- Email: [email protected]
PLLA scaffolds via subcritical CO2 augments the production of tissue engineered intestine
Biomaterials 103 (2016) 150e159
Contents lists available at ScienceDirect
Biomaterials journal homepage: www.elsevier.com/locate/biomaterials
HB-EGF embedded in PGA/PLLA scaffolds via subcritical CO2 augments the production of tissue engineered intestine Yanchun Liu a, Tyler Nelson b, Barrett Cromeens a, Terrence Rager a, John Lannutti c, Jed Johnson d, Gail E. Besner a, * a
Center for Perinatal Research, The Research Institute at Nationwide Children's Hospital, and the Department of Pediatric Surgery, Columbus, OH, USA Department of Biomedical Engineering, The Ohio State University, Columbus, OH, USA Department of Materials Science and Engineering, The Ohio State University, Columbus, OH, USA d Nanofiber Solutions, Inc., Columbus, OH, USA b c
a r t i c l e i n f o
a b s t r a c t
Article history: Received 8 February 2016 Received in revised form 10 May 2016 Accepted 17 June 2016 Available online 21 June 2016
The ability to deliver sustained-release, biologically active growth factors through custom designed tissue engineering scaffolds at sites of tissue regeneration offers great therapeutic opportunity. Due to the short in vivo half-lives of most growth factors, it is challenging to deliver these proteins to sites of interest where they may be used before being degraded. The application of subcritical CO2 uses gasphase CO2 at subcritical pressures ranging from 41 to 62 bar (595e913 PSI) which avoids foaming by reducing the amount of CO2 dissolved in the polymer and maintains completely reversible plasticization. In the current study, heparin-binding EGF-like growth factor (HB-EGF) was embedded into polyglycolic acid (PGA)/Poly-L-latic acid (PLLA) scaffolds via subcritical CO2 exposure for the production of tissue engineered intestine (TEI). PGA fiber morphology after subcritical CO2 exposure was examined by scanning electron microscopy (SEM) and the distribution of HB-EGF embedded in the scaffold fibers was detected by HB-EGF immunofluorescent staining. In vivo implantation of HB-EGF-embedded scaffolds confirmed significantly improved TEI structure as a result of local delivery of the trophic growth factor. These findings may be critical for the production of TEI in the treatment of patients with short bowel syndrome in the future. © 2016 Elsevier Ltd. All rights reserved.
Keywords: Polyglycolic acid Stem cell HB-EGF Intestine Tissue engineering
1. Introduction Patients with short bowel syndrome (SBS) lack the adequate intestinal function needed to sustain their nutritional needs. The production of tissue-engineered intestine (TEI) using a patient's own intestinal cells may restore functional intestinal absorptive area while avoiding the complications of current therapeutic options such as small bowel transplantation. The smallest transplantable mucosal units are intestinal “organoids” which are clusters of 20e40 cells isolated from intestinal mucosa. These organoids contain the putative intestinal stem cells (ISCs) and surrounding cells that comprise the stem cell niche. Transplantation of these intestinal organoids has been shown to generate TEI that resembles native intestine in both structure and function with variable results [1e5].
* Corresponding author. Department of Pediatric Surgery, ED383, Nationwide Children's Hospital, 700 Children's Drive, Columbus, OH 43205, USA. E-mail address: [email protected] (G.E. Besner). http://dx.doi.org/10.1016/j.biomaterials.2016.06.039 0142-9612/© 2016 Elsevier Ltd. All rights reserved.
In the 1980s, Thompson et al. first described the augmentation of neomucosal growth after exposure to luminal factors or to systemically administered urogastrone, a peptide that shares an intestinal receptor with epidermal growth factor (EGF) [6,7]. Based on the concept that intestinal growth is regulated by nutrients and enteric secretions, as well as local and systemic growth factors, subsequent studies compared luminal versus systemic administration of trophic peptides. The route of growth factor delivery has been shown to impact mucosal growth, demonstrating that local delivery is optimal [8e12]. This work led to studies of the administration of growth factors to augment the production of TEI [13e16]. Very few studies have examined the effects of locally delivered growth factors on tissue-engineered neomucosa. The application of three-dimensional, porous polymer constructs for tissue engineering applications has received considerable attention as a method of aiding and defining new tissue growth [17,18]. Due to the short in vivo half-lives of most growth factors, it is desirable to deliver the protein to the site of interest where it may be used before it is degraded [19]. For example, one successful approach to
Y. Liu et al. / Biomaterials 103 (2016) 150e159
deliver VEGF within a polymer implant was to produce a porous PLGA matrix with dispersed alginate microspheres [20]. The growth factor was encapsulated within the alginate spheres and was released in an active state upon exposure of the construct to an aqueous buffer. Another mechanism used to release growth factors at the desired site was to encapsulate the protein within the PLGA matrix itself [20]. Supercritical CO2 fluid mixing technology is performed at or below physiological temperatures without any harsh solvents, allowing bioactive factors to be encapsulated and retained in biodegradable polymer constructs without compromising their biological activity [21]. The use of CO2 avoids concerns surrounding halogenated solvents that are often necessary to achieve mixtures of resorbable polymers with other compounds [22e24]. However, the use of supercritical CO2 with bulk polymeric implants carries a substantial penalty: foaming, or the loss of the original implant form [25]. Therefore, an alternative treatment that preserves form is based on the use of gas-phase CO2 at subcritical pressures ranging from 41 to 62 bar (595e913 pounds per square inch; PSI). This avoids foaming by reducing the amount of CO2 dissolved in the polymer and maintains conditions that allow plasticization to be completely reversible. Subcritical CO2 has been used successfully to embed the anti-cancer drug paclitaxel into polylactic acid with proven sustained chemotherapeutic activity [26]. Dormer et al. reported the use of subcritical CO2 with sinter microspheres for the delivery of transforming growth factor (TGF)-beta3 and bone morphogenetic protein (BMP)-2 to human bone marrow stromal cells in vitro, and found that subcritical CO2 had only slight adverse effects, as well as some desirable effects, on protein availability and bioactivity [27]. In the current study, heparin-binding EGF-like growth factor (HB-EGF) was embedded into polyglycolic acid (PGA)/Poly-L-latic acid (PLLA) scaffolds via subcritical CO2 exposure in an effort to augment the growth of TEI. We have previously shown that HB-EGF promotes intestinal epithelial cell (IEC) proliferation and migration [28e30], decreases IEC apoptosis [31,32], and protects ISC from injury [33], supporting the use of HB-EFG in the current studies. The release kinetics and bioactivity of HB-EGF released from PGA/PLLA scaffolds were characterized, and the effects of embedding scaffolds with HB-EGF were tested in in vitro cell cultures and in vivo scaffold implantation. 2. Materials and methods 2.1. Materials for scaffold preparation PGA Biofelt (2 mm thickness and 60 mg/cm3 density) was purchased from Biomedical Structures (Warwick, RI). Poly-L-latic acid (PLLA) and fluorescein isothiocyanate (FITC) conjugated bovine serum albumin (BSA) were from Sigma-Aldrich (St. Louis, MO). Chloroform and collagen type I were from Fisher Scientific (Pittsburgh, PA). HB-EGF was from R&D Systems (Minneapolis, MN). 2.2. Preparation of scaffolds Scaffold preparation is depicted in Fig. 1. Flat scaffolds (0.8 1.6 cm) were cut directly from PGA Biofelt and used for scanning electron microscopy (SEM), in vitro cell culture, and measurement of surface area as depicted by FITC-BSA distribution. Tubular scaffolds (0.5 3.0 cm) were prepared by wrapping PGA flat sheets around stainless steel mandrels for suture retention strength (SRS) tests. Similarly, tubular scaffolds (0.5 1.0 cm) were prepared for in vitro release kinetics, bioactivity, compression tests, and in vivo implantation. All flat and tubular scaffolds were coated with 5% PLLA in chloroform. Once the solvent was completely
PGA Biofelt
151
Untreated plain PGA (PGA)
Coat with PLLA (5%)
Wash with 100% ethanol
Coat with collagen
Coat with HB-EGF
Subcritical CO2
Group 1 (PGA/PLLA) Group 2 (PGA/PLLA/HB-EGF) Group 3 (PGA/PLLA/HB-EGF/CO2)
Fig. 1. Scaffold preparation for in vitro and in vivo studies. Three groups of scaffolds were fabricated: PGA coated with PLLA (PGA/PLLA), PGA coated with PLLA and embedded with HB-EGF (PGA/PLLA/HB-EGF), and PGA coated with PLLA and embedded with HB-EGF followed by subcritical CO2 exposure (PGA/PLLA/HB-EGF/CO2). In some in vitro studies, untreated plain PGA scaffolds were used as a control.
evaporated, scaffolds were soaked in 100% ethanol for 30 min in order to improve the hydrophilicity of PGA/PLLA scaffolds (Video clip 1 is available online), followed by 3 washes with PBS. Scaffolds were then soaked in 0.4 mg/ml collagen type I for 30 min followed by 3 washes with PBS, lyophilized, and sterilized with ethylene oxide. All flat and tubular scaffolds were divided into three groups based on the following treatments: Supplementary video related to this article can be found at http://dx.doi.org/10.1016/j.biomaterials.2016.06.039. 1) PGA coated with PLLA (PGA/PLLA), 2) PGA coated with PLLA and embedded with HB-EGF (PGA/PLLA/HB-EGF), and 3) PGA coated with PLLA and embedded with HB-EGF followed by subcritical CO2 exposure (PGA/PLLA/HB-EGF/CO2). Scaffolds from the second and third groups were hydrated with 10 mg per scaffold of HB-EGF in 100 ml of PBS for in vitro studies and 2 mg per scaffold of HB-EGF in 100 ml of PBS for in vivo studies. Additionally, scaffolds from the third group were placed in sterile 50 ml conical bottom tubes with porous caps and exposed to subcritical CO2 in a stainless steel vessel (Parr Instruments Co., Moline, IL) at 900 PSI for 1 h followed by a slow release of CO2 at 15 PSI per minute for 60 min. All prepared scaffolds were stored at 30 C and brought to RT 10e20 min prior to use. 2.3. Scanning electron microscopy (SEM) and confocal microscopy Flat scaffolds from the three groups, in addition to plain PGA scaffolds, were cut to 2 5 mm strips and 4 strips from each group were affixed to a stub and sputter-coated with gold (Emitech K550X, Quorum Technologies Ltd, Ashford, Kent, England). Samples were examined using a scanning electron microscope (Hitachi S4800, Hitachi High Technologies America, Inc., Dallas, TX) at a voltage of 10e20 kV. To visualize the distribution of HB-EGF in the scaffolds, HB-EGF-embedded scaffolds were incubated in 10 mg/ml of goat derived anti-HB-EGF primary antibody (R&D Systems) for 1 h followed by 3 washes with PBS and incubation with donkey anti-goat IgG secondary antibody for 1 h (Invitrogen, Grand Island, NY). After 3 washes with PBS, the scaffolds were examined with a confocal microscope (Zeiss LSM 710, Thornwood, New York).
152
Y. Liu et al. / Biomaterials 103 (2016) 150e159
2.4. Suture retention strength (SRS) and compression tests
The maximum force is the ultimate compressive force.
A Test Resources load frame (Model # SM-50-294-Capacity 50 lb, Northbrook, IL) was used for both tests. The preparation of the samples and the settings of the tests are as previously described, and shown in Fig. 3AeC [34e37]. Four scaffolds from each of the three groups, together with plain PGA scaffolds, were tested. For SRS tests, load was applied at 50 mm/min until failure, as determined by the suture pulling through the wall of the scaffold, and the stressestrain curve recorded. The maximum force required is the SRS. For compression tests, load was applied at 10 mm/min for a total of 2 mm of compression and the stress-strain curve recorded.
2.5. HB-EGF release kinetics and bioactivity Tubular PGA/PLLA/HB-EGF scaffolds (group 2, n ¼ 6) and PGA/ PLLA/HB-EGF/CO2 scaffolds (group 3, n ¼ 6) were individually placed into 2 ml round bottom cryovials (Fisher Scientific, Pittsburgh, PA) prefilled with 1.5 ml of PBS. The tubes were placed on an orbital shaker in a 37 C incubator. On days 1, 3, 5, 7, 9, 11, and 13, the medium from each tube was removed and stored at 4 C. Fresh PBS (1.5 ml) was added to each original tube after medium collection and the tubes were returned to the shaker. The release
Fig. 2. Scanning electron microscopy (SEM) and immunofluorescent staining of HB-EGF on scaffolds. (A) Original PGA biofelt; (B) PGA/PLLA scaffold; (C) PGA/PLLA/HB-EGF scaffold after lyophilization; (D) PGA/PLLA/HB-EGF/CO2 scaffold; (E) magnified view of indicated area of panel C; (F) magnified view of indicated area of panel D; (G) HB-EGF immunofluorescent staining of a PGA/PLLA/HB-EGF scaffold after lyophilization demonstrating the presence and distribution of HB-EGF (green) and the PGA fibers identified by their auto-fluorescence (blue); (H) HB-EGF immunofluorescent staining of a PGA/PLLA/HB-EGF/CO2 scaffold. Scale bar: 40 mm for A-D, 5 mm for E and F, 30 mm for G and H.
