Paracrine Activity from Adipose-Derived Stem Cells on In Vitro Wound Healing in Human Tympanic Membrane Keratinocytes

Abstract

Stem cell therapies for tympanic membrane repair have shown initial experimental success using mesenchymal stem cells in rat models to promote healing; however, the mechanisms providing this benefit are not known. We investigated in vitro the paracrine effects of human adipose-derived stem cells (ADSCs) on wound healing mechanisms for human tympanic membrane-derived keratinocytes (hTM) and immortalized human keratinocytes (HaCaT). ADSC conditioned media (CM_ADSC) were assessed for paracrine activity on keratinocyte proliferation and migration, with hypoxic conditions for ADSC culture used to generate contrasting effects on cytokine gene expression. Keratinocytes cultured in CM_ADSC showed a significant increase in cell number compared to serum-free cultures and further significant increases in hypoxic CM_ADSC. Assessment of ADSC gene expression on a cytokine array showed a range of wound healing cytokines expressed and under stringent hypoxic and serum-free conditions was upregulated (VEGF A, MMP9, Tissue Factor, PAI-1) or downregulated (CXCL5, CCL7, TNF-α). Several of these may contribute to the activity of conditioned media on the keratinocytes with potential applications in TM perforation repair. VEGFA protein was confirmed by immunoassay to be increased in conditioned media. Together with gene regulation associated with hypoxia in ADSCs, this study has provided several strong leads for a stem cell–derived approach to TM wound healing.

Introduction

The skin is a protective barrier against the external environment and is routinely subjected to abrasive or penetrating injuries. Wound healing commences immediately after injury with an orchestrated activation of intracellular and intercellular pathways to reestablish tissue integrity and homeostasis [1]. The tympanic membrane is a specialized skin-like structure, which is protected from casual trauma by its anatomical location deep within the bony ear canal, but it remains subject to injury from air pressure, iatrogenic or accidental perforation by objects inserted into the ear canal, or most commonly as a consequence of otitis media. The tympanic membrane also has a robust injury response, but failure to heal occurs in up to 15% of patients, leaving a chronic perforation. The mechanisms of healing and failure-to-heal are of considerable interest, but relatively poorly defined. Unlike cutaneous wounds, which have a source of regenerative tissue at the base [2], TM perforations are full-thickness wounds and rely entirely on tissues at their lateral margin for regenerative responses. TM regeneration is very effective, with up to 90% of TM perforations healing spontaneously without intervention [3,4]. Keratinocytes lead the wound healing process by migrating across the perforation and proliferating to restore the epidermal layer, followed by the underlying granulation tissue and mucosal epithelial tissues to restore the middle and inner layers, respectively [5,6]. Where healing fails, chronic wounds may attract surgical treatment, but with increasing knowledge of TM biology it is envisaged that nonsurgical treatment with bioactive molecules may soon prove suitable in many cases [7], thus reducing cost and providing greater access to treatment outside of a surgical facility. A number of molecular entities have been tested as suitable treatments, but without a clear choice of factors emerging; research is ongoing in this area.

One avenue relatively unexplored is stem cell stimulated wound healing. One of the few reports of stem cell healing for TM is the delivery of human mesenchymal stem cells (MSCs) into the middle ear, which promoted healing of acute and chronic perforations in rat ears [8]. Recent work showed that delivery of mouse bone marrow MSC on hyaluronic acid scaffold to perforated mouse tympanic membrane was able to promote healing of acute perforation [9] compared to scaffold alone. The molecular mechanism of stem cell mediated healing was however not discovered in either study. MSCs applied in other wound healing models have been shown to produce mediators that enhance wound healing locally through improved cell proliferation and viability [10,11], tissue remodeling by reepithelialization [8,12 –16], angiogenesis [12,14,15,17,18], and inflammation [14,19], in a paracrine manner. These paracrine mediators may ultimately be able to reactivate the wound healing cascade in chronic wound healing [20,21].

We propose that paracrine mechanisms might prove useful to promote wound healing in TM perforations and explain some of the wound healing activity derived from MSC in rats. In the present study, we evaluated in vitro models for paracrine activity from human adipose-derived MSCs on human tympanic membrane keratinocytes and the HaCaT keratinocyte cell line. Hypoxic conditioning of the MSC was an effective stimulator of proliferative and chemotactic activity. We further studied the effect of hypoxic conditioning on the MSC wound healing transcriptome and identified several candidate factors for future study.

Materials and Methods

Culture of primary hTM keratinocytes and HaCaT cells

Primary hTM keratinocytes derived from normal human tympanic membrane explants were used as previously described [22]. The St John of God HealthCare Ethics Committee approved collection of excess tympanic membrane tissue from patients undergoing otological procedures at St John of God Hospital (Subiaco, Australia). The HaCaT cell line was supplied by the Burns Injury Research Unit, School of Surgery, University of Western Australia. Both cell types were cultured in Dulbecco's modified Eagle's medium (DMEM) containing 4.5 g/L d-glucose (Gibco), supplemented with 10% fetal bovine serum (FBS; Bovogen), 1% Penicillin-Streptomycin (Gibco), and incubated in a humidified cell culture incubator at 37°C, with 5% carbon dioxide (CO₂) until confluent. Cells were passaged every 3–4 days.

Culture of adipose-derived stem cell

Human adipose-derived stem cells (ADSCs; Lonza) were cultured in DMEM containing 1 g/L d-glucose supplemented with 10% FBS and 1% Penicillin-Streptomycin, with spent medium replaced every 2–3 days.

Collection of ADSC-conditioned medium

ADSCs were cultured to 80%–90% confluence in 75 cm² tissue culture flasks, rinsed with 1 × phosphate-buffered saline (PBS), pH 7.2 (Gibco), and incubated with 10 mL of serum-free (S−) DMEM (1 g/L d-glucose) for 8 h before replacement with fresh serum-free medium. Two flasks were incubated under standard conditions (37°C, 5% CO₂) and two under hypoxic conditions (GENbox Jar; BioMérieux) created by an anaerobic atmosphere generator sachet (BioMérieux) at 37°C. An indicator strip (BioMérieux) was included in the sealed chamber to confirm gas composition (<0.1% O₂ and 15% CO₂). All four flasks were incubated for 48 h. Conditioned media were then collected using a sterile 10 mL syringe (BD) and filtered through a 0.2 μm filter membrane (PALL) to remove cell debris. The media collected are referred to as Nx ADSC conditioned media (CM_ADSC) and Hx CM_ADSC, respectively.

