Classical triad of tissue engineering: cells, supporting scaffold and growth factors
During the past two decades, major obstacles have been tackled, and tissue engineering is currently being used clinically in some applications, while in others it is just taking its first baby steps.
Materials in tissue engineering can be manufactured with different techniques. The most common ones are molding and 3D printing. Scaffolds can also be porous granules inserted into a hollow container or biodegradable mesh, which provides the shape and size of the construct. The scaffold acts as extracellular matrix, which later is resorbed and replaced by proper extracellular matrix synthetized by cells.
9.2.1 Used Materials
A large variety of materials have been introduced to aid tissue engineering both in soft and hard tissues. They can be used as scaffolds to keep the cells and growth factors in the desired area or only to give size and shape to the construct (a mesh filled with the scaffold material).
In soft tissue engineering , collagen and freeze-dried cadaveric human dermis are currently most used material as scaffolds in oral and maxillofacial surgery . They will resorb in due time leaving the cells to replace the missing tissue. In some applications, it is important to use a laminated scaffold to enable the growth of a 3D structure with different tissue types.
In bone tissue engineering , the most commonly used materials are biodegradable granules of beta-tricalcium phosphate (ß-TCP), hydroxyapatite (HA), and bioactive glass (BAG) [4–8]. The mesh is usually made out of titanium or bioresorbable materials such as polylactic acid (PLA) or its composites. The scaffold materials as well as the bioresorbable mesh will degrade, and the space is filled with newly formed bone. The speed of degradation depends on the used materials.
The mesh is often shaped manually or with indirect printing, where a template of the defect (and required reconstruction) is manufactured and the mesh shaped to surround it. It can also be manufactured with the aid of direct 3D printing.
9.2.2 3D Printing
With 3D printing , it is possible to produce custom-made scaffolds and mesh for tissue engineering. By using of computer-aided design and manufacturing (CAD-CAM), the scaffolds will be very precise and accurate. As facial structures are quite complex, this has brought huge advantages to managing different materials. Most used techniques are inkjet printing, laser-assisted printing, extrusion printing, and stereolithography. All these require top-class imaging and manipulation the data as well as fine algorithms. CAD-CAM in medicine consists of four consecutive phases: (1) CT 3D imaging data; (2) data conversion, CAD; (3) planning of surgery/manufacturing of implant; and (4) actual surgery accordingly . After the scaffold has been manufactured, it is possible to seed the scaffold by cells and add growth factors, if needed.
In tissue engineering , both differentiated primary cells and stem cells can be used as a source.
Primary cells are cells taken from living tissue. They have undergone only a few divisions and, hence, represent the cells in the original tissue. The main obstacle when using these cells is the limited number that can be obtained.
9.3.1 Stem Cells
Stem cells are cells which can make perfect copies of themselves (self-renewal) or differentiate to specialized cell types. Several cell types are gathered under the same umbrella called “stem cells”: (1) human embryonic stem cells (hESCs) ; (2) human-induced pluripotent stem cells (hiPSCs) , which are in fact reprogrammed somatic cells; and (3) adult stem cells , which cover several types of cells of hematopoietic and mesenchymal origin. Mesenchymal stem cells (MSCs) and tissue-specific progenitors reside in the human body in most tissues throughout an individual’s life and generally have a limited expansion and differentiation.
Some of the most promising research in regenerative medicine is focusing on the use and applicability of stem cells. The main stem cells used in tissue engineering are tissue derived, so called adult stem cells , which can be extracted from most adult tissues. They can be transplanted back to the same individual (autologous transplantation) avoiding risks of disease transfer or immunological reactions. They can also be transplanted to another individual (allogenic transplantation) .
Mesenchymal stem cells (MSCs) are of great interest to both clinicians and researchers, due to their potential in tissue engineering. Their ease of isolation, manipulation and potential for differentiation are qualities that have gathered interest. Despite their frequent use in research, detailed standardized criteria are called for as to the identification of these cells from their various locations, as they are not all the same. The International Society for Cellular Therapy has published a set of guidelines attempting to standardize the expansion of these cells, but more standardization is still needed [16–18].
