Acta Biomaterialia xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Acta Biomaterialia journal homepage: www.elsevier.com/locate/actabiomat
Performance of an in situ formed bioactive hydrogel dressing from a PEG-based hyperbranched multifunctional copolymer Yixiao Dong a, Waqar U. Hassan a, Robert Kennedy a, Udo Greiser a, Abhay Pandit c, Yolanda Garcia b,⇑, Wenxin Wang a,⇑ a b c
The Charles Institute of Dermatology, School of Medicine and Medical Science, University College Dublin, Dublin, Ireland Anatomy Department, National University of Ireland, Galway, Ireland Network of Excellence for Functional Biomaterials, National University of Ireland, Galway, Ireland
a r t i c l e
i n f o
Article history: Received 31 July 2013 Received in revised form 11 December 2013 Accepted 17 December 2013 Available online xxxx Keywords: In situ cross-link PEG based hyperbranched polymer Hydrogel dressing Adipose derived stem cells Wound healing
a b s t r a c t Hydrogel dressings have been widely used for wound management due to their ability to maintain a hydrated wound environment, restore the skin’s physical barrier and facilitate regular dressing replacement. However, the therapeutic functions of standard hydrogel dressings are restricted. In this study, an injectable hybrid hydrogel dressing system was prepared from a polyethylene glycol (PEG)-based thermoresponsive hyperbranched multiacrylate functional copolymer and thiol-modified hyaluronic acid in combination with adipose-derived stem cells (ADSCs). The cell viability, proliferation and metabolic activity of the encapsulated ADSCs were studied in vitro, and a rat dorsal full-thickness wound model was used to evaluate this bioactive hydrogel dressing in vivo. It was found that long-term cell viability could be achieved for both in vitro (21 days) and in vivo (14 days) studies. With ADSCs, this hydrogel system prevented wound contraction and enhanced angiogenesis, showing the potential of this system as a bioactive hydrogel dressing for wound healing. Ó 2013 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.
1. Introduction Wound dressing-based treatments are the main theapeutic approach for both acute and chronic wounds. Specific dressings have been developed over the years for different types and physiological conditions of wounds [1,2]. Traditional dressings were mainly used to absorb the exudates, to keep the wound bed dry and to prevent bacterial contamination, while modern wound dressings focus more on creating an optimum moist environment to promote wound repair [3,4]. An appropriate amount of wound exudate not only supplies the wound bed with nutrients, but provides favorable conditions for the recruitment and migration of certain cells [5–7]. Among different modern wound dressings, hydrogel dressings made from swellable polymeric hydrophilic materials can maintain a moist wound environment, provide autolytic debridement and also cool the wound surface, leading to pain relief for patients [8–10]. Current commercially available hydrogel dressings include Nu-gelÒ wound dressing, PurilonÒ gel dressing, SAF-GelÒ hydrating dermal wound dressing, Curagel™ hydrogel wound dressing, and CarrasynÒ gel wound dressing [11]. Although
⇑ Corresponding authors. Tel.: +353 1 7166341. E-mail addresses: [email protected] (Y. Garcia), [email protected] (W. Wang).
significant progress has been made in the development of modern wound dressings, promoting the healing process is still restricted, especially when compared to cellular skin substitutes. Stem cells such as bone marrow-derived stem cells (BMSCs) [12–15] and adipose-derived stem cells (ADSCs) [16–19] have been widely studied for wound-healing applications. Most of these studies have focused on developing skin substitutes in which the stem cells differentiate into epidermal cell phenotypes, regulate the level of cytokines and growth factors around the wound site and accelerate wound healing [15,20,21]. In the past few years, increasing evidence has emerged demonstrating that the paracrine effect of stem cells can play key roles in promoting the healing process. Conditioned media for both BMSCs and ADSCs have been reported to enhance angiogenesis, epithelialization, and affect recruitment or proliferation of macrophages and endothelial progenitor cells during the healing process [22–24]. Furthermore, in comparison to BMSCs, ADSCs removed via liposuction are much easier to access because the surgical approach is considerably less painful and the yield of ADSCs from adipose tissue is generally 40-fold higher [25]. In addition, the isolated cells can be cryopreserved, maintaining all their properties intact for up to 6 months [26], which provides a good potential for ADSCs to become integrated into an off-the-shelf product. Ideally, the next step is to develop a hydrogel dressing with specific therapeutic functions that is also able to deliver
1742-7061/$ - see front matter Ó 2013 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.actbio.2013.12.045
Please cite this article in press as: Dong Y et al. Performance of an in situ formed bioactive hydrogel dressing from a PEG-based hyperbranched multifunctional copolymer. Acta Biomater (2014), http://dx.doi.org/10.1016/j.actbio.2013.12.045
2
Y. Dong et al. / Acta Biomaterialia xxx (2014) xxx–xxx
therapeutic agents such as growth factors or stem cells (i.e. preventing unwanted inflammatory reactions and providing an ideal wound-healing environment). Recently, in our group, an in situ cross-linked hydrogel (i.e. P-SH-HA) cell-delivery system was prepared from a polyethylene glycol (PEG)-based hyperbranched multifunctional copolymer and a modified extracellular matrix (ECM) biopolymer, hyaluronic acid [27]. A thermoresponsive hyperbranched copolymer, poly(ethylene glycol) methyl ether methacrylate-co-2-(2-methoxyethoxy) ethyl methacrylateco-poly(ethylene glycol) diacrylate (PEGMEMA-MEO2MA-PEGDA), with multiacrylate functionality was developed by a one-step in situ deactivation enhanced atom transfer radical polymerization (DE-ATRP) approach. As shown in Scheme 1, at room temperature, the polymer solution can be easily mixed with cell suspension and thiolated hyaluronic acid (HA-SH), while at body temperature physical cross-linked hydrogel forms rapidly on the wound site. Furthermore, a chemical gelation between polymer and HA-SH occurs within a short time to achieve a stable hydrogel with enhanced mechanical properties. Hyaluronic acid (HA) is a major polysaccharide component of the ECM, and plays important roles in regulating cellular behaviors, angiogenesis and wound healing; moreover recent reports have shown that HA hydrogel can retain stem cell potency in 3-D culture [28–30]. Therefore, the HA-SH was used as chemical cross-linker for gel formation. As an injectable system, this hydrogel can be easily applied to any wound size, shape or cavity, is less invasive than other approaches and minimizes patient discomfort. Upon gelation, a 3-D water-containing polymer network forms via physical and chemical cross-linking that can mimic the tissue microenvironment for cell growth. In addition, its relatively slow degradable and non-adherent properties provide this hydrogel dressing with the advantage of easy removal at dressing changes without pain or causing further trauma. In this study, we evaluated the performance of this injectable bioactive hydrogel dressing system. The cell viability, proliferation and metabolic activity have been analyzed in vitro; and cell retention, wound closure, inflammatory response and angiogenesis were evaluated in vivo by using a rat dorsal full-thickness wound model.
