Regeneration is often thought to belong to animal species phylogenetically distant from humans. However, human bodies are constantly rebuilding themselves, which means that we all have the chance to improve our health, if we can harness the regenerative capacity of our bodies. After the reconstitution of the hematopoietic system by bone marrow infusion (Thomas et al., 1957), the importance of stem cell content was elucidated by lengthy studies on clinical outcomes. Later on, brilliant scientists isolated stem cells capable to regenerate the human epidermis in vitro (Rheinwatd and Green, 1975) for subsequent transplantation to the human body, managing to heal very severe burns. This step marked the beginning of a new era, sketching out the idea that, by harnessing their power into the clinic, stem cells could be used to tackle a wide range of diseases. Indeed, many human tissues and organs possess the ability to self-renew due to specific stem cells, which generate progenitors, producing terminally differentiated cells. The cultures of these specific stem cells have been shown to fully restore some severely damaged tissues. Unfortunately, the clinical translation was delayed by many hurdles, as several human tissues do not regenerate, stem cells of some tissues have not been found yet, or we are still unable to stimulate them properly. A goal in regenerative medicine is to find ways to engineer tissues replacement and/or kick-start tissue regeneration in the body. The epithelia were the first cultured tissues because most of them are easily accessible and highly renewing. However, epithelia are highly specialized and some of them are characterized by various differentiation lineages, multiple signaling, and specific microenvironments on several matrices (Figure 1). Thus, their regeneration requires different and multidisciplinary medical expertise as they cover the whole body both internally and externally. To evaluate the applications of the first type of cultured stem cell in tissue regeneration, namely the epithelia, an analysis of the past clinical successes and failures can address some pending issues and support new applications.

The skin is the largest organ in our body and ensures not only a barrier against the environment and microorganisms but has also many other functions such as immune surveillance and tactile sensitivity. Its integrity is fundamental for survival; therefore, it renews constantly and comes with stem cells, similar to other lining epithelia. Its stem cells reside in the hair follicles, from where they can generate hair follicles, sebaceous glands, or the epidermis; however, some of them reside in the interfollicular epidermis.

The expression of several markers characterizes the interfollicular cells: cytokeratin (CK) 19 (Michel et al., 1996), β1-integrin (Jensen et al., 1999), α6-integrin (Li et al., 1998), desmoglein-3 (DSG3) (Wan et al., 2003) and high levels of melanoma chondroitin sulfate proteoglycan (MCSP) (Legg et al., 2003). Instead, C200 and CK15 identify hair follicle bulge cells (Ohyama et al., 2006; Inoue et al., 2009). However, these markers do not appear to be specific for the stem cells population. In 1999, the p63 transcription factor was essential in stratified epithelial regeneration (Yang et al., 1999). Two years later, Pellegrini et al. (2001) demonstrated that only small cells expressing high levels of p63 in the nuclei could be identified as epidermal and corneal stem cells in humans. Similarly, Yes-associated protein (YAP), a transcriptional activator involved in the Hippo pathway, was explicitly shown in the nuclei of human stem cells of the epidermis (De Rosa et al., 2019). A YAP target named survivin, more expressed in the hair follicles than in the interfollicular epidermis, is considered an additional stem cells marker (Marconi et al., 2007; Dallaglio et al., 2009).

The epidermal stem cell isolation occurred by chance, while Howard Green’s vision drove investigations into the discovery of a new outstanding technology: from a small skin biopsy, a large number of human epidermal keratinocytes were obtained by cultivating them on a feeder layer of lethally irradiated 3T3 cells (Rheinwatd and Green, 1975). This seminal study progressively defined a real microenvironment, providing adhesion, general and specific proliferative stimuli, as well as differentiation signals to obtain an appropriate polarization with the stratification of this vital tissue.

This technique has been considered the most successful approach compared with the cultivation of epidermal keratinocytes without feeder layer cells or on a feeder layer mitotically arrested by mitomycin C (Llames et al., 2015). The latter, for example, produced microchimerism in treated patients (Hultman et al., 1996), likely due to in vivo re-activation of temporarily arrested cells.

The experience gained from cell therapy for skin regeneration has led to the development of other stem cell-based treatments, such as corneal epithelium replacement, aimed at the restoration of visual acuity (Pellegrini et al., 1997; Rama et al., 2010).

The cornea is a unique avascular tissue that functions as a refractive component of the ocular surface system. It is composed of epithelium, stroma, and endothelium (DelMonte and Kim, 2011), and its main features, namely its mechanical strength and transparency, are prerequisites that allow vision. In fact, corneal trauma and burn (among others), causing the loss of corneal transparency and the consequent corneal opacity, represents the fourth leading cause of blindness in the world (Pascolini and Mariotti, 2012).

