Basic Quality Controls Used in Skin Tissue Engineering

Reconstruction of skin defects is often a challenging effort due to the currently limited reconstructive options. In this sense, tissue engineering has emerged as a possible alternative to replace or repair diseased or damaged tissues from the patient’s own cells. A substantial number of tissue-engineered skin substitutes (TESSs) have been conceived and evaluated in vitro and in vivo showing promising results in the preclinical stage. However, only a few constructs have been used in the clinic. The lack of standardization in evaluation methods employed may in part be responsible for this discrepancy. This review covers the most well-known and up-to-date methods for evaluating the optimization of new TESSs and orientative guidelines for the evaluation of TESSs are proposed.


Introduction
The skin is the largest organ of the body and performs many important physiological functions. This organ provides protection against a wide range of agents, acts as a sensory organ and plays key roles in the hydroelectrolytic balance and termoregulation [1]. The skin is frequently affected by several pathological conditions such as metabolic or genetic disorders, infectious diseases, primary or metastatic cancers, traumatic and burns injuries, etc. Some of these conditions can lead to severe structural defects or even skin loss and therefore the surgical repair in these patients is frequently needed.
Currently, the most common clinical solution for patients with large skin defects is the use of skin autograft (skin grafts obtained from healthy anatomic locations) or skin allografts (healthy skin grafts obtained from donors). These techniques have several drawbacks and limitations [2] and therefore more efficient alternatives are urgently needed. Consequently, the generation of tissue-engineered skin substitutes (TESSs) by tissue engineering (TE) emerged as a promising alternative for these patients. Some of these engineered skin models showed promising ex vivo, in vivo and even clinical results. However, as in other applications, technical improvements and optimization of these skin models are still necessary in order to elaborate more efficient, functional and biomimetic TESSs for these patients worldwide, a process in which quality controls of the products generated, play a crucial role.
The aim of this review is to provide conceptual and technical information concerning skin tissue engineering with special focus on the ex vivo and in vivo quality controls used to determine the potential clinical usefulness of TESSs. First, a general overview of the skin structure, function and regeneration is provided followed by the currently used surgical skin repair techniques. Secondly, the tissue engineering strategies used to generate skin substitutes will be discussed. Finally, a comprehensive review of the ex vivo and in vivo quality controls, which are frequently used to evaluate TESSs, are reported.

Regeneration of the Skin
Multiple diseases and different kinds of traumatic injuries can affect the skin structure and function. One of the most important causes of significant skin loss is traumatic injury [16]. It is estimated that over 6000 people are hospitalized due to severe burns each year in Spain and approximately 200 of these patients unfortunately die [16]. On the other hand, surgical resection of different kinds of neoplasms could affect the structure of the skin and produces a broad spectrum of structural damage. Other reasons for skin loss are trauma and chronic ulcerations secondary to diabetes mellitus, pressure, and venous stasis [17,18].
The repair of this tissue system is carried out through a series of consecutive steps, among which we can mention: (a) performing hemostasis in order to seal the possible rupture. Its main purpose is to prevent infection; (b) tissue regeneration in order to replace those cells that have been lost in the process; (c) wound resolution with the objective of resolution of the different functions of the skin and remodeling of the new matrix [19].
After a structural damage of the skin, keratinocytes undergo a different process, known as keratinocyte activation cycle. Here, injured keratinocytes release IL-1, which triggers local blood vessel formation and immune response to the site of injury. This acute local inflammatory process carried out by keratinocytes, endothelial and immune cells jointly will eventually allow a partial closure of the wound [20]. Additional healing implicates the release a new basement membrane and TGF-β by fibroblasts. TGF-β plays an important role in this procedure shifting keratinocytes to the basal phenotype, suppressing hyperproliferation, activating the standard keratinocyte differentiation program, stimulating the production of extracellular matrix (ECM) components, and inducing normal keratinocyte stratification [21][22][23]. Unfortunately, some types of wounds fail to heal and require surgical interventions or bioactive dressing materials to stimulate and accelerate the healing process [2]. All these cellular and/or molecular processes are directly related to the histological and/or molecular quality controls discussed in the followed sections of this review.

Current Surgical Strategies for Skin Repair
After skin injury or loss, the preferred method for repair is the direct surgical repair. In the case of severe damage with large loss of substance there are others surgical options (Table 1). Flap surgery is a usual reconstructive technique that implicates moving healthy tissue (skin, fat or even muscle) from one location of the body to another adjacent damaged area. This flap generally remains partly connected to the body and its blood supply via a pedicle [24,25]. For this reason, wounds often cure without severe complications and the final cosmetic result is usually satisfactory (Figure 2). Another repair option is a skin graft. It is a surgical technique that allows a definitive coverage of wounds by transplanting tissue from one unaffected area of the body. It is important to consider that the effectiveness of these treatments, like in other disciplines [26,27], will depend on its structural and pro-regenerative properties. An ideal skin graft does not yet exist, but it must meet some important criteria, such as: i) being able to control infection; ii) avoiding fluid and temperature loss; iii) exhibiting adequate histocompatibility; iv) having a good and stable adhesion to the wound site; v) being able to respond adequately to anatomical and biomechanical needs; vi) not being toxic or triggering an immune response; vii) supporting epithelial and stromal regeneration and differentiation; viii) being costeffective; ix) being instantly available for clinical use [28]. Autograft dressing (either meshed or unmeshed) is a graft obtained in the same individual and it is considered the best available technique for wound closure [29], which is able to provide skin integrity (cells and ECM) without rejection risk ( Figure 3). However, this method is not always available and has several well-known disadvantages, such as creation a second surgery injury, restricted availability, significant contraction, shrinkage and scarring [30,31]. A method to overcome the limitations of the use of skin autograft are the use of skin allografts. They can be obtained from cadavers or living donors and are often used as a temporary prevention of wound contamination and fluid loss [32]. Unfortunately, this technique has the disadvantage that not enough tissue is available, due to large demand worldwide, few banks to collect and store these grafts, and many safety and ethical issues. Moreover, even after standardized sterilization and rigorous screening quality controls of these grafts, the transmission of viral diseases has not been completely eliminated [33]. Finally, the use of acellular xenogeneic dermal-derived grafts is being explored, but more research is still needed to elucidate their therapeutic efficacy. Where a lost skin area is unable to be repaired using these conventional surgical strategies, tissue engineered skin substitutes contribute promoting wound healing. Composite synthetic or biological dressings are often used to speed wound healing in chronic or burn wounds but they do not offer permanent treatment [24,25] and subsequent surgical interventions are often needed. These composites are commercially available in different sizes for use in skin repair. In general, these matrices are generated from highly purified bovine or pig ECM molecules, such as collagen or elastin, providing high levels of biocompatibility and degradability ( Figure 4).