2. Materials and methods 2.1. Materials and reagents The monomers of poly(ethylene glycol) methyl ether methacrylate (PEGMEMA, Mn = 475 g mol1), 2-(2-methoxyethoxy) ethyl methacrylate (MEO2MA), and poly(ethylene glycol) diacrylate (PEGDA, Mn = 258 g mol1) were purchased from Sigma–Aldrich. Ethyl 2-bromoisobutyrate (EtBriB, 98%, Aldrich), bis(2-dimethylaminoethyl) methylamine (99%, Aldrich), copper(II) chloride (CuCl2, 97%, Aldrich), L-ascorbic acid (L-AA, 99%, Aldrich), butanone (99%, HPLC grade, Aldrich), diethyl ether (99%) and hexane (95%, Aldrich) were used as received. Thiolated hyaluronic acid (HA-SH, HyStem™, Glycosan) was purchased from BioTime Inc., Alameda, CA. 2.2. Polymer synthesis and hydrogel fabrication The thermoresponsive copolymer of PEGMEMA-MEO2MA-PEGDA was synthesized by the copolymerization of PEGMEMA, MEO2MA and PEGDA via an in situ DE-ATRP approach as previously described [27]. Briefly, PEGMEMA (15 equiv), MEO2MA (65 equiv), PEGDA (20 equiv), EtBriB (1 equiv), CuCl2 (0.25 equiv) and bis(2dimethylaminoethyl) methylamine (0.25 equiv) were all added into a two-necked round-bottom flask (the volume ratio of total monomers to butanone solvent was adjusted to 1:2). The mixture was completely dissolved, purged with argon for 30 min to remove oxygen, and then L-AA (0.125 equiv) was added to the polymerization solution under argon. The reaction was performed at 50 °C in an oil bath for about 6 h and monitored by gel permeation chromatography (GPC). The copolymer products were purified and characterized by GPC, 1H nuclear magnetic resonance (NMR) spectrometry and UV spectrophotometry as previously described [27]. Briefly, the monomer was removed by dropping the solution into a large excess of hexane/diethyl ether solution (1:1 v/v). The precipitated mixture was dissolved in deionized water and dialyzed (MWCO 6–8 kDa) for 4 days in the dark at 4 °C to remove residual monomer and copper. The molecular weight (Mw) of the copolymer was 72 kDa with a polydispersity index of 3.97.
Scheme 1. Schematic of in situ formed bioactive hydrogel dressing prepared from a PEG-based hyperbranched multifunctional copolymer of PEGMEMA-MEO2MA-PEGDA in combination with thiolated hyaluronic acid (HA-SH) and rat ADSCs. At room temperature, the polymer solution can be easily mixed with cell suspension and HA-SH solution; while at body temperature physical cross-linked hydrogel forms rapidly to seal the wound bed; chemical cross-linking occurs within a few minutes between the vinyl group on the polymer and the thiol group on the HA-SH to achieve a stable hydrogel which provides a 3-D microenvironment to support rADSC growth and secretion in the wound.