MSCs are currently the focus of clinical and scientific research due to their exceptional abilities for their immunomodulatory properties. However, MSCs cannot be considered truly hypoimmunogenic; rejection of allogeneic MSCs, immunosuppressive potential and immunogenicity are influenced by levels of systemic or local inflammatory cytokines. High immunosuppressive potential simply allows MSCs to suppress inflammation and delay allogeneic rejection through suppression more efficiently than other allogeneic cell types. Many believe that the therapeutic effect of MSCs is due to a “hit-and-run” mechanism mediated by the production of extracellular vesicles (EVs) or exosomes or secretion of trophic and immunomodulatory factors. Nevertheless, a comprehensive understanding of MSC mechanism of action in cell therapy is still under investigation, and many questions remain to be answered .
Adult stem cell-based therapies are already clinically available; additionally, more than 3000 trials associated with stem cells are currently registered in the World Health Organization International Clinical Trials Registry Platform (apps.who.int/trialsearch/). The majority of these are adult stem cell-based therapies, but the registry also includes the first pluripotent-based (hESC/hiPSC) clinical trials, associated with eye diseases such as macular dystrophy or degeneration. Albeit that the technology is in place to produce a wider range of therapies, especially safety issues are not completely understood, subsequently there is a cautious transition from bench to bedside and application of technologies .
The cells can also be printed together with certain biomaterials combined with growth factors, if needed. The first bioprinting technique was inkjet printing, which is fast and cost-effective. The resolution is high but the quality of vertical structure is poor. Cell density is also low. It has been used to engineer blood vessels, bone, cartilage, and neurons .
When using laser-assisted printing, the cost is high, while cell viability is excellent (>95%). The resolution is high and the vertical structure quality is fair. This technique has been used in blood vessel, bone, skin, and fat regeneration. The high cost has prevented this technique becoming very common. The printers are mostly complex and cumbersome compared to other types of printers. All in all, this technique is still immature for clinical work [21, 22].
Stereolithography is also relatively cheap and fast. The resolution is high, and the quality of vertical structure is good with good cell density and cell viability (>85%). It has been used in blood vessel and cartilage engineering [22, 23].
Bioprinting by extrusion, which is modified inkjet printing, can be used when biomaterial is very viscous. It is of moderate cost but slow. Cell viability is also moderate (40–80%) and resolution moderate, but vertical structure is of good quality with high cell density. Currently, most commonly used bioprinters are based on extrusion technology .
9.4 Growth Factors
Bone morphogenetic factors (BMPs) play a central role in bone and cartilage development, as well as in adult homeostasis of bone metabolism, but are also known to play crucial roles in all organ systems. Therefore, it has even been suggested that BMPs should be named body morphogenetic proteins . Currently, the BMP family is comprised of several members from BMP-2 to BMP-18.
Though BMPs were initially discovered to induce bone formation, BMP-3 for instance has been shown to be a negative regulator of bone density. Some BMPs may have redundant roles in bone formation, as conditional deletion of BMP-7 from limb has no noticeable effect . Further, the osteogenic potential of BMPs has reported contradictory results in in vitro studies. In two- and three-dimensional cultures of AT-MSCs, supplementation of BMP-6, BMP-7, and vascular endothelial growth factor (VEGF) and their combinations showed no significant enhancement of the osteogenic differentiation , while others reported a synergistic effect of BMP-6 and VEGF on the osteogenic differentiation of the same cells [27, 28]. In a study using periodontal ligament cells, supplementation with BMP-2 or BMP-6 showed no enhanced osteogenesis .
BMP-2 is best studied in the context of osteogenesis and has been indicated to have potential in bone formation; however, the reports have been contradictory both in vitro and in vivo. Nevertheless, studies suggest that source of serum supplementation of the in vitro cultures may play a role, with human serum supplemented media showing accelerated biochemical responsiveness to BMP-2 compared with fetal bovine serum (FBS) in AT-MSC cultures . Further, the species origin in which the growth factor is produced is also important for its response, with BMP-2 of mammalian origin showing enhanced response in vitro compared with BMP-2 of bacterial origin . These results may give one explanation to the existing contradiction in the reported BMP-2 studies on AT-MSCs in vitro.
