Biomaterials 57 (2015) 133e141

Contents lists available at ScienceDirect

Biomaterials journal homepage: www.elsevier.com/locate/biomaterials

Biomimetic and synthetic esophageal tissue engineering Todd Jensen a, 1, Alex Blanchette a, 1, Stephanie Vadasz a, 1, Apeksha Dave a, 1, Michael Canfarotta a, 1, Wael N. Sayej b, 2, Christine Finck a, c, *, 1, 3 a

Department of Vascular Biology, University of Connecticut Health Center, 263 Farmington Avenue MC3501, Farmington, CT 06030, USA Department of Gastroenterolgy, Connecticut Children's Medical Center, 282 Washington Street, Hartford, CT 06106, USA c Department of Surgery, Connecticut Children's Medical Center, 282 Washington Street, Hartford, CT 06106, USA b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 28 January 2015 Received in revised form 30 March 2015 Accepted 2 April 2015 Available online 28 April 2015

Background/Purpose: A tissue-engineered esophagus offers an alternative for the treatment of pediatric patients suffering from severe esophageal malformations, caustic injury, and cancer. Additionally, adult patients suffering from carcinoma or trauma would benefit. Methods: Donor rat esophageal tissue was physically and enzymatically digested to isolate epithelial and smooth muscle cells, which were cultured in epithelial cell medium or smooth muscle cell medium and characterized by immunofluorescence. Isolated cells were also seeded onto electrospun synthetic PLGA and PCL/PLGA scaffolds in a physiologic hollow organ bioreactor. After 2 weeks of in vitro culture, tissueengineered constructs were orthotopically transplanted. Results: Isolated cells were shown to give rise to epithelial, smooth muscle, and glial cell types. After 14 days in culture, scaffolds supported epithelial, smooth muscle and glial cell phenotypes. Transplanted constructs integrated into the host's native tissue and recipients of the engineered tissue demonstrated normal feeding habits. Characterization after 14 days of implantation revealed that all three cellular phenotypes were present in varying degrees in seeded and unseeded scaffolds. Conclusions: We demonstrate that isolated cells from native esophagus can be cultured and seeded onto electrospun scaffolds to create esophageal constructs. These constructs have potential translatable application for tissue engineering of human esophageal tissue. © 2015 Elsevier Ltd. All rights reserved.

Keywords: Tissue engineering Esophagus Bioreactor Synthetic scaffold Orthotopic implantation

1. Introduction Esophageal atresia occurs in 1:3000e1:5000 live births in the United States [1]. For many of these patients a primary anastomosis, either immediate or delayed, remains a feasible option. However, in long gap esophageal atresia, the most severe form, esophageal replacement by gastrointestinal transposition may be necessary [2]. Besides esophageal atresia, caustic injuries to the esophagus also add to the demand for esophageal replacement, with an estimated 5000e15,000 caustic ingestions occurring per year in the United States. This can be a devastating injury depending on the length of the esophagus that becomes necrotic [3]. Finally, esophageal

* Corresponding author. Department of Surgery, Connecticut Children's Medical Center, 282 Washington Street, Hartford, CT 06106, USA. E-mail address: cfi[email protected] (C. Finck). 1 Tel.: þ1 (860) 679 7845; fax: þ1 (860) 679 1201. 2 Tel.: þ1 (860) 545 9560; fax: þ1 (860) 545 9561. 3 Tel.: þ1 (860) 545 8477; fax: þ1 (860) 545 9545. http://dx.doi.org/10.1016/j.biomaterials.2015.04.004 0142-9612/© 2015 Elsevier Ltd. All rights reserved.

carcinoma occurs in 500,000 adults annually [4]. Surgical resection remains the mainstay of treatment, often leading to complete esophageal resection. While a variety of reconstructive options are available to replace the esophagus, these procedures are associated with substantial short and long-term complications such as stenosis and dysmotility, and there is a consensus among most pediatric surgeons that better postoperative outcomes are achieved through preservation of native esophageal tissue [5,6]. Unfortunately, conventional organ transplant is not always a suitable option, especially in the pediatric population where there is already a shortage of appropriately sized donor tissue [7]. A tissueengineered esophagus made from either natural or synthetic biomaterials may therefore offer a real alternative to conventional treatments for severe esophageal disease. In order for a tissue-engineered construct to achieve functional properties, it must mimic the architecture of the native tissue [8]. De-cellularization of native tissue is widely recognized in many organs including the lung, kidney, heart, and blood vessels [9e12]. The process of de-cellularization removes cellular elements while