Please cite this article in press as: Dong Y et al. Performance of an in situ formed bioactive hydrogel dressing from a PEG-based hyperbranched multifunctional copolymer. Acta Biomater (2014), http://dx.doi.org/10.1016/j.actbio.2013.12.045
Y. Dong et al. / Acta Biomaterialia xxx (2014) xxx–xxx
The content of vinyl functional group of the copolymer was 12 mol.% calculated by 1H NMR analysis and the lower critical solution temperature of this copolymer was determined to be 32 °C. For hydrogel fabrication, copolymer of PEGMEMA-MEO2MAPEGDA was dissolved in Dulbecco’s Modified Eagle’s Medium (DMEM) serum-free culture medium, mixed gently by pipetting into the ADSC suspension (in serum-free medium) and commercially available HA-SH (HyStem™). The final concentration of the copolymer was 5% (w/v), HA-SH was 0.5% (w/v) and ADSCs was 1 million cells ml1. Gelation occurs within 8 min at 37 °C through Michael-type addition between vinyl groups on copolymer and thiol groups on the HA-SH. 2.3. Cell viability, proliferation and metabolic activity analysis of ADSC-embedded hydrogel Rat ADSCs (rADSCs) were extracted from rat adipose tissue. The tissue was digested using 0.025% (w/v) type I collagenase for 1 h with shaking at 37 °C, and the reaction was stopped by addition of complete medium (DMEM, 10% FBS, 1% P/S). The stromal fraction was collected by centrifugation for 5 min at 1,200 rpm, resuspended and filtered on a cell strainer (70 lm mesh size, BD Sciences, Switzerland). After 24 h of incubation at 37 °C in a humidified atmosphere of 5% CO2, cells were washed in order to eliminate the contaminant cells (e.g. blood cells and adipocytes). Complete culture medium was changed every 2–3 days and cells were maintained as subconfluent. The cells’ cell-surface protein profile is positive for CD90 and CD44 and negative for CD31 and CD34 as demonstrated in flow-cytometry based experiments (FACSCanto, BD Biosciences); meanwhile, adipogenic, chondrogenic and osteogenic differentiation assays were used to confirm the stemness (Fig. S1, supporting information). For the present in vitro and in vivo studies, only rADSCs of passages 2–4 were used. Cell-embedded hydrogels (40 ll each) were prepared on a hydrophobic Teflon tape surface, incubated at 37 °C for 20 min for complete gelation, and then transferred into each well of a 12-well culture plate with 2 ml DMEM culture medium (10% FBS, 1% P/S). Hydrogels were incubated at 37 °C in a humidified atmosphere of 5% CO2 for 3, 7, 14 and 21 days with medium change every 2–3 days. For each time point, LIVE/DEADÒ (Molecular Probes), alamarBlueÒ (Invitrogen) and PicoGreenÒ (Molecular Probes) assays were used to determine the viability, metabolic activity and proliferation of the embedded rADSCs, respectively. LIVE/DEAD and PicoGreen assay were performed with the recommended procedure. For quantitative analysis of the LIVE/DEAD result, 15 micrographs of 100 magnification were taken randomly from three hydrogel samples at each time point using an Olympus 1X81 inverted microscope. Live and dead cells were counted using ImageProÒ Plus software (Media Cybernetics, USA), and the cell viability was calculated from the ratio between the number of live cells and the total cell number in the field. For alamarBlue assay, cell-embedded hydrogels were rinsed three times with Hanks’ Balanced Salt Solution, followed by incubation with 10% (v/v) alamarBlue in complete medium for 12 h. The reduced alamarBlue absorbance was determined by a Varioskan Flash Plate Reader and calculated using a standard approach. 2.4. In vivo model All procedures were conducted under an animal license (No. B100/4342) authorized by the Irish Department of Health and Children and were approved by the Animal Care and Research Ethics Committee of the National University of Ireland, Galway (No. 009/10(B)). Animal care followed the Standard Operating Procedures of the Animal Facility at the National Centre for Biomedical Engineering Science, NUIG.
3
Twenty-one male Sprague–Dawley rats (CDÒ IGS, CharlesRiver, UK) with body weights ranging between 275 and 325 g were used in this study. Three time periods were designated as 3, 7 and 14 days (n = 7 wounds per treatment per time point). The animals were anaesthetized with isoflurane/oxygen (5% for induction and 1–2% for maintenance). Two full-thickness square wounds (1 cm 1 cm) were created on each side of the dorsal midline at least 1 cm apart. Photographs were taken of each wound. Four different treatment groups were randomized: no treatment (NA); hydrogel alone (120 ll gelling mixture solution) (H); hydrogel + rADSCs (0.12 million cells in 120 ll gelling mixture solution) (H + C); cells alone (0.12 million cells in 30 ll serum-free medium) (C). Hydrogels with/without cells were made in situ as described above. The cells-alone group was prepared by pipetting 0.12 million rADSCs in 30 ll serum-free medium onto the wound surface. The wounds were covered by commercially available transparent film dressing (Tegaderm™ Film, 3 M), followed by medical gauze to prevent disturbance of the wounds. Analgesia was granted by the used of Buprenorphine (0.03 mg kg1 every 8 h, Vetergesic) and inflammation due to local contamination was prevented by the use of broad-spectrum antibiotics (Enrofloxacin: 5 mg kg1 once a day) for the first 3 days. 2.5. Harvesting and processing of tissue Animals were killed by CO2 asphyxiation at each time point. Gauze and film dressing were removed carefully and photographs were taken of each wound. All samples were identified by animal number, time point and positions (AL, anterior left; AR, anterior right; PL, posterior left; PR, posterior right) to avoid analytic bias. Samples were fixed in 10% formalin for 24 h, followed by automatic paraffin embedding (Leica ASP300), blocked and sectioned perpendicularly to the wound surface in 5 lm consecutive sections. The samples were stained with haematoxylin eosin (H&E), collagen type IV and CD68 immunohistochemical staining using standard protocols developed in our laboratory for subsequent stereological analysis [31]. 2.6. Analysis for in vivo study 2.6.1. Wound closure The wound areas were measured from the photographs using ImagePro Plus image analysis software. The percentage of the wound area at different time points was performed by comparing them with the wound area at day 0. 2.6.2. Wound contraction and epithelialization Wound contraction (C) and epithelialization (E) was measured on the 12.5 images with H&E staining. The distance between dermal edges at both side of the wound was measured as wound length (Wt). The percentage of epithelialization (E%) was determined by the total epithelial layer over the wound at both side (E1 + E2) divided by the original wound length (W0) (Eq. (1)). The original wound size was 1 cm and roughly 10% contraction was caused by the fixing and processing; therefore W0 was taken as 0.9 cm for the calculations here. The percentage of wound contraction (C%) was determined by the reduction of wound length divided by the original wound length (Eq. (2)). All the parameters are identified in Fig. 3a.
