In the research of esophageal tissue engineering, researchers have conducted a large number of innovative research studies on the construction of single-layer scaffolds to repair the mucosal layer or the muscle layer. Single-layer scaffolds are classified according to the choice of materials, mainly including acellular matrix or (and) polymer scaffolds. The acellular matrix includes the small intestinal submucosa (SIS), urinary bladder submucosa (UBS), esophageal mucosa, etc. Polymer materials include polylactide (PLA), poly (l-lactide-co-ε-caprolactone) (PLGA), poly (3-hydroxybutyrate-co-3-hydroxyvalerate) (PHBV), polylactide-poly (ɛ-caprolactone) (PLA-PCL), polyurethane (PU), etc. (Badylak et al., 2000; Nieponice et al., 2014; Kuppan et al., 2016; Tan et al., 2016; Dorati et al., 2017; Kuppan et al., 2017; Luc et al., 2018) (Table 1).

From the perspective of single-layer scaffold structure, the classification mainly includes acellular matrix, membranes grafted with biomolecules, electrospinning scaffolds, micro-pattern scaffolds, etc. (Hou et al., 2016; Wei et al., 2018; Zhuravleva et al., 2019; Chaitin et al., 2021; Levenson et al., 2021) (Table 2). In order to accurately simulate the inner ring and outer longitudinal structure of the muscle layer, electrospinning scaffolds and micro-pattern scaffolds are mainly used (Gong et al., 2013; Kuppan et al., 2017). Our team has conducted a large number of in vitro and in vivo animal studies on single-layer scaffolds. The original idea was to graft collagen on the surface of polymer materials such as PLGA and poly (l-lactide-co-caprolactone) (PLLC) to improve their biocompatibility (Zhu et al., 2005; Zhu et al., 2006). Then, the polycaprolactone/silk fibroin (PCL/SF) electrospinning scaffold is made by electrospinning technology, and the electrospinning fiber pore size can simulate the structure of the extracellular matrix (ECM) (Lv et al., 2014). At this stage, the micro-pattern membrane technology is used to construct a new type of esophageal bionic scaffold, which has successfully constructed a micro-pattern PU scaffold and double-layer scaffold of the esophageal acellular matrix (Hou et al., 2019; Wang X. et al., 2020).

The abovementioned studies carried out the structural bionic design of single-layer scaffolds from the perspectives of material selection and construction methods and achieved a series of achievements in the damage repair of the esophageal mucosa and muscularis. Studies have shown that if cells are introduced into scaffolds, they will work synergistically with scaffolds in the microenvironment in vivo to further enhance functional repair of tissues. For example, Badylak et al. made artificial defects in the dog’s esophagus and used ECM derived from SIS or UBS to repair the esophageal defect. After 35 days, the scaffold was partially covered by the squamous epithelium, and only scattered skeletal muscle cells surrounded collagen connective tissue (Badylak et al., 2000). Xie et al. proved that, after implantation in dogs, SIS alone is not completely endothelialized. When bone marrow mesenchymal stem cells (MSCs) are combined with SIS, the results show that the defect site is completely endothelialized, the muscle layer is regenerated, and the new microvessels are dense (Tan et al., 2013) (Figure 2).

Therefore, researchers have introduced cells, such as ECs, SMCs, and stem cells, into biomimetic single-layer scaffolds to construct tissue engineering scaffolds. The ECs or SMCs are difficult to be widely used due to the large damage to the donor when they are acquired and the limited ability of proliferation and differentiation after cell expansion. On the contrary, stem cells have the advantages of large differentiation potential, strong proliferation ability, convenient and easy acquisition from the body, and the ability to differentiate into specific cells in tissue. So, they are widely used as seed cells in the field of tissue engineering. For example, Ivo et al. combined MSCs and decellularized esophageal muscle tissue to repair the esophagus in situ in pigs, which showed new muscle tissue compared with the decellularized esophageal muscle layer alone (Marzaro et al., 2020) (Figure 2). Aho et al. Using PU material, autologous adipose-derived mesenchymal stem cells were seeded to form a cell-span esophageal implant (CEI). After resection of the patient’s esophageal cancer, in situ repair was performed using CEI and followed by esophagogastroduodenoscopy (EGD). After the patient’s death, histological examination revealed esophageal luminal epithelialization and partial muscle regeneration 7.5 months after scaffold implantation (Aho et al., 2021) (Figure 3).