Total RNA isolation and complementary DNA synthesis

Total RNA was extracted from ADSCs and hTM keratinocytes according to manufacturer's instructions (FavorPrep^™ Tissue Total RNA Mini Kit; Favorgen Biotech Corp), then RNA concentration and purity were measured by spectrophotometry (Epoch Take3; BioTek Instruments, Inc.). Synthesis of complementary DNA (cDNA) was then performed (RT² First Strand Kit; Qiagen) according to manufacturer's instructions with 0.5 μg of RNA.

Real-time polymerase chain reaction of human wound healing array

Human Wound Healing polymerase chain reaction (PCR) array was performed according to manufacturer's instruction (RT² Profiler; Qiagen) on the cDNA of ADSC_Hx, ADSC_Nx, and hTM keratinocyte S−. Briefly, cDNA, SYBR green master mix, and water were combined to a final volume of 2,700 μL. A volume of 25 μL per well was added to the 96-well array plate and run on a CFX Connect^™ Real-Time PCR Detection System (Bio-Rad) under the following conditions: 95°C for 10 min to activate HotStart DNA Taq Polymerase and 40 cycles of denaturation at 95°C for 15 s and annealing at 60°C for 1 min. Threshold cycles (C_t) were determined using the exponential growth phase and the baseline signal from fluorescence versus cycle number plots. Relative messenger RNA (mRNA) expression levels were determined using the Livak method (2^−ΔΔCt) with beta-2-microglobulin (B2M) selected as the most stable housekeeping gene. Each experiment was performed in duplicate array plates repeated for all ADSC_Hx, ADSC_Nx, and hTM keratinocyte S⁻.

Gene ontology analysis of hypoxia-induced gene expression changes in cultured ADSC

Gene ontology analysis was performed using the g:Profiler platform for enrichment analyses to identify biological functions affected by hypoxic conditions in cultured ADSC (http://biit.cs.ut.ee/gprofiler/welcome.cgi) [23]. Query lists of gene coding for secreted proteins of potential paracrine effect on keratinocytes were uploaded and analyzed for up- and downregulated genes in separate analyses. Benjamini–Hochberg false discovery rate (FDR) was chosen as the statistical method for multiple testing correction.

Total protein assay

Total protein concentration in both Hx and Nx CM_ADSC were quantified using a Bradford Protein Assay (Quick Start; Bio-Rad) according to manufacturer's protocol. Bovine Serum Albumin standards were used over a range of 1.25–10 μg/mL. After incubating the CM_ADSC with dye reagent, the absorbance was measured at 595 nm using a spectrophotometer (Epoch; BioTek Instruments, Inc.) and protein concentrations of the CM_ADSC read from the standard curve.

Quantification of specific protein levels

Secreted VEGF-A, IL6, IL1B, FGF2, HB-EGF, and FGF7 proteins in CM_ADSC were quantified using enzyme-linked immunosorbent assay (ELISA, Quantikine kits; R&D Systems) according to the manufacturer's protocol. Briefly, protein standard curves were prepared by serial dilution. In duplicates, samples were loaded into microplate wells precoated with monoclonal antibodies. After incubation for 2–3 h, wells were washed to remove unbound proteins and then enzyme-linked polyclonal antibody was added to the wells and incubated. Final washing was done to remove unbound antibody-enzyme reagent, then substrate containing hydrogen peroxide and tetramethylbenzidine was added and incubated for 30 min before stopping the reaction by acidification. Optical density was measured at 450 nm on the spectrophotometer with a wavelength correction at 570 nm. Secreted protein concentrations were expressed in ng per mg of total protein, calculated from either linear standard curves or four parameter logistic (4-PL) curve fit by interpolating sample values.

MTS assay for proliferation

Cell proliferation was analyzed by colorimetric assay (CellTiter 96^® AQ_ueous One; Promega Corporation). Briefly, keratinocytes were serum-starved for 8 h before experimentation, resuspended in test medium [5,000–10,000 cells in 10% serum (S+), serum free (S−), or 50:50 mix of conditioned media and S− media], and seeded in 96-well plates for a 24-h or 48-h incubation at 37°C, 5% CO₂. Following incubation, 20 μL of colorimetric solution was added to each well, and plates were incubated for 2 h. Absorbance was measured at 490 nm wavelength using the Epoch colorimetric plate reader (BioTek Instruments, Inc.). Samples were read in triplicate, averaged, and corrected for background optical density. The effect on proliferation was expressed as a fraction of the S− control treatment for each cell, respectively.

Scratch test assay for migration

A scratch wound assay was performed on hTM keratinocytes and HaCaT cells to assess cell migration when cultured in Hx CM_ADSC, Nx CM_ADSC, and S⁺. Initially, 500,000 keratinocytes were seeded in 12-well culture plates with S+ media and grown to confluence. Following 8 h preincubation in S− medium, cell monolayers were wounded with a sterile 200 μL pipette tip with light vacuum applied to clear debris from the wound. Cells were then washed with PBS and incubated with the corresponding test medium. Microscopic images were taken immediately after scratch wounding (t₀) and after 10 h on an Olympus BX60 microscope with a DP70 digital camera (Olympus). Wound gaps were measured using ImageJ (1.48v) software. Each gap was measured 45 times. All measurements were repeated in triplicate and averaged. The migration of keratinocytes into the gap was calculated as: \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\usepackage{upgreek}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} \begin{align*}&{ \% \,{ \rm{Gap}} \,{ \rm{closure}} = [ ( { \rm{Gap}} \,{ \rm{lengt}}{{ \rm{h}}_{ \rm{t}}}{ = }{ _{10 \;{ \rm{h}}}}} \; \\ & \quad - \;{ \rm{Gap}} \,{ \rm{lengt}}{{ \rm{h}}_{ \rm{t}}}{ _ = }{ _{0 \;{ \rm{h}}}} ) / { \rm{Gap}} \,{ \rm{lengt}}{{ \rm{h}}_{ \rm{t}}}{ _ = }{ _{0 \;{ \rm{h}}}} ] \; \times 100 \% \end{align*} \end{document}

The results are expressed as a fraction of the S− control treatment as above.

Statistics

Quantitative data are presented as mean ± standard error of the mean (SEM). Microsoft Excel was used for all data analysis. Statistical methods used include:

• One-way analysis of variance (ANOVA) for proliferation and migration experiments

• Dunnett's test as a post hoc test for individual treatment group comparisons against S− control group for both proliferation and migration experiments

• Student t-test for ELISA and PCR data analysis

Statistical level of confidence was set at P < 0.05 unless otherwise stated.