E% ¼
ðE1 þ E2 Þ 100 W0
ð1Þ
C% ¼
ðW 0 W t Þ 100 W0
ð2Þ
Please cite this article in press as: Dong Y et al. Performance of an in situ formed bioactive hydrogel dressing from a PEG-based hyperbranched multifunctional copolymer. Acta Biomater (2014), http://dx.doi.org/10.1016/j.actbio.2013.12.045
4
Y. Dong et al. / Acta Biomaterialia xxx (2014) xxx–xxx
Fig. 1. Cell viability, proliferation and metabolic activity analysis of rADSCs embedded in P-SH-HA hydrogels at 3, 7, 14 and 21 days. (a) Representative fluorescent images of embedded rADSCs stained with calcein AM (green) for live cells and ethidum homodimer-1 (red) for dead cells by LIVE/DEADÒ stain kit (scale bars in all cases represent 100 lm). (b) Percentage of live cells to total cell number calculated from LIVE/DEAD staining micrographs (mean ± SD, n = 3). (c) Cellular proliferation assay by PicoGreenÒ assay. Total DNA content in the hydrogel is significant increased from 7 days (mean ± SD, n = 3, P < 0.05). (d) Percentage of deduced alamarBlueÒ absorbance at 550 and 595 nm. Total cellular metabolic activity were reduced after 14 days (mean ± SD, n = 3, P < 0.05).
inverted microscope. The volume fractions of rADSCs were determined by stereological analysis using a grid mask (40 40 lm2).
Fig. 2. Percentage of wound area at 3, 7 and 14 days compared to original wound area measured by ImageProÒ Plus (Media Cybernetics, USA). ⁄Significant differences between each time point and groups (n = 7, P < 0.05).
2.6.3. Cell retention rADSCs used for in vivo study were incubated with CellTracker CM-Dil (Invitrogen) at a concentration of 4 lmol l1 for 30 min at 37 °C. The paraffin slide samples of the C and H + C groups were further stained by Hoechst 33258 (Invitrogen), and fluorescent micrographs were taken at 100 magnification (five fields of view per slide, three slides per treatment) with an Olympus 1X-81
2.6.4. Determination of inflammation Inflammatory cells, including neutrophils, lymphocytes and macrophages, were detected by micrographs at 600 magnification taken from H&E-stained slides. In addition, ED-1 (mouse monoclonal anti-CD 68 antibody, Abcam) was used to label macrophages by standard immunohistochemical staining (AlexaFluor 488 Goat anti-mouse IgG (H + L from Invitrogen) as secondary antibody. Nuclei were stained with Hoechst 33258 (Sigma). A grid mask (20 20 lm2) was applied to images with 200 magnification (three fields of view per slide, three slides per treatment) and the numbers of grid intersections that touch or traverse the nucleus of CD68+ cells were tagged. The volume fraction (VV) of each treatment was expressed as the fraction of tagged intersections (PV) in relation to the total number of intersections (PT) in the grid:
Vv ¼
PV PT
ð3Þ
2.6.5. Angiogenesis Neoangiogenesis of the wound was determined by collagen IV immunohistochemical staining (rabbit polyclonal antibody to collagen type IV, ab6586, Abcam plc, and AlexaFluor 488 goat antirabbit IgG (H + L) as secondary antibody, Invitrogen). Stereological analysis was performed as described elsewhere [31]. Surface
Please cite this article in press as: Dong Y et al. Performance of an in situ formed bioactive hydrogel dressing from a PEG-based hyperbranched multifunctional copolymer. Acta Biomater (2014), http://dx.doi.org/10.1016/j.actbio.2013.12.045
Y. Dong et al. / Acta Biomaterialia xxx (2014) xxx–xxx
density (SV) and length density (LV) were measured using a cycloid grid with known total length of cycloid lines (L) and counting the number of intersections (I) of those lines with fluorescently labelled blood vessels. Horizontal positioning grid was used for SV and the vertical positioning grid for LV (Eqs. (4) and (5), TS presents the thickness of sections). Wound volumes (V), surface area (SA) and total length (LT) were calculated according to Eqs. (6)–(8) (WA presents the wound area determined from H&E-stained slides).
I L 2 ðI=LÞ LV ¼ TS V ¼ TS WA
SV ¼ 2
5
day 14, but this trend does not show any statistical significant difference between each group because of the high standard deviation (Fig. 3b). For epithelialization measurement, it was found that the wounds were fully closed with complete re-epithelialization at 14 days in the ‘‘no treatment’’ and ‘‘cell alone’’ treatment groups; while the re-epithelialization only occurred at the edges of the wounds in the hydrogel treatment groups (with or without cells). However, because of the large skin contraction, no statistical significant difference was noted for re-epithelialization between all the groups (Fig. 3c) at each time point.