Therefore, single-layer bionic scaffold combined with stem cells to repair the mucosal layer or muscle layer is the current main research direction. On the basis of mucosal layer or muscle layer repair, higher clinical requirements for esophageal repair are proposed, such as full-thickness or circumferential defect, which requires simultaneous repair of the mucosal layer, submucosal layer, and muscle layer of the esophagus. Therefore, it is particularly important to design a multi-layer functional bionic scaffold.

Researchers have studied full-thickness or circumferential defects by designing lamellar or tubular bionic scaffolds. According to the research methods, it is mainly divided into two categories: scaffolds and scaffolds/cells composite. The scaffolds were prepared by the one-step method or multi-step method. The one-step method is to mix the scaffold material into a whole through melting, electrostatic spinning, temperature-induced sedimentation, etc., and different components complement each other and work in synergy. For example, Tan et al. melted PCL/PLA and stretched it into a directional spinning tubular structure (Tan et al., 2016). The multi-step method is based on the perspective of esophageal structure bionics, combining different layers of scaffold materials through a certain link method. Joshua et al. prepared the silk fibroin double-layer scaffold by solution pouring (Gundogdu et al., 2021). Rossella et al. used electrospinning and temperature-induced sedimentation to construct two double-layer scaffolds (Pisani et al., 2020). Saverio et al. designed a PU electrospinning three-layer scaffold (inner and outer layer pore diameter>10 μm; middle layer <10 μm) (Soliman et al., 2019) (Table 3).

Through the abovementioned research and analysis, it can be seen that the design of multi-layer scaffolds is the guiding ideology of bionics, but these scaffolds still cannot completely induce the structural growth of tissues. Therefore, researchers have constructed scaffolds with cells to enhance the repair function of tissue. Natural materials such as esophageal acellular matrix, SIS, and collagen scaffold are compounded with cells.

Guillaume et al. designed the esophageal mucosal acellular matrix/omentum double-layer scaffold, in which MSCs were cultured on the acellular matrix, and the omentum re-matured in pigs. As a result, it was found that 3 months after the esophageal replacement surgery, a new epithelium and muscle regeneration were visible (Levenson et al., 2021) (Figure 4). Paola et al. used the method of organoid culture to construct a multi-layer esophageal scaffold with cells in vitro. The researchers re-seeded ECs on the acellular matrix of the rat esophageal mucosa and allowed the cells to grow in the lumen of the acellular scaffold to construct the esophageal mucosal layer and co-cultured human or mouse fibroblasts and mouse neural crest cells in vitro. The muscle layer is constructed and then implanted into the rat omentum for in vivo culture to promote angiogenesis and build a multi-layer esophageal structure together with the mucosal layer. This kind of esophageal tissue composed of cells is more complete than the commonly used acellular matrix and other natural materials, but it needs to be verified by animal experiments to prove its positive significance (Urbani et al., 2018) (Figure 4).

In addition to natural materials, synthetic polymer materials such as PLGA, PCL, and PU have also been studied in combination with cells. For example, Christine et al. prepared a PLGA/PCL electrospun tubular esophageal scaffold, the inner cavity of the scaffold was compounded with autologous ECs, and the outer side was compounded with autologous SMCs. The composite scaffolds containing cells were cultured in an in vitro bioreactor for a period of time and then implanted into the mouse esophagus in situ, the esophagus is still viable after 2 weeks, and the cells maintain the phenotype (Jensen et al., 2015) (Figure 4). Dennis et al. combined porcine esophageal mucosal cells and electrospun PU scaffold into a tubular scaffold and implanted it into the whole-peripheral defect of the porcine esophagus, and the results showed that the mucosal layer, submucosa, and muscle layer of the esophagus regenerate simultaneously and have abundant blood vessels (Barron et al., 2018) (Figure 4).

Different from the traditional scaffolds, 3D printing scaffolds have many advantages, such as the flexibility of preparation methods, the customization of irregular tissue damage parts, and the ability to prepare scaffolds with very complex structures (Memic et al., 2017; Matai et al., 2020; Wan et al., 2020; Barros et al., 2021; Wu et al., 2021; Yang et al., 2021). The 3D printing scaffolds have been studied in esophageal repair. For example, Chung et al. used a 3D melt extrusion method to construct a polycaprolactone (PCL) 3D printing scaffold, seeded MSCs on the scaffold to participate in esophageal reconstruction, cells grew along the direction of the scaffold, and implanted it in the defect of the rat esophagus. The results show that the new tissue repaired by the 3D printing scaffold is similar to natural tissue and has obvious advantages compared with electrospun PU scaffolds (Figure 5) (Park et al., 2021). Although 3D printing scaffolds have many advantages, this method also has its own limitations, such as lack of diversity of bio-ink, harsh printing conditions (high temperature or UV curing), and expensive equipment for printing cells.

