Tissue Engineering Part A: Research Advances

This report draws upon data from a variety of sources to estimate the size, scope, and growth rate of the contemporary tissue engineering enterprise. At the beginning of 2001, tissue engineering research and development was being pursued by 3,300 scientists and support staff in more than 70 startup companies or business units with a combined annual expenditure of over $600 million. Spending by tissue engineering firms has been growing at a compound annual rate of 16%, and the aggregate investment since 1990 now exceeds $3.5 billion. At the beginning of 2001, the net capital value of the 16 publicly traded tissue engineering startups had reached $2.6 billion. Firms focusing on structural applications (skin, cartilage, bone, cardiac prosthesis, and the like) comprise the fastest growing segment. In contrast, efforts in biohybrid organs and other metabolic applications have contracted over the past few years. The number of companies involved in stem cells and regenerative medicine is rapidly increasing, and this area represents the most likely nidus of future growth for tissue engineering. A notable recent trend has been the emergence of a strong commercial activity in tissue engineering outside the United States, with at least 16 European or Australian companies (22% of total) now active.

A considerable oxygen gradient exists in vivo, which exerts regulatory effects on tissue development and function. The objective of this study was to evaluate the feasibility of controlling cell proliferation and differentiation by regulating oxygen tension in a tissue-engineered bioreactor model. The effects of oxygen tension on proliferation and differentiation of first-trimester human trophoblast cells (known as ED₂₇ cells) were studied in a fiber-bed perfusion bioreactor system in which cells were grown in polyethylene terephthalate (PET) nonwoven fibrous matrix. By varying the oxygen tension between 2% and 20%, differential responses of trophoblasts in their proliferation and differentiation activities were observed. There was no significant difference in the rates of glucose consumption and lactate production, and lactate dehydrogenase (LDH) level in the culture media for both 2% and 20% oxygen tension cultures, indicating that cell metabolic activities were not limited by low oxygen tension. However, 2% oxygen stimulated cell proliferation but impeded the secretion of a functional hormone, 17β-estradiol. In contrast, 20% oxygen tension reduced cell proliferation, but yielded higher hormone secretion. A step change in oxygen tension from 2% to 20% caused cells in the bioreactor to increase 17β-estradiol secretion and shifted cell cycle from proliferation to differentiation, which were verified with the expression levels of cyclin B1 and p27^kip1. However, no significant response to a change from 6% to 20% oxygen tension was observed. It is concluded that changes in oxygen tension can be an effective strategy to control cell cycle and long-term tissue development. This work also demonstrated the important role of oxygen tension in regulating placental trophoblast tissue development and the feasibility of using the bioreactor under well-controlled physiological environment for tissue engineering applications.

One of the basic requirements for the production of tissue-engineered constructs is an effective means of storing both the constructs and the cells that will be used to make them. This paper reports on the cryopreservation of porcine aortic smooth muscle and endothelial cells intended for the production of model vascular constructs. We first determined the cell volume, nonosmotic volume, and the permeability parameters for water and the cryoprotectant dimethyl sulfoxide (Me₂SO) in these cells at 2–4° and 22°C. The following results were obtained:

Table unavailable in HTML format.

Using a cell culture assay, both cell types were shown to tolerate threefold changes in cell volume, in either direction, without significant injury. Although these data suggested that single-step methods for the introduction and removal of 10% w/w Me₂SO should be effective, an additional mannitol dilution step was adopted in order to reduce the time required for removal of the Me₂SO. Following cooling at 0.3, 1, or 10°C/min and storage at less than –160°C, the survival of porcine aortic smooth muscle cell suspensions, measured by a cell culture assay, was inversely related to cooling rate; at 0.3°C/min, recovery was >80%. The survival rate for aortic endothelial cells was directly related to cooling rate over the range tested and was >80% at 10°C/min.

Nonneural derived nerve conduits fail to support regeneration over larger gaps due to lacking viable Schwann cells. Thus, tissue engineering of nerves is focusing on implantation of viable Schwann cells into suitable scaffolds. We established grafts made from acellular muscles and veins, respectively, seeded with cultured Schwann cells. As timing of revascularization is crucial to determine Schwann cell survival and depending axonal regeneration we studied establishment of vascular architecture in a rat sciatic nerve model (2-cm gap) after 3, 5, 7, and 10 days postoperatively, using albumin bound Evans blue. Additionally, macrophage recruitment was immunohistochemically assessed. Engineered grafts showed a delayed revascularization, starting between day 5 and 7 in comparison to normal autografts, that revascularized by day 3. Macrophage recruitment in autologous nerve grafts was evident by day 3. The engineered groups revealed no macrophage invasion until day 7. As Schwann cells survive up to 7 days in autologous grafts without blood supply, depending purely on diffusion, establishment of vascular structure between day 5 and 7 is rapid enough to support Schwann cell survival in engineered grafts. As these grafts are lacking Wallerian degeneration delayed macrophage invasion may not impair degeneration-dependent regeneration, but presence of macrophage derived or induced growth factors may be decreased.