Results

Cell proliferation and migration assays were performed using Hx and Nx CM_ADSC on hTM and HaCaT keratinocytes with serum-containing (S+) and serum-free (S−) media as positive and negative controls. Figure 1 shows that ADSC cultures were viable before and after conditioning in hypoxic or normoxic conditions.

FIG. 1.

Human ADSC before and after hypoxic and normoxic conditioning for 48 h. Scale bars = 200 μm. ADSC, adipose-derived stem cell.

Effect of CM_ADSC on keratinocyte proliferation

Representative images of proliferation responses in hTM and HaCaT keratinocytes over 48 h are shown in Fig. 2. In hTM keratinocyte cultures, serum produced a substantial proliferative response at 24 and 48 h compared to serum-free cultures (Fig. 3). Hx CM_ADSC media produced a significantly greater proliferative response (P < 0.05) than serum-free media and quantitatively as much as 83% and 95% of the serum response at 24 and 48 h, respectively. Nx CM_ADSC however only showed significant (P < 0.05) proliferative response at 48 h, up to 79% of the serum response. When comparing Hx and Nx CM_ADSC, there was a significant difference (P < 0.05) between the proliferative effects on hTM keratinocytes at both 24 and 48 h.

FIG. 2.

Left columns: hTMt in test medium (S+, Hx and Nx CM_ADSC, S−) during proliferation test at 48 h; right columns: HaCaTs in test medium during proliferation test at 48 h. Scale bars = 200 μm. CM_ADSC, ADSC conditioned media.

FIG. 3.

Effect of S+, Hx and Nx, CM_ADSC on the proliferation of hTM and HaCaT keratinocytes over 24 and 48 h. Graph shows representative data of relative absorbance values normalized against the negative (S−) control. hTM keratinocytes and HaCaT keratinocytes had a greater proliferative response to Hx CM_ADSC than Nx CM_ADSC. Test media marked with asterisk (*) were significantly different to S− media (P < 0.05).

In HaCaT keratinocytes, the proliferative effect of serum was significant (P < 0.05) at 24 h, but not at 48 h compared to serum-free controls (Fig. 3). Proliferative responses of HaCaT keratinocytes to Hx CM_ADSC was significant at both 24 and 48 h, quantitated at 106% and 109% of serum response, respectively. Nx CM_ADSC treatments were not significant at 24 h, but at 48 h responses were significantly greater than serum-free controls of up to 9% compared to serum-free controls. Both CM_ADSC treatments produced a proliferative response larger than the serum treatment effect at 48 h. When comparing Hx and Nx CM_ADSC, there was a significant difference (P < 0.05) between the proliferative effects on HaCaT at both 24 and 48 h.

Effect of CM_ADSC on keratinocyte migration

Migration assays were performed on hTM and HaCaT keratinocytes over 10 h to compare the effect of Hx and Nx CM_ADSC measured against serum-free control media (Figs. 4 and 5). Representative images of wound gaps in all test media are presented in Fig. 4, and the quantitative data are presented as the normalized percentage gap closure in Fig. 5.

FIG. 4.

Representative image of migration test over 10 h; left two columns: hTM keratinocyte wound gaps in various test medium; right two columns: HaCaT keratinocyte wound gaps in various test medium. Scale bars = 500 μm.

FIG. 5.

Effect of S+, Hx and Nx, CM_ADSC on the migration of hTM and HaCaT keratinocytes over 10 h. Migration data over 10 h are represented as normalized percentage gap closure of negative control (S−) wells (at 100%). All test media marked with an asterisk (*) were significantly different (P < 0.05) from the corresponding S− controls.

Hx CM_ADSC and Nx CM_ADSC produced a significantly greater (P < 0.05) gap closure effect on hTM keratinocytes compared to S− media controls. Hx CM_ADSC effects on hTM keratinocytes were significantly higher than Nx CM_ADSC (P < 0.05). Gap closures were on average 27% and 10%, respectively, greater than in the absence of paracrine activity. In these experiments, hTM keratinocytes were found to maintain a migration capacity in the absence of serum (P < 0.05).

In HaCaT keratinocytes, significant differences in migration were observed for both Hx and Nx CM_ADSC (P < 0.05). Gap closures were on average 37% in serum controls, 98% in Hx CM_ADSC and 43% in Nx CM_ADSC greater than in the S− media controls. HaCaT cells cultured in S− media migrated less well and maintained a 35% gap even after 28 h, while all other test wells were closed at this time.

Molecular identification of paracrine activity

PCR array for wound healing gene expression in ADSCs and hTM keratinocytes

PCR Wound Healing arrays were used to determine which of the 84 genes present on the array were expressed in ADSCs, as well as to perform a quantitative comparison of the effects of hypoxic and normoxic culture conditions. B2M was a stable housekeeping gene for these arrays, so was used to normalize all values, and C_t thresholds were manually set to the same value across all samples.

Gene expression

Among the 84 genes present on the array (Supplementary Table S1; Supplementary Data are available online at www.liebertpub.com/scd), 67 were expressed in ADSCs and 17 genes were not detected (C_t values >35). Among the 17 genes not detected were some prominent regulators of keratinocyte growth, which could then be excluded as candidates for paracrine activity from ADSCs on keratinocytes: EGF, TGFα, HGF, FGF10 (KGF2), IFNγ, and IGF-1 (Table 1).

Table 1.