Skin Tissue Engineering
During recent years, several technological advances in TE field have been made, allowing new functional and clinically efficient TESSs to repair critical skin defects. The TE is a relatively novel scientific area that combines the principles and methods of engineering with biological structural bases, such as histology, with the aim to generate biologically functional engineered substitutes for the repair or replacement of injured human tissues or organs [34][35][36].
To search an efficient skin substitute to be used in the replacement of current grafts has spurred a number of researchers over the past 30 years. The closest early attempt to develop what might be called a synthetic skin or skin substitute was devised by Curtis (1951). This was fabricated from a gel composed by partially hydrolyzed casein, sodium lactate and sodium lauryl sulfate [37]. Since Green and Rheinwald's (1975) technical contribution, which described how to isolate and subculture a large amount of human keratinocytes [38], several clinical trials tested the efficacy of keratinocyte-based sheets in the treatment of large skin loss [39,40]. Autologous epidermal sheets served as permanent wound coverage with reasonable cosmetic result and without any risk of rejection. However, this method has disadvantages such as needing 2-3 weeks to obtain a sufficient number of cells and high cost [41]. Furthermore, graft-take depends on several factors like wound preparation, intrinsic status, patient underlying diseases, and operator expertise [7]. Initially, some success was reported, but these grafts were never as good as the use of the standard split-thickness skin, probably due to the lack of a dermal cells and ECM components [42]. The allografts have the advantages to produce a temporary prevention of dehydration and contamination, promoting angiogenesis and less pain. Nevertheless their availability is also limited and an immune rejection and inflammation at the wound site may be important disadvantages.
In this regard, a wide range of synthetic-based engineered analogues to promote keratinocyte growth have been studied in animals and humans [12,[43][44][45]. Furthermore, these analogues can be seeded with cells or be completely acellular, and both systems have been widely used in skin TE [46,47]. Some of the earliest versions have been the xenogenic composites made of purified bovine collagen and those generated with shark cartilagederived chondroitin-6-sulfate, both with an outer silicone covering. Significant progress has also been made with the decellularization technique which allows to be generated natural and tissue specific acellular matrices for biomedical use [48], including skin repair. Through chemical, physical or biological procedures it is possible to remove the cellular contents from the tissues efficiently reducing the immunogenicity and thus reduce the risk of graft rejection [48,49]. However, the risk of viral or DNA traces transmission is small but not inconsiderable, even when sourced from accredited skin banks. In this context, peracetic acid, a widely used chemical decellularization agent, has the promising ability to get rid of such viral contents [50]. These decellularized skin allografts have been effectively used alone or in combination with cultured autologous keratinocytes [51].
The treatment of large and deep skin injuries is still a challenge in surgical dermatology, and researchers worldwide have been working on the development of full-thickness TESSs [2]. These models are generated through the combination of dermal and epidermal cells with an artificial matrix or scaffolds, often composed by collagen [52]. These engineered substitutes, mainly composed by a biomaterial containing dermal fibroblasts and keratinocytes showed promising morphofunctional properties like signs of epidermal differentiation, the establishment of dermoepidermal junctions, and the synthesis of some ECM molecules [2,53].
Regarding the biofabrication of bilayered skin constructs, the most common approach is the use of a hydrogel. It serves as a physical platform for dermal fibroblasts growth, on top of which keratinocytes are seeded and allowed to form an epidermal-like layer and mature in vitro [54]. To generate these substitutes a small healthy skin biopsy from the patient is required which is subsequently used to isolate and expand the dermal fibroblast and keratinocytes [54]. Each cell type must be expanded with specific culture medium, for which in the case of human keratinocytes the epidermal growth factor (EGF) and molecular cues-adhesive glycoproteins, functionalized culture flask or irradiated cell-based feeder layer, etc.-are needed to promote efficient growth and expansion [55]. Different skin substitutes have been generated, and in most of the cases dermal fibroblast are usually encapsulated or seeded on top a hydrogel to create an engineered stroma that has the aim to structurally and biologically mimics the dermis [56,57]. Most of the skin substitutes available use collagen type I as 3D scaffold. This is the main ECM molecule of the dermis, and collagen-based hydrogels are widely used in TE. However, it is important to take into account that the collagen used is xenogeneic (mostly bovine), tends to shrink, is rapidly biodegraded in vivo, and has poor biomechanical properties [24]. For these reasons, other biomaterials have been used in order to generate TESS with a higher level of efficiency and structurally more stable, such as collagen-chitosan/fibrin glue [58] (its main purpose is the proliferation and migration of fibroblasts, it has been difficult for the cells to epidermals grow and reach confluence), collagen-GAG [59] sponges (Collagen-GAG and collagen funnel-shaped collagen sponges have an upper surface layer and a bulk porous layer and in this way allow cells to dislodge homogeneously and improve cell viability). The incorporation of GAG increased the mechanical property and cell viability of collagen sponges, gelatin [60], thiol-norbornene cross-linked pectin hydrogels 202 decellularized dermis [61], fibrin [62], scaffolds based on synthetic biomaterial [63] (these scaffolds were able to support the growth of different cell types, including keratinocytes, fibroblasts, and endothelial cells, as well as the production of extracellular matrix, which ultimately leads to the production of new collagen), a self-assembly technique [64] or hybrid natural hydrogels composed of human fibrin and a small amount agarose [65]. Concerning the cell sources used in skin TE, most studies demonstrated that keratinocytes and fibroblast are suitable for the generation of functional and proregenerative TESSs. However, these cells are not always available and keratinocytes are difficult to expand in culture. For these reasons, researchers started to explore alternative cell sources for skin TE. In this sense, mesenchymal stem cells, which can be obtained in higher amount and has the capability to differentiate in different linages, emerged as a promising alternative for stem cell-based therapies in dermatology [66] and skin TE. These cells have been used as an alternative epithelial cell source for generation of bioengineered human skin substitutes with possible practical utility [67][68][69] and seem to improve skin regeneration on its own or in combination with a scaffold [70]. Although this different kinds of stem cell, currently used in a wide range of TE protocols, is a promising alternative in this field more research is required to demonstrate their potential clinical usefulness. More details about the use of stem cells in skin TE can be found in other review articles [71][72][73].
In order to generate a stratified epidermal layer, thus an efficient barrier, the TESSs are first fully immersed into the culture media for approximately 15 days. This step supports the formation of a monolayer of keratinocytes on top. After this period, TESSs are subjected, to another 15 days, to an air-liquid culture technique to induce a correct stratification and maturation of the engineered epidermis [24] (Figure 5). This well-established practice was successfully applied to generate bilayered skin grafts based on woven/non-woven fibers, porous freeze-dried scaffold, nanofibrous matrix, and even 3D printed hydrogels [54]. A schematic representation of the ex vivo and in vivo main features of the human fibrinagarose skin substitute is shown in the Figure 5. This human skin model is being clinically evaluated in Andalusia, Spain [65]. These 3D models represent an important technical advance which solve the problems associated to the use of cell-sheets technique or 2D skin cell cultures. Indeed, the new generation of TESSs has demonstrated that they promote the recapitulation ex vivo of several morphofunctional features of native skin, such as barrier function, biomechanical resilience, keratinocyte stratification, the establishment of efficient cell-cell molecular interactions, and the synthesis of basal membrane and essential ECM molecules (collagens, glycoproteins and proteoglycans). On the other hand, a three-layer TESSs comprising a hypodermis better reproduced the skin histoarchitecture [74]. In addition, recent essays have been performed to develop an ideal skin substitute that accelerates wound healing via fast development of new vessels [29], the inclusion of adipose tissue-derived microvascular fragments [75], induced pluripotent stem cell (iPSC)-derived endothelial cells into the dermal component [76] and the incorporation of antimicrobial properties by using antibiotic-loaded nanoparticles [48]. However, it is still a challenge to generate TESSs that has immunological properties, an adequate degree of pigmentation, a predefined vascular network and the presence of glands and/or hair follicles precursors.
During the last years the three dimensional (3D) printing technology was applied to generate TESSs with success. It uses a computer-aided design to fabricate functional tissues and organs via a layer-by-layer positioning of biomaterials and living cells [77,78]. It is an efficient alternative to emulate the complex histological 3D structure of the native skin [79]. This methodology was used to combine gelatin methacryloyl and alginate hydrogels to generate and subsequently incorporate layers of endothelial cells [80] or even microchannels [81] into a multilayered skin construct. Currently, several groups are introducing/optimizing the 3D printing technology for the generation of more biomimetic and efficient skin equivalent with promising preclinical results. However, more research is still needed to be able to transfer this methodology to the clinical practice [82,83].

