinically apparent luminal stenosis of the graft. 

Hori et al. (2001) used a collagen sponge scaffold derived from homogenized pig skin over a silicone tube to reconstruct a segment of previously bypassed jejunum in a dog model.³⁸ The silicone tube was removed endoscopically at four weeks. Histological examination showed a regenerated submucosal layer lined by intestinal mucosa with a villus pattern. There was a thin muscularis layer and a firmly adherent serosa. Two out of six dogs had significant stenosis at the graft site.

Small intestinal replacement employing tissue engineering is a promising approach. There are, however, specific areas of concern that must be addressed before clinical trials take place: ensuring sufficient muscularis generation for active peristalsis and demonstration in vivo of acceptable levels of absorptive, secretory and immunological function.

LARGE INTESTINE
Total colectomy may be associated with significant long-term functional problems including bile salt malabsorption and increased stool frequency.³⁹ Ileal pouch reconstruction is by no means a panacea with a long-term risk of pouchitis of about 50%.⁴⁰ Some of these problems may be amenable to tissue engineering solutions. Using a modified version of the small bowel technique described above, implanted colon-seeded scaffolds into the omenta of rats has been carried out.⁴¹ Four weeks later, the rats were randomised to have formation of an end ileostomy, or ileostomy and an interposition segment of tissue engineered colon just proximal to the stoma. These rats demonstrated significantly less weight loss and longer stool transit times than the standard experimental animals. Furthermore, such animals had a decreased stool water content, evidence of greater bile salt absorption and a more normal short chain fatty acid metabolism than controls. Histologically, the tissue engineered colon pouches were larger than at initial anastomosis with a normal large bowel architecture. Whilst large scale production of tissue engineered colon tubes has many problems to overcome, this experiment has demonstrated by ‘proof-of-principle’ that functional colon can be tissue engineered.

LIVER
The use of tissue engineering in hepatology is twofold. It is recognised that there is a huge shortfall of cadaveric liver donors, compared with the requirements and that living sharedorgan donation is still not sufficient to meet these demands. Additionally, there is no reliable liver support device available to keep a patient alive when acute liver failure supervenes. It is known that in acute liver failure only 20% of those patients managed by medical means will survive and that the only reasonable expectation of cure lies with orthotopic transplantation.⁴²

Tissue engineering may contribute in two ways - the creation of a liver support device to keep the patient alive whilst definitive treatment (transplantation or regeneration) occurs and secondly, the creation of a functional implantable liver replacement in lieu of orthotopic transplantation. Extracorporeal liver support systems, generically termed bio-artificial livers, generally incorporate a system of isolated hepatocytes in a synthetic suspension through which plasma is driven.⁴³ Such systems function reasonably well, but there are problems due to the short functional life-span of the hepatocytes; current liver cell cultures remain stable for only up to eight hours before replacement is required. This raises issues, cost being one of them, regarding the suitability of such devices for long-term liver replacement. As there is a shortage of donor livers for orthotopic transplantation, a similar dearth of organs to provide hepatocytes for use in bio-artificial livers exists; consequently most devices use porcine hepatocytes. Using pig cells raises the possibility of transmission of zoonoses, particularly pig endogenous retrovirus, although to date there have been no cases of in vivo transmission.⁴⁴

Initial results of these devices has proved promising. A phase one clinical trial reported the experience of a porcine bio-artificial liver in three groups of patients requiring liver support.⁴⁵ Sixteen out of 18 patients with fulminant hepatic failure survived (15 were bridged to a successful orthotopic transplantation and one recovered native liver function). All three patients who suffered primary failure of an orthotopic transplantation were bridged to re-transplantation and two out of ten patients with decompensation of chronic liver failure not suitable for transplantation were bridged to successful orthotopic transplantation, the remainder succumbing to their underlying disease.

The other method of replicating hepatic function is by in vivo transplantation of hepatocytes. Several methods and sites for such transplantation have been described: intraportal injection with lodgement in the native liver, subcapsular injection in both kidney and spleen and transplantation into pre-implanted polyvinyl alcohol sponges.^46-48 Notwithstanding the potential shortfall of donor human hepatocytes, this technique appears promising. A clinical case, whereby intraportal injection of human hepatocytes under cyclosporin immunosuppression allowed regeneration of the native liver and recovery from fulminant hepatic failure, has been reported and illustrates the technical feasibility of this technique.⁴⁶

It has not yet been defined where the optimal recipient site of engraftment should be, although adjunctive procedures required to provide adequate hepatotrophic stimulation for heterotopically transplanted liver cells have been identified.⁴⁹