134

T. Jensen et al. / Biomaterials 57 (2015) 133e141

preserving the native extracellular matrix (ECM), surface topography, and biomechanical properties of the native tissue. Theoretically, these acellular scaffolds have reduced immunogenicity and can support cell adhesion, proliferation, chemotaxis, and differentiation [9e12]. However, due to the lack of appropriately sized cadaveric donors, generating size-specific scaffolds using electrospinning of natural polymers may lend itself more to personalized medicine. A tissue-engineered scaffold can be constructed using natural polymers already FDA approved for use in skin, bone and cartilage applications [13]. Previous studies have been published utilizing either synthetic poly (lactic-co-glycolic acid) (PLGA) or polycaprolactone (PCL)-based scaffolding for esophageal tissue engineering [14e19]. In addition, recent studies have utilized isolated epithelial cells or organoid units generated from fetal esophageal tissue [14e16]. Our study is novel in that it is the first to enrich both smooth muscle cells and epithelial cells from native tissue and seed them in 3 dimensions on the appropriate surfaces to mimic a normal esophagus. The esophagus is a hollow conduit with stratified squamous epithelium, submucosa, and outer longitudinal and inner circular muscle layers. Optimally, a tissue-engineered scaffold should incorporate into native tissue, enable the passage of food, propagate peristalsis, and withstand the acidic pH of gastric fluid. The ability to synthesize a tubular scaffold that supports the adhesion and proliferation of biomimetic cells lends itself to the creation of a relevant tissue-engineered esophagus. Identification of an appropriate cell source is essential. Ideally cells will be autologous, phenotypically resemble esophageal cells, and ultimately enable restoration of function. One method is to expand cells from the native organ in culture and use this population to seed a scaffold for implantation. Herein we describe the isolation of allogenic cells from a rat esophagus which were then seeded onto synthetic matrix in a physiological bioreactor. Segments of these re-seeded scaffolds were orthotopically implanted into a rat and demonstrated integration with the native tissue. A schematic of the steps required to generate a synthetic scaffold for implantation is presented in Fig. 1. 2. Methods 2.1. Scaffolding 2.1.1. Electrospinning PLGA and PLGA/PCL scaffolds The following procedure was adapted from a previously published protocol [20]. PLGA (Mw ¼ 50e75 kg/mol, 85:15 lactide/ glycolide) (Evonik Wallingford, CT) was dissolved in a 80/20 mix of dichloromethane (DCM) (Fisher Scientific, Pittsburgh, PA) and dimethylformamide (DMF) (Fisher Scientific, Pittsburgh, PA) to

result in a concentration of 25% w/v. The solution was loaded into a 5 mL plastic syringe with a 22 gauge needle and was pumped at a flow rate of 1.0 ml/h using a syringe pump. A high voltage power source was used to apply 15 kV between the needle and a flat piece of aluminum foil. Scaffolding was either spun onto glass coverslips or was spun onto tin foil for generating three dimensional scaffolds. Three-dimensional scaffolds were created by rolling the flat PLGA sheet around a 5 French nasogastric tube and sealing the ends with clinical grade TISSEEL Fibrin Sealant (Baxter Healthcare, Westlake Village, CA). This size nasogastric tube was chosen as it was appropriately sized for an adult Sprague Dawley rat. 2.1.2. Resection of esophageal tissue 10e20 week-old SpragueeDawley rats (Charles River, Wilmington, MA) were euthanized in accordance with University of Connecticut Health Center IACUC approved protocol (ACC#1006210416). A midline incision was made from the lower abdomen to the throat and a median sternotomy was performed. A 5 French nasogastric catheter was inserted through the mouth and advanced to the terminal end of the esophagus. The entire length of the esophagus was carefully dissected and rinsed in phosphatebuffered saline (PBS) with Primocin (InvivoGen, San Diego, CA 1:500). This tissue was then either de-cellularized or utilized for harvesting native cells. 2.1.3. Esophagus de-cellularization for morphologic comparison with synthetic scaffolds The following procedure was adapted from a previously described protocol [21]. De-cellularization of harvested esophagi was completed in a small animal hollow organ bioreactor from HART (Harvard Apparatus Regenerative Technology). Both ends of the organs were fixed to cannulas using 5e0 Prolene sutures (Ethicon). All de-cellularization steps were completed at 37 C and esophagi were subjected to rotational mechanical agitation at a speed of 3 revolutions per minute. The lumen of the esophagi was filled with 4% sodium deoxycholate (SDC) (SigmaeAldrich) and the entire organ was bathed in 4% SDC for 24 h. Esophagi were then treated with 0.4 mg/mL DNase-I (SigmaeAldrich) with 1M NaCl in PBS for 12 h. Tissues were rinsed for 8 h in deionized water (DI) containing Primocin (InvivoGen, San Diego, CA). De-cellularized esophageal scaffolds were used in this study exclusively as a control for SEM. 2.2. Cell source 2.2.1. Isolation of epithelial and smooth muscle cells The following procedure was adapted from a previously described protocol [22]. Esophageal tissue was harvested from

Fig. 1. Schematic of steps required to generate a synthetic esophageal scaffold for In vivo implantation. Epithelial cells were seeded in the lumen of the synthetic scaffold and the smooth muscle cells were seeded on the abluminal surface.