Wound Healing Array Gene Expression of Adipose-Derived Stem Cells Regulated by Hypoxic Environment

Gene	P	Fold Δ	Secreted	Gene	P	Fold Δ	Secreted
F3	0.028	61.49	N	MAPK3	0.317	1.92	Y
MMP9	0.089	21.55	Y	FGF2	0.144	1.90	Y
VEGFA	0.048	15.06	Y	CTSV	0.281	1.79	N
WISP1	0.105	13.35	Y	COL3A1	0.229	1.78	Y
TAGLN	0.094	11.48	N	FGF7	0.018	1.77	Y
SERPINE1	0.112	11.29	Y	RAC1	0.396	1.66	N
COL4A3	0.036	9.01	Y	RHOA	0.421	1.58	N
IL1B	0.002	8.11	Y	MMP2	0.376	1.55	Y
MIF	0.098	7.46	Y	ITGA6	0.409	1.52	N
PLAT	0.120	7.02	Y	PLAU	0.428	1.47	Y
CTGF	0.047	6.44	Y	MAPK1	0.442	1.47	Y
COL5A1	0.022	5.80	Y	COL1A2	0.437	1.41	Y
COL4A1	0.024	5.47	Y	COL1A1	0.416	1.38	Y
IL6	0.006	5.30	Y	ITGAV	0.472	1.27	N
TIMP1	0.127	4.91	Y	WNT5A	0.451	1.12	Y
COL5A3	0.047	4.34	Y	COL14A1	0.156	1.05	Y
PTGS2	0.136	3.98	Y	HBEGF	0.036	1.04	Y
EGFR	0.007	3.87	N	ITGA3	0.350	−1.06	N
ITGA2	0.160	3.63	N	ITGA4	0.268	−1.08	N
ITGA1	0.001	3.33	N	CSF3	0.235	−1.09	Y
COL5A2	0.080	3.25	Y	TGFBR3	0.249	−1.15	N
CSF2	0.027	3.12	Y	PDGFA	0.236	−1.30	Y
PLAUR	0.140	3.11	Y	CTSK	0.049	−2.10	Y
ANGPT1	0.123	3.10	Y	CXCL2	0.087	−2.51	Y
TGFB1	0.186	3.00	Y	CCL2	0.008	−2.67	Y
ITGB5	0.210	2.99	N	TNF	0.153	−4.05	Y
ITGB1	0.252	2.90	N	CCL7	0.062	−4.73	Y
MMP1	0.260	2.70	Y	CXCL1	0.033	−5.24	Y
ITGA5	0.184	2.68	N	CXCL5	0.023	−5.47	Y
PTEN	0.218	2.47	N	ACTC1			N
ACTA2	0.062	2.39	N	IL4			Y
IL6ST	0.005	2.23	N	IL10			Y
PLG	0.273	2.22	Y	CDH1			N
CXCL11	0.211	2.22	Y	TGFA			Y
VTN	0.355	2.17	Y	IGF1			Y
ITGB3	0.284	2.14	N	CD40LG			N
STAT3	0.263	2.11	N	CTSG			N
CTNNB1	0.173	2.09	N	EGF			Y
B2M	-	1.00		F13A1			Y
				FGA			Y
				FGF10			Y
				IFNG			N
				IL2			Y
				ITGB6			Y

B2M, beta-2-microglobulin.

Increase fold Δ >2-folds

Not regulated by hypoxia

Decrease fold Δ <2-folds

Housekeeping Gene

Not expressed

Bold text shows P < 0.05.

Expression of wound healing genes in hTM keratinocytes was also examined in the same manner, and a largely similar set was found (Table 2), but with notable differences being absence of gene expression in keratinocytes for cell adhesion molecule integrin β3, collagens alpha 1A2, 4A1, 4A3, 5A1, 5A3, and 14A1, WISP1 and WNT5A, and presence of gene expression for integrin β6, TGFα, E-cadherin, MMP7, and TNF.

Table 2.

Wound Healing Gene Expression Profile of Hx Adipose-Derived Stem Cell, Nx ADSC, and hTM Keratinocytes

Gene	Hx	Nx	hTMt	Gene	Hx	Nx	hTMt	Gene	Hx	Nx	hTMt
ACTA2	+	+	+	EGFR	+	+	+	MAPK1	+	+	+
ACTC1	−	−	−	F13A1	−	−	−	MAPK3	+	+	+
ANGPT1	+	+	+	F3	+	+	+	MIF	+	+	+
CCL2	+	+	+	FGA	−	−	−	MMP1	+	+	+
CCL7	+	+	+	FGF10	−	−	−	MMP2	+	+	+
CD40LG	−	−	−	FGF2	+	+	+	MMP7	−	−	+
CDH1	−	−	+	FGF7	+	+	+	MMP9	+	+	+
COL14A1	+	+	−	HBEGF	+	+	+	PDGFA	+	+	+
COL1A1	+	+	+	HGF	−	−	−	PLAT	+	+	+
COL1A2	+	+	−	IFNG	−	−	−	PLAU	+	+	+
COL3A1	+	+	+	IGF1	−	−	−	PLAUR	+	+	+
COL4A1	+	+	−	IL10	−	−	−	PLG	+	+	+
COL4A3	+	−	−	IL1B	+	+	+	PTEN	+	+	+
COL5A1	+	+	−	IL2	−	−	−	PTGS2	+	+	+
COL5A2	+	+	+	IL4	−	−	−	RAC1	+	+	+
COL5A3	+	+	−	IL6	+	+	+	RHOA	+	+	+
CSF2	+	+	+	IL6ST	+	+	+	SERPINE1	+	+	+
CSF3	+	+	+	ITGA1	+	+	+	STAT3	+	+	+
CTGF	+	+	+	ITGA2	+	+	+	TAGLN	+	+	+
CTNNB1	+	+	+	ITGA3	+	+	+	TGFA	−	−	+
CTSG	−	−	−	ITGA4	+	+	+	TGFB1	+	+	+
CTSK	+	+	+	ITGA5	+	+	+	TGFBR3	+	+	+
CTSV	+	+	+	ITGA6	+	+	+	TIMP1	+	+	+
CXCL1	+	+	+	ITGAV	+	+	+	TNF	−	+	+
CXCL11	+	+	+	ITGB1	+	+	+	VEGFA	+	+	+
CXCL2	+	+	+	ITGB3	+	+	−	VTN	+	+	+
CXCL5	+	+	+	ITGB5	+	+	+	WISP1	+	+	−
EGF	−	−	−	ITGB6	−	−	+	WNT5A	+	+	−

Hx ADSC, Nx ADSC, and hTM keratinocytes expressed (+) or not expressed (−) where highlighted boxes indicate variations and bold represents secreted genes.

ADSC, adipose-derived stem cell.

Hypoxia in ADSCs

Thirty-eight expressed genes were upregulated (2- to 62-fold) by hypoxia in ADSCs, 12 genes were downregulated (2- to 5-fold), and 17 were expressed, but showed less than twofold difference between the culture conditions. Taking into consideration that we were looking for secreted proteins, genes encoding for nonsecreted proteins were removed from the list of comparison between Hx and Nx ADSCs (Table 1). Genes not regulated by hypoxia, hence not likely related to the hypoxia-induced growth responses, included prominent keratinocyte growth regulators HB-EGF and FGF7. FGF2 was only 1.9-fold higher in Hx. Genes reduced by hypoxia were chemokines and cytokines, including prominently the keratinocyte growth factor TNFα.