ð4Þ ð5Þ
3.4. Cell retention
ð6Þ
The cell retention ratio in the wound bed was assessed by labeling the rADSCs with CellTracker CM-Dil. The marker in this kit, which conjugates with intracellular proteins, reflects the cell survival rate in the hydrogels. After each time point, paraffin-fixed
SA ¼ SV V
ð7Þ
LT ¼ LV V
ð8Þ
2.7. Statistical analysis Comparisons between multiple groups were evaluated by oneway or two-way ANOVA and Tukey’s post-analysis method. A P-value of
Contents lists available at ScienceDirect
Acta Biomaterialia journal homepage: www.elsevier.com/locate/actabiomat
Performance of an in situ formed bioactive hydrogel dressing from a PEG-based hyperbranched multifunctional copolymer Yixiao Dong a, Waqar U. Hassan a, Robert Kennedy a, Udo Greiser a, Abhay Pandit c, Yolanda Garcia b,⇑, Wenxin Wang a,⇑ a b c
The Charles Institute of Dermatology, School of Medicine and Medical Science, University College Dublin, Dublin, Ireland Anatomy Department, National University of Ireland, Galway, Ireland Network of Excellence for Functional Biomaterials, National University of Ireland, Galway, Ireland
a r t i c l e
i n f o
Article history: Received 31 July 2013 Received in revised form 11 December 2013 Accepted 17 December 2013 Available online xxxx Keywords: In situ cross-link PEG based hyperbranched polymer Hydrogel dressing Adipose derived stem cells Wound healing
a b s t r a c t Hydrogel dressings have been widely used for wound management due to their ability to maintain a hydrated wound environment, restore the skin’s physical barrier and facilitate regular dressing replacement. However, the therapeutic functions of standard hydrogel dressings are restricted. In this study, an injectable hybrid hydrogel dressing system was prepared from a polyethylene glycol (PEG)-based thermoresponsive hyperbranched multiacrylate functional copolymer and thiol-modified hyaluronic acid in combination with adipose-derived stem cells (ADSCs). The cell viability, proliferation and metabolic activity of the encapsulated ADSCs were studied in vitro, and a rat dorsal full-thickness wound model was used to evaluate this bioactive hydrogel dressing in vivo. It was found that long-term cell viability could be achieved for both in vitro (21 days) and in vivo (14 days) studies. With ADSCs, this hydrogel system prevented wound contraction and enhanced angiogenesis, showing the potential of this system as a bioactive hydrogel dressing for wound healing. Ó 2013 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.
1. Introduction Wound dressing-based treatments are the main theapeutic approach for both acute and chronic wounds. Specific dressings have been developed over the years for different types and physiological conditions of wounds [1,2]. Traditional dressings were mainly used to absorb the exudates, to keep the wound bed dry and to prevent bacterial contamination, while modern wound dressings focus more on creating an optimum moist environment to promote wound repair [3,4]. An appropriate amount of wound exudate not only supplies the wound bed with nutrients, but provides favorable conditions for the recruitment and migration of certain cells [5–7]. Among different modern wound dressings, hydrogel dressings made from swellable polymeric hydrophilic materials can maintain a moist wound environment, provide autolytic debridement and also cool the wound surface, leading to pain relief for patients [8–10]. Current commercially available hydrogel dressings include Nu-gelÒ wound dressing, PurilonÒ gel dressing, SAF-GelÒ hydrating dermal wound dressing, Curagel™ hydrogel wound dressing, and CarrasynÒ gel wound dressing [11]. Although
⇑ Corresponding authors. Tel.: +353 1 7166341. E-mail addresses: [email protected] (Y. Garcia), [email protected] (W. Wang).
significant progress has been made in the development of modern wound dressings, promoting the healing process is still restricted, especially when compared to cellular skin substitutes. Stem cells such as bone marrow-derived stem cells (BMSCs) [12–15] and adipose-derived stem cells (ADSCs) [16–19] have been widely studied for wound-healing applications. Most of these studies have focused on developing skin substitutes in which the stem cells differentiate into epidermal cell phenotypes, regulate the level of cytokines and growth factors around the wound site and accelerate wound healing [15,20,21]. In the past few years, increasing evidence has emerged demonstrating that the paracrine effect of stem cells can play key roles in promoting the healing process. Conditioned media for both BMSCs and ADSCs have been reported to enhance angiogenesis, epithelialization, and affect recruitment or proliferation of macrophages and endothelial progenitor cells during the healing process [22–24]. Furthermore, in comparison to BMSCs, ADSCs removed via liposuction are much easier to access because the surgical approach is considerably less painful and the yield of ADSCs from adipose tissue is generally 40-fold higher [25]. In addition, the isolated cells can be cryopreserved, maintaining all their properties intact for up to 6 months [26], which provides a good potential for ADSCs to become integrated into an off-the-shelf product. Ideally, the next step is to develop a hydrogel dressing with specific therapeutic functions that is also able to deliver
1742-7061/$ - see front matter Ó 2013 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.actbio.2013.12.045
Please cite this article in press as: Dong Y et al. Performance of an in situ formed bioactive hydrogel dressing from a PEG-based hyperbranched multifunctional copolymer. Acta Biomater (2014), http://dx.doi.org/10.1016/j.actbio.2013.12.045
2
Y. Dong et al. / Acta Biomaterialia xxx (2014) xxx–xxx
therapeutic agents such as growth factors or stem cells (i.e. preventing unwanted inflammatory reactions and providing an ideal wound-healing environment). Recently, in our group, an in situ cross-linked hydrogel (i.e. P-SH-HA) cell-delivery system was prepared from a polyethylene glycol (PEG)-based hyperbranched multifunctional copolymer and a modified extracellular matrix (ECM) biopolymer, hyaluronic acid [27]. A thermoresponsive hyperbranched copolymer, poly(ethylene glycol) methyl ether methacrylate-co-2-(2-methoxyethoxy) ethyl methacrylateco-poly(ethylene glycol) diacrylate (PEGMEMA-MEO2MA-PEGDA), with multiacrylate functionality was developed by a one-step in situ deactivation enhanced atom transfer radical polymerization (DE-ATRP) approach. As shown in Scheme 1, at room temperature, the polymer solution can be easily mixed with cell suspension and thiolated hyaluronic acid (HA-SH), while at body temperature physical cross-linked hydrogel forms rapidly on the wound site. Furthermore, a chemical gelation between polymer and HA-SH occurs within a short time to achieve a stable hydrogel with enhanced mechanical properties. Hyaluronic acid (HA) is a major polysaccharide component of the ECM, and plays important roles in regulating cellular behaviors, angiogenesis and wound healing; moreover recent reports have shown that HA hydrogel can retain stem cell potency in 3-D culture [28–30]. Therefore, the HA-SH was used as chemical cross-linker for gel formation. As an injectable system, this hydrogel can be easily applied to any wound size, shape or cavity, is less invasive than other approaches and minimizes patient discomfort. Upon gelation, a 3-D water-containing polymer network forms via physical and chemical cross-linking that can mimic the tissue microenvironment for cell growth. In addition, its relatively slow degradable and non-adherent properties provide this hydrogel dressing with the advantage of easy removal at dressing changes without pain or causing further trauma. In this study, we evaluated the performance of this injectable bioactive hydrogel dressing system. The cell viability, proliferation and metabolic activity have been analyzed in vitro; and cell retention, wound closure, inflammatory response and angiogenesis were evaluated in vivo by using a rat dorsal full-thickness wound model.