Allogeneic hepatocytes were transplanted into a polyvinyl alcohol sponge that had previously been implanted into the greater omentum of experimental rats - a technique previously shown to be able to support a whole liver-equivalent cell mass.⁴⁸ A portocaval shunt exposed the sponge (in the systemic circulation) to portal blood flow which appears to be vital for hepatotrophic stimulation and significantly increased the volume of graft attachment compared with controls.⁴⁹ Concomitant transplantation of Islets of Langerhans cell clusters increased this still further and appears to be the optimal adjunctive regimen for hepatocyte transplantation.⁵⁰

PANCREAS
Pancreatic replacement focuses on two separate areas; injection of isolated clusters of islets cells and the production of an artificial pancreas. It is notable that all pancreatic tissue engineering research is targeted at the endocrine pancreas, more specifically insulin secretion, and no attempts are made to address the function of the exocrine pancreas. 

The most successful group, to date, are the Edmonton group who have recently published long-term follow-up of 17 patients who have completed their transplant protocol; 11 are now insulin independent.⁵¹ Their technique involves intraportal injection of allogeneic cadaveric-derived islet cells. A non-steroidal immunosuppression regimen is used to minimise the side-effect profile but these have still been substantial with evidence of immunosuppression induced renal failure, anaemia and leucopaenia. Nevertheless, they have demonstrated a good clinical outcome in a reasonable number of patients previously blighted by labile diabetes. In attempts to overcome the problems of immunosuppression, several groups are investigating techniques with which to defeat the human immune system and introduce islet cell clusters safe from immunogenic reflection. One such method is by encapsulating them to prevent host recognition of the implant as foreign - immunoisolation. Several methods have been tried including alginate capsules, hydrogels and semi-permeable membranes but, unfortunately, as well as inhibiting immune recognition, they generate a fibrous foreign body reaction that inhibits oxygen and nutrient diffusion and diminishes islet function and longevity.⁵² A novel tissue engineering solution used autologous chondrocytes cultured in vitro to produce a membrane that can encapsulate the islet clusters without inhibiting insulin production.⁵² It is hoped that when implanted it will both avoid immune rejection and provide normoglycaemia.

By harnessing hydrogel technology, Chae et al. (2001) have constructed a biohybrid artificial pancreas.⁵³ The hydrogel compound N-isopropylacrylamide (NiPAAm) mixed with a little with acrylic acid exhibits thermosensitive reversible gelation at temperatures between 30-34°C. ⁵⁴ By suspending islet cell clusters in a NiPAAm matrix in liquid form it can be injected into a subcutaneously implanted biohybrid artificial pancreas. When the insulin secretion from the islets wanes, surface cooling of the pouch renders its contents liquid once more and suitable for aspiration followed by repeat injection of a fresh islet preparation.

This list is by no means exhaustive and virtually every human tissue is the subject of tissue engineering research. Urologists may in due course look forward to tissue engineering neobladders, and artificial sphincters are already being used in clinical trials for the treatment of vesico-ureteric reflux.^55,56 Orthopaedic surgeons use autologous chondrocyte culture systems to repair osteochondral defects.7 The treatment of cardiac failure and cardiomyopathy by implantation of tissue engineered muscle cells is an exciting prospect for cardiothoracic surgeons.⁵⁷ It is likely that in time some, if not all, of these tissue engineering projects will find a place in the routine practice of most physicians and surgeons.

CURRENT PROBLEMS
It is easy to be engulfed by a tide of enthusiasm for these innovative techniques that promise to revolutionise surgical practice, but it is important to recognise the enormity of some of the problems that, as yet, have not been satisfactorily addressed.

Time: Many of the tissue culture techniques described herein are, by surgical standards, inordinately slow; for example L’Heureux’s artificial artery took 12 weeks of culture. Such techniques are not relevant when a surgical solution is required urgently. The ideal would be to manufacture these tissue constructs in advance of need and have them ready to use ‘off-the-shelf’. This raises two further interlinked problems: immunogenicity and preservation.

Immunogenicity: The current successful tissue engineering products such as cartilage repair systems utilise autologous cells in elective situations. Unless individuals pre-donate cells to generate a unique bank of ‘spare-part organs’, the problems of the host immune response to allogeneic or xenogeneic tissues must be overcome. Current immunosuppressive regimen are costly and are associated with significant deleterious effects. Approaches such as the autologous chondrocyte wrapping, described above, seem promising. Perhaps an advanced donation of cartilage would be sufficient for an individual to maintain a stock of immunoisolatory chondrocyte wrappers to cover allogeneic tissue transplants.