An additional epithelium extensively used for regeneration is the oral mucosa. It is composed by a non-cornified squamous epithelium endowed with a great regenerative capacity that is necessary for the reaction to daily damages caused by chewing. Given its easy accessibility and the possibility for multiple biopsies, the oral mucosa is widely used in many surgical reconstructive approaches (Yarington, 1980; Lin et al., 2003; Grimsby and Baker, 2014).

Several studies have been conducted to identify molecular markers for oral mucosa stem cells. Oral mucosa epithelium differs depending on the oral cavity area (buccal, gingiva, hard palate, tongue, etc.), and each study investigated stem cell markers from different oral mucosa sites. It was reported that buccal and gingival cells, expressing high levels of the neurotrophin receptor p75, had greater in vitro proliferative capacity and were typically slow cycling in vivo (Nakamura et al., 2007). Two additional markers were associated with oral mucosa stem cells, namely p63 (α isoform) and Bmi1 transcription factors. They were enriched in holoclone cells and in young passages of in vitro serial cultivation (Corradini et al., 2016).

Many proteins investigated as putative oral mucosa stem cells, were previously reported as skin basal or stem cell markers including integrins, CK 5, 15 and 19, p63, MCSP, CD44H, p75, ABCG2 (Jones and Klein, 2013).

The high regenerative capacity of oral mucosa epithelium has also been investigated, in human and mouse, highlighting an essential role of SOX2 and PITX1 transcriptional networks (Iglesias-Bartolome et al., 2018).

Among the applications of cultured oral mucosa grafts, some were proposed as wound dressings for the treatment of superficial lesion of the oral cavity or chronic ulcers in the digestive tract to promote wound healing.

In the following paragraphs, we are reporting the use of oral mucosal epithelium for engraftment in ophthalmology, where the cultured tissue has shown the same optical transparency of the corneal epithelium (Nishida et al., 2004), and in urology.

The salivary glands (SGs), anatomically connected with the oral mucosa, are essential structures responsible for saliva production. This fluid is mainly involved in fundamental functions such as digestion, regeneration of oral and esophageal mucosa, and protection from bacterial infection and dental caries (Khan et al., 2020). Many pathological conditions such as infections, autoimmune diseases (e.g., Sjogren’s syndrome), metabolic disorders or consequences of radiation therapy (RT) targeting head and neck cancers can drastically damage SGs, with severe repercussions for patients’ quality of life. Until now, no determined therapeutic approaches are available to treat salivary dysfunctions like xerostomia (dry mouth syndrome); hence current treatments are mainly symptomatic. For this reason, there are many ongoing studies aimed at developing novel regenerative strategies.

The SG presents several progenitor populations that are able to regenerate the tissue, depending on the extent and the location of the damage (Porcheri and Mitsiadis, 2019), as highlighted by studies mainly based on animal models. Among the many putative molecular markers investigated to identify the SG stem-progenitor cells, c-kit, SOX2, CK5 and ASCL3 seem to have a major role. C-kit + cells can proliferate and differentiate in vivo and in vitro and, more importantly, to restore the salivary secretion when transplanted in murine irradiated SGs (Lombaert et al., 2008a). Analogously, the expression of the transcription factor ASCL3 seems to regenerate both acinar and ductal cells (Bullard et al., 2008). However, when ASCL3 is ablated, the SG appear smaller than the control but retain the presence of a population of CK-5 + basal progenitor cells, highlighting a compensatory mechanism among different stem-progenitor populations (Arany et al., 2011). Finally, a subset of SOX2-expressing acinar cells (SG secreting cells) was found to replace acinar cells after SG irradiation through a SOX2 nerve-dependent mechanism (Emmerson et al., 2018).

Currently, the regenerative strategies aimed at restoring the SG function include cell-based therapy, gene-therapy and bioengineering strategies. Among cell-based therapies, starting from 2008, Coppes’s group demonstrated the ability of rodent SG-specific epithelial cell transplantation to restore SG after RT (Lombaert et al., 2008a; Nanduri et al., 2013). These cells can be safely cryopreserved for an extended period (Nanduri et al., 2013), and their stem cell pool can be enriched (Nanduri et al., 2013). Preliminary results in humans, derived from the isolation and expansion of SG stem-progenitor cells, would suggest a similar restoring ability. However, further studies are needed to verify the clinical applicability of these stem-progenitor cells (Feng et al., 2009; Pringle et al., 2016).