Quality Controls in Skin Tissue Engineering
Based on the wide range of TESSs models developed within the last years, it is very important to correctly demonstrate that these substitutes are suitable for their use in skin repair. All these aspects make quality control methods an indispensable part of TE procedures.
Most TESSs are first evaluated ex vivo and, based on these findings, some of these substitutes could prove suitable for further in vivo preclinical evaluation [84,85]. Currently, there is a wide range of methods available to characterize TESSs generated ex vivo and tested in vivo, which include functional analyses, histology (light and electron microscopy), molecular biology and biomechanical testing [86,87]. Furthermore, the performance of all these preclinical studies is a prerequisite to obtain the approval by the specialized governmental agencies for future clinical use of the TESSs [88].
In the following sections a comprehensive review of the ex vivo and in vivo quality control methods used to characterize engineering TESSs is provided.

Ex Vivo Quality Controls
The ex vivo characterization of TESSs [89], such as in other biomedical applications [90], should demonstrate that the TESSs generated are composed by viable and functional cells. Furthermore, the combination of these cells with an adequate biomaterial should allow to be reproduced the main histological structure and some of the key functions of the skin ( Table 2).   Haematoxylin-eosin staining (HE); surface electrical capacitance (SEC); glycosaminoglycan (GAG); fibroblasts (FBs); keratinocytes (KCs); 3-