Preservation: Assuming that the problems of immunogenicity can be overcome with ‘wrappers’ or other techniques, a method of preserving these engineered artificial tissues for ‘off-the-shelf’ use must be identified. Ever since the discovery of cryoprotective agents like glycerol and its successful preservation of human gametes and embryos, science has tried to freeze ever more complex tissues.⁵⁸ Unfortunately, cryopreservation results in formation of extracellular ice crystals which, on thawing, disrupt the intercellular bonds and signalling mechanisms, rendering them non-functional.⁵⁹ To a certain extent, high dose cryoprotective agents overcome this by inducing amorphous solidification (vitrification) but such doses of cryoprotective agents are toxic to human tissue.^{59, 60} Both techniques require low-temperature storage and relatively ‘high-tech’ equipment. Thus, they are unlikely to be useful on a worldwide basis, where perhaps the need is greatest. A potential solution lies in heavy sugar technology. Several lower order biological systems such as seeds, fungi and brine shrimp routinely survive dry dormancy unharmed and it is thought that this is related to high concentrations of heavy sugars such as trehalose.⁶¹ Its applicability to the preservation of engineered tissue is being evaluated.

Figure 6: To many this was a pinnacle of achievement in tissue engineering - the first three dimensional realistic construct ‘grown’ from individual cells; to others it was an abomination and this typifies the problems that surround the perception of tissue engineering amongst the general public.

Vascularisation and Organogenesis: A fundamental limitation to the size of any implantable living construct is its blood supply. Complex tissues can receive nutrients and oxygen (and eliminate waste) by diffusion over only a few micrometres. Thus, any attempts at fabricating complex organoids must address the problems of angiogenesis. Incorporation of angiogenetic factors such as vascular endothelial growth factor is one potential solution, encouraging accelerated ingrowth of vessels from the host.⁶² However, even under the influence of vascular endothelial growth factor, this still occurs too slowly to sustain cellular survival. Alternatively, incorporation of preformed branching capillary networks into a tissue construct scaffold and anastomosis to a host vessel would appear feasible. Two techniques have been investigated to create these capillary networks. ‘Solid free form fabrication’ lays down layer after layer of a PGLA liquid copolymer using a modified ink jet printer.⁶³ When dry, the powder is leached out leaving a 3-D network of branched channels and spaces throughout the scaffold. This can subsequently be seeded with engineered cells such as hepatocytes and excellent synthetic activity has been demonstrated from hepatocytes organised in this manner.⁶⁴ A second method employs micromachining of silicone wafers to produce capillary networks down to 10 microns in diameter which are then seeded with endothelial cells. Proliferation and extracellular matrix deposition follows; the cell-sheets can then be lifted from the silicon mold and folded into three-dimensional constructs.⁶⁵

Ethics: Many of the tissue engineered projects outlined above utilise cells and tissue from allogeneic and xenogeneic sources, both of which generate intense ethical debate. Similarly, embryonic stem cells, whilst representing an immensely useful tool in the genesis of many tissues, remain an emotive issue for many areas of the public. It would seem a pointless exercise engineering tissues if the intended recipients are unhappy to receive them. The ethical and moral dilemmas surrounding this work need to be clarified and resolved before its acceptance by the non-medical public.

Isolated cellular transplantation of both pancreatic islets and hepatic cells have both been used on a limited clinical basis but the rest of tissue engineered cells and tissues are still experimental. In the foreseeable future, improvements in hepatocyte culture may deliver a serviceable liver assist device into routine practice, perhaps in the next 5-10 years. The small diameter arterial conduit will take at least another decade to develop and if the eventual aim of a non-immunogenic ‘offthe-shelf’ vessel is to be realised then 20-years is a more realistic time frame. Somewhere between these two extremes lie the other tissues; intestine, both small and large, is at least a decade away but a reliable breast reconstruction from cultured adipocytes could enter clinical trials within 10 years.

SUMMARY
Tissue engineering has made enormous advances in the last decade and the first generation of clinical products are now in the medical marketplace. There are still significant hurdles to overcome before the general surgeon will be able to use a tissue engineering solution for a problem in everyday practice. It is almost inevitable, however, that the surgeon of 2030 will be widely versed in the application of the results of tissue engineering.

