Scientific Papers

Mesenchymal stem cell-derived secretomes-enriched alginate/ extracellular matrix hydrogel patch accelerates skin wound healing | Biomaterials Research

Preparation of fibroblast-derived, decellularized ECM

Human lung fibroblasts (WI-38, CCL-75; ATCC) were cultivated on a tissue culture plate (TCP) at a density of 2 × 104 cells/cm2 in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (FBS), 100 U/mL penicillin and 100 µg/mL streptomycin (1% P/S) under a normal culture condition (5% CO2, 37℃). The medium was replenished every 2–3 day. On day 10, those confluent cells were rinsed twice with phosphate buffered saline (PBS) and subjected to decellularization via dispensing a solution of 0.25% Triton-X 100 and 20mM NH4OH into the plates, followed by PBS washing and subsequent treatment with 50 U/mL DNase I (18,047 − 019; Invitrogen) and 100 µL /mL RNase A (12,091 − 039; Invitrogen) for 2 h at 37℃. After several washing with PBS, the decellularized extracellular matrix ECM was transferred in a conical tube and kept at -20℃ in deionized water (DW) for further use. The entire process was performed in a sterile condition.

Preparation of hMSC-derived secretomes in concentrated conditioned medium

Human bone marrow-derived mesenchymal stem cells (hMSC; PT-2501, Lonza) were cultivated on the TCP at the density of 2 × 104 cells/cm2 in DMEM containing 10% FBS and 1% P/S. At 80% confluence, those cells were washed with PBS and incubated for 24 h in fresh serum-free medium (SFM) at 37℃. The medium was collected and centrifuged at 1,000 rpm for 5 min to eliminate the cell debris. Next, the concentrated conditioned media (CCM) of the hMSC-derived secretomes was then obtained via another centrifugation at 3,600 rpm for 30 min at 4℃ using an Amicon Ultra-15 Centrifugal Filter Unit with Ultracel-3 membrane 3 kDa molecular weight cut-off (UFC900324, Merck Millipore Ltd., Burlington, MA, USA). The supernatant inside the filter units were carefully collected and stored at -20℃ until further use. The whole process was carried out in a sterile condition. The resulting CCM was profiled using human growth factors microarray and human cytokine array as described in 2.4.

Immunofluorescence (IF) staining

The samples were fixed with 4% paraformaldehyde for 15 min at room temperature (RT) and then rinsed three times with PBS. They were permeabilized with 0.2% Triton-X 100 for 10 min, washed with PBS, and blocked with 3% bovine serum albumin (BSA) for 1 h at RT. They were then incubated with the primary antibodies overnight at 4℃, rinsed with PBS, then incubated with the secondary antibodies for 1 h at RT. After several washes, those samples were mounted onto microscope glass slides using Vectashield® mounting medium containing 4, 6-diamidino-2-phenylindole (DAPI) (H1200; Vector Lab, Burlingame, CA, USA) for the nuclei labeling. Fluorescence images were taken using a confocal laser scanning microscope (LSM 700; Carl Zeiss, Oberkochen, Germany). The primary antibodies and dilution ratios are as follows: rabbit polyclonal anti-collagen type I (ab34710; Abcam, Cambridge, UK, 1:300), mouse monoclonal anti-fibronectin (SC-8422; Santa Cruz Biotechnology, 1:200), rabbit polyclonal anti-collagen type III (ab7778; Abcam, 1:300), rabbit polyclonal anti-laminin (L9393, Sigma-Aldrich, 1:100) and rabbit polyclonal anti-CD31 antibody (ab28364; Abcam, 1:200). The secondary antibodies are Alexa Fluor® 488-conjugated goat anti-rabbit IgG (A-11,008; Invitrogen, Waltham, MA, USA) and Alexa Fluor® 594-conjugated donkey anti-rabbit IgG (A-21,207; Invitrogen), both of which were diluted at 1:300. The F-actin cytoskeleton of HUVEC was observed using ActinRed™ 555 ReadyProbes™ Reagent, an F-actin probe (phalloidin) conjugated to the red-orange fluorescent dye tetramethylrhodamine (TRITC) (R37112; Invitrogen, 1:200).