Assessment of the Cell Viability and Functionality
The assessment of the cell viability, often defined as the quantitative and/or qualitative determination of the number, percentage or fraction of viable and functional cells included in a determined cell culture or engineered substitute. Therefore, it is an important step to confirm that the engineered tissues generated are predominant composed by viable and functional cells [118]. In addition, it is also crucial to confirm the biocompatibility or cytotoxicity of the biomaterials used. From the technical point of view, the methods used can be exactly the same, but they are applied in different experimental contexts [119].
One of the most used methods for assessing cell viability was the dye exclusion method which use the Trypan blue azo dye [120][121][122]. It is based on the principle that viable cells, with an undamaged and functional cell membrane, exclude the dye. However, dead cells or those with irreversible cell-membrane damage (permeable cells) incorporate the dye molecules, and appear stained [120]. This method is mostly carried out on a counting chamber device with a gridded area which allows to be determined the concentration of cells in a determined volume and calculate the cell viability index. These devices are known as haemocytometer or Neubauer chamber slides.
Over time, the cell viability assays have become more complex. Indeed, dye-based colorimetric reactions that depend on the metabolic activity of the cells are gaining favor. These reagents can be applied to determine the metabolic activity of cells in suspension, attached to culture flasks or even those immersed or cultured on top of biomaterials [123]. The first, and probably best known, metabolic dye used for this purpose is the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT). MTT measures mitochondrial metabolism, which is an index of cellular viability rather than proliferation [124]. Similar reagents which allow to spectrophotometrically determine the enzymatic activity of the cells through a colorimetric reaction are the commercially available WST-1 (water-soluble tetrazolium salt) and PrestoBlue ® assays (Thermo Fisher Scientific, Waltham, MA, USA). WST-1 have been often used to determine the cell viability in TE [125], but also the biocompatibility or cytotoxicity of different biomaterials [126].
On the other hand, fluorescein has also been employed in TE to assess the in vitro viability of a variety of cell types and even tissues, including human keratinocytes [100]. This dye was also clinically used to assess skin flap viability [127]. Based on its bipolar side chains, the fluorescein diacetate (FdA) can easily penetrate the cell membrane and is degraded by intracellular esterase's, releasing the fluorescein which remains intracellular in viable cells with intact membranes, allowing the identification of these cells under fluorescence microscopy [120]. This method provides a global, nondestructive measure of the distribution of viable cells within skin constructs [100]. Similarly, live/dead viability staining describes a mixture of two fluorescent dyes that differentially identify live (usually resulting in a green fluorescence) and dead cells (usually with a red fluorescence) [128].
There are few other fluorochromes which can cross the cell membrane of live intact cells, such as Hoechts (HO33342). This fluorescent dye is one of the most popular supravital probe which binds to the DNA. When this reagent is used at low concentration, it will allow the identification of intact or viable cells, which metabolically introduce lower concentrations of the dye, by a bluish fluorescence or damaged/dead cells which exhibit a metachromatic white-yellow emission, due to the high concentration of the dye [129]. The combination of HO with propidium iodide allows to obtain more accurate information of the cell viability [130]. The supravital fluorochrome-based stainings can be used to determine the cell viability and functionality of cells in suspension or cultured within culture flask or biomaterials. However, these dyes can provide reliable quantitative results of cells in suspension when analyzed by flow cytometry [130].
Finally, determination of cell viability by electron-probe X-ray microanalysis (EPXMA) is also used for identifying, localizing, and quantifying the ionic elemental composition related to cell viability, functionality or death (such as potassium, sulfur, sodium, calcium, chlorine, phosphorous and magnesium). This method allows to determine the concentration of these ions in a whole cell or at the intracellular or even organelle levels [131,132]. This is probably the most specific and sensible method available to accurately determine the cell viability [125]. This technique is frequently used within conventional cell cultures, but it can also be applied to evaluate the biocompatibility of biomaterials in TE [133]. However, it is an expensive technique which requires a scanning electron microscope equipped with an EPXMA, which limits it use [122].