Y. Liu et al. / Biomaterials 103 (2016) 150e159
153
Fig. 3. Suture retention strength (SRS) and compression tests. (A) Preparation of samples for SRS test. (B) Example of SRS test. (C) Example of compression test. (D) SRS results demonstrating that PGA/PLLA scaffolds have significantly higher suture retention strength than plain PGA scaffolds. Embedding of PGA/PLLA scaffolds with HB-EGF in the presence or absence of CO2 exposure did not lead to additional changes in the strength of the scaffolds. (E) Compression test results. PGA/PLLA scaffolds have significantly higher ultimate compressive force than plain PGA scaffolds. Embedding of PGA/PLLA scaffolds with HB-EGF in the presence or absence of CO2 exposure did not lead to additional changes in the compressibility of the scaffolds. * indicates statistically significant difference (p < 0.001).
buffer from the other time points was harvested as described above. The concentration of HB-EGF in the eluted buffers was measured using an HB-EGF ELISA assay (R&D Systems, Minneapolis, MN). The amount of HB-EGF released from each condition at each time point was averaged from the 6 samples and plotted over time. To test the bioactivity of released HB-EGF, three additional samples from groups 2 and 3 were prepared at each of three time points (days 1, 7, and 13) using Assay Medium (DMEM containing 0.1% BSA, 0.001% heparin, and 1% insulin-transferrin-selenium-G supplement) (Gibco, Grand Island, NY) as elution buffer. The concentration of HB-EGF was determined by HB-EGF ELISA assay and diluted by 3 fold serial dilution starting from 1000 ng/ml. The diluted HB-EGF was added to a 96-well plate in duplicate. An HBEGF stock solution was diluted and used to produce a standard curve. BALB/c 3T3 mouse fibroblast cells (ATCC, Manassas, VA) were expanded in Growth Medium (DMEM containing 10% bovine calf serum) (Hyclone, Logan, UT) and re-suspended to 1.5 105 cells/ml in Assay Medium. 50 ml of cell suspension was added to the wells of a 96-well plate preloaded with the above samples. After a 72 h incubation at 37 C and 5% CO2, 20 ml of diluted bromodeoxyuridine (BrdU) solution was added followed by an additional 4 h of culture. Proliferating cells were quantified using a BrdU cell proliferation assay kit purchased from Millipore (Billerica, MA). The concentration of HB-EGF from each condition (without or with subcritical CO2 exposure) was calculated using the HB-EGF standard curve. The half maximal effective concentration (EC50) was calculated using 5-parameter asymmetric fit using GraphPad Prism 6 Version 6.05 (GraphPad software, Inc., San Diego, CA) and averaged from the triplicate samples.
2.6. Measurement of BSA distribution in the scaffolds To demonstrate that PLLA coating increases surface area for protein delivery, FITC-BSA was used to simulate HB-EGF-coating of the scaffolds due to the visibility of FITC-BSA by fluorescence microscopy. A solution of 1% FITC-BSA in 10 mM Tris-base was prepared and 100 ml of the solution was pipetted onto 0.8 1.6 cm flat plain PGA and PGA/PLLA scaffolds (n ¼ 3 for each). Scaffolds were examined by confocal microscopy right after lyophilization without the need for washing steps, therefore the distribution of FITC-BSA in the scaffolds accurately represents the amount of protein that scaffolds can accommodate. Z-stack images were captured with a 20 objective lens for a total of 50 intervals at 1 mm per interval with a confocal microscope. Surface areas with BSA were measured in
batch using ImageJ software after the threshold was determined and applied to all images measured.
2.7. In vitro culture of crypt-seeded PGA scaffolds Pregnant adult Lewis rats were purchased from Charles River Laboratories (Ohio, Spencerville) and used to provide pups as cell donors. All procedures were approved by the IACUC of the Research Institute at Nationwide Children's Hospital (protocol #AR1200001). Briefly, 1e3 day old rat pups were euthanized and under sterile conditions the intestines were harvested for isolating ISCcontaining crypts as we have described previously [38]. Intestines were harvested from the ligament of Treitz to the ileocecal junction. After the lumen was opened lengthwise, the intestines were pooled and washed vigorously with Hank's buffered salt solution (HBSS) containing 100 units/ml of penicillin, 100 mg/ml of streptomycin, 0.25 mg/ml of amphotericin B, and 25 mg/ml gentamicin. The intestines were minced into ~5e7 mm pieces and digested in 300 units/ml of collagenase type I and 0.1 g/ml of dispase (both from Worthington Biochemical Corporation, Lakewood, NJ) in serumfree DMEM for 30 min at 37 C with gentle shaking. The digestion was stopped by the addition of 10% fetal bovine serum in DMEM and the suspension (containing a mixed population of intact crypts and villi, fragments of partially digested intestinal tissue, and single cells) was washed with DMEM. The mixed cells were serially filtered through 200 mm, 70 mm and 25 mm sieve membranes. The cells in the 25e70 mm fraction were mainly ISC-containing crypts and were suspended in stem cell culture medium (DMEM containing 10% mesenchymal stem cell-qualified FBS (ThermoFisher, Waltham, MA) at 106 crypts/ml. 100 ml of cell suspension was added onto each flat scaffold (n ¼ 12 scaffolds per group). Each cell-seeded scaffold was placed into an individual well of a 24-well plate and incubated for 1 h. Each well then received stem cell culture medium (1.5 ml/well) for 12 h of static culture followed by dynamic culture with gentle shaking on a shaker. Medium was changed every 2 days. On days 0, 2 and 4, 150 ml of resazurin was added to each well. After an additional 12 h of culture, cell-seeded scaffolds were removed and used for fluorescein diacetate (F-DA) staining. The medium remaining in the well with resazurin was used for quantification of cells on the scaffolds. Briefly, 150 ml of cell culture medium with resazurin was transferred to another 96-well plate in triplicate and relative fluorescence units (RFU) were read with a plate reader. RFU values were proportional to the number of cells on the scaffolds, and were compared with Two-way ANOVA using
154
Y. Liu et al. / Biomaterials 103 (2016) 150e159
GraphPad Prism 6. To visualize cell viability and density, cellseeded scaffolds were rinsed with PBS twice, stained with F-DA (20 mg/ml in PBS) for 5 min, and rinsed with PBS [39]. The density of viable cells (green) was observed with confocal microscopy. 2.8. Production of tissue engineered intestine in vivo All procedures were approved by the IACUC of the Research Institute at Nationwide Children's Hospital (protocol #AR1200001). One to 3 day old rat pups were used as cell donors with the dam (200e250 g and 2e3 month old) of the pups used as recipients. The yield of intestinal cells from each litter of pups (8e10 pups) was used to seed multiple scaffolds, and the dam accommodated multiple scaffolds for side-by-side comparisons. Small intestines harvested from the same litter of pups were pooled, digested, and filtered as described above to concentrate ISCcontaining crypts that were suspended in stem cell culture medium (106 crypts/ml). Next, 200 ml of cell suspension was seeded onto each tubular scaffold (n ¼ 12 scaffolds per group). Each of 12 recipient dams received three cell-seeded scaffolds (one from each group). Under general anesthesia with 2% isoflurane in oxygen, a midline laparotomy was performed on recipient animals. A cellseeded scaffold from group 1 (PGA/PLLA) and group 2 (PGA/PLLA/ HB-EGF) were attached to the anterior surface of the left abdominal wall and a cell-seeded scaffold from group 3 (PGA/PLLA/HB-EGF/ CO2) was attached to the anterior surface of the right abdominal wall, using 7/0 polypropylene suture. Incisions were closed in layers with 5/0 polydioxanone (PDO) monofilament suture. Scaffolds were explanted after 4 weeks of in vivo incubation, fixed in 10% neutral buffered formalin, embedded in paraffin, cut into 5 mm sections, and stained with periodic acid Schiff (PAS). Immunofluorescent staining was conducted to characterize the epithelial cell lineages of TEI using the following primary antibodies: lysozyme (LYSO) (Abcam, Cambridge, MA) to identify Paneth cells, chromogranin A (CHRG) to identify enteroendocrine cells (Bioss Inc., Woburn, MA), villin (Santa Cruz, Dallas, Texas) to identify absorptive enterocytes, Ki67 (Millipore, Billerica, MA) to identify proliferating cells, and desmin (Bioss Inc., Woburn, MA) and alpha smooth muscle actin (a-SMA) (Abcam, Cambridge, MA) to identify intestinal subepithelial myofibroblasts (ISEMFs) and smooth muscle cells [4,40]. All secondary antibodies (goat anti rabbit, goat anti mouse, and donkey anti-goat) were purchased from Invitrogen (Grand Island, NY). Villous height was measured using ImageJ software (version 1.46r, Wayne Rasband, National Institutes of Health, USA) and compared across the groups with one-way ANOVA using Graphpad Prism 6.
superstructure of the scaffolds. 3.2. Suture retention strength (SRS) and compression tests The SRS force of PGA/PLLA scaffolds was significantly higher than that of plain PGA scaffolds, with addition of HB-EGF with or without subcritical CO2 exposure resulting in no change to the SRS forces (Fig. 3D). Similar results were observed in compression tests (Fig. 3E). 3.3. HB-EGF release kinetics and bioactivity HB-EGF was initially released in a bolus fashion from PGA/PLLA/ HB-EGF scaffolds (Fig. 4A). In the first day, ~49% of the initial 10 mg of HB-EGF was released, followed by a slow release of an additional ~26% of HB-EGF at the remaining time points, leading to a total release of ~75.5% over 13 days. On the other hand, only ~23.7% of HB-EGF was released in the first day from PGA/PLLA/HB-EGF/CO2 scaffolds, followed by a constant release of ~10% at each later time point until day 13. The bioactivity of HB-EGF released from the scaffolds as determined by a BrdU cell proliferation assay revealed
3. Results 3.1. Embedding PGA scaffolds with HB-EGF via subcritical CO2 exposure SEM of a plain PGA scaffold is shown in Fig. 2A, demonstrating clean individual PGA fibers. In PGA/PLLA scaffolds, the PLLA bridges the PGA fibers, leading to two networks: large pores defined by a network of PGA fibers and small pores defined by a honey-comblike network of PLLA (Fig. 2B). PGA/PLLA/HB-EGF scaffolds demonstrate loosely attached flakes on the PGA fibers (Fig. 2C and E). Confirmation that these flakes represent attached HB-EGF was obtained by HB-EGF immunofluorescent staining (Fig. 2G). In PGA/ PLLA/HB-EGF/CO2 scaffolds, HB-EGF forms small crystals on the surface of the PGA fibers that are partially embedded into the PGA fibers (Fig. 2D and F), with the distribution of HB-EGF again shown by immunofluorescent staining (Fig. 2H). Compared to untreated plain PGA, subcritical CO2 exposure did not deform the fibers or the
Fig. 4. HB-EGF release kinetics and bioactivity. (A) Release of HB-EGF from scaffolds over a 13 day period. In the absence of subcritical CO2 exposure, an initial large bolus release of HB-EGF occurred, followed by a slow release at the remaining time points. After subcritical CO2 exposure, a more constant release of HB-EGF occurred over the 13 day time period. (B) Bio-activity of HB-EGF released from the scaffolds was examined using a BrdU cell proliferation assay. No significant differences were observed in the half maximal effective concentration (EC50) of released HB-EGF in the presence or absence of subcritical CO2 exposure compared to HB-EGF control at 1, 7 or 13 days after loading onto PGA scaffolds.
Y. Liu et al. / Biomaterials 103 (2016) 150e159
no significant differences in EC50 from released HB-EGF in the presence or absence of subcritical CO2 exposure compared to HBEGF control over 13 days (Fig. 4B). 3.4. Measurement of BSA distribution in the scaffolds The distribution of FITC-BSA on plain PGA scaffolds was limited to the surface of the PGA fibers, whereas the distribution of FITCBSA on PGA/PLLA scaffolds extended to the PLLA bridges between the PGA fibers (Fig. 5A and B). There was a significant increase in surface area occupied by FITC-BSA in PGA/PLLA scaffolds compared to plain PGA scaffolds (Fig. 5C). 3.5. In vitro culture of crypt-seeded PGA scaffolds The density of viable cells was examined in freshly isolated crypts seeded onto PGA/PLLA, PGA/PLLA/HB-EGF, and PGA/PLLA/ HB-EGF/CO2 scaffolds, as demonstrated by F-DA staining. On day 1, the density of viable cells was similar in the three different scaffolds (Fig. 6AeC). On day 3, the density of viable cells was slightly higher on PGA/PLLA/HB-EGF and PGA/PLLA/HB-EGF/CO2 scaffolds compared to PGA/PLLA scaffolds (Fig. 6DeF). By day 5, the density of viable cells on PGA/PLLA/HB-EGF/CO2 scaffolds was markedly increased compared to that on PGA/PLLA/HB-EGF scaffolds, with both being higher than that on PGA/PLLA scaffolds (Fig. 6GeI). Quantification of viable cell numbers using resazurin assay confirmed these findings (Fig. 6J). 3.6. Production of tissue engineered intestine in vivo Freshly isolated crypts were seeded onto PGA/PLLA, PGA/PLLA/ HB-EGF, PGA/PLLA/HB-EGF/CO2 scaffolds, and implanted into the peritoneal cavity of recipient rats for in vivo production of TEI. After 4 weeks of in vivo incubation, PGA/PLLA scaffolds formed a neomucosa with shorter villi and fewer crypts compared to that of PGA/ PLLA/HB-EGF scaffolds (Fig. 7B and C). TEI produced on PGA/PLLA/ HB-EGF/CO2 scaffolds had markedly longer villi and denser crypts (Fig. 7D). The neomucosa produced from all three different scaffolds demonstrated the presence of goblet cells as determined by positive PAS staining, as seen in native rat intestine (Fig. 7A). Quantification of villous height confirmed that the neomucosa produced on PGA/PLLA scaffolds had the shortest villi, while neomucosa produced on PGA/PLLA/HB-EGF/CO2 scaffolds had the longest villi, even exceeding that of native rat intestine (Fig. 7E). Importantly, TEI produced from all three groups contained all elements of native small intestine, including Paneth cells as
155
demonstrated by lysozyme staining (Fig. 8A), enteroendocrine cells as demonstrated by chromogranin A staining (Fig. 8B), absorptive enterocytes as demonstrated by villin staining (Fig. 8C), smooth muscle cells as demonstrated by desmin/a-SMA double staining (Fig. 8D), intestinal subepithelial myofibroblasts (ISEMFs) as demonstrated by positive a-SMA and negative desmin staining (insert on Fig. 8D), and actively proliferating crypt cells as demonstrated by Ki67 staining (Fig. 8E).