Gene expression regulated by hypoxia

Tissue Factor (F3), showed the highest upregulation in this assay with more than 60-fold higher expression in Hx ADSCs than Nx ADSCs. Out of 56 genes coding for secreted proteins, 24 were upregulated by at least twofold; 13 showed similar expression levels; 7 were downregulated between two- to sixfold; and 12 genes did not meet the threshold of detection (C_t values >35 or blank) in the array.

Gene ontology analysis

The results are summarized in Table 3. For a list of regulated genes and all significant gene ontology terms, see Supplementary Figure S1. The genes upregulated by hypoxic conditions are rich in collagens IV and V particularly and, thus, related to basement membrane and fibrillar collagen formation in the extracellular matrix. Among the downregulated genes are multiple chemokine (C-X-C) ligands, showing that ADSC influenced by hypoxia may have a direct impact on the chemotaxis of leukocytes.

Table 3.

Gene Ontology Analyses of Up- and Downregulated Genes of Secreted Proteins in Adipose-Derived Stem Cell During Hypoxic Culture Conditions

Term type	Term name	P
Upregulated genes
BP	Extracellular matrix disassembly	0.0456
BP	Negative regulation of cellular protein metabolic process	0.0273
BP	Regulation of proteolysis	0.0192
CC	Complex of collagen trimers	0.0133
CC	Intracellular	0.0406
CC	Endoplasmic reticulum lumen	0.0117
CC	Collagen type V trimer	0.0234
CC	Collagen trimer	0.0276
MF	Ion binding	0.0500
MF	Integrin binding	0.0094
Downregulated genes
BP	Neutrophil chemotaxis	0.0005
BP	G-protein coupled receptor signaling pathway	0.0001
BP	Regulation of leukocyte migration	0.0001
BP	Chemokine-mediated signaling pathway	0.0001
BP	Regulation of leukocyte chemotaxis	0.0039
BP	Positive regulation of immune system process	0.0034
BP	Response to lipopolysaccharide	0.0071
MF	Chemokine receptor binding	0.0001

Organism: Homo sapiens; query list of significantly up- (19) and downregulated (7) genes; options: significant only, ordered query, no electronic gene ontology annotations, gene list as a statistical background: all genes in the RT² Profiler PCR array that were expressed by ADSC (69 genes); hierarchical filtering: best per parent; gene ontology analyses; significance threshold: Benjamini–Hochberg FDR corrected P value (P < 0.0005).

BP, biological process; CC, cellular component; FDR, false discovery rate; MF, molecular function; PCR, polymerase chain reaction.

Protein concentration in conditioned media

To determine whether the hypoxia-mediated increase in gene expression of VEGFA, IL6, and IL-1B from the PCR array results was translated into increased protein secretion, CM_ADSC were analyzed using ELISA. FGF2, HB-EGF, and FGF7 were included in the analysis due to their potential involvement in TM wound healing mechanism. VEGFA, IL6, IL-1B, FGF2, and FGF7 were all observed at detectable levels in CM_ADSC. HB-EGF was not detected. Figure 6 shows the relative fold change in specific protein levels in Hx CM_ADSC against Nx CM_ADSC. VEGFA protein level was sixfold higher in Hx CM_ADSC than in Nx CM_ADSC (P < 0.05), while IL6 and IL-1B were not significantly higher (P > 0.05).

FIG. 6.

Relative fold change of secreted VEGFA, IL6, IL-1B, FGF2, HB-EGF, and FGF7 in hypoxic and normoxic conditioned media after 48 h culture of ADSC (*P ≤ 0.05).

Discussion

The major finding of this study was that ADSCs produced a paracrine activity to promote wound healing mechanisms in hTM and HaCaT keratinocytes in vitro. Conditioned media from hypoxic ADSC cultures promoted further increase in both proliferation and migration of keratinocyte cell lines compared to normoxic conditioned media. To find active candidates for this paracrine effect, we identified which wound healing genes coding for secreted proteins were expressed by ADSC and were regulated by hypoxic culture conditions. We were also able to exclude some candidates based on a lack of gene expression, or no change with hypoxia. This study is the first to show that ADSC-secreted VEGF might influence keratinocyte growth and wound healing in a paracrine manner and could influence MSC therapy at the tympanic membrane.

ADSCs in TM wound healing and keratinocytes

Recent studies show that human bone marrow-derived MSCs promote healing of chronic TM perforation in rats [8] and that mouse bone marrow-derived MSCs promote healing of acute TM perforation in mice [9]. Neither study identified the mechanism of these effects. In this study, we show for the first time an improved in vitro wound healing of hTM keratinocytes using adipose-derived MSC paracrine activity. These data provide a plausible mechanism for the in vivo findings in rodents and support further investigation into the therapeutic potential of conditioned media or growth factor therapy, rather than the logistically more difficult cell implantation therapy.

Based on the wound-healing literature, keratinocyte proliferation and migration can be regulated by exposure to bioactive molecules from exogenous, paracrine, or autocrine sources. Factors involved include EGF, TGFα, TGFβ1, HB-EGF, FGF2, KGF, IL6, and VEGF A. [16,24 –27]. In TM perforations, however, the molecular mechanisms regulating wound healing are less well studied [5,8,28], and the search for an optimal growth promoting activity is ongoing.

Response to normoxic and hypoxic CM_ADSC

In the present study, ADSCs in a normoxic environment enhanced proliferation and migration in keratinocytes consistent with previous studies on epithelial cells such as skin keratinocytes [29 –32] or endothelial cells [32 –34] and on nonepithelial cells like fibroblasts [16,33]. Conditioned media from hypoxic culture conditions produced a greater regenerative response in hTM and HaCaT keratinocytes, suggesting that in hypoxia ADSCs secreted more or different paracrine factors to that in normoxia. A hypoxic environment stimulates greater production of some paracrine factors from MSC [34] through activation of hypoxia inducible factor alpha (HIF-1α) transcription factor [2,35 –38]. MSCs using glycolysis rather than mitochondrial respiration in hypoxic environments [39] increase their expression of paracrine factors [40 –42]. The migration response in both hTM and HaCaT keratinocytes to Hx CM_ADSC was greater than responses to serum, indicating a high degree of responsiveness and specificity for chemotactic mechanisms by the paracrine mediators.