Profiling of the secretomes in the CCM

To elucidate the secretome components contained in the CCM, the Human Growth Factor Antibody Array Membrane (41 targets, ab134002; Abcam) was employed according to the manufacturer’s instructions. Briefly, once the membrane was blocked with a blocking buffer at RT, it was then incubated with the CCM overnight at 4℃ on a shaker. After washing with a washing buffer, the membrane was incubated with biotin-conjugated anti-cytokines overnight at 4℃ with agitation, followed by another overnight incubation with HRP-conjugated streptavidin at 4℃. The membrane was extensively washed and finally added with a detection buffer. Moreover, the cytokines in the CCM were also examined using the Proteome Profiler™ human cytokines array kit (ARY005B; R&D systems). In brief, the nitrocellulose membrane was blocked with a block buffer for 1 h at RT. Subsequently, a mixture of the CCM and biotinylated detection antibodies was added onto the membrane and then incubated overnight at 4℃ on a shaker. After several washes, the membranes were incubated for 30 min with streptavidin-horseradish and then treated with chemiluminescence detection reagents. Chemiluminescence blot detection was carried out via iBright CL1500 imaging system (Invitrogen, Waltham, MA, USA). The collected data were quantitatively analyzed using iBright analysis software 4.0.1. The results are presented as the relative ratio (%) between the positive dots and the reference dots.

Fabrication of alginate hydrogel patches

Sodium alginate powder (9005-38-3, 500–600 cP; FUJIFILM Wako Chemicals, Osaka, Japan) was purchased and sterilized under UV light overnight before solubilization in autoclaved DW. Alginate solution (3 or 5%, w/v) was produced by continuous stirring using a magnetic bar at 100℃ overnight. Meanwhile, the ECM suspension was prepared in DW via pulverization using an ultrasonicator (Sonic Dismembrator Model 500; Thermo Fisher Scientific, Waltham, MA, USA). Total proteins content in the ECM suspension was quantitatively measured via BCA protein assay (23,225, Thermo Fisher Scientific). The resulting ECM suspension was then homogeneously mixed with 5% alginate solution and stirred 1 h to yield an alginate/ECM solution, where it contains both alginate (3%, final conc.) and ECM (0.1%, w/v). Subsequently, 100 µL of alginate/ECM solution was poured into 12-well plate, respectively and pressed with an 18 mm diameter coverslip to produce a thin membrane. The alginate hydrogel membrane was then crosslinked by dispensing 2 mL of 500 mM calcium chloride at 37℃. The resulting membrane was peeled off from coverslip, washed three times with DW, punched in 12 mm diameter using biopsy punch, and air-dried in a clean bench. To include hMSC-derived CCM, the dried alginate patch was rehydrated with 120 µL of CCM or SFM (a negative control) at 37℃. The overall fabrication process was illustrated in the Scheme 1. Moreover, alginate hydrogel patches were prepared with or without ECM, where they were also added with CCM or SFM. As a result, there are four experimental groups in this study: alginate/SFM (AS), alginate/ECM/SFM (AES), alginate/CCM (AC), and alginate/ECM/CCM (AEC).

Scanning electron microscope observation

To confirm the surface morphology of alginate and alginate/ECM patch, the lyophilized samples were placed on carbon tape, then platinum-coated for 1 min using ion sputter (E-1045, Hitachi, Tokyo, Japan), and observed using scanning electron microscope (SEM; Phenom Pro G6 Desktop SEM, Thermo fisher scientific, Massachusetts, USA).

Characterization of AEC hydrogel patches

We examined the physical and mechanical property of alginate hydrogel patches, particularly paying attention to the difference between AC and AEC hydrogel. The test samples were prepared as a patch type as described above for water contact angles measurement. A contact angle measurement device (Attension Theta flow, Biolin Scientific, Sweden) equipped with a video camera was employed to measure the hydrophilicity and wettability of each patch. Static sessile drop method was performed with a droplet size of 5.0 µL and contact angles were measured immediately after the drop, where they were calculated using Young-Lapace equation. The rheological property of AC and AEC hydrogel sample (10 mm diameter, 1 mm thick) was also examined via an Anton Paar Rheometer (MCR102; Anton Paar, Austria). The rheometer was equipped with a parallel plate (25 mm dia.) and the sample gap size was 0.35 mm. We determined both storage modulus (G’) and loss modulus (G”) while applying 3% shear strain at RT. The collected data were analyzed using RheoCompass software. For the mechanical property, cylindrical patch samples (5 mm diameter, 3 mm height; n = 3, each group) were prepared and tested using Universal Testing machine (Instron® 5566, Instron Corporation, USA), where they were compressed to 60% of the whole thickness under 1.0 mm/min speed. We acquired stress-strain curves and calculated the compressive strength using Bluehill® 2 software.