The use of bioresorbable conduits supplemented with Schwann cells (SCs) is a promising tissue engineering technique to replace nerve grafting. Alginate hydrogel (AH), as a SC tissue engineering matrix, has many advantages over previously used matrices but has not been evaluated for this purpose. In this study, the viability and proliferation of SCs together with SC function in AH was evaluated in vitro. AlamarBlue™ cell assay was used to monitor the viability of SCs in AH and compared to SC viability in collagen gel, fibrin glue, hyaluronic acid, Matrigel^®, and standard culture plate over 5 days in culture. The results showed that the viability and growth of SCs in different matrices over the culture period did not significantly differ to culture plate culture. SC function when suspended in AH was monitored using chick embryo dorsal root ganglia (CDRG) growth assay. Growth of CDRG in AH with or without SCs was compared to CDRG growth without AH matrix. After 3 days in culture, the mean length of neurite sprouting was measured. The results showed that there was neurite growth in AH but was reduced to 43% of control. The neurite growth in AH was, however, enhanced by 170% when SCs were suspended in the gel. In conclusion, AH supported SC viability and function in vitro and may be useful in peripheral nerve tissue engineering in reconstructive procedures.

Cultured epithelium has been used successfully in the treatment of extensive burns. Regenerated epidermis, however, lacks such as hair follicles and sweat glands that are common in mammalian skin. We attempted to determine whether cultured epithelium could be induced to form hair follicles by dermal papillae, which are most important for the morphogenesis and growth of hair follicles. We cultivated adult rat sole keratinocytes, obtained the cultured epithelium, and prepared recombinants consisting of cultured epithelium and fresh dermal papillae with or without the sole dermis. These recombinants were then transplanted underneath the dermis of the dorsal skin of syngeneic rats or athymic mice. Histologic examination revealed that the transplanted cultured epithelium formed the follicular structures with sebaceous gland-like structure following induction of the dermal papillae, especially when supported by the dermis. We concluded that transplanted cultured epithelium of adult rat sole keratinocytes can respond to growth signals from adult dermal papillae.

Sonic hedgehog (Shh) regulates the principal possesses in many developmental stages, including the epithelial–mesenchymal interaction. The extraordinary acceleration of signaling by Shh is responsible for the development of human basal cell carcinomas and trichoepitheliomas; they might originate from the very immature keratinocytes, including the stem cells. We tried to utilize the mitogenic effect of Shh to accelerate the formation of cultured epithelium, which is already used in the medical field practically. To this end, we transfected shh cDNA into a Swiss-3T3 cell line, widely used as a feeder for keratinocytes, and established a Shh expressing cell line. The lethally irradiated Shh expressing feeder cells remarkably accelerated the growth of keratinocyte colonies obtained from the human neonatal foreskin, and the formation of well-stratified cultured epithelium, which is rich in immature small keratinocytes, expressing cytokeratin 14. This acceleration was suppressed by the addition of cyclopamine, a specific inhibitor of Shh signaling. These data indicate that the Shh is a promising mitogen to improve the technology for cultured epithelium formation.

The aim of this study was to determine the influence of two key scaffold design parameters, void fraction (VF) and pore size, on the attachment, growth, and extracellular matrix deposition by several cell types. Disc-shaped, porous, poly(-lactic acid) (L-PLA) scaffolds were manufactured by the TheriForm™ solid free-form fabrication process to generate scaffolds with two VF (75% and 90%) and four pore size distributions (< 38, 38–63, 63–106, and 106–150 μm). Microcomputed tomography analysis revealed that the average pore size was generally larger than the NaCl used, while VF was at or near the designated percentage. The response of three cell types—canine dermal fibroblasts (DmFb), vascular smooth muscle cells (VSMC), or microvascular epithelial cells (MVEC)—to variations in architecture during a 4-week culture period were assessed using histology, metabolic activity, and extracellular matrix deposition as comparative metrics. DmFb, VSMC, and MVEC showed uniform seeding on scaffolds with 90% VF for each pore size, in contrast to the corresponding 75% VF scaffolds. DmFb showed the least selectivity for pore sizes. VSMC displayed equivalent cell proliferation and matrix deposition for the three largest pore sizes. MVEC formed disconnected webs of tissue with sparse extracellular matrix at 90% VF and >38 to 150 μm; however, when cultured on scaffolds with pores formed with salt particles of <38 μm, MVEC formed a multilayered lining on the scaffolds surface. Culture data from scaffolds with a 75% VF suggests that the structural features were unsuitable for tissue formation. Hence, there were limits of acceptable scaffold architecture (VF, pore size) that modulated in vitro cellular responses.