4. Discussion Many tissue engineering applications employ a considerable amount of regenerative efficacy from physicochemical signals embedded within scaffolds. These signals can be natural or material-based, including compounds such as collagens, minerals, synthetic substrates, or small molecules [41,42]. PGA and PLLA have received much attention for scaffold production, and these materials have FDA approval for use in vivo. The mechanical and chemical properties of these polymers are desirable as tissue engineering scaffolds because they can be processed to form porous matrices, have excellent mechanical strength [18,43,44], and are relatively biocompatible [45]. The porosity of the material enables nutrient delivery to developing tissue within the construct and provides a high surface area for new cell growth. In addition, suitable porosity allows for cell penetration of existing biological tissue into the construct, a key factor for acceptance of the implant [46]. The ability to continuously deliver biologically active growth factors to sites of tissue regeneration using custom designed scaffolds offers great therapeutic opportunity. In the current studies, HB-EGF was chosen to be embedded in PGA/PLLA scaffolds due to its multiple proven benefits in the stimulation of intestinal development and in protection of the intestines from injury. HB-EGF is produced and secreted by many different cell types. Although the HB-EGF gene is widely expressed [47], the basal level of its mRNA seems to be relatively low in normal cells [48]. Expression of HBEGF is significantly increased in response to tissue damage [49], hypoxia [50,51] and oxidative stress [52,53], and also during wound healing and regeneration [54e56]. It exerts its mitogenic effects by binding to and activation of EGF receptor subtypes ErbB-1 and ErbB-4 [57]. While the mitogenic function of HB-EGF is mediated through activation of ErbB-1, its migration-inducing function involves the activation of ErbB-4 and N-arginine dibasic convertase (NRDc, Nardilysin), the latter representing a completely HB-EGFspecific receptor [58]. We have demonstrated that administration of HB-EGF improves restitution and reduces apoptosis in intestinal
Fig. 5. Distribution of FITC-BSA on scaffolds. (A) FITC conjugated BSA (green) was located on the surface of PGA fibers in plain PGA scaffolds. (B) FITC-BSA was located on the surface of PGA fibers and also on the PLLA bridges between PGA fibers in PGA/PLLA scaffolds. The PGA fibers are identified by their auto-fluorescence (blue). Scale bar ¼ 100 mm. (C) Surface area was measured in batch using ImageJ software, and averaged from 50 slices of each z-stack capture. *p < 0.001.
156
Y. Liu et al. / Biomaterials 103 (2016) 150e159
Fig. 6. In vitro culture of crypts on PGA scaffolds. Crypts were seeded onto scaffolds and incubated in vitro for 5 days. The density of viable cells was determined by fluorescein diacetate staining (green). The PGA fibers are demonstrated by their auto-florescence (blue). (A, D, and G) crypts cultured on PGA/PLLA scaffolds at 1, 3 or 5 days respectively; (B, E and H) crypts cultured on PGA/PLLA/HB-EGF scaffolds at 1, 3 or 5 days respectively; (C, F and I) crypts cultured on PGA/PLLA/HB-EGF/CO2 scaffolds at 1, 3 or 5 days respectively. Scale bar ¼ 200 mm. (J) Cultures were quantified at the indicated time points using a resazurin assay. Relative fluorescence unit (RFU) values are proportional to the number of viable cells on the scaffolds. *p < 0.01 as determined by two-way ANOVA.
epithelial cells (IEC) [28,32,59], increases migration and decreases apoptosis in intestinal stem cells (ISC) [33], decreases neutrophileendothelial cell interactions and pro-inflammatory cytokine release by immunocytes [60,61], promotes angiogenesis [62,63], promotes vasodilation in terminal mesenteric arterioles [64], preserves microcirculatory blood flow [65e67], enhances neurite outgrowth and neuroprotection, and protects the ENS from injury [68]. In the current study, plain PGA was first coated with 5% PLLA which offers several benefits to the scaffolds. First, it increases surface area for delivery of growth promoting molecules such as
HB-EGF. SEM images demonstrate that PLLA bridges the PGA fibers and forms its own network within the PGA porous structure. This honey-comb-like network is not interconnected, and forms numerous pores with dead ends, helping to trap HB-EGF, and leading to a slow sustained release of HB-EGF from the scaffolds. After subcritical CO2 exposure, loosely attached HB-EGF flakes were replaced by crystals partially embedded into the PGA fibers. The slow depressurization that was applied during subcritical CO2 exposure avoided deformities of the PGA fibers (foaming), allowing preservation of the mechanical properties of the scaffolds. In order to demonstrate that PLLA coating increases the surface areas for
Y. Liu et al. / Biomaterials 103 (2016) 150e159
157
Fig. 7. Histologic examination and villous height measurement. (A) Native rat intestine; (B) TEI produced on PGA/PLLA scaffold; (C) TEI produced on PGA/PLLA/HB-EGF scaffold; and (D) TEI produced on PGA/PLLA/HB-EGF/CO2 scaffold. Periodic acid Schiff (PAS) staining indicates the presence of goblet cells. Scale bar ¼ 200 mm. D) Villous height in native rat intestine and in TEI was measured using ImageJ software (E). *p < 0.05 as determined by one-way ANOVA.
Fig. 8. Characterization of epithelial cell linages in TEI. TEI was successfully produced from all three groups of scaffolds and was subjected to immunofluorescent staining. Representative images from a PGA/PLLA/HB-EGF/CO2 scaffold include: (A) lysozyme (LYSO) staining to demonstrate the presence of Paneth cells; (B) chromonogranin A (CHGA) staining to demonstrate the presence of enteroendocrine cells; (C) villin staining to demonstrate the brush border of enterocytes; (D) desmin and a-smooth muscle actin (a-SMA) double-staining to demonstrate smooth muscle cells. The insert demonstrates intestinal subepithelial myofibroblasts (ISEMFs) located around the crypts which stain positive for aSMA only. (E) Ki67 staining to demonstrate actively proliferating crypt cells; and Scale bar: 200 mm in low magnification and 10 mm in the inserts. DAPI counterstaining demonstrates the presence of cell nuclei in all panels.
protein delivery, FITC-BSA was used to simulate HB-EGF-coating of the scaffolds due to the visibility of FITC-BSA by fluorescence microscopy. FITC-BSA was located on both PGA fibers and on the PLLA bridges between the PGA fibers in PGA/PLLA scaffolds. This significantly increased the surface area for FITC-BSA distribution. Secondly, the PLLA network strengthens the mechanical properties of the scaffolds, with increased SRS and ultimate compressive force compared to plain PGA. Without PLLA coating, plain PGA tubes cannot withstand the compressive force of intra-abdominal pressure, and the lumen collapses during in vivo incubation (data not shown). Therefore PLLA coating plays an important role in fabricating scaffolds for the production of TEI. Additional treatment with 100% ethanol is beneficial because it improves the hydrophilicity of PGA/PLLA scaffolds for the delivery of soluble HB-EGF and intestinal stem cells into the scaffolds, as shown in the hydration test (Video clip 1 is available online). Initial bolus release of HB-EGF occurred in scaffolds without subcritical CO2 exposure due to the loosely attached HB-EGF flakes on the surface of PGA fibers, which were easily washed off once the scaffolds were exposed to elution buffer. This was followed by slow release of HB-EGF, possibly resulting from the physically entrapped HB-EGF in the PLLA network, in which non-interconnected pores entrapped HB-EGF molecules. With subcritical CO2 exposure, a gradual and constant release of HB-EGF occurred. This is likely attributable to swelling of the scaffolds during the gas phase of CO2
exposure, which allows the HB-EGF molecules to be embedded onto the superficial surface of the PGA fibers and into the PLLA pores and bridges. The crystal-like HB-EGF on HB-EGF-embedded PGA fibers with subcritical CO2 exposure might contribute to the relatively faster release at earlier time points, while HB-EGF embedded in PGA fibers and entrapped in the PLLA micro-porous network might contribute to the constant slow release at later time points. Subcritical CO2 exposure did not affect the bioactivity of HB-EGF, consistent with growth factors released from PLGA microspheres sintered with subcritical CO2 [27]. Culturing of freshly isolated ISC-containing crypts on scaffolds demonstrated that cell density was higher on HB-EGF-embedded scaffolds, with or without CO2 exposure, compared to that on non-HB-EGF-embedded scaffolds on day 3. This might be attributed to the faster release of HB-EGF from the scaffolds at earlier time points. Since the medium was changed every two days, the concentration of HB-EGF in the medium on day 5 was reflective of the slow release of HB-EGF from the scaffolds at later time points. HB-EGF-embedded scaffolds in the presence of subcritical CO2 exposure maintained a relatively high concentration of HB-EGF in the medium on day 5. This lead to higher cell density compared to HB-EGF-embedded scaffolds without subcritical CO2 exposure. In vivo incubation of crypt-seeded, non-HB-EGF-embedded scaffolds produced neomucosa with short disorganized villi. In vivo incubation of crypt-seeded, HB-EGF-embedded scaffolds without
158
Y. Liu et al. / Biomaterials 103 (2016) 150e159
subcritical CO2 exposure formed a neomucosa with longer and more organized villi. This may be attributed to an initial bolus release of HB-EGF at the implantation site that stimulated the proliferation and differentiation of the seeded cells, however, an effective concentration of HB-EGF could not be maintained over time. In vivo incubation of crypt-seeded, HB-EGF-embedded scaffolds followed by subcritical CO2 exposure demonstrated neomucosa with the longest and most well organized villi, similar to the mucosa of native intestine. This might be attributed to the constant release of HB-EGF that maintained an effective dose at the implantation site. Immunofluorescent staining confirmed the presence of all epithelial cell lineages in the neomucosa produced in all three scaffold groups. The doses selected for the current in intro and in vivo studies were based on pilot studies. Higher doses of HB-EGF (10 mg) were chosen for in vitro release kinetics studies. This is because a complete medium change was used to elute HB-EGF from the scaffolds, with ELISA used to detect the released HB-EGF. In order to obtain detectable concentrations of HB-EGF at the later time points, higher amounts of HB-EGF (10 mg) were needed initially. This dose was also used for in vivo pilot studies. Dose dependent effects were observed for 2 mg and 10 mg (unpublished data), but the 2 mg dose was efficacious compared to non-HB-EGF-embedded scaffolds. Therefore, the lower efficacious dose (2 mg) was chosen for in vivo implantation studies. Future studies will focus on determination of the optimal dose of HB-EGF for the production of TEI in vivo. 5. Conclusions In conclusion, HB-EGF embedded into PGA/PLLA scaffolds via subcritical CO2 exposure is slowly released in an active form, and can stimulate the proliferation of crypts in vitro and the formation of intestinal neomucosa in vivo. Subcritical CO2 exposure does not deform the PGA/PLLA scaffolds and does not impair the mechanical properties of the scaffolds. The development of biomimetic scaffolds which can release growth factors stimulating a resident population of stem and progenitor cells may improve the formation of new tissue structures such as TEI. The methodology described here may lead to improved methods of producing tissue engineered intestine to treat patients with short bowel syndrome in the future.
[6] [7] [8] [9]
[10]
[11]
[12] [13]
[14]
[15]
[16]
[17] [18]
[19] [20]
[21]
[22]
[23] [24]
Disclosure statement No competing financial interests exist.
[25]
Acknowledgements
[26]
This study was supported by grants from the Research Institute at Nationwide Children's Hospital (82002914), the Ohio Third Frontier, and NIH R43DK107168. The authors thank Cynthia McAllister, Patricia Craig, and Melanie Herring for their assistance in sample preparation for histological analyses.