ACKNOWLEDGEMENTS
I am profoundly grateful to Dr Joseph Vacanti, John Homans Professor of Surgery at Harvard Medical school, Dr Steve Badylak, Director of Pre-Clinical Tissue Engineering at the McGowan Institute for Regenerative Medicine, University of Pittsburgh, and Dr Yoshio Hori, of the Tohuku University Biomedical Engineering Research Organisation, Japan for their generous provision of images to illustrate this review.

REFERENCES
1. Bisceglie V. Uber die antineoplastische Immunitat; heterologe Einpflanzung von Tumoten in Huhner-Embryonen. Ztschr Krebsforsch 1933; 40: 122-40.
2. White SA, James RFL, Swift SM, Kimber RM, Nicholson ML. Human islet cell transplantation: future prospects. Diabet Med 2001; 18: 78-103.
3. Menasché P, Scorsin M, Hagege A et al. Myoblast transplantation for heart failure. Lancet 2001; 357: 279-80.
4. Hadlock T, Sundback C, Koka R, Hunter D, Cheney M, Vacanti JP. A novel, biodegradable polymer conduit delivers neurotrophins and promotes nerve regeneration. Laryngoscope 1999; 109: 1412-6.
5. Merwin GE, Atkins JS, Wilson J, Hench LL. Comparison of ossicular replacement materials in a mouse ear model. Otolaryngol Head Neck Surg 1982; 90(4): 461-9.
6. Boyce ST, Kagan RJ, Meyer NA, Yakuboff KP, Warden GD. The 1999 clinical research award. Cultured skin substitutes combined with Integra to replace native skin autograft and allograft for closure of full thickness burns J Burn Care Rehabil 1999; 20:453-61.
7. Peterson L, Minas T, Brittberg M, Nilsson A, Sjogren-Jansson E, Lindahl A. Two- to 9-year outcome after autologous chondrocyte transplantation of the knee. Clin Orthop 2000; 374 : 212-34.
8. Conte MS. The ideal small arterial substitute: a search for the Holy Grail? FASEB J 1998; 12:43-5.
9. Whittamore A. Infrainguinal bypass. In Rutherford R ed. Vascular Surgery, Philadelphia : WB Saunders 1995, 23-31.
10. Mazzucotelli JP, Moczar M, Zede L, Bambang LS, Loisance D. Human vascular endothelial cells on expanded PTFE pre-coated with an engineered protein adhesion factor. Int J Artif Org 1994; 17:112-7.
11. Weinberg CB, Bell E. A blood vessel model constructed from collagen and cultured vascular cells. Science 1986; 231: 397-400.
12. Patel NH, Butler PEM, Buttery L, Polak JM, Tolley NS. Tissue engineering in ENT surgery. J Laryngol Otol 2002; 116: 165-9.
13. Badylak SF, Lantz GC, Cofey A, Geddes LA. Small intestine submucosa as a large diameter vascular graft in the dog. J Surg Res 1989; 47: 74-80.
14. Hiles MC, Badylak SF, Lantz GC, Kokini K, Geddes LA, Morff RJ. Mechanical properties of xenogenic small-intestine submucosa when used as an aortic graft in the dog. J Biomed Mat Res 1995; 29: 883-91.
15. Roeder R, Wolfe J, Lianakis N, Hinson T, Geddes LA, Obermiller J. Compliance, elastic modulus and burst pressure of small intestine submucosa (SIS), small diameter vascular grafts. J Biomed Mat Res 1999; 47: 65-70.
16. Sandusky GE, Lantz GC, Badylak SF. Healing comparison of small intestine submucosa and ePTFE grafts in the canine carotid artery. J Surg Res 1995; 58: 415-20.
17. Nerem RM, Seliktar D. Vascular tissue engineering. Ann Rev Biomed Eng 2001; 3: 225-43.
18. L’Heureux N, Paquet S, Labbe R, Germain L, Auger FA. A completely biological tissue-engineered human blood vessel. FASEB J 1998; 12:47056.
19. Seliktar D, Black RA, Vito RP, Nerem RM. Dynamic mechanical conditioning of collagen-gel blood vessel constructs induces remodelling in vitro. Ann Biomed Eng 2000; 28:351-62.
20. Kanda K, Matsuda T, Oka T. mechanical stress-induced orientation and ultrastructural change of smooth muscle cells cultured in three-dimensional collagen lattices. Cell Transplant 1994; 3:481-92.
21. Niklason LE, Gao J, Abbott WM et al. Functional arteries grown in vitro. Science 1999; 284: 489-93.