Due to their approval for clinical use, BMP-2 and BMP-7 are probably the most studied growth factors for bone tissue engineering. However, the translation of BMP-2 and BMP-7 to clinical use has been hindered by various problems, including high administration dosage, safety issues, short half-life, and high cost relative to efficacy [31, 32]. The safe administration dosage may be one the reason to the complications and safety issues concerning BMP-2 in off-label use in the cervical spine to increase bone growth and bony fusion. The FDA published a safety communication on recombinant human BMP-2 use in 2015, especially for pediatric patients, with recommendation to cautiousness in the use of BMPs until further safety evidence is available.
9.5 Enhancing Osteogenesis Using Physical Stimulation
The bone is a tissue capable of modifying its structure and mass according to the mechanical loading it is exposed to. This process occurs through the intrinsic balance in the mechanosensitive activation of the osteoblasts and the osteoclasts .
Several studies have provided robust evidence that reduced mechanical loading results in decreased bone mass and that increased mechanical loading can be utilized to enhance bone formation in vivo as well as in clinical cases. Further, animal and human studies have led to the examination of mechanical loading in cell models. Studies have shown that bone cells and their precursors and also MSCs exhibit mechanosensitivity in vitro in response to various mechanical stimuli; however, there is no consensus on the most effective combination of vibration loading parameters for osteoinduction .
Further, osteogenesis can be enhanced in vitro using another potential strategy, by application of conductive polymers as a functional surface coating on biomaterial scaffolds. One of the conductive polymers (CPs), polypyrrole (PPy), has shown great promise in tissue engineering, not only because it can mediate electrical currents, but also because it may enhance the bioactivity of biomaterials in bone applications. Reports show that AT-MSCs cultured on polymer scaffolds coated with PPy induced with electric stimulation supported the viability and proliferation of the AT-MSCs as well as supported osteogenesis [35, 36].
9.6 Experimental Studies
In preclinical studies, regeneration of many tissues in the oral and maxillofacial has been studied. Of these, teeth, salivary glands, and nerves have not yet been explored in clinical applications.
Considering the teeth , many different types of cells are needed, including dental pulp regeneration and dentino- and amelogenesis [37–39]. In the future, it might be possible to regenerate the root of a tooth, but at the moment it seems impossible to guide the shape and color of the crown of a tooth. Hence, dental regeneration at least at the moments, cannot replace dental implants.
9.6.2 Periodontal Ligament
Infection leading to loss of periodontal ligament is very common in adults, and the treatment methods need improving . In long-term cultures, surprisingly, estrogen seems to have properties to retain the stemness of periodontal ligament stem cells ]. New materials such as zein (protein derived from corn) can be electrospun with gelatin and seems to support the growth of periodontal ligaments cells. This, however, is not yet in clinical use .
9.6.3 Salivary Glands
Salivary gland regeneration has been studied quite extensively; however, no clinical studies have been reported yet. Salivary hypofunction can be caused by systemic diseases such as Sjögren’s syndrome, but it may also be due to radiotherapy in the head and neck region. Secretion of saliva is extremely important for the well-being of oral mucosa and teeth; however, no adequate saliva substitutes to replace the hypofunction of the glands have been developed. For tissue-engineered salivary gland, it is necessary also to be able to connect the secretion of saliva to the complex ductal system to deliver the saliva to the oral cavity. However, function of salivary glands including secretion of bioactive molecules in case of radiotherapy might be feasible when using AT-MSCs administrated by systemic routes, i.e., blood stream, by paracrine mechanisms which will provide growth factors helping neovascularization and promoting epithelial proliferation as well as angiogenesis . Dental follicle-derived stem cells have been shown to be able to differentiate to salivary gland and duct cells, which might be a promising future treatment modality .