Swelling property and biodegradation

We prepared alginate and alginate/ECM patch, air-dried them, then weighed (W0). These samples were then immersed in normal saline for 30 min, 1, 2, 3, 4, and 5 h at 37℃. Once we removed the surface water, they were weighed at each time point (Wt). Each sample was repeated five times. The swelling ratio of test sample (n = 5, each group) was calculated using this formula:

$${\text{Swellingratio}} = \frac{{{W_t} – {W_0}}}{{{W_0}}} \times 100\%$$

For biodegradability analysis, they were dipped in normal saline (pH 6.5) added with 2 mg/mL lysozyme (10,837,059,001; Roche, Basel, Switzerland) and incubated at 37℃, where the medium was replaced every other day. Those patch samples were collected at 1, 2, 3, 4, 5, 6, and 7 day (Wt), respectively, then lyophilized and weighed. As a control, the same patch samples were incubated for 24 h at 37℃ in normal saline without lysozyme, lyophilized and weighed (W0). Each experiment was conducted in triplicates. The degradation ratio was calculated using this formula:

$${\text{Degradationratio}} = 100\% – \left( {\frac{{{{\text{W}}_{\text{t}}}{\text{ – }}{{\text{W}}_{\text{0}}}}}{{{{\text{W}}_{\text{0}}}}} \times {\text{100\% }}} \right)$$

Release profile of secretomes from alginate patches

To investigate the release profile of secretomes out of either AC or AEC patch, the release rate and amount was evaluated with time via BCA protein assay. Each patch sample was put in the 500 µL of DW and remained at 37℃ for 96 h. During the time period, we collected 100 µL supernatant at specific time points (2, 4, 8, 12, 24, 48, 72, 96 h) and subsequently replenished 100 µL of fresh DW. For each AC and AEC group, five patches were prepared, respectively and examined in triplicates at each time points. In addition, we opted a specific time point, 72 h and thoroughly evaluated the total amount of released secretomes for 72 h using the collected medium (ten patch samples for each group). The assay was performed in accordance with the manufacturer’s instructions. Quantitative analysis was carried out based on the standard curve. Meanwhile, to determine amount of GFs contained in the AC or AEC patch, each sample was incubated in DW at 37℃ for 3 days. We collected the supernatant and measured the amount of each GF using enzyme-linked immunosorbent assay (ELISA) kit, including human hepatocyte growth factor (HGF; DY294), human Insulin-like growth factor binding protein 1 (IGFBP-1; DY871) and human vascular endothelial growth factor (VEGF) (DVE00, R&D Systems, MN, USA) according to the manufacturer’s protocols. Moreover, the cytokines released from the two patches were also evaluated via IL-6 (DY206) and Interleukin-8 (IL-8) ELISA kits (DY208, R&D Systems). Each group was tested in triplicates from different samples.

Cell scratch assay and cell proliferation test

Human skin fibroblasts (hSFB; BJ, ATCC) were seeded and cultivated on the 12-well plate at the density of 2 × 104 cells/cm2 in DMEM containing 10% FBS and 1% P/S until confluence. A wound scratch was created by scraping the center of plate using a 200 µL pipette tip in a straight line. After a brief wash with PBS, the medium was replaced with SFM (DMEM with 1% P/S). Meanwhile, we also prepared transwell inserts with 8.0 μm pore size (353,097; FALCON), in which four different types of alginate patches were loaded (1 patch/ 1 insert), respectively and being allowed to release the contents included in each patch system. Each group was tested in triplicates. The wound area was closely monitored at specific time points (0, 6, 12, 18, and 24 h) using optical microscopy. The remaining wound area was quantitatively calculated using Image J as a ratio between the wound area at each time point and the one at 0 h. In addition, to assess the proliferation of hSFBs under the same transwell setting, those cells were seeded and cultivated for 24 h. After replacing the medium to SFM, we placed those alginate patches on the transwell insert and allowed the release of the contents out of the patches. Cell proliferation was examined at 3, 7, and 10 day (n = 3, each group) using the cell counting kit-8 (CCK-8) assay (CK04; Dojindo, Kumamoto, Japan). The results were quantitatively presented, based on the standard curve of known number of hSFB.