A myooid is a three-dimensional skeletal muscle construct cultured from mammalian myoblasts and fibroblasts. The purpose was to compare over several weeks in culture the morphology, excitability, and contractility of myooids developed from neonatal and adult rat cells. The hypotheses tested were as follows: (1) baseline forces of myooids correlate with the cross-sectional area (CSA) of the myooids composed of fibroblasts, and (2) peak isometric tetanic forces normalized by total CSA (specific P_o) of neonatal and adult rat myooids are not different. Electrical field stimulation was used to measure the excitability and peak tetanic forces. The proportion of the CSA composed of fibroblasts was greater for neonatal (40%) than adult (17%) myooids. For all myooids the baseline passive force normalized by fibroblast CSA (mean = 5.5 kPa) correlated with the fibroblast CSA (r² = 0.74). A two-element cylindrical model was analyzed to determine the contributions of fibroblasts and myotubes to the baseline force. At each measurement period, the specific P_o of the adult myooids was greater than that of the neonatal myooids. The specific P_o of the adult myooids was ~1% of the control value for adult muscles and did not change with time in culture, while that of neonatal myooids increased.

To date, introduction of gene-modified cells in vivo is still a critical limitation for cell-based gene therapy. In this study, based on tissue engineering techniques, we developed a three-dimensional (3-D) transfection system to be cell-based gene delivery vehicle. Human trophoblast-like ED₂₇ and fibroblastic NIH3T3 cells were used as model cell lines. Cells were seeded onto PET fibrous matrices and plated on polyethylene terephathalate (PET) films as 2-D transfection control. The cell-matrices and cell-films were transfected with pCMV-βgal and pEGFP (green fluorescent protein) reporter gene vectors using LipofectAmine^® reagent. Gene expression on 3-D versus 2-D growth surface were investigated. The effects of seeding method, seeding density, porosity of the PET matrix, and culturing time of the cell-matrix complex on cDNA transfection and expression in the 3-D cell-matrix complex were also investigated. The β-gal assay and GFP detection showed that 3-D transfection promoted a higher gene expression level and longer expression time as compared to 2-D transfection. There existed an optimal initial cell seeding density for gene transfection of 3-D cell-matrix complex. Cells seeded on PET matrices with a lower porosity (∼87%) had higher gene expression activities than cells in the matrices with a higher porosity (∼90%). Also, Higher gene expression levels of β-gal were obtained for the more uniformly seeded matrices that were seeded with a depth-filtration method. The results from this study demonstrate the potential utility of cells seeded onto 3-D fibrous matrices as cell-based gene delivery vehicle for in vitro study of gene expression or in vivo gene therapy.

Two-photon excitation laser scanning microscopy (TPM) was used to image human, porcine, and rabbit nasal septal cartilage. TPM provides optical sections of thick tissue specimens in situ without the use of exogenous dyes or need for tissue fixation. The cartilage tissue was imaged using near-infrared light generated by a mode-locked titanium/sapphire laser that was raster-scanned and coupled to an inverted microscope. Absorption of two photons by endogenous molecules and subsequent fluorescence was filtered to specific spectral bandwidths and detected with photomultiplier tubes. Two-photon stimulated fluorescence was detected with photomultiplier tubes optimized to specific spectral bandwidths. Signal intensity corresponds to the concentration of fluorophores, principally NADH, NADPH, and flavoproteins hence providing a means of redox imaging the cellular metabolic state. Specimens were scanned from the surface to a depth of about 150 μm. Image size was 50 × 50 μm with a diffraction limited pixel size of 0.4 μm. Cell membranes, nuclei, and matrix structures were identified in human, pig, and rabbit tissues. TPM provides a means to study three dimensional chondrocyte structure and matrix organization in situ at substantial depths, and permits longitudinal examination of cultured tissue explants without the need for exogenous dyes, tissue preparation, or fixation.