22. Modena S, Benassuti I, Marchiori M et al. Mastectomy and immediate breast reconstruction: oncological considerations and evaluation of two different methods relating to 88 cases. Eur J Surg Oncol 1995; 21: 36-41.
23. Thuillier P, Baillie R, Sha X, Clarke SD. Cytosolic and nuclear distribution of PPARgamma2 in differentiating 3T3-L1 preadipocytes. J Lipid Res 1998; 39: 2329-38.
24. Yuksel E, Weifeld AB, Cleek R et al. Increased free fat graft survival with longterm local delivery of insulin, IGF-1 and bFGF by PLGA/PEG microspheres. Plast Rec Surg 2000; 105: 1712-20.
25. Shih C, Higuchi T, Himmelstein KJ. Drug delivery from catalysed erodible polymeric matrices of poly-(ortho ester) s. Biomaterials 1984; 5: 237-40.
26. Shenaq SM, Yuksel E. New research in breast reconstruction. Clin Plast Surg 2002; 29: 111-25.
27. Nousia-Arvanitakis S, Angelopoulou-Sakadami N, Metroliou K. Complications associated with total parenteral nutrition in infants with short bowel syndrome. Hepatogastroenterology 1992; 39(2): 169-72.
28. Panis Y, Messing B, Rivet P et al. Segmental reversal of the small bowel as an alternative to intestinal transplantation in patients with small bowel syndrome. Ann Surg 1997; 225: 401-7.
29. Reyes J, Bueno J, Kocoshis S et al. Current status of intestinal transplantation in children. J Pediatr Surg 1998; 33: 243-54.
30. Matsumoto T, Holmes RH, Burdick CO, Hesiterkamp CA 3rd, O’Connell TJ Jr. Replacement of large veins with free inverted segments of small bowel: autografts of submucosal membrane in dogs and clinical use. Ann Surg 1966; 164: 845-8.
31. Tavakkolizadeh A, Ashley SW, Vacanti JP Whang EE. Tissue-engineered intestine: progress towards a functional and physiological mucosa. Transplantation Rev 2001; 15: 178.
32. Tavakkolizadeh A, Stephen AE, Kaihara S et al. Topography of epithelial proliferation and apoptosis in the tissue-engineered intestine. Surg Forum 2000; 51:59.
33. Choi RS, Reigler M, Pothoulakis C et al. Studies of brush border enzymes basement membrane components and electrophysiology of tissue-engineered neointestine. J Pediatr Surg 1998; 33(7): 991-6.
34. Perez A, Grikscheit TC, Blumberg RS, Ashley SW, Vacanti JP, Whang EE. Tissue engineered small intestine: ontogeny of the immune system. Transplantation 2002; 74(5): 619-23.
35. Chen MK, Badylak SF. Small bowel tissue engineering using small intestinal submucosa as a scaffold. J Surg Res 2001; 99: 352-8.
36. Cobb MA, Badylak SF, Janas W, Boop FA. Histology after dural grafting with small intestinal submucosa. Surg Neurol 1996; 46: 389-93.
37. Kropp BP, Eppley BL, Prevel CD et al. Experimental assessment of small intestinal submucosa as a bladder wall substitute. Urology 1995; 46: 396-400.
38. Hori Y, Nakamura T, Matsumoto K, Kurakowa Y, Satomi S, Shimizu Y. Tissue engineering of the small intestine by acellular collagen sponge scaffolding. Int J Artif Organs 2001; 24(1): 50-4.
39. Papa MZ, Karni T, Koller M et al. Avoiding diarrhoea after subtotal colectomy with primary anastomosis in the treatment of colon cancer. J Am Col Surg 1997; 184: 269-72.
40. Meagher AP, Farouk R, Dozois RR, Kelly KA, Pemberton JH. J ileal pouchanal anastomosis for chronic ulcerative colitis: complications and long-term outcome in 1310 patients. Br J Surg 1998; 85: 800-3.
41. Grikshceit TC, Ogilvie JB, Ochoa ER, Alsberg E, Mooney D, Vacanti JP. Tissue-engineered colon exhibits function in vivo. Surgery 2002; 132: 200-4.
42. Stockmann HBAC, Ijzermanns JNM. Prospects for the temporary treatment of acute liver failure. Eur J Gastroenterol Hepatol 2002; 14(2): 195-203.
43. Matsamura KN, Guevara GR, Huston H et al. Clinical experience with a bioartificial liver in hepatic failure: preliminary clinical report. Surgery 1987; 101: 99-103.