T. Jensen et al. / Biomaterials 57 (2015) 133e141

SpragueeDawley rats as previously described and stored in Hanks balanced salt solution (HBSS) (Life Technologies Carlsbad, CA) with Primocin on ice. Briefly, after esophageal harvest, a longitudinal cut was made down the length of the esophagus and the tissue was rinsed with HBSS to remove luminal material. Upon satisfactory rinsing of the esophagus, the tissue was then minced into a fine pulp with sterile razor blades and washed again with cold HBSS with Primocin stored on ice. The tissue was then treated enzymatically with a solution of collagenase (800 u/mL) (Life Technologies Carlsbad, CA) and dispase (0.22 u/mL) (Life Technologies Carlsbad, CA). Epithelial cell medium (EpiCM) (ScienCell Research Labs Carlsbad, CA) or smooth muscle cell medium (SMCM) (ScienCell Research Labs Carlsbad, CA) with 10% FBS and Primocin was used to stop the enzymatic reaction. The solution was centrifuged and cells were seeded onto synthetic scaffolds or tissue culture plastic for in vitro assays, or on three dimensional scaffolds in the HART bioreactor for downstream implantation. Epithelial cell medium (EpiCM) and smooth muscle cell medium (SMCM) are purchased commercially and are made by adding 10 mls of FBS and 5 mls of either Smooth Muscle Cell Supplement or Epithelial Cell Supplement, which are provided with the medium. In addition Primocin is added (1 ml per 500 mls of medium) as an antibiotic/ antimycotic. Cells were isolated from esophagi of non-recipient animals (allogeneic) for all studies. 2.2.2. Seeding of the three dimensional synthetic esophageal scaffold Adequate cell numbers for reseeding were based on surface area calculations for the luminal and abluminal surfaces of the esophagus. For these calculations, the esophagus is assumed to be an annular cylinder with outer diameter Do and inner diameter Di. The number of cells, N, required for seeding is given by the equation: N ¼ DpL/Ac, where L is the length of the scaffold being re-seeded and Ac is the surface area of either the mature epithelial cells (corresponding with the inner diameter Di) or the surface area of the mature smooth muscle cells (corresponding with the outer diameter Do). This method of calculating cell number based on surface area of a cylindrical surface was previously described when performing bronchial airway calculations to determine cell number [23,24]. Once the surface area was calculated based on diameter and length, this was divided by the surface area calculated for each cell type in culture to determine an average number of cells to seed per cell layer. Typical dimensions of the esophagi used in the experiments had length L ¼ 4 cm, inner diameter Di ¼ 1.67 mm, and outer diameter Do ¼ 4.0 mm. Typical dimensions of the mature epithelial cells had surface area AEpi ¼ 280.86 mm2 and mature smooth muscle cells had surface area ASM ¼ 327.93 mm2. Surface area measurements of esophageal epithelial and smooth muscle cells were obtained using the surface area calculation tool in Image Pro Plus (Media Cybernetics, Rockville, MD) on cells in culture. Substituting these values into the equation yields the cell numbers required to seed each side of the esophagus, where Ni ¼ 186,703 cells and No ¼ 159,909 cells. Synthetic scaffolds were UV sterilized for 30 min prior to suturing them onto the cannulas provided in the HART small animal hollow organ bioreactor system. Cells were re-suspended in smooth muscle cell medium in a volume of 1200 ul and 300 ul of the suspension was carefully pipetted over the top of the scaffold 4 times, rotating the scaffold a quarter turn each time to ensure the entire extra-luminal surface is seeded. Cells were re-suspended in the epithelial cell medium in a volume of 1200ul and carefully injected via the luminal port of the bioreactor using a 3 ml syringe in order to seed the intra-luminal surface. The bioreactor (depicted in Fig. 1) fabricated by HART is completely autoclavable and possess various ports in order to generate two separate culturing

135

compartments. This permits SMCM to bath the outside of the scaffold, while EpiCM was injected into the luminal compartment using a syringe. The medium in these compartments did not mix and the medium was changed every other day. The lid of the bioreactor acts similar to a cell culture dish and allows for gas exchange within the incubator. The bioreactor containing the seeded construct was rotated at 3 revolutions per minute using the provided drive unit that is attached to the construct in the bioreactor. The entire unit is placed in a 37  C incubator for 14 days to maintain proper CO2 and O2 levels. 2.2.3. Topographic characterization of scaffolds: scanning electron microscopy Ultrastructure features were determined using scanning electron microscopy (SEM). De-cellularized esophagus and synthetic scaffolds were washed with PBS, fixed in 0.12 M pH 7.4 phosphate buffer containing 2.5% glutaraldehyde and 2.5% paraformaldehyde for 2 h at 4 C and then post-fixed in 1% osmium tetroxide overnight at 4 C. De-cellularized esophagus was dehydrated with ethanol, scaffolds were flash frozen and lyophilized, then all were coated with gold and observed on a Nova NanoSEM 450 NS, FEI, The Netherlands) at the Biosciences Electron Microscopy Facility at the University of Connecticut. 2.2.4. Evaluating Cell viability and proliferation on synthetic scaffolds Various combinations of PCL and PLGA were evaluated in order to identify which combination would induce the greatest cell adherence and proliferation while limiting apoptosis for three dimensional engineering and implantation. The following combinations were tested in this assay: 100% PLGA, 10% PCL/90% PLGA, 20% PCL/80% PLGA, 30% PCL/70% PLGA, 40% PCL/60% PLGA and tissue culture treated glass coverslips without scaffolds as a control. Isolated rat cells were seeded at 50,000 cells per coverslip in a 12-well dish and were incubated for 7 days in EpiCM medium. Medium was changed every other day and scaffolds were fixed with 4% PFA after 7 days and stained via immunofluorescence for Cleaved Caspase 3 (Apoptosis) and Ki-67 (Proliferation). 2.3. Histology & immunofluorescence Cells were fixed in 4% PFA for 10 min at room temperature. Tissue sections (5 mm thick) were stained with hematoxylin and eosin to visualize the cells of the graft and surrounding tissue. In addition, sections were de-paraffinized and antigen retrieval was performed by incubating for 35 min in 10 mM sodium citrate (SigmaeAldrich, St. Louis, MO) and 0.5% Tween-20 (SigmaeAldrich, St. Louis, MO). Cells and tissue sections were permeabilized and blocked in PBS with 0.1% Saponin and 2% FBS for 1 h at room temperature. Primary antibody staining for ECadherin (Santa Cruz sc-7870, 1:200), alpha- Smooth Muscle Actin (Abcam ab32575, 1:500), and S100b (SigmaeAldrich S2644, 1:200) was conducted in PBS with 2% FBS and 0.1% Saponin overnight at 4  C. Secondary antibodies used were of specific isotypes conjugated to Alexa 546 (Life Technologies, Grand Island, NY) at a 1:1000 dilution. Nuclei were stained for 10 min in the dark with DAPI (Life Technologies, Grand Island, NY). Tissue sections were cover slipped using Immuno-mount (Fisher Scientific, Pittsburgh, PA). Tissue sections and cells were imaged on a Zeiss Observer Z1 Inverted Fluorescent Microscope and Carl Zeiss ZEN Blue Software (Carl Zeiss Microimaging, LLC Thornwood, NY). All images were processed using Carl Zeiss ZEN Blue Software (Carl Zeiss Microimaging, LLC Thornwood, NY).