Analysis of collagen deposition and HUVEC assembly in vitro

To evaluate collagen deposition by skin fibroblasts, when treated with the released secretomes, hSFB were seeded on the gelatin-coated coverslip in 12-well plate under the same condition as described in 2.8. Those cells were then washed with PBS and the medium was changed to SFM. Four different alginate patches were located on top of the hSFB using transwell inserts and remained for 3 days (1 patch/ 1 insert). After then, they were fixed and prepared for IF staining of collagen type 1 (Col 1). The samples were tested in triplicates for each group. Collagen deposition by hSFB was quantitatively determined using Image J by measuring the Col 1-positive area (five random images, each group). On the other hand, human umbilical vein endothelial cells (HUVECs) (C2517A; Lonza, Basel, Switzerland) were seeded on gelatin-coated cover glass at the density of 2 × 104 cells/cm2 and cultivated in endothelial cell growth medium (EGM-2 BulletKit, CC-3162; EBM-2 with supplements kit, Lonza). After cultivating HUVECs for 24 h, we washed them with PBS and changed the medium to endothelial cell basal medium (EBM-2, CC-3156; without supplements kit, Lonza). Likewise the collagen deposition test protocols, we followed the same procedure except the use of HUVECs. After 48 h, the cells were fixed and assessed via F-actin and CD31 IF staining. A self-assembly of HUVECs was scrutinized and quantified using image J software, based on the F-actin positive area appeared in the IF images (five random images, each group).

Keratinocyte migration assay in vitro

Human epidermal keratinocyte cells (CB-HK-001; CEFOBio, Gyeonggido, Korea) were cultured in keratinocyte growth medium 2 kit (KGM2, C-20,011, PromoCell, Germany), supplemented with 1% P/S. To investigate the effect of released secretomes on cell migration, firstly, four different alginate patches were incubated (1 patch/ 1 well) for 24 h in KBM2 with 20% KGM2 SupplementMix. Then, the keratinocytes suspended in keratinocyte basal medium 2 (KBM2, C-20,211, PromoCell) without supplement were seeded in the transwell insert with 8 μm pore size (353,097, FALCON) at the density of 1 × 105 cells (Fig. S2A). After 24 h incubation, once the cells in the upper transwell membrane was cleared with cotton swab, only the keratinocytes on the lower surface were fixed with 100% of cold methanol for 10 min at 4℃. The fixed keratinocytes were stained with 1% crystal violet solution (61,135, Sigma-Aldrich, St. Louis, MO, USA) for 20 min at RT, then subjected to air-dry, and confirmation of the migrated keratinocytes using optical microscope. For each insert, three different fields were randomly selected and the number of keratinocytes were counted. This experiment was conducted in triplicates.

Murine full-thickness skin wound model

Therapeutic effects of the secretomes were further investigated using full-thickness skin wound model. For this, the male BALB/c mice (6-week-old) were purchased form Orient Bio (Gapyeong, Korea). They were randomly divided into four groups according to the treatment regimens: AS, AES, AC, and AEC (n = 3 per group). Before surgery, they were anesthetized by gas inhalation using isoflurane in oxygen. Mouse hair was shaved, and the dorsal skin was then scrubbed using alcohol gauze for sterilization. Full-thickness wounds were created by using a biopsy punch (8 mm) under sterile surgical condition. Each patch, prepared as described in 2.5, was placed on the wound site (two wounds per mouse). Subsequently, Tegaderm™ Film and Coban bandage were wrapped around the wounds to keep the patches from detachment and to protect the treatment. These patches and dressing films were replaced every 2–3 day. The gross observation of wound closure was carried out on day 0, 2, 4, 7 and 14 post-treatments. The remaining wound area at specific time point (four random images, each group) was quantitatively calculated to assess wound closure rate using Image J software as a percentage of the wound region normalized to that of day 0. The animal experiments we carried out comply with the National Research Council’s Guide for the Care and Use of Laboratory Animals (8th edition, NIH Publication, 2011). All the animal studies were also approved by the Korea Institute of Science and Technology Animal Care and Use Committee (KIST-IACUC-2022-05-083).