2. Materials and methods 2.1. Materials and reagents The monomers of poly(ethylene glycol) methyl ether methacrylate (PEGMEMA, Mn = 475 g mol1), 2-(2-methoxyethoxy) ethyl methacrylate (MEO2MA), and poly(ethylene glycol) diacrylate (PEGDA, Mn = 258 g mol1) were purchased from Sigma–Aldrich. Ethyl 2-bromoisobutyrate (EtBriB, 98%, Aldrich), bis(2-dimethylaminoethyl) methylamine (99%, Aldrich), copper(II) chloride (CuCl2, 97%, Aldrich), L-ascorbic acid (L-AA, 99%, Aldrich), butanone (99%, HPLC grade, Aldrich), diethyl ether (99%) and hexane (95%, Aldrich) were used as received. Thiolated hyaluronic acid (HA-SH, HyStem™, Glycosan) was purchased from BioTime Inc., Alameda, CA. 2.2. Polymer synthesis and hydrogel fabrication The thermoresponsive copolymer of PEGMEMA-MEO2MA-PEGDA was synthesized by the copolymerization of PEGMEMA, MEO2MA and PEGDA via an in situ DE-ATRP approach as previously described [27]. Briefly, PEGMEMA (15 equiv), MEO2MA (65 equiv), PEGDA (20 equiv), EtBriB (1 equiv), CuCl2 (0.25 equiv) and bis(2dimethylaminoethyl) methylamine (0.25 equiv) were all added into a two-necked round-bottom flask (the volume ratio of total monomers to butanone solvent was adjusted to 1:2). The mixture was completely dissolved, purged with argon for 30 min to remove oxygen, and then L-AA (0.125 equiv) was added to the polymerization solution under argon. The reaction was performed at 50 °C in an oil bath for about 6 h and monitored by gel permeation chromatography (GPC). The copolymer products were purified and characterized by GPC, 1H nuclear magnetic resonance (NMR) spectrometry and UV spectrophotometry as previously described [27]. Briefly, the monomer was removed by dropping the solution into a large excess of hexane/diethyl ether solution (1:1 v/v). The precipitated mixture was dissolved in deionized water and dialyzed (MWCO 6–8 kDa) for 4 days in the dark at 4 °C to remove residual monomer and copper. The molecular weight (Mw) of the copolymer was 72 kDa with a polydispersity index of 3.97.
Scheme 1. Schematic of in situ formed bioactive hydrogel dressing prepared from a PEG-based hyperbranched multifunctional copolymer of PEGMEMA-MEO2MA-PEGDA in combination with thiolated hyaluronic acid (HA-SH) and rat ADSCs. At room temperature, the polymer solution can be easily mixed with cell suspension and HA-SH solution; while at body temperature physical cross-linked hydrogel forms rapidly to seal the wound bed; chemical cross-linking occurs within a few minutes between the vinyl group on the polymer and the thiol group on the HA-SH to achieve a stable hydrogel which provides a 3-D microenvironment to support rADSC growth and secretion in the wound.