[27]
[28]
[29]
References [30] [1] D.C. Chen, J.R. Avansino, V.G. Agopian, V.D. Hoagland, J.D. Woolman, S. Pan, et al., Comparison of polyester scaffolds for bioengineered intestinal mucosa, Cells Tissues Organs 184 (2006) 154e165. [2] D.E. Levin, E.R. Barthel, A.L. Speer, F.G. Sala, X. Hou, Y. Torashima, et al., Human tissue-engineered small intestine forms from postnatal progenitor cells, J. Pediatr. Surg. 48 (2013) 129e137. [3] F.G. Sala, S.M. Kunisaki, E.R. Ochoa, J. Vacanti, T.C. Grikscheit, Tissue-engineered small intestine and stomach form from autologous tissue in a preclinical large animal model, J. Surg. Res. 156 (2009) 205e212. [4] F.G. Sala, J.A. Matthews, A.L. Speer, Y. Torashima, E.R. Barthel, T.C. Grikscheit, A multicellular approach forms a significant amount of tissue-engineered small intestine in the mouse, Tissue Eng. Part A 17 (2011) 1841e1850. [5] B. Wulkersdorfer, K.K. Kao, V.G. Agopian, J.C. Dunn, B.M. Wu, M. Stelzner,
[31]
[32]
[33]
[34]
Growth factors adsorbed on polyglycolic acid mesh augment growth of bioengineered intestinal neomucosa, J. Surg. Res. 169 (2011) 169e178. J.S. Thompson, M.A. Tempero, J.L. Haun, J.A. Vanderhoof, The importance of luminal factors in neomucosal growth, J. Surg. Res. 40 (1986) 126e132. J.S. Thompson, J.G. Sharp, S.K. Saxena, K.G. McCullagh, Stimulation of neomucosal growth by systemic urogastrone, J. Surg. Res. 42 (1987) 402e410. Y. Kato, D. Yu, J.R. Lukish, M.Z. Schwartz, Influence of luminal hepatocyte growth factor on small intestine mucosa in vivo, J. Surg. Res. 71 (1997) 49e53. M.Z. Schwartz, R.B. Storozuk, Influence of epidermal growth factor on intestinal function in the rat: comparison of systemic infusion versus luminal perfusion, Am. J. Surg. 155 (1988) 18e22. C.E. Shin, M.A. Helmrath, R.A. Falcone Jr., J.W. Fox, K.R. Duane, C.R. Erwin, et al., Epidermal growth factor augments adaptation following small bowel resection: optimal dosage, route, and timing of administration, J. Surg. Res. 77 (1998) 11e16. M.Z. Schwartz, Y. Kato, D. Yu, J.R. Lukish, Growth-factor enhancement of compromised gut function following massive small-bowel resection, Pediatr. Surg. Int. 16 (2000) 174e175. Y. Kato, D. Yu, M.Z. Schwartz, Enhancement of intestinal adaptation by hepatocyte growth factor, J. Pediatr. Surg. 33 (1998) 235e239. D.A. Lloyd, T. Ansari, S. Shurey, V. Maquet, P.D. Sibbons, A.R. Boccaccini, et al., Prolonged maintenance of neointestine using subcutaneously implanted tubular scaffolds in a rat model, Transplant. Proc. 38 (2006) 3097e3099. F.G. Rocha, C.A. Sundback, N.J. Krebs, J.K. Leach, D.J. Mooney, S.W. Ashley, et al., The effect of sustained delivery of vascular endothelial growth factor on angiogenesis in tissue-engineered intestine, Biomaterials 29 (2008) 2884e2890. A. Ramsanahie, M.S. Duxbury, T.C. Grikscheit, A. Perez, D.B. Rhoads, J. GardnerThorpe, et al., Effect of GLP-2 on mucosal morphology and SGLT1 expression in tissue-engineered neointestine, Am. J. Physiol. Gastrointest. Liver Physiol. 285 (2003) G1345eG1352. A. Perez, T.C. Grikscheit, R.S. Blumberg, S.W. Ashley, J.P. Vacanti, E.E. Whang, Tissue-engineered small intestine: ontogeny of the immune system, Transplantation 74 (2002) 619e623. R. Langer, J.P. Vacanti, Tissue engineering, Science 260 (1993) 920e926. L.G. Cima, J.P. Vacanti, C. Vacanti, D. Ingber, D. Mooney, R. Langer, Tissue engineering by cell transplantation using degradable polymer substrates, J. Biomech. Eng. 113 (1991) 143e151. J.E. Babensee, L.V. McIntire, A.G. Mikos, Growth factor delivery for tissue engineering, Pharm. Res. 17 (2000) 497e504. M.H. Sheridan, L.D. Shea, M.C. Peters, D.J. Mooney, Bioabsorbable polymer scaffolds for tissue engineering capable of sustained growth factor delivery, J. Control. Release Off. J. Control. Release Soc. 64 (2000) 91e102. J.M. Kanczler, J. Barry, P. Ginty, S.M. Howdle, K.M. Shakesheff, R.O. Oreffo, Supercritical carbon dioxide generated vascular endothelial growth factor encapsulated poly(DL-lactic acid) scaffolds induce angiogenesis in vitro, Biochem. Biophys. Res. Commun. 352 (2007) 135e141. H.M. Burt, J.K. Jackson, S.K. Bains, R.T. Liggins, A.M. Oktaba, A.L. Arsenault, et al., Controlled delivery of taxol from microspheres composed of a blend of ethylene-vinyl acetate copolymer and poly (d,l-lactic acid), Cancer Lett. 88 (1995) 73e79. C. Bitz, E. Doelker, Influence of the preparation method on residual solvents in biodegradable microspheres, Int. J. Pharm. (Amsterdam) 131 (1996) 171e181. C. Thomasin, P. Johansen, R. Alder, R. Bemsel, G. Hottinger, H. Altorfer, et al., A contribution to overcoming the problem of residual solvents in biodegradable microspheres prepared by coacervation, Eur. J. Pharm. Biopharm. 42 (1996) 16e24. T.L. Sproule, J.A. Lee, H. Li, J.J. Lannutti, D.L. Tomasko, Bioactive polymer surfaces via supercritical fluids, J. Supercrit. Fluids 28 (2004) 241e248. H.M. Powell, O. Ayodeji, T.L. Summerfield, D.M. Powell, D.A. Kniss, D.L. Tomasko, et al., Chemotherapeutic implants via subcritical CO2 modification, Biomaterials 28 (2007) 5562e5569. N.H. Dormer, V. Gupta, A.M. Scurto, C.J. Berkland, M.S. Detamore, Effect of different sintering methods on bioactivity and release of proteins from PLGA microspheres, Mater. Sci. Eng. C Mater. Biol. Appl. 33 (2013) 4343e4351. J. Feng, O.N. El-Assal, G.E. Besner, Heparin-binding EGF-like growth factor (HB-EGF) and necrotizing enterocolitis, Semin. Pediatr. Surg. 14 (2005) 167e174. J. Feng, G.E. Besner, Heparin-binding epidermal growth factor-like growth factor promotes enterocyte migration and proliferation in neonatal rats with necrotizing enterocolitis, J. Pediatr. Surg. 42 (2007) 214e220. G. Xia, A.E. Martin, M.P. Michalsky, G.E. Besner, Heparin-binding EGF-like growth factor preserves crypt cell proliferation and decreases bacterial translocation after intestinal ischemia/reperfusion injury, J. Pediatr. Surg. 37 (2002) 1081e1087 discussion -7. J. Feng, O.N. El-Assal, G.E. Besner, Heparin-binding epidermal growth factorlike growth factor reduces intestinal apoptosis in neonatal rats with necrotizing enterocolitis, J. Pediatr. Surg. 41 (2006) 742e747 discussion -7. M.P. Michalsky, A. Kuhn, V. Mehta, G.E. Besner, Heparin-binding EGF-like growth factor decreases apoptosis in intestinal epithelial cells in vitro, J. Pediatr. Surg. 36 (2001) 1130e1135. C.L. Chen, X. Yu, I.O. James, H.Y. Zhang, J. Yang, A. Radulescu, et al., intestinal stem cells from injury in a rat model of necrotizing enterocolitis, Lab. Investig J. Tech. Methods Pathol. 92 (2012) 331e344. L. Boomer, Y. Liu, N. Mahler, J. Johnson, K. Zak, T. Nelson, et al., Scaffolding for
Y. Liu et al. / Biomaterials 103 (2016) 150e159
[35]
[36]
[37]
[38]
[39]
[40]
[41]
[42]
[43]
[44]
[45]
[46]
[47]
[48]
[49]
[50]
[51]
challenging environments: materials selection for tissue engineered intestine, J. Biomed. Mater. Res. Part A 102 (2014) 3795e3802. F.J. Chaparro, M.E. Matusicky, M.J. Allen, J.J. Lannutti, Biomimetic microstructural reorganization during suture retention strength evaluation of electrospun vascular scaffolds, J. Biomed. Mater. Res. Part B Appl. Biomater. (2015), http://dx.doi.org/10.1002/jbm.b.33493. S. Drilling, J. Gaumer, J. Lannutti, Fabrication of burst pressure competent vascular grafts via electrospinning: effects of microstructure, J. Biomed. Mater. Res. Part A 88 (2009) 923e934. B. Rath, J. Nam, T.J. Knobloch, J.J. Lannutti, S. Agarwal, Compressive forces induce osteogenic gene expression in calvarial osteoblasts, J. Biomech. 41 (2008) 1095e1103. Y. Liu, T. Rager, J. Johnson, J. Enmark, G.E. Besner, Enriched intestinal stem cell seeding improves the architecture of tissue-engineered intestine, Tissue Eng. Part C. Methods 21 (2015) 816e824. Y. Liu, X.Z. Shu, S.D. Gray, G.D. Prestwich, Disulfide-crosslinked hyaluronangelatin sponge: growth of fibrous tissue in vivo, J. Biomed. Mater. Res. Part A 68 (2004) 142e149. C.L. Watson, M.M. Mahe, J. Munera, J.C. Howell, N. Sundaram, H.M. Poling, et al., An in vivo model of human small intestine using pluripotent stem cells, Nat. Med. 20 (2014) 1310e1314. N.H. Dormer, C.J. Berkland, M.S. Detamore, Emerging techniques in stratified designs and continuous gradients for tissue engineering of interfaces, Ann. Biomed. Eng. 38 (2010) 2121e2141. M. Singh, C. Berkland, M.S. Detamore, Strategies and applications for incorporating physical and chemical signal gradients in tissue engineering, Tissue Eng. Part B Rev. 14 (2008) 341e366. M.J. Yaszemski, R.G. Payne, W.C. Hayes, R.S. Langer, T.B. Aufdemorte, A.G. Mikos, The ingrowth of new bone tissue and initial mechanical properties of a degrading polymeric composite scaffold, Tissue Eng. 1 (1995) 41e52. R.A. Giordano, B.M. Wu, S.W. Borland, L.G. Cima, E.M. Sachs, M.J. Cima, Mechanical properties of dense polylactic acid structures fabricated by three dimensional printing, J. Biomater. Sci. Polym. Ed. 8 (1996) 63e75. G.G. Giordano, R.C. Thomson, S.L. Ishaug, A.G. Mikos, S. Cumber, C.A. Garcia, et al., Retinal pigment epithelium cells cultured on synthetic biodegradable polymers, J. Biomed. Mater. Res. 34 (1997) 87e93. W.D. Holder Jr., H.E. Gruber, A.L. Moore, C.R. Culberson, W. Anderson, K.J. Burg, et al., Cellular ingrowth and thickness changes in poly-L-lactide and polyglycolide matrices implanted subcutaneously in the rat, J. Biomed. Mater. Res. 41 (1998) 412e421. K.M. Davis-Fleischer, G.E. Besner, Structure and function of heparin-binding EGF-like growth factor (HB-EGF), Front. Biosci. J. Virtual Libr. 3 (1998) d288ed299. P.D. Ellis, K.M. Hadfield, J.C. Pascall, K.D. Brown, Heparin-binding epidermalgrowth-factor-like growth factor gene expression is induced by scrapewounding epithelial cell monolayers: involvement of mitogen-activated protein kinase cascades, Biochem. J. 354 (2001) 99e106. R.K. Cribbs, P.A. Harding, M.H. Luquette, G.E. Besner, Endogenous production of heparin-binding EGF-like growth factor during murine partial-thickness burn wound healing, J. Burn Care Rehabil. 23 (2002) 116e125. K. Jin, X.O. Mao, Y. Sun, L. Xie, L. Jin, E. Nishi, et al., Heparin-binding epidermal growth factor-like growth factor: hypoxia-inducible expression in vitro and stimulation of neurogenesis in vitro and in vivo, J. Neurosci. Off. J. Soc. Neurosci. 22 (2002) 5365e5373. T. Homma, M. Sakai, H.F. Cheng, T. Yasuda, R.J. Coffey Jr., R.C. Harris, Induction of heparin-binding epidermal growth factor-like growth factor mRNA in rat kidney after acute injury, J. Clin. Investig. 96 (1995) 1018e1025.
159
[52] G.D. Frank, M. Mifune, T. Inagami, M. Ohba, T. Sasaki, S. Higashiyama, et al., Distinct mechanisms of receptor and nonreceptor tyrosine kinase activation by reactive oxygen species in vascular smooth muscle cells: role of metalloprotease and protein kinase C-delta, Mol. Cell. Biol. 23 (2003) 1581e1589. [53] Y. Kayanoki, S. Higashiyama, K. Suzuki, M. Asahi, S. Kawata, Y. Matsuzawa, et al., The requirement of both intracellular reactive oxygen species and intracellular calcium elevation for the induction of heparin-binding EGF-like growth factor in vascular endothelial cells and smooth muscle cells, Biochem. Biophys. Res. Commun. 259 (1999) 50e55. [54] S. Kiso, S. Kawata, S. Tamura, Y. Inui, Y. Yoshida, Y. Sawai, et al., Liver regeneration in heparin-binding EGF-like growth factor transgenic mice after partial hepatectomy, Gastroenterology 124 (2003) 701e707. [55] M. Marikovsky, K. Breuing, P.Y. Liu, E. Eriksson, S. Higashiyama, P. Farber, et al., Appearance of heparin-binding EGF-like growth factor in wound fluid as a response to injury, Proc. Natl. Acad. Sci. U. S. A. 90 (1993) 3889e3893. [56] D.W. McCarthy, M.T. Downing, D.R. Brigstock, M.H. Luquette, K.D. Brown, M.S. Abad, et al., Production of heparin-binding epidermal growth factor-like growth factor (HB-EGF) at sites of thermal injury in pediatric patients, J. Investig. Dermatol. 106 (1996) 49e56. [57] T.T. Junttila, M. Sundvall, J.A. Maatta, K. Elenius, Erbb4 and its isoforms: selective regulation of growth factor responses by naturally occurring receptor variants, Trends Cardiovasc. Med. 10 (2000) 304e310. [58] E. Nishi, A. Prat, V. Hospital, K. Elenius, M. Klagsbrun, N-arginine dibasic convertase is a specific receptor for heparin-binding EGF-like growth factor that mediates cell migration, EMBO J. 20 (2001) 3342e3350. [59] Y. Su, J. Yang, G.E. Besner, HB-EGF promotes intestinal restitution by affecting integrin-extracellular matrix interactions and intercellular adhesions, Growth factors 31 (2013) 39e55. [60] D.V. Rocourt, V.B. Mehta, G.E. Besner, Heparin-binding EGF-like growth factor decreases inflammatory cytokine expression after intestinal ischemia/reperfusion injury, J. Surg. Res. 139 (2007) 269e273. [61] D.V. Rocourt, V.B. Mehta, D. Wu, G.E. Besner, Heparin-binding EGF-like growth factor decreases neutrophil-endothelial cell interactions, J. Surg. Res. 141 (2007) 262e266. [62] V.B. Mehta, G.E. Besner, HB-EGF promotes angiogenesis in endothelial cells via PI3-kinase and MAPK signaling pathways, Growth Factors 25 (2007) 253e263. [63] D. Chalothorn, S.M. Moore, H. Zhang, S.W. Sunnarborg, D.C. Lee, J.E. Faber, Heparin-binding epidermal growth factor-like growth factor, collateral vessel development, and angiogenesis in skeletal muscle ischemia, Arterioscler. Thromb. Vasc. Biol. 25 (2005) 1884e1890. [64] Y. Zhou, D. Brigstock, G.E. Besner, Heparin-binding EGF-like growth factor is a potent dilator of terminal mesenteric arterioles, Microvasc. Res. 78 (2009) 78e85. [65] O.N. El-Assal, A. Radulescu, G.E. Besner, Heparin-binding EGF-like growth factor preserves mesenteric microcirculatory blood flow and protects against intestinal injury in rats subjected to hemorrhagic shock and resuscitation, Surgery 142 (2007) 234e242. [66] X. Yu, A. Radulescu, C.L. Chen, I.O. James, G.E. Besner, Heparin-binding EGFlike growth factor protects pericytes from injury, J. Surg. Res. 172 (2012) 165e176. [67] X. Yu, A. Radulescu, N. Zorko, G.E. Besner, Heparin-binding EGF-like growth factor increases intestinal microvascular blood flow in necrotizing enterocolitis, Gastroenterology 137 (2009) 221e230. [68] Y. Zhou, G.E. Besner, Heparin-binding epidermal growth factor-like growth factor is a potent neurotrophic factor for PC12 cells, Neuro Signals 18 (2010) 141e151.