Identity of the paracrine activity from ADSCs

From a wound healing PCR array on ADSCs we identified gene expression of secreted paracrine factors and found genes expressed differently in Hx ADSCs versus Nx ADSCs. Eleven genes expressed but not regulated by hypoxia were set aside as not likely to explain the changes in paracrine activity seen in hypoxia. Among the genes for secreted proteins, 12 were not expressed, 24 were upregulated, and 9 were downregulated in Hx ADSCs. These are all candidates worthy of further consideration, as the differential expression in genes could be associated with the difference in paracrine activity secreted. (Table 4).

Table 4.

Selected Genes Expressed by Adipose-Derived Stem Cells in This Study Categorized Based on Significant Relationship to Keratinocyte Growth

Gene	Known relation to keratinocytes	Expression in ADSCs
Chemotactic
MMP9	Stimulates epithelial cell migration [55 –57]	Upregulated in Hx
	MMP9-deficient mouse model showed arrested epithelial wound healing and mice were unable to remove fibrinogen matrix during wound healing [58]
CXCL1, 2, 5, 7, CCL2, 7	No known relation	Downregulated in Hx
Angiogenic
VEGFA	Binds to VEGFR1 and VEGFR2	Upregulated in Hx
	Keratinocytes express VEGFR1 [59,60]
	Keratinocytes also express and secrete VEGFA [59,61 –63]
	Suggested proliferative signals in promoting repair through autocrine pathway in keratinocytes [60]
SERPINE1	One of the upregulated genes measured in HaCaT when exposed to serum [64]	Upregulated in Hx
	TGFB1 Found to increase expression of SERPINE1 in keratinocytes [65]
	Contributes to wound repair through cellular migration [66]
	Inhibits proliferation while promoting migration [67]
	Stimulates keratinocyte adhesion and wound-initiated planar migration [68]
Inflammatory/immunomodulatory
IL6	Inflammatory cytokine expressed in keratinocytes [69,70]	Upregulated in Hx
	Expression of IL6 receptor suggests an autocrine regulation of keratinocyte growth [70]
	Stimulates proliferation in vitro [69]
	IL6-deficient mouse model displayed hindered cutaneous wound healing process [71]
TNF	TNF and its receptors are expressed by keratinocytes [72]	Downregulated in Hx
	TNF induces MMP9 production in keratinocytes [57,71,72]
	MMP9 regulates keratinocyte migration in a TNF-dependent manner [73]
TGFβ1	Known as keratinocyte growth inhibitor although it is secreted as well by keratinocytes [74,75]	Upregulated in Hx
	Found to increase expression of SERPINE1 in keratinocytes [65]
	Found to stimulate reepithelialization regulated by CTGF and Ras/MEK/ERK signaling pathway [76]
	Stimulate epithelial cell migration but not proliferation [76]
FGF2	Binds to FGFR2b expressed in keratinocytes [77,78]	Expressed in both at similar levels
	Stimulates keratinocyte migration [79]
	Topical application of FGF2 improved healing of traumatic tympanic membrane perforation [47]
FGF7	Known as an important factor in wound healing [80], which acts on keratinocytes in paracrine manner [81]	Expressed in both at similar levels
	Binds to FGFR2b expressed in keratinocytes [77]
	Stimulates keratinocyte migration [82]
	Inhibition of KGF receptor signaling reduces proliferation rate of keratinocytes at wound edge and delayed reepithelialization [59]
TGFα	Secreted by keratinocytes and promotes cell proliferation through autocrine signaling [83]	Not expressed in ADSCs
	Binds to ErbB1 receptor [78] expressed on keratinocytes [77]
	Induces IL6 secretion in keratinocytes indicating paracrine interaction [3]
Coagulation cascade
F3	Regulates cell migration in a coagulation-independent manner in HaCaT [84]	Upregulated the most in Hx

Several important keratinocyte growth regulators from the literature such as TGFα [24] and EGF were not expressed by ADSC. HB-EGF, another important regulator of TM wound healing [43,44], was expressed but not regulated significantly by hypoxia, and secreted protein was not detected. FGF-2, also used to promote TM wound healing in other studies [43,45 –47], was expressed and secreted into the CM_ADSC, but was not regulated by hypoxia in our study. These data suggest that there is significant growth activity from factors other than those already presented in the TM repair literature. Other promoters of keratinocyte growth such as VEGFA and IL6 were upregulated by hypoxia, and secreted VEGF was increased most and so seems the most likely candidate to increase keratinocyte growth response (Table 4). The other cytokines we tested showed no significant effects of hypoxia on protein levels accumulated, so their influence remains uncertain and will require further study.

Consistent with previous findings [17,19,40,48], VEGFA protein levels in CM_ADSC were significantly increased by hypoxia. Of note in this study, VEGFA had been shown to be upregulated in traumatic TM perforations, in both epithelial and fibrous layers of the TM, and was reduced in nonhealing TM [49]. This could represent a novel finding for the importance of VEGFA in the healing of TM perforations and to improve healing by regulating the production of VEGFA and other factors.

The gene enrichment analysis is another approach that may provide insight; it showed a downregulation of neutrophil chemotaxis, and neutrophils are an important cell type in the inflammation phase of wound healing [50]. HIF-1α has been shown to reduce the recruitment of neutrophils in vitro through nestin-1 induction [51] and similar mechanisms may be apparent in keratinocytes. Our finding of ADSC regulated chemokines during hypoxia indicates that the paracrine effect may be directed toward regeneration through tissue repair mechanisms and reduction of inflammation.

Comparison of these data with TM wound healing microarray [52] may provide further context for these genes; however, further experiments such as proteomic analysis of the conditioned media or PCR arrays on keratinocytes posttreatment will be required to support the PCR and ELISA data. A broader transcriptome/proteome approach would also be valuable, particularly together with the rat TM wound healing transcriptome data already available [53] and our limited data set from human TM keratinocyte gene expression array. Once a set of candidates is refined, receptor inhibitors or recombinant proteins could be used to target specific pathways in vitro or in vivo.

Practical applications of this finding include cell-free conditioned media approaches by topical application in vivo or combining the mediator(s) with scaffolds to promote wound healing in a chronic TM perforation [46]. These data may also support use of fat plug myringoplasty [54,55], where direct benefits might come from ADSC present in the graft. The paracrine activity identified in this study requires further characterization and optimization regarding which tissue MSCs to use, what conditions might optimize the activity, and what is the nature of the molecular entities producing the activity. Both adipose- and bone marrow-derived MSC produce cytokines and growth factors. Adipose tissue can be a preferred source for MSC therapies, due to their low risk in isolation, high cell yields, and ready availability in most individuals.