136

T. Jensen et al. / Biomaterials 57 (2015) 133e141

2.3.1. In vivo application: esophageal patch implantation Male and Female Sprague Dawley rats (greater than 10 weeks of age) were placed on a commercial liquid diet (BioServ Flemington, NJ) for 5 days prior to surgery. Some rodents received esophageal scaffolds that were seeded and incubated in the HART bioreactor for 14 days prior to surgery and some rodents received scaffolds without cells. Prior to the start of the procedure, a 5 French nasogastric tube was placed down the oropharynx and into the esophagus of the animal to serve as a guide during implantation. A 2e3 cm incision was made in the ventral region of the neck to expose the trachea and esophagus. A 1 cm incision was made longitudinally in the native esophagus to accept a 1e1.5 cm esophageal patch. The esophageal patch was made by cutting a 1 cm segment of the 3D synthetic scaffold and cutting it longitudinally to create a semicircular patch. The luminal side was attached so that it corresponded to the luminal compartment of the native esophagus. The scaffold was sutured into place with 6e0 prolene suture and the cervical incision was closed with 3e0 prolene suture. The animal received liquid diet for another 5 days prior to transitioning the animal to softened rodent food (with water) and lastly normal rodent diet. The scaffolds were harvested after 14 days of implantation and evaluated histologically. A total of 3 animals were implanted for both seeded and unseeded 20% PCL/80% PLGA scaffolds as well as 1 animal that was implanted with 100% PLGA scaffolding with or without the addition of cells. 3. Results 3.1. Topographic comparison of synthetic and de-cellularized esophageal scaffolds via scanning electron microscopy Scanning electron microscopy (SEM) was performed on seeded and unseeded synthetic scaffolds and compared to de-celluarized rat esophagus. Fiber sizes of de-cellularized rat esophagus are significantly smaller than those of the synthetic scaffolds that were electrospun (Fig. 2 A, B, C). The fiber pattern of the electrospun scaffolds, however, closely mimics that of de-cellularized rat esophagus. After synthetic electrospun scaffolds are seeded and incubated in a HART bioreactor for 14 days, a sheet of cells is present on the surface of the scaffold covering a majority of the surface of the scaffold (Fig. 2 D, E). The surface of the 100% PLGA scaffold is almost

completely covered compared to the 20% PCL/80% PLGA scaffold which contains regions that are not fully covered (Fig. 2 D, E). 3.2. Determination of optimal scaffold ratio that supports viability and proliferation In order to evaluate proliferation and viability of various concentrations of PCL and PLGA, cells isolated from native rat esophagus (as described above) were seeded onto glass cover slips that were coated with combinations of PCL and PLGA. Cells were cultured on these scaffolds for 7 days in EpiCM and were then qualitatively evaluated for cell density, proliferation (Ki-67) and apoptosis (Cleaved Caspase 3). After 7 days in culture, it is clear that cells were dense and proliferative with low levels of apoptosis on tissue culture treated coverslips (Fig. 3A, B). The cell density on PLGA was slightly less than that of the TC control with some degree of proliferation and little to no apoptosis (Fig. 3C&D). As the concentration of PCL increased, the level of proliferation and cell density decreased, with few cells present at 40% PCL and almost no proliferation (Fig. 3 E, G, I, K). The opposite observation was observed with apoptosis, as the PCL concentrations increased the amount of apoptosis increased with many of the cells being apoptotic at 40% PCL concentration (Fig. 3 F, H, J, L). This data demonstrated that the ideal combination that would enhance scaffold properties, while still supporting proliferation and limiting apoptosis was 20% PCL/80% PLGA, which is what was used for three dimensional and implantation studies. 3.3. Characterization of isolated cells on tissue culture plastic and electrospun scaffolds Following dissociation of native rat esophagus tissue, cells were allowed to expand in culture for up to 7 days to obtain sufficient cell numbers for seeding. Cells were seeded on tissue culture treated coverslips as a control (TC) and compared to 100% PLGA scaffolding spun onto coverslips as well as 80% PLGA/20% PCL spun onto coverslips. Cells were maintained in either EpiCM or SMCM for 14 days and evaluated for markers of epithelial cells (E-Cadherin), smooth muscle cells (aSMA), and glial cells (S100b). After 14 days of culture in EpiCM, significant numbers of

Fig. 2. Characterization of de-cellularized and synthetic scaffolds. High magnification SEM images of a de-cellularized rat esophagus (A) compared to the 100% PLGA and 80% PLGA/ 20% PCL electrospun scaffolds (B, C) illustrate a clear difference in fiber size with similar fiber patterns. SEM demonstrates an almost continuous layer of cells on 100% PLGA and 80% PLGA/20% PCL scaffolding (D, E) after 14 days in a HART bioreactor. Scale bars represent 10 mm in (A) and 20 mm in (B-E).