Please cite this article in press as: Dong Y et al. Performance of an in situ formed bioactive hydrogel dressing from a PEG-based hyperbranched multifunctional copolymer. Acta Biomater (2014), http://dx.doi.org/10.1016/j.actbio.2013.12.045
Y. Dong et al. / Acta Biomaterialia xxx (2014) xxx–xxx
The content of vinyl functional group of the copolymer was 12 mol.% calculated by 1H NMR analysis and the lower critical solution temperature of this copolymer was determined to be 32 °C. For hydrogel fabrication, copolymer of PEGMEMA-MEO2MAPEGDA was dissolved in Dulbecco’s Modified Eagle’s Medium (DMEM) serum-free culture medium, mixed gently by pipetting into the ADSC suspension (in serum-free medium) and commercially available HA-SH (HyStem™). The final concentration of the copolymer was 5% (w/v), HA-SH was 0.5% (w/v) and ADSCs was 1 million cells ml1. Gelation occurs within 8 min at 37 °C through Michael-type addition between vinyl groups on copolymer and thiol groups on the HA-SH. 2.3. Cell viability, proliferation and metabolic activity analysis of ADSC-embedded hydrogel Rat ADSCs (rADSCs) were extracted from rat adipose tissue. The tissue was digested using 0.025% (w/v) type I collagenase for 1 h with shaking at 37 °C, and the reaction was stopped by addition of complete medium (DMEM, 10% FBS, 1% P/S). The stromal fraction was collected by centrifugation for 5 min at 1,200 rpm, resuspended and filtered on a cell strainer (70 lm mesh size, BD Sciences, Switzerland). After 24 h of incubation at 37 °C in a humidified atmosphere of 5% CO2, cells were washed in order to eliminate the contaminant cells (e.g. blood cells and adipocytes). Complete culture medium was changed every 2–3 days and cells were maintained as subconfluent. The cells’ cell-surface protein profile is positive for CD90 and CD44 and negative for CD31 and CD34 as demonstrated in flow-cytometry based experiments (FACSCanto, BD Biosciences); meanwhile, adipogenic, chondrogenic and osteogenic differentiation assays were used to confirm the stemness (Fig. S1, supporting information). For the present in vitro and in vivo studies, only rADSCs of passages 2–4 were used. Cell-embedded hydrogels (40 ll each) were prepared on a hydrophobic Teflon tape surface, incubated at 37 °C for 20 min for complete gelation, and then transferred into each well of a 12-well culture plate with 2 ml DMEM culture medium (10% FBS, 1% P/S). Hydrogels were incubated at 37 °C in a humidified atmosphere of 5% CO2 for 3, 7, 14 and 21 days with medium change every 2–3 days. For each time point, LIVE/DEADÒ (Molecular Probes), alamarBlueÒ (Invitrogen) and PicoGreenÒ (Molecular Probes) assays were used to determine the viability, metabolic activity and proliferation of the embedded rADSCs, respectively. LIVE/DEAD and PicoGreen assay were performed with the recommended procedure. For quantitative analysis of the LIVE/DEAD result, 15 micrographs of 100 magnification were taken randomly from three hydrogel samples at each time point using an Olympus 1X81 inverted microscope. Live and dead cells were counted using ImageProÒ Plus software (Media Cybernetics, USA), and the cell viability was calculated from the ratio between the number of live cells and the total cell number in the field. For alamarBlue assay, cell-embedded hydrogels were rinsed three times with Hanks’ Balanced Salt Solution, followed by incubation with 10% (v/v) alamarBlue in complete medium for 12 h. The reduced alamarBlue absorbance was determined by a Varioskan Flash Plate Reader and calculated using a standard approach. 2.4. In vivo model All procedures were conducted under an animal license (No. B100/4342) authorized by the Irish Department of Health and Children and were approved by the Animal Care and Research Ethics Committee of the National University of Ireland, Galway (No. 009/10(B)). Animal care followed the Standard Operating Procedures of the Animal Facility at the National Centre for Biomedical Engineering Science, NUIG.
3
Twenty-one male Sprague–Dawley rats (CDÒ IGS, CharlesRiver, UK) with body weights ranging between 275 and 325 g were used in this study. Three time periods were designated as 3, 7 and 14 days (n = 7 wounds per treatment per time point). The animals were anaesthetized with isoflurane/oxygen (5% for induction and 1–2% for maintenance). Two full-thickness square wounds (1 cm 1 cm) were created on each side of the dorsal midline at least 1 cm apart. Photographs were taken of each wound. Four different treatment groups were randomized: no treatment (NA); hydrogel alone (120 ll gelling mixture solution) (H); hydrogel + rADSCs (0.12 million cells in 120 ll gelling mixture solution) (H + C); cells alone (0.12 million cells in 30 ll serum-free medium) (C). Hydrogels with/without cells were made in situ as described above. The cells-alone group was prepared by pipetting 0.12 million rADSCs in 30 ll serum-free medium onto the wound surface. The wounds were covered by commercially available transparent film dressing (Tegaderm™ Film, 3 M), followed by medical gauze to prevent disturbance of the wounds. Analgesia was granted by the used of Buprenorphine (0.03 mg kg1 every 8 h, Vetergesic) and inflammation due to local contamination was prevented by the use of broad-spectrum antibiotics (Enrofloxacin: 5 mg kg1 once a day) for the first 3 days. 2.5. Harvesting and processing of tissue Animals were killed by CO2 asphyxiation at each time point. Gauze and film dressing were removed carefully and photographs were taken of each wound. All samples were identified by animal number, time point and positions (AL, anterior left; AR, anterior right; PL, posterior left; PR, posterior right) to avoid analytic bias. Samples were fixed in 10% formalin for 24 h, followed by automatic paraffin embedding (Leica ASP300), blocked and sectioned perpendicularly to the wound surface in 5 lm consecutive sections. The samples were stained with haematoxylin eosin (H&E), collagen type IV and CD68 immunohistochemical staining using standard protocols developed in our laboratory for subsequent stereological analysis [31]. 2.6. Analysis for in vivo study 2.6.1. Wound closure The wound areas were measured from the photographs using ImagePro Plus image analysis software. The percentage of the wound area at different time points was performed by comparing them with the wound area at day 0. 2.6.2. Wound contraction and epithelialization Wound contraction (C) and epithelialization (E) was measured on the 12.5 images with H&E staining. The distance between dermal edges at both side of the wound was measured as wound length (Wt). The percentage of epithelialization (E%) was determined by the total epithelial layer over the wound at both side (E1 + E2) divided by the original wound length (W0) (Eq. (1)). The original wound size was 1 cm and roughly 10% contraction was caused by the fixing and processing; therefore W0 was taken as 0.9 cm for the calculations here. The percentage of wound contraction (C%) was determined by the reduction of wound length divided by the original wound length (Eq. (2)). All the parameters are identified in Fig. 3a.