Histological Assessment
Skin histology is considered one of the most solid quality control methods to evaluate skin regeneration and/or pathological processes [86]. Histological analyses can be performed on either frozen sections or paraffin-embedded tissues, being the paraffinembedded tissues the most commonly used [134]. A wide range of diverse histological staining methods are available depending on the structure of interest to be studied [134].
The histological analysis of TESSs must be oriented to demonstrate the establishment of a well-structured and functional epidermis. In the case of the stroma, it will be necessary to demonstrate the homogeneous distribution of functional and viable cells. Stromal cells should be able to positively interact with the biomaterial showing an adequate elongated morphology, proliferation rate and/or ECM synthesis.
The starting point for histological analyses is the haematoxylin-eosin (HE) routine staining. This method allows the staining of cellular and tissue elements through the use of a basic (haematoxylin) and an acid (eosin) dye. This simple method allows to evaluate the general overview of native or engineered models histoarchitecture [12,135] confirming the establishment of an engineered epidermis and dermis. However, HE staining is not an accurate and specific method to evaluate other essential cell or tissue elements, such as the ECM molecular composition or the expression profile of tissue specific or functional proteins (differentiation or cell linage markers) [12]. Therefore, histochemical and immunohistochemical methods should be used.
In general, most of the TESSs have demonstrated the establishment of a relatively well-stratified epithelium composed by viable and functional keratinocytes layers and a new-formed basal membrane. Some of these models achieve a closely similar histological pattern than a healthy human epidermis [94,97,136,137]. In order to specifically determine the expression of epithelial, dermal or ECM molecules in TESSs, immunohistochemical procedures are used.
In the case of the stroma, some authors investigated the functional properties of encapsulated fibroblast (or other cell sources) [164]. To evaluate the engineered dermis some classic histochemical methods can be useful tools. In this sense, fibrillar collagens, the most abundant fibers of the dermis (collagen type I and III), can be identified by histochemical methods, such as Masson, Van Gieson, picrosirius or the integrated histochemical approach called Fontana-Masson picrosirius (FMPS) [8]. Interestingly, these methods can also be applied to evaluate the structural organization of collagen-based TESSs or decellularized matrices. In this context, picrosirius-based techniques increase the natural birefringence of collagen fibers allowing us to determine the collagen organization pattern under polarized light microscopy [8]. Concerning the elastic fibers, they can be easily and specifically identified by orcein, aldehyde fuchsine or Verhoeff stainings. The reticular fibers, which can be observed in the skin of some anatomical locations, can be well-identified by using metal reduction (such as Gomori, Lynch reticulin or Gordon and Sweet methods) or PAS (periodic-acid Schiff) techniques. Both methods demonstrate the carbohydrates associated with these fibers. Based on this basic principle, PAS histochemical method is also a useful alternative to demonstrate glycoproteins, especially those which form part of basal membrane. Although for the evaluation of the basal membrane there exists a number of techniques, such as methenamine-silver technique, it is better to use specific immunohistochemical markers [3].
Regarding the dermal proteoglycans, there are several types in the skin. Proteoglycans play key roles during collagen fibrillogenesis, regulating the cell function (storing or presenting growth factors), providing an adequate hydration rate to the stroma, and participating in the wound healing process [19]. Histochemistry can help to determine the synthesis and deposition of proteoglycans in TESSs ex vivo or even to confirm their presence when they were used as scaffold. The methods available for this propose are the alcian blue (pH 1, 2.5 or 4), safranin O, toluidine blue and many other stainings [5,9,65]. These techniques are sensible enough to provide a general, but not specific, overview of the presence of these molecules, that could be part of the biomaterials used or be produced by the cells within the engineered tissues [24]. However, further molecular-based staining, such as immunohistochemistry, are needed to determine the presence and distribution of specific fibrillar or nonfibrillar ECM molecules.
In this context, immunohistochemistry will allow to accurately demonstrate the synthesis and distribution pattern of ECM molecules within the structural framework of the TESSs (Table 4). To demonstrate the synthesis of a basal membrane ex vivo, the immunohistochemical identification of laminin isoforms and collagen type IV are useful and highly specific alternatives. Furthermore, the identification of integrin α6β4 will help to demonstrate if the basal keratinocytes within engineered epidermis were able or not to establish cell-basal membrane interactions [104]. Integrins are the main adhesion proteins that communicate the cellular cytoskeleton with the ECM. These molecules are bidirectional signal transducers which regulate cell proliferation, differentiation, adhesion and migration. Therefore, integrins are fundamental for keratinocytes migration during development, regeneration and physiological functions [104].  In situ transfected cells [76,177] Human leukocyte antigen (HLA) Transplanted cell "label" [92,117,143,169,173] Immune Response CD68 Macrophages [63,106] CD45 Lymphocytes [63] The analysis of the CKs expression patterns in TE can provide solid evidence about the stratification and maturation of ketatinocytes in TESSs. These analyses are essential to confirm that the engineered epidermis have a similar structural pattern and CKs expression than native and functional human epidermis [77]. Other proteins which are frequently used as epithelial differentiation markers are filaggrin, involucrin [67], loricrin or transglutaminase [53]. These proteins, as described in previous sections, are part of the barrier that protect the internal environment from external harmful agents and minimizing the loss of water and other fundamental components to the outside. Additionally, immunohistochemistry is a valuable option to demonstrate the establishment of different kind of cell-cell interactions, which are important for the barrier function of the epidermis, like desmoplakin, plakoglobin or plakophilins (Table 3). Finally, the cell proliferation of the keratinocytes or fibroblast can be easily assessed by using antibodies against Ki-67, PCNA or cyclins. The quantitative analysis of these proteins allow to define the cell proliferation index, a practical indicator of cell viability [12]. Furthermore, apoptosis can be evaluated by immunohistochemistry (caspases) or TUNEL assay.
Concerning the ECM of the dermis within engineered TESSs, it is probably that fibroblasts are not able to produce a high amount of diverse ECM molecules under ex vivo conditions. Histochemistry could provide some results, but the immunohistochemical analysis of most abundant ECM molecules will help to confirm the normal fibroblast function. In this context, the immunohistochemistry for the identification of the main and more abundant ECM molecules will demonstrate the ability of fibroblast to synthesize ECM molecules for example collagens (type I, III and V), glycoproteins (fibronectin) and proteoglycans (decorin, biglycan, versican, etc.). In relation to the identification of elastic and reticular fibers, histochemical methods are more sensitive, easier, faster and cheaper alternatives than immunohistochemistry. However, the immunohistochemistry will allow to be determined the specific expression of the microfibrils, glycoproteins which form part of the elastic fibers, such as fibrillin I and/or II [174].
The fibroblast can be immunohistochemically identified by using antibodies against their intermediate filament vimentin. When fibroblast acquires a reparative phenotype, often observed during wound healing process, expressed contractile cytoskeletal proteins such as a-smooth muscle actin allowing to identify them by immunohistochemistry. In some TESSs a combination of cells are used, especially within the stroma. In this context, the distribution of endothelial or endothelial-like cells can be assessed by the identification of CD-31 or clotting factor VIII.

Transmission Electron Microscopy
In addition to conventional light microscopy, transmission electron microscopy (TEM) may be used to evaluate the ultrastructural features of TESSs [51]. This kind of microscopy will allow us to clearly identify pathognomonic features at the epidermal, stromal, and ECM levels. In the case of the keratinocytes, TEM will allow to confirm their stratification features, such as the progressive establishment and enhance the cytoskeletal CKs organization and the interdigitated cell-cell interactions. Indeed, TEM is the most accurate method to evaluate the features of the cell-cell desmosomal interactions with their typical thick bundles of CKs filaments, dense plaques and outer and inner zones [178]. In relation to the basal membrane, if epithelial and stromal cells formed a basal membrane ex vivo, TEM will help to reveal the ultrastructural features. In this sense, it will be possible to distinguish the two main layers: the basal lamina or lamina basalis and the reticular lamina or fibroreticularis and the presence of the hemidesmosomes. A mature and well-structured epithelial lamina basalis should be composed by the lamina rara, a narrow space between the cells and the second element the lamina densa [178,179].
At the stromal level, TEM will provide clear images of the main fibers of the ECM molecules, such as the collagen fibers, with their characteristic striated pattern, and elastic fibers, with their elastin protein core and surrounding microfibrils [178]. In relation to the fibroblast, TEM will show their main organelles (packed rough endoplasmic reticulum, Golgi apparatus, and a prominent nuclei and nucleoli) and shape (elongated cells with multiple surface folds and exocytosis vesicles) [178]. All these ultrastructural features are critical for the stability of the dermal-epidermal junction and barrier function.
In general, it is recommended to conduct several techniques to demonstrate accurately the histological and ultrastructural features of engineered tissues ex vivo. The use of one method will provide limited information, for example, it was immunohistochemically demonstrated that mesenchymal stem cells cultured within biomaterials can proliferate and produce diverse ECM, like collagens or laminin ex vivo. However, TEM analysis did not demonstrate the existence of mature and well-organized collagen fibers within the designed tissues [180]. Immunohistochemistry is a reliable and useful method, but it is not uncommon for us to resort to TEM to confirm these findings.
Another emerging technique with some appeal but much less frequently used to perform high-resolution images of tissue structure on the micron scale is the optical coherence tomography (OCT) system [110]. OCT is a relatively recent noninvasive and nondestructive optical imaging technique based on measuring backscattered or backreflected light. OCT generates 2D and 3D tomographic images with a micron resolution. The results show a high degree of correlation with the histological findings regarding structure and layer thicknesses. Imaging can be performed in situ, without removing a tissue specimen and in real time [108]. OCT has been increasingly used in dermatology [181][182][183] for the evaluation of wound healing [184] and other varied fields, including the diagnosis of melanoma [168]. Unfortunately, other smaller and highly specific features, such as the epidermal-dermal junction and cellular features, cannot currently be visualized [185]. More research is needed to determine the usefulness of the OCT in skin TE ex vivo and in vivo.