T. Jensen et al. / Biomaterials 57 (2015) 133e141

137

having the highest level of expression in EpiCM (Fig. 4C, F, I). Epithelial marker E-Cadherin was not highly expressed in the three culture conditions in SMCM as compared to aSMA, which was highly expressed in all three culture conditions in SMCM (Fig. 4 J, K, M, N P, Q). After five days in culture with SMCM, contractile cell clusters were observed (Supplemental Video 1), however, these clusters were not observed in EpiCM. S100b expression was minimal in all 3 culture conditions in SMCM (Fig. 4 L, O, R). Supplementary data related to this article can be found online at http://dx.doi.org/10.1016/j.biomaterials.2015.04.004. 3.4. Characterization of seeded or unseeded scaffolds following 14 days of implantation In vivo An esophageal patch was sutured into the native esophagus as depicted by Fig. 5A. De-cellularized esophageal scaffolds were implanted but were too fragile and resulted in leaks. Scaffolds that were composed of 100% PLGA were also implanted, but became rigid and presented challenges during implantation. The 20% PCL/ 80% PLGA scaffolds were qualitatively more flexible compared to the 100% PLGA scaffold, and combined with the improved cell viability results from the in vitro study (Fig. 3), the 20% PCL/80% PLGA scaffold was used for subsequent studies. The scaffolds and adjacent native esophagus was harvested and longitudinally sectioned (Fig. 5A). Hematoxylin and eosin staining of a longitudinal section of implanted scaffold is illustrated in Fig. 5B, with the implanted scaffold segment outlined by a red box. Hematoxylin and eosin staining of 100% PLGA scaffolds appears fibrous and heterogeneous, while 80% PLGA/20% PCL scaffolds appear denser with higher numbers of blood vessels (Fig. 5 CeF Arrows). Immunofluorescence staining for key markers was performed and high magnification images of the construct region of this large longitudinal section are shown in Fig. 6. E-cadherin expression was present in both seeded and unseeded 80% PLGA/20% PCL scaffolds, but was only observed in seeded 100% PLGA scaffolds (Fig. 6 A, D, G, J). aSMA signal was highest in the 100% PLGA scaffolds, with blood vessel-like structures being clearly outlined, while a diffuse positive signal was noted in the unseeded 80% PLGA/20% PCL scaffold following 14 days in vivo (Fig. 6 B, E H, K). Expression of glial cell marker S100b was low in seeded 80% PLGA/20% PCL scaffolds, while expression was practically absent in the other three implant conditions (Fig. 6C, F, I, L). Normal rat esophagus was stained with ECadherin, aSMA and S100b as a control and for comparison (Fig. 6 M, N, O). 4. Discussion

Fig. 3. . Evaluation of viability and proliferation on various scaffold ratios. Proliferation (Ki-67), Cell Density (Blue Signal) and Apoptosis (Cleaved Caspase 3) were evaluated on a variety of different scaffold combinations and compared to tissue culture treated coverslips in epithelial medium. Apoptosis was the highest at 40% PCL compared to the other combinations (D-L) and density of viable cells began to decrease in scaffolds greater than 20% PCL (D-L). Proliferation was highest at 10% PCL (E), but was still present in 20% PCL (F). In future experiments we chose the 20% PCL/80% PLGA combination because it supported proliferation with minimal apoptosis while providing scaffold integrity for implantation. Total Magnification 200X. Scale bars represent 100 mm.

cells expressed E-Cadherin on TCP as well as 100% PLGA and 20% PCL/80% PLGA (Fig. 4 A, D, G). Cells cultured in EpiCM did not express significant amounts of aSMA on TCP or 80% PLGA/20% PCL, however, low levels of expression was present in cells cultured on 100% PLGA (Fig. 4 B, E, H). Expression was present in all three culture conditions for S100b, with 80% PLGA/20% PCL

The goal of this study was to test the ability of electrospun scaffolds to support esophageal viability and phenotype. We utilized a dual chamber small animal hollow organ bioreactor to culture the esophageal cells on synthetic scaffolds. This enabled us to recapitulate an inner epithelial cell layer and an outer muscular layer. The use of a small animal model in this study enabled feasibility testing that can then be applied in the future to a larger animal model. Use of an autologous, synthetic esophageal scaffold could profoundly impact our ability to treat diseases such as esophageal atresia or caustic injury to the esophagus. Current treatments for long gap esophageal atresia in neonatals include tubes fashioned from the gastric fundus or small and large intestine [2,25e29]. These procedures are fraught with complications such as leakage and stenosis. New therapeutic options are therefore desperately needed. A recent publication describes the use of a de-cellularized esophagus as a basis to engineer new esophageal tissue for

138

T. Jensen et al. / Biomaterials 57 (2015) 133e141

Fig. 4. Characterization of Harvested Esophageal Cells Cultured on Tissue Culture Plastic (TC) (Control) compared to two types of Electrospun Scaffolds. Qualitatively, epithelial phenotype was optimally supported in EpiCM on 80% PLGA/20% PCL (A, D, G). The EpiCM still supported some muscle cell phenotype (B, E, H). In addition, the 80% PLGA/20% PCL qualitatively supported the most ganglion cell phenotype (C, F, I). In contrast, in SMCM, the 100% PLGA and the 80% PLGA/20% PCL qualitatively had the most smooth muscle actin (K, N, Q). In SMCM, the 80% PLGA/20% PCL and 100% PLGA qualitatively supported the most ganglion cell phenotype (L, O, R). Secondary antibody controls (SeU) demonstrate no nonspecific binding of the secondary antibodies used for analysis. 200X Magnification. Scale bars represent 100 mm.