Different cell therapy approaches involve the use of non-epithelial cell types to activate regenerative mechanisms on residual SG after RT or recreate the surrounding environment. Among them, human adipose-derived mesenchymal stem cells (hAdMSCs) have shown to reduce cell apoptosis and tissue fibrosis and differentiate in SG endothelial cells (Kojima et al., 2011; Lim et al., 2013). Moreover, in 2017, the first in man double-blinded, Phase I/II Clinical Trial was performed to evaluate safety and feasibility of this type of cells to treat radiation-induced xerostomia (no results yet) (Grønhøj et al., 2017). Similarly, Bone Marrow (BM)-derived cells were found to increase microvessel density and enhance saliva secretions by inducing epithelial repair (Lombaert et al., 2008b).

Besides cell-based therapy, promising results came from gene therapy with adenoviral vectors to deliver hAQP1 and KGF genes. Notably, the administration of hAQP1 gene in irradiated parotid glands led to an increase in saliva production in 5 out of 11 treated patients, who maintained stable results up to 5 years (Baum et al., 2012; Alevizos et al., 2017). Instead, the administration of KGF genes in mice before irradiation prevents salivary hypofunction with no effect on tumor growth, suggesting a putative future clinical application (Zheng et al., 2011).

Finally, bioengineering and TE strategies are under evaluation for the regeneration of damaged SGs. Despite a plethora of different substrates tested, the optimal scaffold for in vivo SG regeneration has not yet been identified (Mitroulia et al., 2019). Nevertheless, a promising proof-of-concept for regenerative organ replacement comes from bioengineering, as in 2013, Ogawa and colleagues demonstrated the regeneration of an utterly functional SG through the orthotopic implantation in mice, of a bioengineered SG germ. The bioengineered graft went through successful morphogenesis; it perfectly integrated with the recipient salivary duct and exerted the main SG functions (Ogawa et al., 2013).

In conclusion, until now, no SG regeneration strategies have yet become a real therapeutic alternative for the patient suffering from xerostomia, despite the progress obtained in the last years. Since most of the studies are based on animal models, further work is needed to validate the results in humans.

In contrast to the previously described epidermal, corneal and mucosal human epithelia, an effective approach to reconstruct the airway epithelium has not yet been clinically established. Some of the problems that limit airway regeneration include difficulties observed by several groups to effectively grow primary human respiratory epithelial cells (Widdicombe et al., 2005; Walters et al., 2013; Butler et al., 2016) to maintain the stem cell pool and their potential to regenerate all specialized cell types, as well as the reconstruction of a long, full-thickness respiratory tract.

In addition, the airway structure and its epithelium drastically change from the nasal cavities to the alveoli, and different stem/progenitor cells populations have been proposed for the various respiratory tracts, mainly relying on lineage tracing experiments and airway injury models (Basil et al., 2020; Parekh et al., 2020). The basal cells, characterized by the expression of CK5 and CK14 and the transcription factor p63, have been identified as the primary multipotent stem cells population of the tracheobronchial pseudostratified epithelium due to their capacity to self-renew and differentiate into the supra-basal cell types (Rock et al., 2009; Rock et al., 2010). Deep into distal airways, SCGB1A1 + club cells act as progenitors for the goblet and ciliated cells (Zuo et al., 2018; Parekh et al., 2020). Following an injury, these cells can also dedifferentiate acquiring a basal cell morphology and regenerate alveolar type I and II cells (Barkauskas et al., 2013; Tata et al., 2013). Meanwhile, the regeneration of the alveolar epithelium depends on the presence of cuboidal alveolar type II SFTPC + cells (AT2) which can self-renew and give rise to the gas-exchanging AT1 cells (Barkauskas et al., 2013) other than secrete surfactants proteins.

The complexity of the several compartments of the respiratory system highlights the necessity to consider a specific reconstructive approach for each district.

A wide range of approaches has been tested for tracheal and main bronchi reconstructions. The clinical use of synthetic prosthesis is characterized by high morbidity and mortality, so that solid substitutes are no longer considered suitable for long-term tracheal replacement, and the use of porous prosthesis is limited to laryngeal replacement (Etienne et al., 2018).

Decellularized aortic and tracheal allografts are biocompatible and do not require immunosuppression, although their clinical applications are characterized by a high number of re-interventions and high morbidity. Indeed, the absence of a continuous respiratory epithelium and poor mechanical properties require stent application with increased risks of chronic inflammation, infection, and damage to the surrounding tissues (Etienne et al., 2018).