Molecular Biology
ECM components and DNA content can be quantified through a variety of as stated [186]. These techniques make it possible to complete all the semiquantitative information obtained by the histological examination and help to identify similarities and deficiencies between the tissue that has been designed and the native tissues [107]. For example, the participation of fibroblasts in the formation of the dermoepidermal junction has been confirmed by the reverse transcription polymerase chain reaction (RT-PCR). This technique shows dynamic interactions between fibroblasts and keratinocytes during in vitro maturation, as well as the marked changes that occur after in vivo transplantation [70,117]. Furthermore, semiquantitative analysis of specific epithelial or stromal proteins, used as markers for the identification of cells and ECM molecules, can be easily performed by Western blot. Another useful technique is the determination of gene expression profile by microarray technology, which allows to evaluate several genes simultaneously and quantitatively. These findings can orient prospective designs of TESSs for more physiologic characteristics.
The highly specific information provided by the molecular biology is useful to confirm the gene or protein profile of TESSs as compared to native skin or pathological conditions. However, these methods should be used synergistically to support, confirm and complete the histological findings, but not as the unique quality control method in TE, since the histological pattern of the generated substitute must always be demonstrated.

Biomechanical Characterization
An ideal TESS should be identical to normal human skin, not only biologically, but also biomechanically [173]. Nowadays, overall biomechanical properties of any engineered construct can be measured through the application of different forces under specific conditions (Table 5). In this context, conventional techniques to evaluate the biomechanical functionality of artificial skin substitutes include tensile, compression, and shear stress tests [78,116]. Ue, the immediate elastic elongation; Uf, the total elongation; Uv, the viscous (creep) elongation; Ur, the immediate recovery; Ur/Ue, the elastic ratio; Ur/Uf, the relative elastic recovery; and Uv/Ue, the viscous ratio Lafrance et al. [193] hKCs seeded on our anchorage based a human type supplemented with elastin DE, I+III CG + GAGS

Indentation method Deflection (A in mm)
Lafrance et al. [194] FBs + type I bovine collagen  The most common method of loading for characterizing materials is uniaxial loading by tensile test. For the tensile test, the specimens are aligned with their longer length in a parallel direction to the tensile force ( Figure 6). In order to be accurate with the measurements and the results obtained, a series of variables must be taken under consideration. A fundamental variable is the distance between the clamps that must be constant and a constant strain should be applied. Another criteria to be considered is the shape and size of the samples since they can affect the results of the tensile test [198]. The results of the uniaxial tests are plotted on a stress-strain curve, which is represented by the response of the material to applied forces. From this curve, important information on the material's load capacity can be obtained and we can differentiate three regions: the elastic region, in which the material returns to the nondeformed state when forces are applied; the plastic region, in which the material deforms permanently, and the failure region where the tensile strength is reached and the fibers begin to break sequentially until they all break completely. In addition, there are parameters, such as Young's modulus, maximum strength or tensile strength, and elongation at break or strain at break, which are calculated from the stress-strain curve after the tensile test. Young's modulus (E) is the most common parameter since it indicates how rigid the material is and it is calculated as the tangent modulus of the initial linear portion of the stress-strain curve of each sample. The ultimate strength or tensile strength (break σ) has as its ultimate goal the measurement of the maximum stress that a plastic sample can withstand while being stretched before breaking. This can be calculated using the following formula: σ (stress) = F/A, where F: normal force acting perpendicular to area and A: area. Finally, the elongation at break or deformation at break (ε-break) is the ratio between the increase in length and the initial length after failure of the test specimen. Elongation is calculated as the relative increase in length. ε= (∆L/L) x 100, where ∆L: final length and L: initial length.

Functional Evaluation
One critical function of skin is to form an effective barrier to protect the body from penetration of infectious agents and loss of water and necessary nutrients. The skin barrier function can be measured using skin biophysical instrumentation by noninvasive methods (surface hydration [93,96,199] transepidermal water loss [102,200] and invasive methods (water permeation [201,202], niacinamide flux [201] in vitro and in vivo [145]).

•
Surface hydration: human stratum corneum electrical conductance depend on their water content. This is the reason why capacitance and/or conductance methods are commonly used to measure water content of TESSs and development of barrier function in vivo.

•
Transepidermal water loss (TEWL). TEWL is the amount of water vapor evaporating from a fixed surface of the skin per unit time. It is measured using sensors that detect changes in water vapor density. Niacinamide flux: Permeability of niacinamide has been revealed as a sensitive invasive method to measure the barrier function in cultured skin substitutes.
Although restoration skin barrier is essential for the general well-being of the body, this function was not often measured directly in most reviewed articles.