T. Jensen et al. / Biomaterials 57 (2015) 133e141

139

Fig. 5. Histology of seeded and unseeded electrospun scaffolds after 14 days In vivo. Representative H&E sections of seeded and unseeded electrospun scaffolds after 14 days of implantation demonstrate feasibility of synthetic scaffold implant. Blood vessels are present in all scaffold conditions (arrows). 200X Magnification. e¼epithelium; m¼muscle; l¼lumen.

implantation [30], however, the donor availability for such neonatal applications as well as size compatibility are both major drawbacks to this approach. Many protocols that have been described for decellularizing donor esophageal tissue take an extended period of time to remove all the cellular contents and therefore cannot be immediately ready to accept new cells. We were able to decellularize 2 rat esophagi and re-seed them with esophageal cells and autologously transplant them into two animals (data not shown). We found the scaffolding to be extremely fragile resulting in perforation and failure necessitating euthanization. The use of artificial materials that can be sized for each individual patient is a more feasible approach to treat esophageal disease, injuries and defects. Materials such as Teflon and polypropylene have been tested [31], but can often lead to development of fibrosis. The materials we use in this study, poly (lactic-co-glycolic acid) (PLGA) and polycaprolactone (PCL), are FDA approved and utilized in sutures, prosthetic devices, patches and other medical devices without complications [32e36]. Even though both materials hydrolize and therefore degradable over time at different rates, it has been established in previous work that epithelial cells will generate their own ECM once they have reached confluence [37]. Therefore, it is reasonable to postulate that ECM can be secreted by the new host cells over time and will replace the material that degrades. Our original electrospun studies focused on the use of 100% PLGA; however, these scaffolds became brittle with recellularization, making them unsuitable for esophageal function. We tested various combinations of PLGA with PCL ranging from 60% PLGA/40% PCL to 90% PLGA/10% PCL. These scaffolds were much more flexible and maintained cellular viability. The combination with the highest cell density that still demonstrated some proliferation was observed on 80% PLGA/20% PCL, which was used in subsequent studies. The two polymers utilized in our three dimensional scaffold have been studied in detail and have very different degradation rates, which can affect the scaffold over longer periods of time. The PCL polymer used for our scaffolding had a very high molecular weight between 43,000e50,000 which does greatly increase the degradation time over time compared to PCL polymers with lower molecular weights between 15,000 and 20,000 [38]. Furthermore, the PLGA polymer utilized in our study had a PLA/PGA ratio of 85:15. The proportion of each type of acid does affect the degradation rate as well as the hydrophilic

properties of the scaffold. Previous studies describe that the greater the proportion of PLA compared to PGA will result in a slower degradation rate in the human body [39]. Although the degradation rate was not evaluated in this particular study, prolonged implantation of synthetic scaffolds is an ongoing investigation. Previous studies have described the mechanical properties of each polymer and when combined, how they will affect the overall scaffold's mechanical properties. The elastic modulus is 0.4 GPa with electrospun PCL [39], indicative of a more elastic and flexible material, similar to a rubber band. The elastic modulus is 5e6 times higher for electrospun PLGA (2.5 GPa), indicating it is much more rigid and inflexible. These mechanical values clearly explain why our 100% PLGA scaffolding was too brittle for implantation. This underscored the need for addition of PCL to the PLGA scaffold to increase elasticity and flexibility. Over time the combination scaffold should become more flexible as the PLGA degrades faster than PCL, which will complement the function of the esophagus. In addition, previous studies have shown that murine smooth muscle cells have improved viability over time on PCL compared to PLGA. Therefore the addition of PCL to our PLGA should help maintain viability in our scaffolds. Lastly, the inflammatory response generated by each type of polymer is important in promoting angiogenesis and nutrient delivery to cells. Previous work demonstrates a more robust inflammatory response to PCL than to PLGA with implantation in a rat model [38]. Therefore, in our experiments, the combination of PCL and PLGA polymers for esophageal scaffolding should contribute to improved viability with implantation compared to PLGA alone [38]. The optimal cell source to engineer an esophagus would come from the patient (i.e., be autologous) and would utilize both epithelial and muscle cells. When mucosal epithelial cells from oral biopsy [40] or esophageal organoid units following digestion of rat esophagi [15,41] were implanted, the scaffolds did not support morphogenesis. These investigators found that in order to create a functional esophageal replacement, both epithelial and muscle cells must be seeded prior to implantation. We adapted protocols from previous studies [15,41] to isolate native esophageal cells and cultured them on a synthetic scaffold in a bioreactor with commercial epithelial media injected into the lumen and commercial smooth muscle media bathing the abluminal surface. Each cell type was successfully enriched in their respective medium following 14 days in culture. We also observed contractile clusters in our smooth

140

T. Jensen et al. / Biomaterials 57 (2015) 133e141

Fig. 6. Characterization of seeded and unseeded electrospun scaffolds after 14 days In vivo. After 14 days of implantation, seeded scaffolds maintained expression of epithelial, smooth muscle and glial cell phenotypes (A-C, G-I). Normal rat esophagus was also stained for comparison (M-O). Secondary antibody controls demonstrate little non-specific binding of the secondary antibodies used for imaging (P-R). 200X Magnification.

muscle cell medium dishes after 5 days, which suggests that generation of a tissue-engineered esophagus that can undergo peristalsis is feasible. In addition, we found that only a small amount of tissue is needed to obtain an adequate numbers of cells for reseeding, making it feasible to endoscopically obtain autologous cells through biopsy in larger animal models and in human patients. Theoretically utilizing cells that are specific to the patient and are already differentiated to the preferred cell types (epithelial cells, smooth muscle cells and glial cells) will save time and efficiency as opposed to differentiating stem cells to the appropriate phenotype [42e44].