Subcutaneous transplantation for in vivo safety

For investigation of in vivo safety of our AEC patch, the male BALB/c mice (6-week-old) obtained from Orient Bio (Gapyeong, Korea). They were anesthetized by gas inhalation using isoflurane in oxygen before surgery. After hair shaved, the skin was wiped with alcohol swab and scrubbed using gauze with povidone iodine. Minimal incision was created using scissors and AEC patches were transplanted in subcutaneous. Two AEC patches were transplanted each mouse (n = 2) on two different back parts. The incised area was sutured and covered with Tegaderm Film. The mice were euthanized by using CO2 inhalation at day 3. And the remained AEC patch was collected with around tissue. Behavioral abnormalities or death of mice were checked, and inflammatory reactions or adverse reactions were confirmed through histological staining.

Histological analysis of wound tissues

On day 7 and 14, the mice were euthanized by using CO2 inhalation, then followed by the excision of the skin wound tissues. Such tissue samples were fixed with 10% formalin, embedded in paraffin blocks, and sectioned in 5 μm thickness across the tissues. These thin sections were deparaffinized using xylene and rehydrated in a series of alcohol solutions. They were then subjected to hematoxylin and eosin (H&E) staining to evaluate the degree of skin tissue regeneration, where four different alginate patches were administered, respectively. Collagen deposition and the degree of maturity were also examined via Herovici (KTHERPT; StatLab, Mckinney, TX, USA) staining. Based on the randomly selected, high resolution images (8–10, each group) as obtained using an inverted microscope (Axio Vert.A1; Carl Zeiss, Oberkochen, Germany), we quantitatively assessed cell recruitment and neovascularization at 7 day as well as epidermal thickness and mature collagen deposition at 14 day using Image J.

Immunohistochemical analysis

To further understand the wound healing efficacy of four different treatment regimens via immunohistochemistry, the tissue sections were deparaffinized, rehydrated, then peroxidase blocked, and followed by antigen retrieval using microwave heating in citrate buffer (pH 6). After the blocking with 1% BSA, we incubated them with primary antibodies at 4℃ overnight. They were then rinsed with PBS several times and subsequently treated with secondary antibodies for 1 h at RT. Finally, those samples were counter-stained with NucBlue™ Live ReadyProbes™ Reagent (R37605, Invtirogen) and mounted using VECTASHIELD® Antifade Mounting Medium (H-1000; Vector Lab). The primary antibodies we used are as follows: rabbit monoclonal anti-cytokeratin 10 (ab76318; Abcam, 1:500), mouse monoclonal alpha smooth muscle actin (α-SMA) (A2547; Sigma-Aldrich, 1:400) and rabbit polyclonal anti-CD31 antibody (ab28364; Abcam, 1:100), mouse monoclonal vimentin antibody (E-5) (SC-373,717; Santa Cruz Biotechnology, Dallas, TX, USA) and rabbit polyclonal anti-alpha smooth muscle actin (ab5694; Abcam, 1:100). The secondary antibodies are as follows: Alexa Fluor® 488-conjugated goat anti-rabbit IgG (A-11,008; Invitrogen), Alexa Fluor® 594-conjugated goat anti-mouse IgG (A-11,005; Invitrogen), Alexa Fluor® 488-conjugated goat anti-mouse IgG (A-11,001; Invitrogen) and Alexa Fluor® 594-conjugated donkey anti-rabbit IgG (A-21,207; Invitrogen). All secondary antibodies were diluted at 1:200. The stained samples were observed using confocal laser scanning microscope (Carl Zeiss). Based on the positive area as shown in the high-power field images (3–5 random ones, each group), we attempted to quantitatively assess various wound healing parameters, such as keratinocyte migration, mature blood vessel formation, and activated myofibroblasts using image J.


All the data are presented as mean ± standard deviation, along with individual data points. Statistical analysis was performed using a two-tailed (α = 0.05) Student’s t-test for the two experimental groups. One-way analysis of variance (ANOVA) with a post-hoc Tukey’s multiple comparison test was also carried out for more than three test groups. Two-way ANOVA with a post-hoc Tukey’s multiple comparison test was performed for more than three test groups with two or more variables. Statistically significant differences are denoted by *p < 0.05, **p < 0.01, ***p < 0.001, or ****p < 0.0001.

Source link