In Vivo Quality Controls
In vivo studies are crucial and must be focused on demonstrating the therapeutic efficacy or failure of the use of TESSs in models of skin repair (Table 3). If positive results were obtained, the TESSs should support a proper epithelial maturation with a collagenrich and vascularized stroma. Furthermore, these quality controls should help to accurately elucidate if the TESSs were biointegrated, biodegraded, encapsulated by the host connective tissue, or rejected by the host immunological system.

Macroscopic Evaluation
Macroscopic aspects of the wound healing process are important because they provide information related to the success or failure of the graft used. Some important aspects include the TESSs take rates, color (pigmentation, and signs of vascularization), elasticity, surface smoothness, if results are comparable to the surrounding skin. Furthermore, these evaluations should confirm the absence of signs of infection, contraction, inflammation or necrosis [97,128,136,145,160]. While it is true that most studies reviewed supply a macroscopic description of the wound healing process, these descriptions remain unsystematic. For this reason, the generalized use of a standard checklist would improve the quality of macroscopic evaluation and the comparison between studies [120].

Histological and Ultrastructural Analyses
Almost all studies reviewed employed routine histological analyses to determine the efficacy of TESSs in skin repair. In general, HE staining is employed for general evaluation of the histological pattern and host tissue response to the grafted engineered tissues. This method will confirm the epidermal growth (implanted cells and/or from the host), the degree of stratification as well as some undesired results such as partial epithelization, epithelial lost and/or inflammatory infiltration. In the case of the stroma, this routinely used method provides useful information concerning the presence and thickness of granulation tissue, the fate of the biomaterial (biodegradation, integration, encapsulation or rejection), extracellular matrix remodeling, degree of vascularization, and if newly formed rete ridges, papillary plexus, and appendages have started to be regenerated.
Alternative methods, such as Masson's trichrome [116,128], picrosirius, PAS and orcein stain [156] were used for more detailed distinction, between different tissue/biomaterial components [65,152,153]. These histochemical methods allow to confirm the progressive synthesis, remodeling of the host and/or newly-formed ECM of the dermis. Picrosirius under polarized microscopy will allow to determine the degree of organization and parallelism of the collagen network. Furthermore, these techniques may help to determine if an abnormal synthesis or fibrotic response to the grafted tissues occur, often associated to synthetic and/or slowly degrading biomaterials [207]. In a regenerative microenvironment, the fibroblast which expressed vimentin, can acquire contractile properties expression a-smooth muscle actin (SMA), which allows to identify them.
In addition with the assessment of the histological pattern, specific set of proteins are often analyzed by immunohistochemistry (Table 3). One or more antibodies have been used to characterize the proliferation and differentiation of the epidermis (basal, intermediate and superficial CKs). Moreover, a set of antibodies can be used to demonstrate if the newlyformed epidermis was or not repopulated by melanocytes (Melan A, S-100), Merkel (CD20) and/or Langerhans (CD1a) cells. The components of the cell-cell and dermoepidermal junction are evaluated through the identification of some specific markers. Cell-cell interactions can be demonstrated by immunohistochemistry against some desmosomal (e.g., desmoplakin, plakoglobin) or hemidesmosomal (α6β4 integrin) proteins. The basal membrane can be identified by histochemistry (PAS, silver stains) or immunohistochemistry (collagen IV, VII or laminin).
Regarding the degree and quality of the blood, lymphatic and nerve supply, these crucial elements can be assessed histologically. Blood vessels can be easily identified and classified by routine or histochemical methods. However, more specific analyses can be done by using endothelial (Von Willebrand Factor, CD31), basal membrane (laminin or collagen type IV) or SMA [14], which also allows to differentiate them from lymphatic vessels, positive for D2-40 protein [5]. Actually, it is also possible to differentiate vascular elements of the superficial and/or deep vascular plexus by using a-SMA and smoothelin [14]. With respect to the nerve supply, it is well-known that peripheral nerves can progressively reinnervate distal target organs, including the skin. Well-structured nerve fascicles can be identified with conventional histological techniques or myelin histochemical methods, as well as encapsulated nerve endings. However, free nerve endings or small regenerating fascicles are more difficult to identify. The Schwann cells can be identified by several markers, being the S-100 protein the most frequently used. In addition, neuronal axons are positive for neurofilaments, GAP-43 and PGP 9.5 proteins [135,139].
On the other hand, TEM has been used for a precise examination of highly specific epidermal and/or dermal features. As mentioned above, through this methodology it is possible to observe the ultrastructural details at the intracellular, intercellular and extracellular levels. TEM can be used to confirm, or as an advanced evaluation, of routine histology or histochemistry, but this will be of great scientific value if used to confirm highly specific immunohistochemical findings. In this context, TEM will allow a comprehensive characterization of the epidermal layers, the cell-cell or cell-ECM interactions. In the case of the basal membrane, the histochemistry or immunohistochemistry will provide information about if this structure was formed or not, but only TEM will provide pathognomonic images of its main elements (lamina basalis, lamina reticularis). Indeed, some authors used this technique for this specific propose [41].
Concerning the dermal ECM, histochemistry and immunohistochemistry will demonstrate the synthesis and organization of the main fibers and non-fibrillary ECM molecules while TEM will confirm the presence and organization of the collagen, reticular or elastic fibers [63,65,67,111]. In this regard, Lamme et al. used TEM to distinguish newlysynthesized collagen from the collagen-based scaffold used, and it was also a useful way to identify the myofibroblasts with their characteristic stress fibers [140]. Unfortunately, TEM only allows to evaluate very small samples losing the complete context of the whole healing process. For this reason, TEM is most suitable to answer specific research questions and it is less appropriate to take general conclusions about the whole wound healing process, where conventional histological techniques still represent a better option [86].