5. Conclusion Electrospun synthetic PLGA/PCL tubes show promise as potential scaffolds for esophageal tissue engineering. Morphologically the fiber size is larger than a de-cellularized esophagus however the orientation of the fibers is very similar. Autologous cells could be isolated from native esophageal tissue and the scaffolds were able to support viability and proliferation. Utilizing native esophageal tissue as a source to generate epithelial and smooth muscle cells for engineering circumvents potential issues with utilizing stem cells or other similar approaches. Furthermore this approach

T. Jensen et al. / Biomaterials 57 (2015) 133e141

allows for patient specificity and aims to generate an autologous graft, theoretically preventing rejection responses. In future studies, transgenic rats will be used to evaluate the contribution of the donor and host cells to incorporation of the engineered esophageal scaffold. Acknowledgments We acknowledge Harvard Apparatus Regenerative Technologies (HART) for providing us the small animal hollow organ bioreactor used in these studies. In addition HART provided us a CAD rendered image and technical support during our experiments. In addition, we would like to thank Xuanhao Sun and Marie Cantino at the Biosciences Electron Microscopy Facility of the University of Connecticut for assistance with SEM. Funds from NFF grant #1126100 were used to purchase the FEI Nova Nano SEM. In addition, we acknowledge the assistance of University of Connecticut undergraduates Asiyah Khan and Keval Vyas for their assistance in analyzing data for this manuscript. Lastly, we acknowledge the input and expertise in generating the electrospun scaffolding from Eric James and Dr. Lakshmi S. Nair from the Institute for Regenerative Engineering at UConn Health. Funding for this work was provided by Connecticut Children's Medical Center Strategic Research Funding. References [1] Clark DC. Esophageal atresia and tracheoesophageal fistula. Am Fam Physician 1999;59. 910e6, 9-20. [2] Spitz L. Gastric transposition for esophageal substitution in children. J Pediatr Surg 1992;27:252e7. discussion 7-9. [3] Freedman SF, White JA. Caustic ingestion. J La State Med Soc 1989;141:13e5. [4] Napier KJ, Scheerer M, Misra S. Esophageal cancer: a review of epidemiology, pathogenesis, staging workup and treatment modalities. World J Gastrointest Oncol 2014;6:112e20.  L, Morini F, Conforti A. Long-gap esophageal atresia: trac[5] Bagolan P, Valfre tion-growth and anastomosis - before and beyond. Dis Esophagus 2013;26: 372e9. [6] Chian KS, Leong MF, Kono K. Regenerative medicine for oesophageal reconstruction after cancer treatment. Lancet Oncol 2015;16:e84e92. [7] Gangemi A, Tzvetanov IG, Beatty E, Oberholzer J, Testa G, Sankary HN, et al. Lessons learned in pediatric small bowel and liver transplantation from livingrelated donors. Transplantation 2009;87:1027e30. [8] Kuppan P, Sethuraman S, Krishnan UM. Tissue engineering interventions for esophageal disordersepromises and challenges. Biotechnol Adv 2012;30: 1481e92. [9] Badylak SF, Taylor D, Uygun K. Whole-organ tissue engineering: decellularization and recellularization of three-dimensional matrix scaffolds. Annu Rev Biomed Eng 2011;13:27e53. [10] Gilbert TW, Sellaro TL, Badylak SF. Decellularization of tissues and organs. Biomaterials 2006;27:3675e83. [11] Ackbar R, Ainoedhofer H, Gugatschka M, Saxena AK. Decellularized ovine esophageal mucosa for esophageal tissue engineering. Technol Health Care 2012;20:215e23. [12] Keane TJ, Londono R, Carey RM, Carruthers CA, Reing JE, Dearth CL, et al. Preparation and characterization of a biologic scaffold from esophageal mucosa. Biomaterials 2013;34:6729e37. [13] Franco RA, Nguyen TH, Lee BT. Preparation and characterization of electrospun PCL/PLGA membranes and chitosan/gelatin hydrogels for skin bioengineering applications. J Mater Sci Mater Med 2011;22:2207e18. [14] Beckstead BL, Pan S, Bhrany AD, Bratt-Leal AM, Ratner BD, Giachelli CM. Esophageal epithelial cell interaction with synthetic and natural scaffolds for tissue engineering. Biomaterials 2005;26:6217e28. [15] Grikscheit T, Ochoa ER, Srinivasan A, Gaissert H, Vacanti JP. Tissue-engineered esophagus: experimental substitution by onlay patch or interposition. J Thorac Cardiovasc Surg 2003;126:537e44. [16] Chung EJ, Ju HW, Park HJ, Park CH. Three-layered scaffolds for artificial esophagus using poly(3-caprolactone) nanofibers and silk fibroin: an experimental study in a rat model. J Biomed Mater Res A 2014 Oct 8. http:// dx.doi.org/10.1002/jbm.a.35347 [Epub ahead of print].