The mixed use of polymer synthetic materials and natural materials has gradually become the focus of research. For example, Jonathan et al. used electrospinning technology to make PLGA fiber layers on the SIS acellular matrix to form a double-layer esophageal scaffold, the results showed that human esophageal smooth muscle cell culture experiments and subcutaneous embedding presented good biocompatibility (Syed et al., 2019), but further research is needed for in vivo repair.

Our group’s previous study used micro-pattern technology to construct a three-layer scaffold, which corresponds to the inner ring muscle (S1), outer longitudinal muscle (S2), and mucosal layer (S3) of the esophagus. After inoculating MSCs on the composite scaffold, it was implanted into the esophageal defect. The results showed that there was new esophageal tissue, including the muscle layer and mucosal layer. However, the PU material can still be found in the tissue 180 days after implantation, which may affect the speed of muscle regeneration (Wang X. et al., 2020) (Figure 6).

At present, there are two main construction methods of esophageal scaffold, one-step construction and multi-step construction. The advantage of one-step construction is that it is relatively simple, only need one or more types of biomaterials are required, and generally, the binding force between the support layers is strong and stable. However, the disadvantage is that the material and function are relatively single, so it is very difficult to accurately simulate the multi-layer structure of the esophagus and give specific functions to each layer of the scaffold. Since the multi-layer scaffold prepared by the multi-step construction method is flexible, it can provide different materials (natural materials, synthetic materials, or both) and cells (ECs, SMCs, or stem cells) for each layer of the scaffold, so as to more accurately mimic the structure and function of the esophagus. Therefore, it is the current development direction to prepare an accurate bionic multi-layer esophageal scaffold by multi-step construction.

For multi-layer scaffolds prepared by multi-step construction, the firmness between the scaffolds is another key issue. If the adhesion between scaffolds is poor, the multi-layer scaffolds will fall off or shift, which will seriously affect the repair effect of the esophagus. The connection modes between the layers of multi-layer scaffolds include the solution casting method (Gundogdu et al., 2021), temperature-induced precipitation method (Dorati et al., 2017), solvent volatilization method of electrospinning (Chung et al., 2015), and glue bonding (Deng et al., 2019). The first three methods are not universally applicable because solutions or solvents may dissolve the active components such as protein, growth factor in the scaffold, and too high or low temperature is not conducive to the introduction of proteins and cells into the scaffold. Glue bonding does not affect the design of each layer of the scaffold, as long as the scaffold prepared separately is combined, which is a simple combination method with universal applicability. Generally, it is relatively easy for the glue to adhere to objects in a dry environment, but it remains a great challenge for repair in a wet environment (exudate or blood at the injury) or dynamic adhesion (human movement).

At present, the tissue glue used in the clinical treatment of esophageal anastomotic fistula is mainly cyanoacrylate (superglue, highly toxic, and rarely used) and fibrin glue (fibrin glue, frequently used, but with low adhesion ability) (Rao et al., 2018). The adhesion strength and adhesion energy of fibrin glue are about 10 kPa and 10 Jm⁻², respectively (Deng et al., 2019). As the esophagus is a soft tissue with peristalsis and swallowing functions, higher requirements are put forward for the glue used to bond the multi-layer esophageal scaffold (>>10 kPa and >>10 Jm⁻²). New adhesives, such as nano-clay/multi-walled carbon nanotubes/isopropylacrylamide hydrogel (adhesive strength 7 kPa) (Deng et al., 2019), sodium p-styrene sulfonate/chloromethane quaternized dimethylaminoethyl acrylate hydrogel (adhesive strength 25 kPa, adhesive energy 50 Jm^-2) (Rao et al., 2018), chitosan/double-bonded phenylalanine hydrogel (adhesive strength 14 kPa) (Sharma et al., 2019), aldehyde functionalized hyaluronic acid/3,3′-dithiobis (propionyl hydrazide) hydrogel (adhesive strength 120 kPa) (Sigen et al., 2021), polyethylene glycol/lysozyme hydrogel (adhesive strength 32 kPa) (Tan et al., 2019), folic acid/polydimethyl diallyl ammonium chloride hydrogel (adhesive strength 150 kPa) (Gao et al., 2021), and hyaluronic acid/catechol/horseradish peroxidase hydrogels (17 kPa) (Wang D. et al., 2020) (Figure 7), are used. The hydrogels mentioned above can only meet one of the requirements of adhesion or cytocompatibility. Therefore, it is the research direction of adhesiveness of hydrogel to satisfy high adhesion and biocompatibility in a complex environment.