Molecular Biology
Histological study continues to be the most used and most informative diagnostic method. It provides global information and reports on some specific items. However, histological techniques are not able to quantify and evaluate some molecular and metabolic processes. In this context, molecular biology may proffer a valuable, fast and quantitative tool to evaluate the in vivo performance of tissue-engineered constructs. These procedures have increased the ability of analyze and describe basic molecular factors related to wound healing of skin [159].
Histology provides global information about the processes and some specific elements. However, the information provided may be lacking in terms of the quantification and evaluation of some molecular and metabolic processes and therefore molecular biology plays a fundamental role in the analysis and description of these processes. In fact, the histological examination is easy and accessible to perform. Molecular biology solves some of these drawbacks and generates a valuable, impartial and moderately early tool to evaluate the in vivo performance of tissue engineering constructs [159]. A practical tool in molecular analysis has been the Western blotting, a powerful technique used to detect specific protein molecules from among a combination of proteins, evaluate the size of a protein of interest and measure the amount of protein expression. It is based on an electrophoretically-separated sample using antibodies [208]. Western blot helps us to reaffirm the histological findings and makes a semiquantitative analysis of the expression of various markers, such as epithelial or dermal markers.
The proteomic analysis by Western blot will confirm all those results that have been obtained by a conventional histological examination and will contribute with a semiquantitative analysis of the expression of the different markers, among which the epithelial markers stand out.
PCR allows highly specific genes to be seen, which differentiates it from immunohistochemistry and the Western blot technique that recognize proteins. These genes can be related to differentiation, proliferation, signaling pathways, etc. Scherer et al. used quantitative RT-PCR to analyze the RNA expression levels in explanted mouse skin of the proangiogenic vasculature endothelial growth factor, the urokinase plasminogen activator receptor involved in cell migration, the inflammatory cytokine interleukin 1b, and matrix metalloproteinases 3 and 9 involved in extracellular matrix remodelling [157].
Finally, the gene expression microarrays technique allows us to determine the expression of many genes at a quantitative level. Klingenberg et al. used gene expression microarrays to evaluate changes in the gene expression profile of human cells in a cultured skin substitute after grafting onto mouse full-thickness wound [107]. Lammers et al. used this methodology to analyze changes in biological processes that occurred in a collagen-based acellular skin construct after implantation in a rat full-thickness wound model [159].
In conclusion, regardless of the great value that genetic analyzes provide, it is recommended that they be accompanied by histological analyses, generating an overall structural analysis.

Biomechanical Characterization
Mechanical properties of TESSs are largely related to its collagen-fiber architecture and kinematics and they are elementary for a correct functioning. It is remarkable that this aspect of artificial skin is often not assessed [209].
Yannas et al. determined the peeling force of their construct after transplantation, but the method used for this determination was not mentioned [210]. Pandit et al. tested the mechanical strength (ultimate tensile strength, stiffness, and failure strength) using an Instron tester to evaluate 1 × 4 cm skin strips in uniaxial tension and calculated the ultimate tensile strength, modulus of elasticity, and strain-to-failure [144]. Biomechanical analyses provide highly valuable information which could help to understand the histological and even molecular results. However, the high number of large samples needed may explain the limited use of biomechanical analyses in this field [86].

Conclusions
To conclude, diverse evaluation methods are available for the complete and comprehensive characterization of TESSs. Fortunately, there is more than one method available for the assessment of ex vivo and in vivo TESSs parameters. Therefore, it is important to choose those methods that permit a complete evaluation, in the most efficient and informative approach, of the TESSs generated ( Figure 6).
Based on the studies included in this review, we can conclude that there is a large heterogeneity in the characterization of TESSs. Hence, it is necessary to improve the preclinical quality controls in this field defining some minimal criteria to ensure a complete, homogeneous and efficient preclinical characterization of these biomedical products to ensure their future clinical translation.
According to the information available, the most successful TESSs are those in which high levels of biomimicry and functionality are achieved. It is essential to confirm these features. In this review we found that histology represents one of the pillars of TESSs ex vivo and in vivo quality controls and for this reason we think that the tables presented in this review can be an invaluable help to know the quality requirements presented by the different constructs.
Therefore, for characterization of TESSs, we recommend a descriptive histological analysis with routine stainings or trichrome methods, and then selection of some key histochemical and immunohistochemical methods that efficiently demonstrate the level of epidermal and/or dermal biomimicry achieved.
In the case of the in vivo studies, the skin regeneration due to the use of TESSs should be demonstrated, but it is also important to evaluate the stromal remodeling, neovascularization and reinnervation processes, and, not least, the host immunological response. Even better and highly specific features can be demonstrated by the use of transmission electron microscopy, especially at the intracellular and extracellular level.
Despite the versatility offered by the histological analyses, there are always some limitations, and it is advisable to complement these results with molecular biology and functional tests. Molecular biology will provide highly valuable semiquantitative information about certain proteins or genes whereas functional tests will demonstrate the suitability of the TESSs generated for future in vivo preclinical studies. Finally, the TESSs should have adequate biomechanical properties to facilitate their implantation and suturing and avoid their contraction or rupture. In this sense, biomechanical characterization, an essential part in many TE applications, has been poorly studied in this field and thus there is also an important need for biomechanical evaluation of TESSs to approximate a successful clinical function.
In conclusion, it is evident that more standardized criteria for the characterization and validation of TESSs are necessary, especially when the aim of these models is clinical practice, where they are still urgently needed.  Institutional Review Board Statement: Ethical review and approval were waived for this study as it is a review study where it was not necessary to consult the medical records of the patients, whose images have been incorporated and for which written informed consent has been obtained.
Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.

Data Availability Statement: Not Applicable.
Acknowledgments: Thanks to Ariane Ruyffelaert (University of Granada) for her contribution in the critical revision of the manuscript and its grammar.

Conflicts of Interest:
The authors declare no conflict of interest.