In cartilage tissue engineering , committed chondrocytes, ESCs, and MSCs have been used. Based on the results and availability of cells, MSCs seem to be a viable choice for this application. The regenerated cartilage in the joint will have to bear large contact area strains and stresses. It must also allow growth of functional tissue by providing appropriate cell-scaffold interactions .
To enable long-term survival of cells inside the scaffold, the scaffold must be either porous or woven. These properties will challenge the appropriate strength of the scaffold. In oral and maxillofacial surgery, the need for cartilage is usually in the temporomandibular joint (TMJ).
Mäenpää et al.  studied the regeneration of TMJ discs in rabbits. The bilayer scaffold disc comprised of a non-woven mat of resorbable PLA and a PLA membrane plate. AT-MSCs were seeded in the discs and cultured in parallel in control and chondrogenic medium for 6 weeks. Relative expression of the genes, aggrecan, type I collagen, and type II collagen, normally present in the TMJ disc extracellular matrix, increased in the discs in the chondrogenic medium. They concluded that the PLA discs seeded with AT-MSCs have potential in the development of a tissue-engineered TMJ disc. The same group later used these discs in ten rabbit TMJs. The original TMJ disc was bilaterally removed, and the AT-MSC-seeded PLA disc was used to replace the removed original disc on one side. On the other side, the cell-seeded PLA disc was pretreated in chondrogenic differentiation media. Unfortunately, the cone beam computed tomography and histology showed that most of the discs had dislocated and caused sclerotic changes and condylar hypertrophy in the joints. The pretreated discs seemed to function slightly better than the non-pretreated discs. No signs of foreign body reaction, infection, or inflammation could be seen. The authors conclude that better disc design and fixation technique might lead to better results .
To be able to regenerate mandibular condyle, it must be realized that both the bone and the cartilage must be produced and bound together. Chondrocytes and osteoblasts can be harvested or differentiated from abovementioned many sources: the properties of the scaffold needed are different for the bone and for the cartilage. The growth factors used need to differ as well. However, if this could be safely and predictably performed, this approach would give great relief to patients suffering from major TMJ disorders and diseases .
Nasal cartilage has also been a target for tissue engineering. Chang et al.  used autologous chondrocytes injected in fibrin glue to rabbits’ dorsal nasal bones. The histological result was identical to that of normal auricular cartilage. The concentration of fibrinogen and thrombin as well as chondrocytes plays a crucial role in the formation of the cartilage. If cartilage cells are not available, bone marrow- and umbilical cord-derived stem cells have been studied. The umbilical cord-derived cells seemed to produce more type I collagen and aggrecan compared to bone marrow-derived cells, a finding which warrants further studies also in a sandwich-type construct for osteochondral reconstruction [49, 50].
9.7 Clinical Work
There are two main objectives in maxillofacial reconstruction: surgery should provide form and function of oromaxillofacial area. As facial skeleton has a very complex structure , reconstruction should restore volume, shape, bone continuity, and symmetry of bone skeleton. On the other hand, soft and hard tissues in this area enable several functions like articulation, mimics, mastication, swallowing and breathing. When the reconstruction is carried out, esthetic and reconstructive aims need to be met.
Clinically, the applications have been mainly in bone regeneration as well as in epithelial defect repair. Currently, the aim is also to avoid all animal-derived materials and replace them with synthetic or human-derived materials, such as recombinant human BMP (rhBMP) and human serum.
9.7.1 Soft Tissue Regeneration
Currently, surgeons often use collagen sponges and freeze-dried cadaveric human dermis to replace missing soft tissue . However, tissue-engineered approaches have also been introduced, and they will be elucidated below.
18.104.22.168 Oral Mucosa
Oral mucosal grafts have also been used in clinical work. EVPOME®, a tissue-engineered ex vivo-produced full-thickness mucosal graft, has shown the ability to create keratinized mucosal surface epithelium when grafted on an intact periosteal bed. EVPOME® is manufactured in the laboratory from the patient’s own keratinocytes (from biopsies of the palate) seeded and cultured on an acellular dermal matrix. It has proven to be safe and has potential to augment keratinized mucosa around the teeth .