Conclusion

In summary, this study has identified a wound healing activity from ADSCs that promotes growth and migration of tympanic membrane keratinocytes. The activity was regulated by hypoxia and was associated with expression of a number of secreted proteins that could be major effectors of keratinocyte proliferation and migration. These findings not only revealed the paracrine effects of ADSCs potentially affecting TM keratinocyte growth but also encourage further investigations for stem cells or their conditioned media as possible therapeutics for chronic TM wound healing.

Footnotes

Acknowledgments

The HaCaT cell line was kindly supplied by the Burns Injury Research Unit (School of Surgery, University of Western Australia). The work was funded, in part, by grants from the Garnett Passe and Rodney Williams Memorial Foundation (R.J.D.), National Health and Medical Research Council (R.J.D., M.D.A.), the Gift of Hearing (R.J.M), and Raine Foundation (M.v.U.).

Author Disclosure Statement

No competing financial interests exist.

References

Gurtner

, Werner

, Barrandon

and Longaker

. (2008). Wound repair and regeneration. Nature, 453:314–321.

Mole

, Blancher

, Copley

, Pollard

, Gleadle

, Ragoussis

and Ratcliffe

. (2009). Genome-wide association of hypoxia-inducible factor (HIF)-1alpha and HIF-2alpha DNA binding with expression profiling of hypoxia-inducible transcripts. J Biol Chem, 284:16767–16775.

Aragane

, Yamada

, Schwarz

, Poppelmann

, Luger

, Tezuka

and Schwarz

. (1996). Transforming growth factor-alpha induces interleukin-6 in the human keratinocyte cell line HaCaT mainly by transcriptional activation. J Invest Dermatol, 106:1192–1197.

Santa Maria

, Atlas

and Ghassemifar

. (2007). Chronic tympanic membrane perforation: a better animal model is needed. Wound Repair Regen, 15:450–458.

Santa Maria

, Redmond

, Atlas

and Ghassemifar

. (2010). Histology of the healing tympanic membrane following perforation in rats. Laryngoscope, 120:2061–2070.

Makino

, Amatsu

, Kinishi

and Mohri

. (1990). Epithelial migration in the healing process of tympanic membrane perforations. Eur Arch Otorhinolaryngol, 247:352–355.

Kaftan

, Herzog

, Miehe

and Hosemann

. (2006). Topical application of transforming growth factor-beta1 in acute traumatic tympanic membrane perforations: an experimental study in rats. Wound Repair Regen, 14:453–456.

Rahman

, Olivius

, Dirckx

, Von Unge

and Hultcrantz

. (2008). Stem cells and enhanced healing of chronic tympanic membrane perforation. Acta Otolaryngol, 128:352–359.

Goncalves

, Bas

, Goldstein

and Angeli

. (2016). Effects of cell-based therapy for treating tympanic membrane perforations in mice. Otolaryngol Head Neck Surg, 154:1106–1114.

10.

Kim

, Park

, Sung

, Yang

, Park

, Kwak

and Park

. (2007). Wound healing effect of adipose-derived stem cells: a critical role of secretory factors on human dermal fibroblasts. J Dermatol Sci, 48:15–24.

11.

Weil

, Markel

, Herrmann

, Abarbanell

and Meldrum

. (2009). Mesenchymal stem cells enhance the viability and proliferation of human fetal intestinal epithelial cells following hypoxic injury via paracrine mechanisms. Surgery, 146:190–197.

12.

Javazon

, Keswani

, Badillo

, Crombleholme

, Zoltick

, Radu

, Kozin

, Beggs

, Malik

and Flake

. (2007). Enhanced epithelial gap closure and increased angiogenesis in wounds of diabetic mice treated with adult murine bone marrow stromal progenitor cells. Wound Repair Regen, 15:350–359.

13.

Gnecchi

, Zhang

, Ni

and Dzau

. (2008). Paracrine mechanisms in adult stem cell signaling and therapy. Circ Res, 103:1204–1219.

14.

Lau

, Paus

, Tiede

, Day

and Bayat

. (2009). Exploring the role of stem cells in cutaneous wound healing. Exp Dermatol, 18:921–933.

15.

Hocking

and Gibran

. (2010). Mesenchymal stem cells: paracrine signaling and differentiation during cutaneous wound repair. Exp Cell Res, 316:2213–2219.

16.

Walter

, Wright

, Fuller

HR.

, MacNeil

and Johnson

. (2010). Mesenchymal stem cell-conditioned medium accelerates skin wound healing: an in vitro study of fibroblast and keratinocyte scratch assays. Exp Cell Res, 316:1271–1281.

17.

, Chen

, Scott

and Tredget

. (2007). Mesenchymal stem cells enhance wound healing through differentiation and angiogenesis. Stem Cells, 25:2648–2659.

18.

Matsuda

, Falkenberg

, Woods

, Choi

, Morrison

and Dilley

. (2013). Adipose-derived stem cells promote angiogenesis and tissue formation for in vivo tissue engineering. Tissue Eng Part A, 19:1327–1335.

19.

Chen

, Tredget

, Wu

and Wu

. (2008). Paracrine factors of mesenchymal stem cells recruit macrophages and endothelial lineage cells and enhance wound healing. PLoS One, 3:e1886.

20.

Falanga

. (1993). Chronic wounds: pathophysiologic and experimental considerations. J Invest Dermatol, 100:721–725.

21.

Spandow

, Hellstrom

and Dahlstrom

. (1996). Structural characterization of persistent tympanic membrane perforations in man. Laryngoscope, 106(Pt 1):346–352.

22.

Redmond

, Levin

, Heel

, Atlas

and Marano

. (2011). Phenotypic and genotypic profile of human tympanic membrane derived cultured cells. J Mol Histol, 42:15–25.

23.

Reimand

, Kull

, Peterson

, Hansen

and Vilo

. (2007). g:Profiler—a web-based toolset for functional profiling of gene lists from large-scale experiments. Nucleic Acids Res, 35:W193–W200.

24.

Teh

, Redmond

, Shen

, Atlas

, Marano

and Dilley

. (2013). TGF-alpha/HA complex promotes tympanic membrane keratinocyte migration and proliferation via ErbB1 receptor. Exp Cell Res, 319:790–799.