141

[17] Chen L, Lv J, Yu X, Kang C, Zhu Y. Epithelium constitution for esophageal tissue engineering using electrospinning technology. Sheng Wu Gong Cheng Xue Bao 2013;29:1796e807. [18] Leong MF, Chian KS, Mhaisalkar PS, Ong WF, Ratner BD. Effect of electrospun poly(D,L-lactide) fibrous scaffold with nanoporous surface on attachment of porcine esophageal epithelial cells and protein adsorption. J Biomed Mater Res A 2009;89:1040e8. [19] Zhu Y, Leong MF, Ong WF, Chan-Park MB, Chian KS. Esophageal epithelium regeneration on fibronectin grafted poly(L-lactide-co-caprolactone) (PLLC) nanofiber scaffold. Biomaterials 2007;28:861e8. [20] Kolluru PV, Lipner J, Liu W, Xia Y, Thomopoulos S, Genin GM, et al. Strong and tough mineralized PLGA nanofibers for tendon-to-bone scaffolds. Acta Biomater 2013;9:9442e50. [21] Ozeki M, Narita Y, Kagami H, Ohmiya N, Itoh A, Hirooka Y, et al. Evaluation of decellularized esophagus as a scaffold for cultured esophageal epithelial cells. J Biomed Mater Res Part A 2006;79:771e8. [22] Levin DE, Sala FG, Barthel ER, Speer AL, Hou X, Torashima Y, et al. A “living bioreactor” for the production of tissue-engineered small intestine. Methods Mol Biol 2013;1001:299e309. [23] Trout L, Gatzy JT, Ballard ST. Acetylcholine-induced liquid secretion by bronchial epithelium: role of Cle and HCO3e transport. Am J Physiol 1998;275:L1095e9. [24] Martens CJ, Ballard ST. Effects of secretagogues on net and unidirectional liquid fluxes across porcine bronchial airways. Am J Physiol Lung Cell Mol Physiol 2010;298:L270e6. [25] Bax NM, Van Renterghem KM. Ileal pedicle grafting for esophageal replacement in children. Pediatr Surg Int 2005;21:369e72. cs L, Cseke L. Complications following esophageal surgery. [26] Horv ath OP, Luka Recent Results Cancer Res 2000;155:161e73. [27] Yamataka A, Miyano T, Segawa O, Nozawa M. Tubed latissimus dorsi musculocutaneous flaps for thoracic esophageal replacement: an experimental study. J Pediatr Surg 1994;29:590e3. [28] Dantas RO, Aprile LR, Aben-Athar CG, Miranda AL. Esophageal striated muscle contractions in patients with Chagas' disease and idiopathic achalasia. Braz J Med Biol Res 2002;35:677e83. [29] Dantas RO, Aprile LR. Esophageal striated muscle contractions in patients with gastroesophageal reflux symptoms. Dig Dis Sci 2002;47:2586e90. €qvist S, Jungebluth P, Lim ML, Haag JC, Gustafsson Y, Lemon G, et al. [30] Sjo Experimental orthotopic transplantation of a tissue-engineered oesophagus in rats. Nat Commun 2014;5:3562. [31] Smith G. Mesh repairs in hiatal surgery. The case for mesh repairs in hiatal surgery. Ann R Coll Surg Engl 2007;89:481e3. [32] Lü JM, Wang X, Marin-Muller C, Wang H, Lin PH, Yao Q, et al. Current advances in research and clinical applications of PLGA-based nanotechnology. Expert Rev Mol Diagn 2009;9:325e41. [33] Kai D, Liow SS, Loh XJ. Biodegradable polymers for electrospinning: towards biomedical applications. Mater Sci Eng C Mater Biol Appl 2014;45:659e70. [34] Ulery BD, Nair LS, Laurencin CT. Biomedical applications of biodegradable polymers. J Polym Sci B Polym Phys 2011;49:832e64. [35] Hiep NT, Lee BT. Electro-spinning of PLGA/PCL blends for tissue engineering and their biocompatibility. J Mater Sci Mater Med 2010;21:1969e78. [36] Pham QP, Sharma U, Mikos AG. Electrospinning of polymeric nanofibers for tissue engineering applications: a review. Tissue Eng 2006;12:1197e211. [37] Chan BP, Leong KW. Scaffolding in tissue engineering: general approaches and tissue-specific considerations. Eur Spine J 2008;17(Suppl. 4):467e79. [38] Sung HJ, Meredith C, Johnson C, Galis ZS. The effect of scaffold degradation rate on three-dimensional cell growth and angiogenesis. Biomaterials 2004;25:5735e42. [39] Gunatillake PA, Adhikari R. Biodegradable synthetic polymers for tissue engineering. Eur Cell Mater 2003;5:1e16. discussion. [40] Wei RQ, Tan B, Tan MY, Luo JC, Deng L, Chen XH, et al. Grafts of porcine small intestinal submucosa with cultured autologous oral mucosal epithelial cells for esophageal repair in a canine model. Exp Biol Med 2009;234:453e61. [41] Spurrier RG, Speer AL, Hou X, El-Nachef WN, Grikscheit TC. Murine and human tissue-engineered esophagus form from sufficient stem/progenitor cells and do not require microdesigned biomaterials. Tissue Eng Part A 2015;21: 906e15. [42] DeWard AD, Cramer J, Lagasse E. Cellular heterogeneity in the mouse esophagus implicates the presence of a nonquiescent epithelial stem cell population. Cell Rep 2014;9:701e11. [43] Kalabis J, Oyama K, Okawa T, Nakagawa H, Michaylira CZ, Stairs DB, et al. A subpopulation of mouse esophageal basal cells has properties of stem cells with the capacity for self-renewal and lineage specification. J Clin Invest 2008;118:3860e9. [44] Kantarcioglu M, Caliskan B, Demirci H, Karacalioglu O, Kekilli M, Polat Z, et al. The efficacy of mesenchymal stem cell transplantation in caustic esophagus injury: an experimental study. Stem Cells Int 2014;2014:939674.

Biomimetic and synthetic esophageal tissue engineering.

A tissue-engineered esophagus offers an alternative for the treatment of pediatric patients suffering from severe esophageal malformations, caustic in...
3MB Sizes 1 Downloads 13 Views