On top of esthetics, lips provide several other functions. These include eating (closure of the oral cavity), breathing (by nose), articulation, and sensation (hot, cold, etc.). They also play an important role in facial expressions.
Regeneration of lips is very difficult especially when there is a large defect which cannot be repaired with local flaps. To regenerate an avulsed lip, both the surface structures (underlying muscles and surface structures such as skin and mucosa) have to be considered .
For tissue-engineered construct to be successful, it needs adequate blood supply to enable survival of cells. A sufficient fixation of different tissues plays also a crucial role in the healing and regeneration process.
Already approximately one million patients, mainly burn patients and patients with diabetic ulcers, have received tissue-engineered human skin. Dermagraft® (PGA + neonatal fibroblasts) was one of the first commercially available products for skin defects . Currently, allogenic skin grafts are available as off-the-shelf products. However, they are somewhat immunogenic and may transfer diseases, although the risk is minimal after extensive testing required for these products. Skin grafts can be manufactured to replace only epidermis or dermis of both. Unfortunately, hair follicles or sweat glands are not yet included in these grafts.
9.7.2 Hard Tissue Regeneration
Our own research group has used tissue engineering to produce cartilage to the nasal septum. The two operations, in which a resorbable Chronos® sheet was seeded with patients’ own ASCs, were successful. However, one of the patients continued her nose picking with artificial nails, and after the initial healing period, the graft was lost .
Bone transplants are the second most used tissues in clinical work after blood transfusions . However, if autologous bone is used, usually another surgical site is required which causes more morbidity to the patient as well as extends the length of the operation. Bone banks provide solution for this is some cases as allogeneic bone can quite safely be used even though there is a small risk of immunologic reactions and disease transfer. Bone grafts usually resorb partly; hence, in oral and maxillofacial area, it might in some cases be difficult to predict how much bone needs to be transplanted.
Sinus lift is one of the most common procedures to enable placement of dental implants in the edentulous maxilla. Traditionally it is carried out by using autologous bone harvested in the craniomaxillofacial skeleton or iliac crest. However, it was one of the first applications where bone regeneration was attempted by tissue engineering. The used carries for cells and/or growth factors are resorbable fleeces, HA, bovine bone, and, naturally, autologous and allogenic bone.
Schimming and Schmelzeisen  used periosteal cells on a resorbable (polyglactin 910 combined with polydioxanone) fleece in 27 patients for augmentation of edentulous posterior maxilla. They used good manufacturing practice (GMP)-class expanded periosteal cells from mandibular angle and the fleece was soaked with cell suspension. Bovine thrombin in FBS was used to seal the cells in the fleece. Cells were cultured for nearly 2 months after which they were transplanted in the sinus floor. One patient had to be dropped out due to an infection. In 18 patients the result was excellent; however, an unsuccessful result was seen in eight patients (30%) needing further supplementary autologous bone transplantation.
Meijer et al.  augmented sinus floors or walls prior to dental implant insertion in six patients. BM-MSCs were harvested from the iliac crest and cultured for a week on porous HA in an osteogenic culture medium, containing also xenogeneic materials such as FBS. The cells were then transplanted and the augmentation effect studied 4 months after augmentation. Of the 11 biopsies taken, bone formation was observed only in three patients (50%). It can be speculated that inadequate vascular supply might have been the reason for failures.
Other Small Local Defects
In oral and maxillofacial surgery , large bone defects, caused by cysts, are often filled with autologous bone, bovine bone, or synthetic materials such as hydroxyapatite (HA), ß-TCP, or BAG. In a study published by Stoor et al.  21 bony cavities in 20 patients were filled with BAG S53P4, some even in the presence of infection (65%). The authors state that the use of this material provides and infection-free and reliable bone regeneration. When cells have been used, autogenous osteoblasts seeded in biomaterials have shown to be an excellent choice to fill these defects compared to iliac crest bone grafts . Unfortunately, the need for a GMP-class facility to produce the tissue-engineered filling materials is very labor-intensive and not very cost-effective hindering their use to become more widespread. According to current legislation, iliac crest bone graft can be obtained simultaneously during the same operation when the cyst is removed, hence lowering the costs markedly.