Contents lists available at ScienceDirect
Biomaterials journal homepage: www.elsevier.com/locate/biomaterials
HB-EGF embedded in PGA/PLLA scaffolds via subcritical CO2 augments the production of tissue engineered intestine Yanchun Liu a, Tyler Nelson b, Barrett Cromeens a, Terrence Rager a, John Lannutti c, Jed Johnson d, Gail E. Besner a, * a
Center for Perinatal Research, The Research Institute at Nationwide Children's Hospital, and the Department of Pediatric Surgery, Columbus, OH, USA Department of Biomedical Engineering, The Ohio State University, Columbus, OH, USA Department of Materials Science and Engineering, The Ohio State University, Columbus, OH, USA d Nanofiber Solutions, Inc., Columbus, OH, USA b c
a r t i c l e i n f o
a b s t r a c t
Article history: Received 8 February 2016 Received in revised form 10 May 2016 Accepted 17 June 2016 Available online 21 June 2016
The ability to deliver sustained-release, biologically active growth factors through custom designed tissue engineering scaffolds at sites of tissue regeneration offers great therapeutic opportunity. Due to the short in vivo half-lives of most growth factors, it is challenging to deliver these proteins to sites of interest where they may be used before being degraded. The application of subcritical CO2 uses gasphase CO2 at subcritical pressures ranging from 41 to 62 bar (595e913 PSI) which avoids foaming by reducing the amount of CO2 dissolved in the polymer and maintains completely reversible plasticization. In the current study, heparin-binding EGF-like growth factor (HB-EGF) was embedded into polyglycolic acid (PGA)/Poly-L-latic acid (PLLA) scaffolds via subcritical CO2 exposure for the production of tissue engineered intestine (TEI). PGA fiber morphology after subcritical CO2 exposure was examined by scanning electron microscopy (SEM) and the distribution of HB-EGF embedded in the scaffold fibers was detected by HB-EGF immunofluorescent staining. In vivo implantation of HB-EGF-embedded scaffolds confirmed significantly improved TEI structure as a result of local delivery of the trophic growth factor. These findings may be critical for the production of TEI in the treatment of patients with short bowel syndrome in the future. © 2016 Elsevier Ltd. All rights reserved.
Keywords: Polyglycolic acid Stem cell HB-EGF Intestine Tissue engineering
1. Introduction Patients with short bowel syndrome (SBS) lack the adequate intestinal function needed to sustain their nutritional needs. The production of tissue-engineered intestine (TEI) using a patient's own intestinal cells may restore functional intestinal absorptive area while avoiding the complications of current therapeutic options such as small bowel transplantation. The smallest transplantable mucosal units are intestinal “organoids” which are clusters of 20e40 cells isolated from intestinal mucosa. These organoids contain the putative intestinal stem cells (ISCs) and surrounding cells that comprise the stem cell niche. Transplantation of these intestinal organoids has been shown to generate TEI that resembles native intestine in both structure and function with variable results [1e5].
* Corresponding author. Department of Pediatric Surgery, ED383, Nationwide Children's Hospital, 700 Children's Drive, Columbus, OH 43205, USA. E-mail address: [email protected] (G.E. Besner). http://dx.doi.org/10.1016/j.biomaterials.2016.06.039 0142-9612/© 2016 Elsevier Ltd. All rights reserved.
In the 1980s, Thompson et al. first described the augmentation of neomucosal growth after exposure to luminal factors or to systemically administered urogastrone, a peptide that shares an intestinal receptor with epidermal growth factor (EGF) [6,7]. Based on the concept that intestinal growth is regulated by nutrients and enteric secretions, as well as local and systemic growth factors, subsequent studies compared luminal versus systemic administration of trophic peptides. The route of growth factor delivery has been shown to impact mucosal growth, demonstrating that local delivery is optimal [8e12]. This work led to studies of the administration of growth factors to augment the production of TEI [13e16]. Very few studies have examined the effects of locally delivered growth factors on tissue-engineered neomucosa. The application of three-dimensional, porous polymer constructs for tissue engineering applications has received considerable attention as a method of aiding and defining new tissue growth [17,18]. Due to the short in vivo half-lives of most growth factors, it is desirable to deliver the protein to the site of interest where it may be used before it is degraded [19]. For example, one successful approach to
Y. Liu et al. / Biomaterials 103 (2016) 150e159
deliver VEGF within a polymer implant was to produce a porous PLGA matrix with dispersed alginate microspheres [20]. The growth factor was encapsulated within the alginate spheres and was released in an active state upon exposure of the construct to an aqueous buffer. Another mechanism used to release growth factors at the desired site was to encapsulate the protein within the PLGA matrix itself [20]. Supercritical CO2 fluid mixing technology is performed at or below physiological temperatures without any harsh solvents, allowing bioactive factors to be encapsulated and retained in biodegradable polymer constructs without compromising their biological activity [21]. The use of CO2 avoids concerns surrounding halogenated solvents that are often necessary to achieve mixtures of resorbable polymers with other compounds [22e24]. However, the use of supercritical CO2 with bulk polymeric implants carries a substantial penalty: foaming, or the loss of the original implant form [25]. Therefore, an alternative treatment that preserves form is based on the use of gas-phase CO2 at subcritical pressures ranging from 41 to 62 bar (595e913 pounds per square inch; PSI). This avoids foaming by reducing the amount of CO2 dissolved in the polymer and maintains conditions that allow plasticization to be completely reversible. Subcritical CO2 has been used successfully to embed the anti-cancer drug paclitaxel into polylactic acid with proven sustained chemotherapeutic activity [26]. Dormer et al. reported the use of subcritical CO2 with sinter microspheres for the delivery of transforming growth factor (TGF)-beta3 and bone morphogenetic protein (BMP)-2 to human bone marrow stromal cells in vitro, and found that subcritical CO2 had only slight adverse effects, as well as some desirable effects, on protein availability and bioactivity [27]. In the current study, heparin-binding EGF-like growth factor (HB-EGF) was embedded into polyglycolic acid (PGA)/Poly-L-latic acid (PLLA) scaffolds via subcritical CO2 exposure in an effort to augment the growth of TEI. We have previously shown that HB-EGF promotes intestinal epithelial cell (IEC) proliferation and migration [28e30], decreases IEC apoptosis [31,32], and protects ISC from injury [33], supporting the use of HB-EFG in the current studies. The release kinetics and bioactivity of HB-EGF released from PGA/PLLA scaffolds were characterized, and the effects of embedding scaffolds with HB-EGF were tested in in vitro cell cultures and in vivo scaffold implantation. 2. Materials and methods 2.1. Materials for scaffold preparation PGA Biofelt (2 mm thickness and 60 mg/cm3 density) was purchased from Biomedical Structures (Warwick, RI). Poly-L-latic acid (PLLA) and fluorescein isothiocyanate (FITC) conjugated bovine serum albumin (BSA) were from Sigma-Aldrich (St. Louis, MO). Chloroform and collagen type I were from Fisher Scientific (Pittsburgh, PA). HB-EGF was from R&D Systems (Minneapolis, MN). 2.2. Preparation of scaffolds Scaffold preparation is depicted in Fig. 1. Flat scaffolds (0.8 1.6 cm) were cut directly from PGA Biofelt and used for scanning electron microscopy (SEM), in vitro cell culture, and measurement of surface area as depicted by FITC-BSA distribution. Tubular scaffolds (0.5 3.0 cm) were prepared by wrapping PGA flat sheets around stainless steel mandrels for suture retention strength (SRS) tests. Similarly, tubular scaffolds (0.5 1.0 cm) were prepared for in vitro release kinetics, bioactivity, compression tests, and in vivo implantation. All flat and tubular scaffolds were coated with 5% PLLA in chloroform. Once the solvent was completely
PGA Biofelt
151
Untreated plain PGA (PGA)
Coat with PLLA (5%)
Wash with 100% ethanol
Coat with collagen
Coat with HB-EGF
Subcritical CO2
Group 1 (PGA/PLLA) Group 2 (PGA/PLLA/HB-EGF) Group 3 (PGA/PLLA/HB-EGF/CO2)
Fig. 1. Scaffold preparation for in vitro and in vivo studies. Three groups of scaffolds were fabricated: PGA coated with PLLA (PGA/PLLA), PGA coated with PLLA and embedded with HB-EGF (PGA/PLLA/HB-EGF), and PGA coated with PLLA and embedded with HB-EGF followed by subcritical CO2 exposure (PGA/PLLA/HB-EGF/CO2). In some in vitro studies, untreated plain PGA scaffolds were used as a control.
evaporated, scaffolds were soaked in 100% ethanol for 30 min in order to improve the hydrophilicity of PGA/PLLA scaffolds (Video clip 1 is available online), followed by 3 washes with PBS. Scaffolds were then soaked in 0.4 mg/ml collagen type I for 30 min followed by 3 washes with PBS, lyophilized, and sterilized with ethylene oxide. All flat and tubular scaffolds were divided into three groups based on the following treatments: Supplementary video related to this article can be found at http://dx.doi.org/10.1016/j.biomaterials.2016.06.039. 1) PGA coated with PLLA (PGA/PLLA), 2) PGA coated with PLLA and embedded with HB-EGF (PGA/PLLA/HB-EGF), and 3) PGA coated with PLLA and embedded with HB-EGF followed by subcritical CO2 exposure (PGA/PLLA/HB-EGF/CO2). Scaffolds from the second and third groups were hydrated with 10 mg per scaffold of HB-EGF in 100 ml of PBS for in vitro studies and 2 mg per scaffold of HB-EGF in 100 ml of PBS for in vivo studies. Additionally, scaffolds from the third group were placed in sterile 50 ml conical bottom tubes with porous caps and exposed to subcritical CO2 in a stainless steel vessel (Parr Instruments Co., Moline, IL) at 900 PSI for 1 h followed by a slow release of CO2 at 15 PSI per minute for 60 min. All prepared scaffolds were stored at 30 C and brought to RT 10e20 min prior to use. 2.3. Scanning electron microscopy (SEM) and confocal microscopy Flat scaffolds from the three groups, in addition to plain PGA scaffolds, were cut to 2 5 mm strips and 4 strips from each group were affixed to a stub and sputter-coated with gold (Emitech K550X, Quorum Technologies Ltd, Ashford, Kent, England). Samples were examined using a scanning electron microscope (Hitachi S4800, Hitachi High Technologies America, Inc., Dallas, TX) at a voltage of 10e20 kV. To visualize the distribution of HB-EGF in the scaffolds, HB-EGF-embedded scaffolds were incubated in 10 mg/ml of goat derived anti-HB-EGF primary antibody (R&D Systems) for 1 h followed by 3 washes with PBS and incubation with donkey anti-goat IgG secondary antibody for 1 h (Invitrogen, Grand Island, NY). After 3 washes with PBS, the scaffolds were examined with a confocal microscope (Zeiss LSM 710, Thornwood, New York).
152
Y. Liu et al. / Biomaterials 103 (2016) 150e159
2.4. Suture retention strength (SRS) and compression tests
The maximum force is the ultimate compressive force.
A Test Resources load frame (Model # SM-50-294-Capacity 50 lb, Northbrook, IL) was used for both tests. The preparation of the samples and the settings of the tests are as previously described, and shown in Fig. 3AeC [34e37]. Four scaffolds from each of the three groups, together with plain PGA scaffolds, were tested. For SRS tests, load was applied at 50 mm/min until failure, as determined by the suture pulling through the wall of the scaffold, and the stressestrain curve recorded. The maximum force required is the SRS. For compression tests, load was applied at 10 mm/min for a total of 2 mm of compression and the stress-strain curve recorded.
2.5. HB-EGF release kinetics and bioactivity Tubular PGA/PLLA/HB-EGF scaffolds (group 2, n ¼ 6) and PGA/ PLLA/HB-EGF/CO2 scaffolds (group 3, n ¼ 6) were individually placed into 2 ml round bottom cryovials (Fisher Scientific, Pittsburgh, PA) prefilled with 1.5 ml of PBS. The tubes were placed on an orbital shaker in a 37 C incubator. On days 1, 3, 5, 7, 9, 11, and 13, the medium from each tube was removed and stored at 4 C. Fresh PBS (1.5 ml) was added to each original tube after medium collection and the tubes were returned to the shaker. The release
Fig. 2. Scanning electron microscopy (SEM) and immunofluorescent staining of HB-EGF on scaffolds. (A) Original PGA biofelt; (B) PGA/PLLA scaffold; (C) PGA/PLLA/HB-EGF scaffold after lyophilization; (D) PGA/PLLA/HB-EGF/CO2 scaffold; (E) magnified view of indicated area of panel C; (F) magnified view of indicated area of panel D; (G) HB-EGF immunofluorescent staining of a PGA/PLLA/HB-EGF scaffold after lyophilization demonstrating the presence and distribution of HB-EGF (green) and the PGA fibers identified by their auto-fluorescence (blue); (H) HB-EGF immunofluorescent staining of a PGA/PLLA/HB-EGF/CO2 scaffold. Scale bar: 40 mm for A-D, 5 mm for E and F, 30 mm for G and H.