With regard to the tissue engineering strategies, in 2008, the first bioengineered trachea was implanted in a patient affected by severe bronchomalacia (Macchiarini et al., 2008), and despite the complications, other subjects were treated with a similar therapeutic approach (Hamilton et al., 2015). However, morbidity and a high rate of lethality were observed during the follow-up period (Gonfiotti et al., 2014; Elliott et al., 2017). The recurrence of stenosis, infections, and inflammation within the transplanted bioengineered constructs was- linked to the lack of correct vascularization and epithelialization (Pepper et al., 2019; Niermeyer et al., 2020). Previous studies have highlighted the need for appropriate preclinical data to optimize the scaffold, for the identification and culture of all cell types to manufacture grafts, and the lack of studies on the interactions between cells and cell-scaffolds (Niermeyer et al., 2020).

The inner epithelium of the human intestine, with a surface of > 30m², is the second-largest tissue of our body (Gehart and Clevers, 2019). It is involved simultaneously in two main functions: metabolites absorption and barrier protection against potentially noxious microorganisms or environmental insults. It undergoes continuous mechanical stress that causes the death of 1011 intestinal epithelial cells every day (Barker, 2014): the life-time of a mature intestinal epithelial cell is therefore very short (3–5 days) (Darwich et al., 2014). This elevated cell loss is compensated by a high self-renewal rate, triggered by a stem cell population residing within invaginations of the intestinal epithelial wall, the crypts. The crypt is composed of continuously dividing crypt base columnar cells (CBCs), interspaced with Paneth cells, an essential source of niche factors. The CBCs have been identified through several markers, including Lgr5, Ascl2, Prom1, Olfm4 and Smoc2 (Kim et al., 2017). In addition to the CBC, another population of quiescent reserve stem cells has been identified at position + 4 from the bottom of the crypt and is characterized by Bmi1, Tert, Hopx and Lrig1 (Gehart and Clevers, 2019). The stem cells give rise to the transient amplifying cells that rapidly divide and become mature, while being pushed out from the crypt to reach the villus’ tip, where they eventually undergo apoptosis, exfoliating into the lumen. Six types of mature intestinal cells can be distinguished in absorptive (enterocytes and M cell) and secretory (goblet, Paneth, enteroendocrine and tuft cell, the latter involved in immune regulation).

The high regenerative capacity of this epithelium, with stem cells able to self-renew constantly throughout life, makes this tissue ideal for regenerative medicine approaches. At the same time, however, the continuous proliferation of intestinal stem cells accumulates mutations that can promote cancer growth. The most effective treatments for cancers remain nowadays radiotherapy and chemotherapy, even though they induce severe damages to the normal tissues and their side effects are more evident in rapidly renewing tissues such as hematopoietic system and gastrointestinal tract (Yu, 2013). Lgr5+ CBC are particularly susceptible to radiation, and they are replenished by reserve stem cells, which appears to be radioresistant given their quiescence (Kim et al., 2017). Although intestinal cells, following irradiation, showed high plasticity in regenerating the epithelium, the administration of mesenchymal stem cells demonstrated a reduction in the inflammatory response, facilitating epithelial regeneration (Qi et al., 2020).

The intestine is very difficult to regenerate due to its complex structure, pH conditions and its enteric nervous associated motility. Use of decellularised extracellular matrix scaffolds shows promises in regenerating the intestinal trait in preclinical studies, even though they promoted substantial contractions upon implantation (Hussey et al., 2017).

The breakthrough in intestinal regeneration has been developing intestinal organoids (Sato et al., 2009). Lgr5+ cells in organoids guarantee the self-renewal of these “mini-guts,” which can exert both digestive and absorptive functions. Intestinal organoids have been used to study cancer progression and mutagenesis (Drost et al., 2015; van de Wetering et al., 2015; Fumagalli et al., 2017) and study CFTR function in cystic fibrosis (Dekkers et al., 2016). A recent study demonstrated for the first time the success obtained following transplantation of human colon organoids into receiving mice (Sugimoto et al., 2018). Human intestinal organoids could be instrumental for developing novel approaches in regenerative medicine with the aid of a phenotypic landscape (Shin et al., 2019; Lukonin et al., 2020) to understand signaling pathway and biomaterials to favor implantation.

The number of attempts involving epithelial regeneration clearly shows that these tissues’ absence represents a huge medical problem (Kucharzewski et al., 2019; Table 1). Even a partial disruption of the epithelial tissues, the most common being a penetrating infection, dramatically increases the likelihood of health impairment and illustrates the importance of epithelia in body functions.

It is worth noting that humans can live after partial heart impairment, with limited pancreatic activity or partial brain loss, but not with a partial absence of epithelia in any part of the body. As signaling and secretory tissues, these barriers are associated with a critical immunosurveillance, making the therapeutic use of allogenic tissues, only a temporary solution under immunosuppression or a wound dressing.