Continuity defects are caused mainly due to tumor ablation of trauma. This can include only the bone, sometimes teeth and some soft tissue. If the soft tissue coverage and blood supply to it is adequate, it is possible to tissue-engineer the transplant directly in the defect site. However, if there is a major loss of soft tissue, the construct needs to be transplanted first to an ectopic site and after maturation transplanted again to the defect site either as a microvascular flap or a pedicled flap.
22.214.171.124 On-Site Regeneration
Zétola and his coworkers  published a case report in 2010 of a repair of a mandibular defect repair after resection of an ameloblastoma using recombinant human morphogenetic protein-2 associated with collagen sponge, autogenous bone chips, and synthetic HA and β-TCP blocks. No cells were transplanted. Titanium reconstruction plate and titanium scaffold filled the abovementioned combination was implanted into defect area. Collagen with rhBMP-2 was superposed above open titanium mesh to allow muscle cells and periosteum to migrate to defect area. After 7 months, the patient had a stable occlusion. Control CT showed good bone formation directed to the center of the defect. The authors concluded that the reported reconstruction technique gave a satisfactory result with less invasive surgery and with minimum morbidity. However, studies with larger number of patients are required to indicate the treatment modality as a routine in cases of bone continuity defects.
In 2015, Park et al.  reported a case study where a large continuity defect after resection of ameloblastoma in the angle of the mandible was reconstructed with iliac bone and autologous BM-MSCs. The iliac bone served as a scaffold, fixed with titanium plates and screws, with cancellous bone removed. The gap was then filled with cultured BM-MSCs and fibrin glue covered with collagen membranes. Later three dental implants were placed in the graft resulting in uneventful healing.
Stoor et al.  used direct CAD–CAM technique and tissue engineering to repair mandibular defects in 14 patients immediately at the time of ablation surgery. Most of the patients had squamous cell carcinoma or ameloblastoma. The surgery was simulated and patient-specific implant (PSI) designed on virtual model. The PSI was a combination of scaffold and reconstruction plate with screw holes. The scaffold was filled with β-TCP and autologous bone. In four patients with ameloblastoma or drug-induced osteonecrosis cases, BMP-2 soaked in a sponge was placed to cover the cage to improve the bone formation. Finally, PSI was covered with collagen membrane or sponge (ten patients) and either radial for arm or anterolateral thigh (ALT) microvascular flap (12 patients). The follow-up was between 9 and 24 months. The overall recovery of the patients was favorable considering how demanding the patients were. Nine patients had an uneventful recovery. The main reasons for failure were infection and dehiscence of the mucosa or the microvascular flap. In these cases, extreme caution should be exercised to avoid soft tissue injury or dehiscence during the surgery and follow-up.
It is noteworthy that all these reports can be estimated having been successful due to sufficient coverage of the regenerate with vascular soft tissue enabling oxygen and nutrient supply to the healing area.
126.96.36.199 Ectopic Prefabrication
One of the first clinical papers was published by Warnke et al. . They reconstructed a mandibular continuity defect using vascularized custom-made bone flap with indirect technique in which patient’s CT data was uploaded to CAD software, and the defect reconstructed in the mandible was virtually simulated. A virtual implant to repair the defect was designed and converted into solid 3D Teflon replica, which was used as a model when manually shaping titanium mesh around it. The shaped mesh was filled with bovine bone mineral blocks combined with growth factor rhBMP-7, bovine collagen type-1, and autologous iliac crest bone marrow. The filled mesh was implanted into patient’s back muscle (latissimus dorsi). A microvascular flap was raised 7 weeks later, and 4 weeks after the implantation, the patient was able to use her mandible and was satisfied with the aesthetic outcome. The authors concluded that ectopic bone formation is possible and causes less burden to the patient compared to conventional bone grafts.