44. Paradis K, Langford G, Long Z et al. The XEN 111 study group, Otto E. Search for cross-species transmission of porcine endogenous retrovirus in patients treated with living pig tissue. Science 1999; 285: 1236-41.
45. Watanabe FD, Mullon CJ, Hewitt WR et al. Clinical experience with a bioartificial liver in the treatment of severe liver failure. A phase 1 clinical trial. Ann Surg 1997; 225: 484-94.
46. Fisher RA, Bu D, Thompson M et al. Defining hepatocellular chimerism in  liver failure patient bridged with hepatocyte infusion. Transplantation 2000; 69: 303-7.
47. Papalois A, Arkadpolous N, Pataryas TH, Papadimitriou J, Golem B. Combined hepatocyte-islet transplantation: an allograft model. Transpl Int 1994; 7(suppl): S432-5.
48. Uyama S, Kaufmann PM, Takeda T, Vacanti JP. Delivery of whole liverequivalent hepatocyte mass using polymer devices and hepatotrophic stimulation. Transplantation 1993; 55: 932-5.
49. Kaufmann PM, Sano K, Uyama S, Takeda T, Vacanti JP. Heterotopic hepatocyte transplantation: assessing the impact of hepatotrophic stimulation. Transplant Proc 1994; 24: 2240-1.
50. Kauffman PM, Sano K, Uyama S et al. Evaluation of methods of hepatotrophic stimulation in rat heterotopic hepatocyte transplantation using polymers. J Paediatr Surg 1999; 34(7): 1118-23.
51. Ryan EA, Lakey JRT, Paty BW et al. Successful islet transplantation: continued insulin reserve provides long-term glycemic control. Diabetes 2002; 51: 2148-57.
52. Pollok J-M, Begemann J-F, Kaufmann PM et al. Long-term insulinsecretory function of islets of Langerhans encapsulated with a layer of confluent chondrocytes for immunoisolation. Paediatr Surg Int 1999; 15: 164-7.
53. Chae SY, Kim SW, Bae YH. Bioactive polymers for biohybrid artificial pancreas. J Drug Target 2001; 9: 473-84.
54. Yoshida R, Kaneko Y, Sakai K et al. Positive thermosensitive pulsatile drug release using negative thermosensitive hydrogels. J Control Rel 1994; 32: 97-102.
55. Oberpenning F, Meng J, Yoo JJ, Atala A. De Novo reconstitution of a functional mammalian urinary bladder by tissue engineering. Nat Biotechnol 1999; 17: 149-55.
56. Atala A, Guzman LF, Retik AB. Urethral repair using a novel inert collagen matrix. Tissue Eng 1998; 4: 481.
57. Al Attar N, Razak AB, Scorsin M. Cellular transplantation: new horizons in the surgical management of heart failure. J R Coll Surg Edinb 2002; 47(6): 749-52.
58. Polgie C, Smith AY, Parkes AS. Revival of spermatazoa after vitrification and dehydration at low temperatures. Nature 1949; 164:666.
59. Brockbank KGM. Stabilization of tissue-engineered products for transportation and extended shelf-life. Ann NY Acad Sci 2002; 961: 265-7.
60. Song YC, Khiribadi BS, Lightfoot F, Brockbank KG, Taylor MJ. Vitreous cryopreservation maintains the function of vascular grafts. Nat Biotechnol 2000; 18: 296-9.
61. Bieganski RM, Fowler A, Morgan JR, Toner M. Stabilization of active recombinant retroviruses in an amorphous dry state with trehalose. Biotechnol Pro 1998; 14:615-20.
62. Richardson TP, Peters MC, Ennett AB, Mooney DJ. Polymeric system for dual growth factor delivery. Nat Biotechnol 2001; 19: 1029-34.
63. Sachs EM, Cima MJ, Williams P, Brancazio D, Cornie J. Three dimensional printing. J Eng Ind 1992; 114: 481.
64. Kim SS, Utsunomiya H, Koski JA et al. Survival and function of hepatocytes on a novel three dimensional synthetic biodegradable polymer scaffold with an intrinsic network of channels. Ann Surg 1998; 2287: 8-13.
65. Kaihara S, Borenstein J, Koka R et al. Silicon micromachining to tissue engineer branched vascular channels for liver fabrication. Tissue Eng 2000; 6(2): 105-17.

© British Crown copyright 2004/DSTL - published with the permission of the Controller of Her Majesty’s Stationery Office

Copyright: 3 March 2004