E% ¼
ðE1 þ E2 Þ 100 W0
ð1Þ
C% ¼
ðW 0 W t Þ 100 W0
ð2Þ
Please cite this article in press as: Dong Y et al. Performance of an in situ formed bioactive hydrogel dressing from a PEG-based hyperbranched multifunctional copolymer. Acta Biomater (2014), http://dx.doi.org/10.1016/j.actbio.2013.12.045
4
Y. Dong et al. / Acta Biomaterialia xxx (2014) xxx–xxx
Fig. 1. Cell viability, proliferation and metabolic activity analysis of rADSCs embedded in P-SH-HA hydrogels at 3, 7, 14 and 21 days. (a) Representative fluorescent images of embedded rADSCs stained with calcein AM (green) for live cells and ethidum homodimer-1 (red) for dead cells by LIVE/DEADÒ stain kit (scale bars in all cases represent 100 lm). (b) Percentage of live cells to total cell number calculated from LIVE/DEAD staining micrographs (mean ± SD, n = 3). (c) Cellular proliferation assay by PicoGreenÒ assay. Total DNA content in the hydrogel is significant increased from 7 days (mean ± SD, n = 3, P < 0.05). (d) Percentage of deduced alamarBlueÒ absorbance at 550 and 595 nm. Total cellular metabolic activity were reduced after 14 days (mean ± SD, n = 3, P < 0.05).
inverted microscope. The volume fractions of rADSCs were determined by stereological analysis using a grid mask (40 40 lm2).
Fig. 2. Percentage of wound area at 3, 7 and 14 days compared to original wound area measured by ImageProÒ Plus (Media Cybernetics, USA). ⁄Significant differences between each time point and groups (n = 7, P < 0.05).
2.6.3. Cell retention rADSCs used for in vivo study were incubated with CellTracker CM-Dil (Invitrogen) at a concentration of 4 lmol l1 for 30 min at 37 °C. The paraffin slide samples of the C and H + C groups were further stained by Hoechst 33258 (Invitrogen), and fluorescent micrographs were taken at 100 magnification (five fields of view per slide, three slides per treatment) with an Olympus 1X-81
2.6.4. Determination of inflammation Inflammatory cells, including neutrophils, lymphocytes and macrophages, were detected by micrographs at 600 magnification taken from H&E-stained slides. In addition, ED-1 (mouse monoclonal anti-CD 68 antibody, Abcam) was used to label macrophages by standard immunohistochemical staining (AlexaFluor 488 Goat anti-mouse IgG (H + L from Invitrogen) as secondary antibody. Nuclei were stained with Hoechst 33258 (Sigma). A grid mask (20 20 lm2) was applied to images with 200 magnification (three fields of view per slide, three slides per treatment) and the numbers of grid intersections that touch or traverse the nucleus of CD68+ cells were tagged. The volume fraction (VV) of each treatment was expressed as the fraction of tagged intersections (PV) in relation to the total number of intersections (PT) in the grid:
Vv ¼
PV PT
ð3Þ
2.6.5. Angiogenesis Neoangiogenesis of the wound was determined by collagen IV immunohistochemical staining (rabbit polyclonal antibody to collagen type IV, ab6586, Abcam plc, and AlexaFluor 488 goat antirabbit IgG (H + L) as secondary antibody, Invitrogen). Stereological analysis was performed as described elsewhere [31]. Surface
Please cite this article in press as: Dong Y et al. Performance of an in situ formed bioactive hydrogel dressing from a PEG-based hyperbranched multifunctional copolymer. Acta Biomater (2014), http://dx.doi.org/10.1016/j.actbio.2013.12.045
Y. Dong et al. / Acta Biomaterialia xxx (2014) xxx–xxx
density (SV) and length density (LV) were measured using a cycloid grid with known total length of cycloid lines (L) and counting the number of intersections (I) of those lines with fluorescently labelled blood vessels. Horizontal positioning grid was used for SV and the vertical positioning grid for LV (Eqs. (4) and (5), TS presents the thickness of sections). Wound volumes (V), surface area (SA) and total length (LT) were calculated according to Eqs. (6)–(8) (WA presents the wound area determined from H&E-stained slides).
I L 2 ðI=LÞ LV ¼ TS V ¼ TS WA
SV ¼ 2
5
day 14, but this trend does not show any statistical significant difference between each group because of the high standard deviation (Fig. 3b). For epithelialization measurement, it was found that the wounds were fully closed with complete re-epithelialization at 14 days in the ‘‘no treatment’’ and ‘‘cell alone’’ treatment groups; while the re-epithelialization only occurred at the edges of the wounds in the hydrogel treatment groups (with or without cells). However, because of the large skin contraction, no statistical significant difference was noted for re-epithelialization between all the groups (Fig. 3c) at each time point.
ð4Þ ð5Þ
3.4. Cell retention
ð6Þ
The cell retention ratio in the wound bed was assessed by labeling the rADSCs with CellTracker CM-Dil. The marker in this kit, which conjugates with intracellular proteins, reflects the cell survival rate in the hydrogels. After each time point, paraffin-fixed
SA ¼ SV V
ð7Þ
LT ¼ LV V
ð8Þ
2.7. Statistical analysis Comparisons between multiple groups were evaluated by oneway or two-way ANOVA and Tukey’s post-analysis method. A P-value of