Y. Liu et al. / Biomaterials 103 (2016) 150e159
153
Fig. 3. Suture retention strength (SRS) and compression tests. (A) Preparation of samples for SRS test. (B) Example of SRS test. (C) Example of compression test. (D) SRS results demonstrating that PGA/PLLA scaffolds have significantly higher suture retention strength than plain PGA scaffolds. Embedding of PGA/PLLA scaffolds with HB-EGF in the presence or absence of CO2 exposure did not lead to additional changes in the strength of the scaffolds. (E) Compression test results. PGA/PLLA scaffolds have significantly higher ultimate compressive force than plain PGA scaffolds. Embedding of PGA/PLLA scaffolds with HB-EGF in the presence or absence of CO2 exposure did not lead to additional changes in the compressibility of the scaffolds. * indicates statistically significant difference (p < 0.001).
buffer from the other time points was harvested as described above. The concentration of HB-EGF in the eluted buffers was measured using an HB-EGF ELISA assay (R&D Systems, Minneapolis, MN). The amount of HB-EGF released from each condition at each time point was averaged from the 6 samples and plotted over time. To test the bioactivity of released HB-EGF, three additional samples from groups 2 and 3 were prepared at each of three time points (days 1, 7, and 13) using Assay Medium (DMEM containing 0.1% BSA, 0.001% heparin, and 1% insulin-transferrin-selenium-G supplement) (Gibco, Grand Island, NY) as elution buffer. The concentration of HB-EGF was determined by HB-EGF ELISA assay and diluted by 3 fold serial dilution starting from 1000 ng/ml. The diluted HB-EGF was added to a 96-well plate in duplicate. An HBEGF stock solution was diluted and used to produce a standard curve. BALB/c 3T3 mouse fibroblast cells (ATCC, Manassas, VA) were expanded in Growth Medium (DMEM containing 10% bovine calf serum) (Hyclone, Logan, UT) and re-suspended to 1.5 105 cells/ml in Assay Medium. 50 ml of cell suspension was added to the wells of a 96-well plate preloaded with the above samples. After a 72 h incubation at 37 C and 5% CO2, 20 ml of diluted bromodeoxyuridine (BrdU) solution was added followed by an additional 4 h of culture. Proliferating cells were quantified using a BrdU cell proliferation assay kit purchased from Millipore (Billerica, MA). The concentration of HB-EGF from each condition (without or with subcritical CO2 exposure) was calculated using the HB-EGF standard curve. The half maximal effective concentration (EC50) was calculated using 5-parameter asymmetric fit using GraphPad Prism 6 Version 6.05 (GraphPad software, Inc., San Diego, CA) and averaged from the triplicate samples.
2.6. Measurement of BSA distribution in the scaffolds To demonstrate that PLLA coating increases surface area for protein delivery, FITC-BSA was used to simulate HB-EGF-coating of the scaffolds due to the visibility of FITC-BSA by fluorescence microscopy. A solution of 1% FITC-BSA in 10 mM Tris-base was prepared and 100 ml of the solution was pipetted onto 0.8 1.6 cm flat plain PGA and PGA/PLLA scaffolds (n ¼ 3 for each). Scaffolds were examined by confocal microscopy right after lyophilization without the need for washing steps, therefore the distribution of FITC-BSA in the scaffolds accurately represents the amount of protein that scaffolds can accommodate. Z-stack images were captured with a 20 objective lens for a total of 50 intervals at 1 mm per interval with a confocal microscope. Surface areas with BSA were measured in
batch using ImageJ software after the threshold was determined and applied to all images measured.
2.7. In vitro culture of crypt-seeded PGA scaffolds Pregnant adult Lewis rats were purchased from Charles River Laboratories (Ohio, Spencerville) and used to provide pups as cell donors. All procedures were approved by the IACUC of the Research Institute at Nationwide Children's Hospital (protocol #AR1200001). Briefly, 1e3 day old rat pups were euthanized and under sterile conditions the intestines were harvested for isolating ISCcontaining crypts as we have described previously [38]. Intestines were harvested from the ligament of Treitz to the ileocecal junction. After the lumen was opened lengthwise, the intestines were pooled and washed vigorously with Hank's buffered salt solution (HBSS) containing 100 units/ml of penicillin, 100 mg/ml of streptomycin, 0.25 mg/ml of amphotericin B, and 25 mg/ml gentamicin. The intestines were minced into ~5e7 mm pieces and digested in 300 units/ml of collagenase type I and 0.1 g/ml of dispase (both from Worthington Biochemical Corporation, Lakewood, NJ) in serumfree DMEM for 30 min at 37 C with gentle shaking. The digestion was stopped by the addition of 10% fetal bovine serum in DMEM and the suspension (containing a mixed population of intact crypts and villi, fragments of partially digested intestinal tissue, and single cells) was washed with DMEM. The mixed cells were serially filtered through 200 mm, 70 mm and 25 mm sieve membranes. The cells in the 25e70 mm fraction were mainly ISC-containing crypts and were suspended in stem cell culture medium (DMEM containing 10% mesenchymal stem cell-qualified FBS (ThermoFisher, Waltham, MA) at 106 crypts/ml. 100 ml of cell suspension was added onto each flat scaffold (n ¼ 12 scaffolds per group). Each cell-seeded scaffold was placed into an individual well of a 24-well plate and incubated for 1 h. Each well then received stem cell culture medium (1.5 ml/well) for 12 h of static culture followed by dynamic culture with gentle shaking on a shaker. Medium was changed every 2 days. On days 0, 2 and 4, 150 ml of resazurin was added to each well. After an additional 12 h of culture, cell-seeded scaffolds were removed and used for fluorescein diacetate (F-DA) staining. The medium remaining in the well with resazurin was used for quantification of cells on the scaffolds. Briefly, 150 ml of cell culture medium with resazurin was transferred to another 96-well plate in triplicate and relative fluorescence units (RFU) were read with a plate reader. RFU values were proportional to the number of cells on the scaffolds, and were compared with Two-way ANOVA using
154
Y. Liu et al. / Biomaterials 103 (2016) 150e159
GraphPad Prism 6. To visualize cell viability and density, cellseeded scaffolds were rinsed with PBS twice, stained with F-DA (20 mg/ml in PBS) for 5 min, and rinsed with PBS [39]. The density of viable cells (green) was observed with confocal microscopy. 2.8. Production of tissue engineered intestine in vivo All procedures were approved by the IACUC of the Research Institute at Nationwide Children's Hospital (protocol #AR1200001). One to 3 day old rat pups were used as cell donors with the dam (200e250 g and 2e3 month old) of the pups used as recipients. The yield of intestinal cells from each litter of pups (8e10 pups) was used to seed multiple scaffolds, and the dam accommodated multiple scaffolds for side-by-side comparisons. Small intestines harvested from the same litter of pups were pooled, digested, and filtered as described above to concentrate ISCcontaining crypts that were suspended in stem cell culture medium (106 crypts/ml). Next, 200 ml of cell suspension was seeded onto each tubular scaffold (n ¼ 12 scaffolds per group). Each of 12 recipient dams received three cell-seeded scaffolds (one from each group). Under general anesthesia with 2% isoflurane in oxygen, a midline laparotomy was performed on recipient animals. A cellseeded scaffold from group 1 (PGA/PLLA) and group 2 (PGA/PLLA/ HB-EGF) were attached to the anterior surface of the left abdominal wall and a cell-seeded scaffold from group 3 (PGA/PLLA/HB-EGF/ CO2) was attached to the anterior surface of the right abdominal wall, using 7/0 polypropylene suture. Incisions were closed in layers with 5/0 polydioxanone (PDO) monofilament suture. Scaffolds were explanted after 4 weeks of in vivo incubation, fixed in 10% neutral buffered formalin, embedded in paraffin, cut into 5 mm sections, and stained with periodic acid Schiff (PAS). Immunofluorescent staining was conducted to characterize the epithelial cell lineages of TEI using the following primary antibodies: lysozyme (LYSO) (Abcam, Cambridge, MA) to identify Paneth cells, chromogranin A (CHRG) to identify enteroendocrine cells (Bioss Inc., Woburn, MA), villin (Santa Cruz, Dallas, Texas) to identify absorptive enterocytes, Ki67 (Millipore, Billerica, MA) to identify proliferating cells, and desmin (Bioss Inc., Woburn, MA) and alpha smooth muscle actin (a-SMA) (Abcam, Cambridge, MA) to identify intestinal subepithelial myofibroblasts (ISEMFs) and smooth muscle cells [4,40]. All secondary antibodies (goat anti rabbit, goat anti mouse, and donkey anti-goat) were purchased from Invitrogen (Grand Island, NY). Villous height was measured using ImageJ software (version 1.46r, Wayne Rasband, National Institutes of Health, USA) and compared across the groups with one-way ANOVA using Graphpad Prism 6.
superstructure of the scaffolds. 3.2. Suture retention strength (SRS) and compression tests The SRS force of PGA/PLLA scaffolds was significantly higher than that of plain PGA scaffolds, with addition of HB-EGF with or without subcritical CO2 exposure resulting in no change to the SRS forces (Fig. 3D). Similar results were observed in compression tests (Fig. 3E). 3.3. HB-EGF release kinetics and bioactivity HB-EGF was initially released in a bolus fashion from PGA/PLLA/ HB-EGF scaffolds (Fig. 4A). In the first day, ~49% of the initial 10 mg of HB-EGF was released, followed by a slow release of an additional ~26% of HB-EGF at the remaining time points, leading to a total release of ~75.5% over 13 days. On the other hand, only ~23.7% of HB-EGF was released in the first day from PGA/PLLA/HB-EGF/CO2 scaffolds, followed by a constant release of ~10% at each later time point until day 13. The bioactivity of HB-EGF released from the scaffolds as determined by a BrdU cell proliferation assay revealed
3. Results 3.1. Embedding PGA scaffolds with HB-EGF via subcritical CO2 exposure SEM of a plain PGA scaffold is shown in Fig. 2A, demonstrating clean individual PGA fibers. In PGA/PLLA scaffolds, the PLLA bridges the PGA fibers, leading to two networks: large pores defined by a network of PGA fibers and small pores defined by a honey-comblike network of PLLA (Fig. 2B). PGA/PLLA/HB-EGF scaffolds demonstrate loosely attached flakes on the PGA fibers (Fig. 2C and E). Confirmation that these flakes represent attached HB-EGF was obtained by HB-EGF immunofluorescent staining (Fig. 2G). In PGA/ PLLA/HB-EGF/CO2 scaffolds, HB-EGF forms small crystals on the surface of the PGA fibers that are partially embedded into the PGA fibers (Fig. 2D and F), with the distribution of HB-EGF again shown by immunofluorescent staining (Fig. 2H). Compared to untreated plain PGA, subcritical CO2 exposure did not deform the fibers or the
Fig. 4. HB-EGF release kinetics and bioactivity. (A) Release of HB-EGF from scaffolds over a 13 day period. In the absence of subcritical CO2 exposure, an initial large bolus release of HB-EGF occurred, followed by a slow release at the remaining time points. After subcritical CO2 exposure, a more constant release of HB-EGF occurred over the 13 day time period. (B) Bio-activity of HB-EGF released from the scaffolds was examined using a BrdU cell proliferation assay. No significant differences were observed in the half maximal effective concentration (EC50) of released HB-EGF in the presence or absence of subcritical CO2 exposure compared to HB-EGF control at 1, 7 or 13 days after loading onto PGA scaffolds.
Y. Liu et al. / Biomaterials 103 (2016) 150e159
no significant differences in EC50 from released HB-EGF in the presence or absence of subcritical CO2 exposure compared to HBEGF control over 13 days (Fig. 4B). 3.4. Measurement of BSA distribution in the scaffolds The distribution of FITC-BSA on plain PGA scaffolds was limited to the surface of the PGA fibers, whereas the distribution of FITCBSA on PGA/PLLA scaffolds extended to the PLLA bridges between the PGA fibers (Fig. 5A and B). There was a significant increase in surface area occupied by FITC-BSA in PGA/PLLA scaffolds compared to plain PGA scaffolds (Fig. 5C). 3.5. In vitro culture of crypt-seeded PGA scaffolds The density of viable cells was examined in freshly isolated crypts seeded onto PGA/PLLA, PGA/PLLA/HB-EGF, and PGA/PLLA/ HB-EGF/CO2 scaffolds, as demonstrated by F-DA staining. On day 1, the density of viable cells was similar in the three different scaffolds (Fig. 6AeC). On day 3, the density of viable cells was slightly higher on PGA/PLLA/HB-EGF and PGA/PLLA/HB-EGF/CO2 scaffolds compared to PGA/PLLA scaffolds (Fig. 6DeF). By day 5, the density of viable cells on PGA/PLLA/HB-EGF/CO2 scaffolds was markedly increased compared to that on PGA/PLLA/HB-EGF scaffolds, with both being higher than that on PGA/PLLA scaffolds (Fig. 6GeI). Quantification of viable cell numbers using resazurin assay confirmed these findings (Fig. 6J). 3.6. Production of tissue engineered intestine in vivo Freshly isolated crypts were seeded onto PGA/PLLA, PGA/PLLA/ HB-EGF, PGA/PLLA/HB-EGF/CO2 scaffolds, and implanted into the peritoneal cavity of recipient rats for in vivo production of TEI. After 4 weeks of in vivo incubation, PGA/PLLA scaffolds formed a neomucosa with shorter villi and fewer crypts compared to that of PGA/ PLLA/HB-EGF scaffolds (Fig. 7B and C). TEI produced on PGA/PLLA/ HB-EGF/CO2 scaffolds had markedly longer villi and denser crypts (Fig. 7D). The neomucosa produced from all three different scaffolds demonstrated the presence of goblet cells as determined by positive PAS staining, as seen in native rat intestine (Fig. 7A). Quantification of villous height confirmed that the neomucosa produced on PGA/PLLA scaffolds had the shortest villi, while neomucosa produced on PGA/PLLA/HB-EGF/CO2 scaffolds had the longest villi, even exceeding that of native rat intestine (Fig. 7E). Importantly, TEI produced from all three groups contained all elements of native small intestine, including Paneth cells as
155
demonstrated by lysozyme staining (Fig. 8A), enteroendocrine cells as demonstrated by chromogranin A staining (Fig. 8B), absorptive enterocytes as demonstrated by villin staining (Fig. 8C), smooth muscle cells as demonstrated by desmin/a-SMA double staining (Fig. 8D), intestinal subepithelial myofibroblasts (ISEMFs) as demonstrated by positive a-SMA and negative desmin staining (insert on Fig. 8D), and actively proliferating crypt cells as demonstrated by Ki67 staining (Fig. 8E).