The cornea has been considered a typical exception, as an immune-privileged site. However, this is not entirely true (Henderson et al., 2001; Inomata et al., 2020). Indeed, human corneal endothelial cells are a potential target of immune attack after corneal transplantation, and the limbus, containing corneal stem and progenitor cells, are vascularized and immunoreactive to allogeneic components (Lahdou et al., 2013).

The importance of immunological reactions can be related to the use of allogeneic or xenogenic cells from some matrix sources, as well as to the inflammatory reactions driven by some materials used for tissue regeneration. This was the case with tracheal and many urethral or intestine scaffolds, which produced significant adverse reactions in in vivo studies.

The introduction of a fully biocompatible support, such as fibrin glue or amniotic membrane, resulted in a significant improvement of different processes: avoided the shrinking of the epithelium, thus enabling the coverage of large wound surfaces in a short time, in addition to safe long-distance transport.

The easy access to some epithelia does not guarantee the development of successful clinical applications (Figure 2). However, the accessibility to epithelia has been the starting point for much research in the field. The knowledge achieved has laid the groundwork for some very effective treatments for several pathologies, such as burns, skin pigmentation disorders, ocular defects, and some genetic conditions (Figure 2).

Additional progress will come from multidisciplinary interactions, providing different perspectives and different solutions supporting treatments’ efficiency. Examples can be the development of alternative diagnostic methods aimed at the precise grading of some pathologies, the pharmaco-toxicology of drugs on cultured tissues or in the whole treatment, the functionalization/patterning of surfaces by nanotechnology, and the analysis of predictive markers for responses to the therapeutic interventions.

Many conclusions can be drawn from these different applications: altogether, these experiences suggest several improvements for the therapeutic use of advanced therapy medicinal products. Without being exhaustive, the list of problems includes the need for standardization, which is common to most therapies. The non-homogeneous selection of pathologies included in the preliminary clinical trials produced significant intra- and inter-pathology heterogeneity, associated with differences in the corresponding microenvironments and significantly contributed to variable cell behavior. Within these microenvironments, the extracellular matrix and paracrine signaling can profoundly influence the way cells engraft, grow, differentiate, or persist.

Additional common issues were the incomplete cell characterization of the cultured tissues and some lack of comparison with their in vivo counterparts, frequently impairing the clinical outcome and related evaluations. Therefore, a more exhaustive cell characterization, together with specific stem cell identification, are instrumental for self-renewal and long-term tissue maintenance.

The nature of stem cells could change under homeostatic conditions or during tissue repair. The essential characteristics of the stem cells, like the specific location and marker expression, quiescence, asymmetric division, and unidirectional differentiation, do not apply to all tissues, as shown in the airways. In a pragmatic vision, the focus on the search for stem cells should move from their physical identification to their function, meaning the tissue ability to restore all cell types, over the life-time. Preclinical and clinical studies must support the validation of the selected markers for their role in these adult stem cells’ structural and functional regeneration capabilities.

The standardization of methods does not mean having one single procedure; rather, it means that scientific experts would define the available golden standard time by time, by selecting the most characterized and safe method from a scientific and regulatory perspective.

This approach would enable bridging new methods to the golden standard, as the most characterized procedure, to highlight different cell/tissue behavior and define the best risk/benefit ratio. Agreements on methods and controls should also be based on long-term follow-up data of previous treatments, which implies a shift from estimating probabilities to relying on certainties, without censuring the possibility of medical innovation (Sugarman et al., 2018). The World Medical Association in the declaration of Helsinki, literally reported: “in the treatment of an individual patient, where proven interventions do not exist, or other known interventions have been ineffective, the physician, after seeking expert advice, with informed consent from the patient or a legally authorized representative, may use an unproven intervention if, in the physician’s judgment, it offers hope of saving life, re-establishing health, or alleviating suffering.” This intervention should subsequently be made the object of research, designed to evaluate its safety and efficacy. In all cases, new information must be recorded and, where appropriate, made publicly available (The World Medical Association Inc, 2008).

EM, DA, FM, GG, EA, AM, BM, VG, VS, and GP contributed to the analysis of literature, writing and correction, and figures. LL contributed to the analysis of literature, revision of the manuscript, and figures. EM and DA collected and summarized the whole work. GP also provided the final revision. All authors contributed to the article and approved the submitted version.

EM, FM, AM, VG, VS, and GP are employees of Holostem Terapie Avanzate, producing an Advanced Therapy Medicinal Product for corneal restoration. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.