4. Discussion Many tissue engineering applications employ a considerable amount of regenerative efficacy from physicochemical signals embedded within scaffolds. These signals can be natural or material-based, including compounds such as collagens, minerals, synthetic substrates, or small molecules [41,42]. PGA and PLLA have received much attention for scaffold production, and these materials have FDA approval for use in vivo. The mechanical and chemical properties of these polymers are desirable as tissue engineering scaffolds because they can be processed to form porous matrices, have excellent mechanical strength [18,43,44], and are relatively biocompatible [45]. The porosity of the material enables nutrient delivery to developing tissue within the construct and provides a high surface area for new cell growth. In addition, suitable porosity allows for cell penetration of existing biological tissue into the construct, a key factor for acceptance of the implant [46]. The ability to continuously deliver biologically active growth factors to sites of tissue regeneration using custom designed scaffolds offers great therapeutic opportunity. In the current studies, HB-EGF was chosen to be embedded in PGA/PLLA scaffolds due to its multiple proven benefits in the stimulation of intestinal development and in protection of the intestines from injury. HB-EGF is produced and secreted by many different cell types. Although the HB-EGF gene is widely expressed [47], the basal level of its mRNA seems to be relatively low in normal cells [48]. Expression of HBEGF is significantly increased in response to tissue damage [49], hypoxia [50,51] and oxidative stress [52,53], and also during wound healing and regeneration [54e56]. It exerts its mitogenic effects by binding to and activation of EGF receptor subtypes ErbB-1 and ErbB-4 [57]. While the mitogenic function of HB-EGF is mediated through activation of ErbB-1, its migration-inducing function involves the activation of ErbB-4 and N-arginine dibasic convertase (NRDc, Nardilysin), the latter representing a completely HB-EGFspecific receptor [58]. We have demonstrated that administration of HB-EGF improves restitution and reduces apoptosis in intestinal
Fig. 5. Distribution of FITC-BSA on scaffolds. (A) FITC conjugated BSA (green) was located on the surface of PGA fibers in plain PGA scaffolds. (B) FITC-BSA was located on the surface of PGA fibers and also on the PLLA bridges between PGA fibers in PGA/PLLA scaffolds. The PGA fibers are identified by their auto-fluorescence (blue). Scale bar ¼ 100 mm. (C) Surface area was measured in batch using ImageJ software, and averaged from 50 slices of each z-stack capture. *p < 0.001.
156
Y. Liu et al. / Biomaterials 103 (2016) 150e159
Fig. 6. In vitro culture of crypts on PGA scaffolds. Crypts were seeded onto scaffolds and incubated in vitro for 5 days. The density of viable cells was determined by fluorescein diacetate staining (green). The PGA fibers are demonstrated by their auto-florescence (blue). (A, D, and G) crypts cultured on PGA/PLLA scaffolds at 1, 3 or 5 days respectively; (B, E and H) crypts cultured on PGA/PLLA/HB-EGF scaffolds at 1, 3 or 5 days respectively; (C, F and I) crypts cultured on PGA/PLLA/HB-EGF/CO2 scaffolds at 1, 3 or 5 days respectively. Scale bar ¼ 200 mm. (J) Cultures were quantified at the indicated time points using a resazurin assay. Relative fluorescence unit (RFU) values are proportional to the number of viable cells on the scaffolds. *p < 0.01 as determined by two-way ANOVA.
epithelial cells (IEC) [28,32,59], increases migration and decreases apoptosis in intestinal stem cells (ISC) [33], decreases neutrophileendothelial cell interactions and pro-inflammatory cytokine release by immunocytes [60,61], promotes angiogenesis [62,63], promotes vasodilation in terminal mesenteric arterioles [64], preserves microcirculatory blood flow [65e67], enhances neurite outgrowth and neuroprotection, and protects the ENS from injury [68]. In the current study, plain PGA was first coated with 5% PLLA which offers several benefits to the scaffolds. First, it increases surface area for delivery of growth promoting molecules such as
HB-EGF. SEM images demonstrate that PLLA bridges the PGA fibers and forms its own network within the PGA porous structure. This honey-comb-like network is not interconnected, and forms numerous pores with dead ends, helping to trap HB-EGF, and leading to a slow sustained release of HB-EGF from the scaffolds. After subcritical CO2 exposure, loosely attached HB-EGF flakes were replaced by crystals partially embedded into the PGA fibers. The slow depressurization that was applied during subcritical CO2 exposure avoided deformities of the PGA fibers (foaming), allowing preservation of the mechanical properties of the scaffolds. In order to demonstrate that PLLA coating increases the surface areas for
Y. Liu et al. / Biomaterials 103 (2016) 150e159
157
Fig. 7. Histologic examination and villous height measurement. (A) Native rat intestine; (B) TEI produced on PGA/PLLA scaffold; (C) TEI produced on PGA/PLLA/HB-EGF scaffold; and (D) TEI produced on PGA/PLLA/HB-EGF/CO2 scaffold. Periodic acid Schiff (PAS) staining indicates the presence of goblet cells. Scale bar ¼ 200 mm. D) Villous height in native rat intestine and in TEI was measured using ImageJ software (E). *p < 0.05 as determined by one-way ANOVA.
Fig. 8. Characterization of epithelial cell linages in TEI. TEI was successfully produced from all three groups of scaffolds and was subjected to immunofluorescent staining. Representative images from a PGA/PLLA/HB-EGF/CO2 scaffold include: (A) lysozyme (LYSO) staining to demonstrate the presence of Paneth cells; (B) chromonogranin A (CHGA) staining to demonstrate the presence of enteroendocrine cells; (C) villin staining to demonstrate the brush border of enterocytes; (D) desmin and a-smooth muscle actin (a-SMA) double-staining to demonstrate smooth muscle cells. The insert demonstrates intestinal subepithelial myofibroblasts (ISEMFs) located around the crypts which stain positive for aSMA only. (E) Ki67 staining to demonstrate actively proliferating crypt cells; and Scale bar: 200 mm in low magnification and 10 mm in the inserts. DAPI counterstaining demonstrates the presence of cell nuclei in all panels.
protein delivery, FITC-BSA was used to simulate HB-EGF-coating of the scaffolds due to the visibility of FITC-BSA by fluorescence microscopy. FITC-BSA was located on both PGA fibers and on the PLLA bridges between the PGA fibers in PGA/PLLA scaffolds. This significantly increased the surface area for FITC-BSA distribution. Secondly, the PLLA network strengthens the mechanical properties of the scaffolds, with increased SRS and ultimate compressive force compared to plain PGA. Without PLLA coating, plain PGA tubes cannot withstand the compressive force of intra-abdominal pressure, and the lumen collapses during in vivo incubation (data not shown). Therefore PLLA coating plays an important role in fabricating scaffolds for the production of TEI. Additional treatment with 100% ethanol is beneficial because it improves the hydrophilicity of PGA/PLLA scaffolds for the delivery of soluble HB-EGF and intestinal stem cells into the scaffolds, as shown in the hydration test (Video clip 1 is available online). Initial bolus release of HB-EGF occurred in scaffolds without subcritical CO2 exposure due to the loosely attached HB-EGF flakes on the surface of PGA fibers, which were easily washed off once the scaffolds were exposed to elution buffer. This was followed by slow release of HB-EGF, possibly resulting from the physically entrapped HB-EGF in the PLLA network, in which non-interconnected pores entrapped HB-EGF molecules. With subcritical CO2 exposure, a gradual and constant release of HB-EGF occurred. This is likely attributable to swelling of the scaffolds during the gas phase of CO2
exposure, which allows the HB-EGF molecules to be embedded onto the superficial surface of the PGA fibers and into the PLLA pores and bridges. The crystal-like HB-EGF on HB-EGF-embedded PGA fibers with subcritical CO2 exposure might contribute to the relatively faster release at earlier time points, while HB-EGF embedded in PGA fibers and entrapped in the PLLA micro-porous network might contribute to the constant slow release at later time points. Subcritical CO2 exposure did not affect the bioactivity of HB-EGF, consistent with growth factors released from PLGA microspheres sintered with subcritical CO2 [27]. Culturing of freshly isolated ISC-containing crypts on scaffolds demonstrated that cell density was higher on HB-EGF-embedded scaffolds, with or without CO2 exposure, compared to that on non-HB-EGF-embedded scaffolds on day 3. This might be attributed to the faster release of HB-EGF from the scaffolds at earlier time points. Since the medium was changed every two days, the concentration of HB-EGF in the medium on day 5 was reflective of the slow release of HB-EGF from the scaffolds at later time points. HB-EGF-embedded scaffolds in the presence of subcritical CO2 exposure maintained a relatively high concentration of HB-EGF in the medium on day 5. This lead to higher cell density compared to HB-EGF-embedded scaffolds without subcritical CO2 exposure. In vivo incubation of crypt-seeded, non-HB-EGF-embedded scaffolds produced neomucosa with short disorganized villi. In vivo incubation of crypt-seeded, HB-EGF-embedded scaffolds without
158
Y. Liu et al. / Biomaterials 103 (2016) 150e159
subcritical CO2 exposure formed a neomucosa with longer and more organized villi. This may be attributed to an initial bolus release of HB-EGF at the implantation site that stimulated the proliferation and differentiation of the seeded cells, however, an effective concentration of HB-EGF could not be maintained over time. In vivo incubation of crypt-seeded, HB-EGF-embedded scaffolds followed by subcritical CO2 exposure demonstrated neomucosa with the longest and most well organized villi, similar to the mucosa of native intestine. This might be attributed to the constant release of HB-EGF that maintained an effective dose at the implantation site. Immunofluorescent staining confirmed the presence of all epithelial cell lineages in the neomucosa produced in all three scaffold groups. The doses selected for the current in intro and in vivo studies were based on pilot studies. Higher doses of HB-EGF (10 mg) were chosen for in vitro release kinetics studies. This is because a complete medium change was used to elute HB-EGF from the scaffolds, with ELISA used to detect the released HB-EGF. In order to obtain detectable concentrations of HB-EGF at the later time points, higher amounts of HB-EGF (10 mg) were needed initially. This dose was also used for in vivo pilot studies. Dose dependent effects were observed for 2 mg and 10 mg (unpublished data), but the 2 mg dose was efficacious compared to non-HB-EGF-embedded scaffolds. Therefore, the lower efficacious dose (2 mg) was chosen for in vivo implantation studies. Future studies will focus on determination of the optimal dose of HB-EGF for the production of TEI in vivo. 5. Conclusions In conclusion, HB-EGF embedded into PGA/PLLA scaffolds via subcritical CO2 exposure is slowly released in an active form, and can stimulate the proliferation of crypts in vitro and the formation of intestinal neomucosa in vivo. Subcritical CO2 exposure does not deform the PGA/PLLA scaffolds and does not impair the mechanical properties of the scaffolds. The development of biomimetic scaffolds which can release growth factors stimulating a resident population of stem and progenitor cells may improve the formation of new tissue structures such as TEI. The methodology described here may lead to improved methods of producing tissue engineered intestine to treat patients with short bowel syndrome in the future.
[6] [7] [8] [9]
[10]
[11]
[12] [13]
[14]
[15]
[16]
[17] [18]
[19] [20]
[21]
[22]
[23] [24]
Disclosure statement No competing financial interests exist.
[25]
Acknowledgements
[26]
This study was supported by grants from the Research Institute at Nationwide Children's Hospital (82002914), the Ohio Third Frontier, and NIH R43DK107168. The authors thank Cynthia McAllister, Patricia Craig, and Melanie Herring for their assistance in sample preparation for histological analyses.
[27]
[28]
[29]
References [30] [1] D.C. Chen, J.R. Avansino, V.G. Agopian, V.D. Hoagland, J.D. Woolman, S. Pan, et al., Comparison of polyester scaffolds for bioengineered intestinal mucosa, Cells Tissues Organs 184 (2006) 154e165. [2] D.E. Levin, E.R. Barthel, A.L. Speer, F.G. Sala, X. Hou, Y. Torashima, et al., Human tissue-engineered small intestine forms from postnatal progenitor cells, J. Pediatr. Surg. 48 (2013) 129e137. [3] F.G. Sala, S.M. Kunisaki, E.R. Ochoa, J. Vacanti, T.C. Grikscheit, Tissue-engineered small intestine and stomach form from autologous tissue in a preclinical large animal model, J. Surg. Res. 156 (2009) 205e212. [4] F.G. Sala, J.A. Matthews, A.L. Speer, Y. Torashima, E.R. Barthel, T.C. Grikscheit, A multicellular approach forms a significant amount of tissue-engineered small intestine in the mouse, Tissue Eng. Part A 17 (2011) 1841e1850. [5] B. Wulkersdorfer, K.K. Kao, V.G. Agopian, J.C. Dunn, B.M. Wu, M. Stelzner,
[31]
[32]
[33]
[34]
Growth factors adsorbed on polyglycolic acid mesh augment growth of bioengineered intestinal neomucosa, J. Surg. Res. 169 (2011) 169e178. J.S. Thompson, M.A. Tempero, J.L. Haun, J.A. Vanderhoof, The importance of luminal factors in neomucosal growth, J. Surg. Res. 40 (1986) 126e132. J.S. Thompson, J.G. Sharp, S.K. Saxena, K.G. McCullagh, Stimulation of neomucosal growth by systemic urogastrone, J. Surg. Res. 42 (1987) 402e410. Y. Kato, D. Yu, J.R. Lukish, M.Z. Schwartz, Influence of luminal hepatocyte growth factor on small intestine mucosa in vivo, J. Surg. Res. 71 (1997) 49e53. M.Z. Schwartz, R.B. Storozuk, Influence of epidermal growth factor on intestinal function in the rat: comparison of systemic infusion versus luminal perfusion, Am. J. Surg. 155 (1988) 18e22. C.E. Shin, M.A. Helmrath, R.A. Falcone Jr., J.W. Fox, K.R. Duane, C.R. Erwin, et al., Epidermal growth factor augments adaptation following small bowel resection: optimal dosage, route, and timing of administration, J. Surg. Res. 77 (1998) 11e16. M.Z. Schwartz, Y. Kato, D. Yu, J.R. Lukish, Growth-factor enhancement of compromised gut function following massive small-bowel resection, Pediatr. Surg. Int. 16 (2000) 174e175. Y. Kato, D. Yu, M.Z. Schwartz, Enhancement of intestinal adaptation by hepatocyte growth factor, J. Pediatr. Surg. 33 (1998) 235e239. D.A. Lloyd, T. Ansari, S. Shurey, V. Maquet, P.D. Sibbons, A.R. Boccaccini, et al., Prolonged maintenance of neointestine using subcutaneously implanted tubular scaffolds in a rat model, Transplant. Proc. 38 (2006) 3097e3099. F.G. Rocha, C.A. Sundback, N.J. Krebs, J.K. Leach, D.J. Mooney, S.W. Ashley, et al., The effect of sustained delivery of vascular endothelial growth factor on angiogenesis in tissue-engineered intestine, Biomaterials 29 (2008) 2884e2890. A. Ramsanahie, M.S. Duxbury, T.C. Grikscheit, A. Perez, D.B. Rhoads, J. GardnerThorpe, et al., Effect of GLP-2 on mucosal morphology and SGLT1 expression in tissue-engineered neointestine, Am. J. Physiol. Gastrointest. Liver Physiol. 285 (2003) G1345eG1352. A. Perez, T.C. Grikscheit, R.S. Blumberg, S.W. Ashley, J.P. Vacanti, E.E. Whang, Tissue-engineered small intestine: ontogeny of the immune system, Transplantation 74 (2002) 619e623. R. Langer, J.P. Vacanti, Tissue engineering, Science 260 (1993) 920e926. L.G. Cima, J.P. Vacanti, C. Vacanti, D. Ingber, D. Mooney, R. Langer, Tissue engineering by cell transplantation using degradable polymer substrates, J. Biomech. Eng. 113 (1991) 143e151. J.E. Babensee, L.V. McIntire, A.G. Mikos, Growth factor delivery for tissue engineering, Pharm. Res. 17 (2000) 497e504. M.H. Sheridan, L.D. Shea, M.C. Peters, D.J. Mooney, Bioabsorbable polymer scaffolds for tissue engineering capable of sustained growth factor delivery, J. Control. Release Off. J. Control. Release Soc. 64 (2000) 91e102. J.M. Kanczler, J. Barry, P. Ginty, S.M. Howdle, K.M. Shakesheff, R.O. Oreffo, Supercritical carbon dioxide generated vascular endothelial growth factor encapsulated poly(DL-lactic acid) scaffolds induce angiogenesis in vitro, Biochem. Biophys. Res. Commun. 352 (2007) 135e141. H.M. Burt, J.K. Jackson, S.K. Bains, R.T. Liggins, A.M. Oktaba, A.L. Arsenault, et al., Controlled delivery of taxol from microspheres composed of a blend of ethylene-vinyl acetate copolymer and poly (d,l-lactic acid), Cancer Lett. 88 (1995) 73e79. C. Bitz, E. Doelker, Influence of the preparation method on residual solvents in biodegradable microspheres, Int. J. Pharm. (Amsterdam) 131 (1996) 171e181. C. Thomasin, P. Johansen, R. Alder, R. Bemsel, G. Hottinger, H. Altorfer, et al., A contribution to overcoming the problem of residual solvents in biodegradable microspheres prepared by coacervation, Eur. J. Pharm. Biopharm. 42 (1996) 16e24. T.L. Sproule, J.A. Lee, H. Li, J.J. Lannutti, D.L. Tomasko, Bioactive polymer surfaces via supercritical fluids, J. Supercrit. Fluids 28 (2004) 241e248. H.M. Powell, O. Ayodeji, T.L. Summerfield, D.M. Powell, D.A. Kniss, D.L. Tomasko, et al., Chemotherapeutic implants via subcritical CO2 modification, Biomaterials 28 (2007) 5562e5569. N.H. Dormer, V. Gupta, A.M. Scurto, C.J. Berkland, M.S. Detamore, Effect of different sintering methods on bioactivity and release of proteins from PLGA microspheres, Mater. Sci. Eng. C Mater. Biol. Appl. 33 (2013) 4343e4351. J. Feng, O.N. El-Assal, G.E. Besner, Heparin-binding EGF-like growth factor (HB-EGF) and necrotizing enterocolitis, Semin. Pediatr. Surg. 14 (2005) 167e174. J. Feng, G.E. Besner, Heparin-binding epidermal growth factor-like growth factor promotes enterocyte migration and proliferation in neonatal rats with necrotizing enterocolitis, J. Pediatr. Surg. 42 (2007) 214e220. G. Xia, A.E. Martin, M.P. Michalsky, G.E. Besner, Heparin-binding EGF-like growth factor preserves crypt cell proliferation and decreases bacterial translocation after intestinal ischemia/reperfusion injury, J. Pediatr. Surg. 37 (2002) 1081e1087 discussion -7. J. Feng, O.N. El-Assal, G.E. Besner, Heparin-binding epidermal growth factorlike growth factor reduces intestinal apoptosis in neonatal rats with necrotizing enterocolitis, J. Pediatr. Surg. 41 (2006) 742e747 discussion -7. M.P. Michalsky, A. Kuhn, V. Mehta, G.E. Besner, Heparin-binding EGF-like growth factor decreases apoptosis in intestinal epithelial cells in vitro, J. Pediatr. Surg. 36 (2001) 1130e1135. C.L. Chen, X. Yu, I.O. James, H.Y. Zhang, J. Yang, A. Radulescu, et al., intestinal stem cells from injury in a rat model of necrotizing enterocolitis, Lab. Investig J. Tech. Methods Pathol. 92 (2012) 331e344. L. Boomer, Y. Liu, N. Mahler, J. Johnson, K. Zak, T. Nelson, et al., Scaffolding for
Y. Liu et al. / Biomaterials 103 (2016) 150e159
[35]
[36]
[37]
[38]
[39]
[40]
[41]
[42]
[43]
[44]
[45]
[46]
[47]
[48]
[49]
[50]
[51]
challenging environments: materials selection for tissue engineered intestine, J. Biomed. Mater. Res. Part A 102 (2014) 3795e3802. F.J. Chaparro, M.E. Matusicky, M.J. Allen, J.J. Lannutti, Biomimetic microstructural reorganization during suture retention strength evaluation of electrospun vascular scaffolds, J. Biomed. Mater. Res. Part B Appl. Biomater. (2015), http://dx.doi.org/10.1002/jbm.b.33493. S. Drilling, J. Gaumer, J. Lannutti, Fabrication of burst pressure competent vascular grafts via electrospinning: effects of microstructure, J. Biomed. Mater. Res. Part A 88 (2009) 923e934. B. Rath, J. Nam, T.J. Knobloch, J.J. Lannutti, S. Agarwal, Compressive forces induce osteogenic gene expression in calvarial osteoblasts, J. Biomech. 41 (2008) 1095e1103. Y. Liu, T. Rager, J. Johnson, J. Enmark, G.E. Besner, Enriched intestinal stem cell seeding improves the architecture of tissue-engineered intestine, Tissue Eng. Part C. Methods 21 (2015) 816e824. Y. Liu, X.Z. Shu, S.D. Gray, G.D. Prestwich, Disulfide-crosslinked hyaluronangelatin sponge: growth of fibrous tissue in vivo, J. Biomed. Mater. Res. Part A 68 (2004) 142e149. C.L. Watson, M.M. Mahe, J. Munera, J.C. Howell, N. Sundaram, H.M. Poling, et al., An in vivo model of human small intestine using pluripotent stem cells, Nat. Med. 20 (2014) 1310e1314. N.H. Dormer, C.J. Berkland, M.S. Detamore, Emerging techniques in stratified designs and continuous gradients for tissue engineering of interfaces, Ann. Biomed. Eng. 38 (2010) 2121e2141. M. Singh, C. Berkland, M.S. Detamore, Strategies and applications for incorporating physical and chemical signal gradients in tissue engineering, Tissue Eng. Part B Rev. 14 (2008) 341e366. M.J. Yaszemski, R.G. Payne, W.C. Hayes, R.S. Langer, T.B. Aufdemorte, A.G. Mikos, The ingrowth of new bone tissue and initial mechanical properties of a degrading polymeric composite scaffold, Tissue Eng. 1 (1995) 41e52. R.A. Giordano, B.M. Wu, S.W. Borland, L.G. Cima, E.M. Sachs, M.J. Cima, Mechanical properties of dense polylactic acid structures fabricated by three dimensional printing, J. Biomater. Sci. Polym. Ed. 8 (1996) 63e75. G.G. Giordano, R.C. Thomson, S.L. Ishaug, A.G. Mikos, S. Cumber, C.A. Garcia, et al., Retinal pigment epithelium cells cultured on synthetic biodegradable polymers, J. Biomed. Mater. Res. 34 (1997) 87e93. W.D. Holder Jr., H.E. Gruber, A.L. Moore, C.R. Culberson, W. Anderson, K.J. Burg, et al., Cellular ingrowth and thickness changes in poly-L-lactide and polyglycolide matrices implanted subcutaneously in the rat, J. Biomed. Mater. Res. 41 (1998) 412e421. K.M. Davis-Fleischer, G.E. Besner, Structure and function of heparin-binding EGF-like growth factor (HB-EGF), Front. Biosci. J. Virtual Libr. 3 (1998) d288ed299. P.D. Ellis, K.M. Hadfield, J.C. Pascall, K.D. Brown, Heparin-binding epidermalgrowth-factor-like growth factor gene expression is induced by scrapewounding epithelial cell monolayers: involvement of mitogen-activated protein kinase cascades, Biochem. J. 354 (2001) 99e106. R.K. Cribbs, P.A. Harding, M.H. Luquette, G.E. Besner, Endogenous production of heparin-binding EGF-like growth factor during murine partial-thickness burn wound healing, J. Burn Care Rehabil. 23 (2002) 116e125. K. Jin, X.O. Mao, Y. Sun, L. Xie, L. Jin, E. Nishi, et al., Heparin-binding epidermal growth factor-like growth factor: hypoxia-inducible expression in vitro and stimulation of neurogenesis in vitro and in vivo, J. Neurosci. Off. J. Soc. Neurosci. 22 (2002) 5365e5373. T. Homma, M. Sakai, H.F. Cheng, T. Yasuda, R.J. Coffey Jr., R.C. Harris, Induction of heparin-binding epidermal growth factor-like growth factor mRNA in rat kidney after acute injury, J. Clin. Investig. 96 (1995) 1018e1025.
159
[52] G.D. Frank, M. Mifune, T. Inagami, M. Ohba, T. Sasaki, S. Higashiyama, et al., Distinct mechanisms of receptor and nonreceptor tyrosine kinase activation by reactive oxygen species in vascular smooth muscle cells: role of metalloprotease and protein kinase C-delta, Mol. Cell. Biol. 23 (2003) 1581e1589. [53] Y. Kayanoki, S. Higashiyama, K. Suzuki, M. Asahi, S. Kawata, Y. Matsuzawa, et al., The requirement of both intracellular reactive oxygen species and intracellular calcium elevation for the induction of heparin-binding EGF-like growth factor in vascular endothelial cells and smooth muscle cells, Biochem. Biophys. Res. Commun. 259 (1999) 50e55. [54] S. Kiso, S. Kawata, S. Tamura, Y. Inui, Y. Yoshida, Y. Sawai, et al., Liver regeneration in heparin-binding EGF-like growth factor transgenic mice after partial hepatectomy, Gastroenterology 124 (2003) 701e707. [55] M. Marikovsky, K. Breuing, P.Y. Liu, E. Eriksson, S. Higashiyama, P. Farber, et al., Appearance of heparin-binding EGF-like growth factor in wound fluid as a response to injury, Proc. Natl. Acad. Sci. U. S. A. 90 (1993) 3889e3893. [56] D.W. McCarthy, M.T. Downing, D.R. Brigstock, M.H. Luquette, K.D. Brown, M.S. Abad, et al., Production of heparin-binding epidermal growth factor-like growth factor (HB-EGF) at sites of thermal injury in pediatric patients, J. Investig. Dermatol. 106 (1996) 49e56. [57] T.T. Junttila, M. Sundvall, J.A. Maatta, K. Elenius, Erbb4 and its isoforms: selective regulation of growth factor responses by naturally occurring receptor variants, Trends Cardiovasc. Med. 10 (2000) 304e310. [58] E. Nishi, A. Prat, V. Hospital, K. Elenius, M. Klagsbrun, N-arginine dibasic convertase is a specific receptor for heparin-binding EGF-like growth factor that mediates cell migration, EMBO J. 20 (2001) 3342e3350. [59] Y. Su, J. Yang, G.E. Besner, HB-EGF promotes intestinal restitution by affecting integrin-extracellular matrix interactions and intercellular adhesions, Growth factors 31 (2013) 39e55. [60] D.V. Rocourt, V.B. Mehta, G.E. Besner, Heparin-binding EGF-like growth factor decreases inflammatory cytokine expression after intestinal ischemia/reperfusion injury, J. Surg. Res. 139 (2007) 269e273. [61] D.V. Rocourt, V.B. Mehta, D. Wu, G.E. Besner, Heparin-binding EGF-like growth factor decreases neutrophil-endothelial cell interactions, J. Surg. Res. 141 (2007) 262e266. [62] V.B. Mehta, G.E. Besner, HB-EGF promotes angiogenesis in endothelial cells via PI3-kinase and MAPK signaling pathways, Growth Factors 25 (2007) 253e263. [63] D. Chalothorn, S.M. Moore, H. Zhang, S.W. Sunnarborg, D.C. Lee, J.E. Faber, Heparin-binding epidermal growth factor-like growth factor, collateral vessel development, and angiogenesis in skeletal muscle ischemia, Arterioscler. Thromb. Vasc. Biol. 25 (2005) 1884e1890. [64] Y. Zhou, D. Brigstock, G.E. Besner, Heparin-binding EGF-like growth factor is a potent dilator of terminal mesenteric arterioles, Microvasc. Res. 78 (2009) 78e85. [65] O.N. El-Assal, A. Radulescu, G.E. Besner, Heparin-binding EGF-like growth factor preserves mesenteric microcirculatory blood flow and protects against intestinal injury in rats subjected to hemorrhagic shock and resuscitation, Surgery 142 (2007) 234e242. [66] X. Yu, A. Radulescu, C.L. Chen, I.O. James, G.E. Besner, Heparin-binding EGFlike growth factor protects pericytes from injury, J. Surg. Res. 172 (2012) 165e176. [67] X. Yu, A. Radulescu, N. Zorko, G.E. Besner, Heparin-binding EGF-like growth factor increases intestinal microvascular blood flow in necrotizing enterocolitis, Gastroenterology 137 (2009) 221e230. [68] Y. Zhou, G.E. Besner, Heparin-binding epidermal growth factor-like growth factor is a potent neurotrophic factor for PC12 cells, Neuro Signals 18 (2010) 141e151.