25.

Ando

and Jensen

. (1993). Epidermal growth factor and insulin-like growth factor I enhance keratinocyte migration. J Invest Dermatol, 100:633–639.

26.

Clymer

, Schwaber

and Davidson

. (1996). The effects of keratinocyte growth factor on healing of tympanic membrane perforations. Laryngoscope, 106(Pt 1):280–285.

27.

Haase

, Evans

, Pofahl

and Watt

. (2003). Regulation of keratinocyte shape, migration and wound epithelialization by IGF-1- and EGF-dependent signalling pathways. J Cell Sci, 116(Pt 15):3227–3238.

28.

Teh

, Marano

, Shen

, Friedland

, Dilley

and Atlas

. (2013). Tissue engineering of the tympanic membrane. Tissue Eng Part B Rev, 19:116–132.

29.

Yuan

, Lei

, Fu

, Sheng

, Cai

, and Sun

. (2008). Promotive effect of adipose-derived stem cells on the wound model of human epidermal keratinocytes in vitro [in Chinese]. Zhonghua Wai Ke Za Zhi, 46:1575–1578.

30.

Altman

, Yan

, Matthias

, Bai

, Rios

, Mathur

, Song

and Alt

. (2009). IFATS collection: human adipose-derived stem cells seeded on a silk fibroin-chitosan scaffold enhance wound repair in a murine soft tissue injury model. Stem Cells, 27:250–258.

31.

Lee

, Lee

and Cho

. (2011). Effects of human adipose-derived stem cells on cutaneous wound healing in nude mice. Ann Dermatol, 23:150–155.

32.

Huang

, Hsu

, Chang

, Wang

, Deng

, Dai

, Chen

, Chan

, Chen

and Huang

. (2012). Adipose-derived stem cells seeded on acellular dermal matrix grafts enhance wound healing in a murine model of a full-thickness defect. Ann Plast Surg, 69:656–662.

33.

Lee

, Jin

, Song

, Seo

and Cho

. (2012). Paracrine effects of adipose-derived stem cells on keratinocytes and dermal fibroblasts. Ann Dermatol, 24:136–143.

34.

Kinnaird

, Stabile

, Burnett

, Lee

, Barr

, Fuchs

and Epstein

. (2004). Marrow-derived stromal cells express genes encoding a broad spectrum of arteriogenic cytokines and promote in vitro and in vivo arteriogenesis through paracrine mechanisms. Circ Res, 94:678–685.

35.

Andrikopoulou

, Zhang

, Sebastian

, Marti

, Liu

, Milner

and Harmon

. (2011). Current Insights into the role of HIF-1 in cutaneous wound healing. Curr Mol Med, 11:218–235.

36.

Hong

, Hu

, Esquivel

, Liang

, Rennert

, McArdle

, Paik

, Duscher

, Gurtner

, Lorenz

and Longaker

. (2014). The role of hypoxia-inducible factor in wound healing. Adv Wound Care (New Rochelle), 3:390–399.

37.

Ahluwalia

and Tarnawski

. (2012). Critical role of hypoxia sensor—HIF-1alpha in VEGF gene activation. Implications for angiogenesis and tissue injury healing. Curr Med Chem, 19:90–97.

38.

Semenza

, Jiang

, Leung

, Passantino

, Concordet

, Maire

and Giallongo

. (1996). Hypoxia response elements in the aldolase A, enolase 1, and lactate dehydrogenase A gene promoters contain essential binding sites for hypoxia-inducible factor 1. J Biol Chem, 271:32529–32537.

39.

Mylotte

, Duffy

, Murphy

, O'Brien

, Samali

, Barry

and Szegezdi

. (2008). Metabolic flexibility permits mesenchymal stem cell survival in an ischemic environment. Stem Cells, 26:1325–1336.

40.

Rehman

, Traktuev

, Li

, Merfeld-Clauss

, Temm-Grove

, Bovenkerk

, Pell

, Johnstone

, Considine

and March

. (2004). Secretion of angiogenic and antiapoptotic factors by human adipose stromal cells. Circulation, 109:1292–1298.

41.

, Pan

, Neo

, Yong

and Bian

. (2008). Cyclooxygenase-2 mediates the delayed cardioprotection induced by hydrogen sulfide preconditioning in isolated rat cardiomyocytes. Pflugers Arch, 455:971–978.

42.

Mirotsou

, Jayawardena

, Schmeckpeper

, Gnecchi

and Dzau

. (2011). Paracrine mechanisms of stem cell reparative and regenerative actions in the heart. J Mol Cell Cardiol, 50:280–289.

43.

Hakuba

, Taniguchi

, Shimizu

, Sugimoto

, Shinomori

and Gyo

. (2003). A new method for closing tympanic membrane perforations using basic fibroblast growth factor. Laryngoscope, 113:1352–1355.

44.

Alzahrani

and Saliba

. (2015). Hyaluronic acid fat graft myringoplasty vs fat patch fat graft myringoplasty. Eur Arch Otorhinolaryngol, 272:1873–1877.

45.

Santa Maria

, Redmond

, Atlas

and Ghassemifar

. (2011). Keratinocyte growth factor 1, fibroblast growth factor 2 and 10 in the healing tympanic membrane following perforation in rats. J Mol Histol, 42:47–58.

46.

Kanemaru

, Umeda

, Kitani

, Nakamura

, Hirano

and Ito

. (2011). Regenerative treatment for tympanic membrane perforation. Otol Neurotol, 32:1218–1223.

47.

Lou

and Wang

YBZ

. (2013). Healing outcomes of large (>50%)traumatic membrane perforations with inverted edges following no intervention, edge approximation and fibroblast growth factor application; a sequential allocation, three-armed trial. Clin Otolaryngol, 37:446–451.

48.

Hsiao

, Lokmic

, Peshavariya

, Abberton

, Dusting

, Lim

and Dilley

. (2013). Hypoxic conditioning enhances the angiogenic paracrine activity of human adipose-derived stem cells. Stem Cells Dev, 22:1614–1623.

49.

Cho

, Lee

, Shin

, Park

and Chon

. (2010). The importance of vascular endothelial growth factor in the healing of acute tympanic membrane perforation. Am J Otolaryngol, 31:309–314.

50.

Maxson

, Lopez

, Yoo

, Danilkovitch-Miagkova

and Leroux

. (2012). Concise review: role of mesenchymal stem cells in wound repair. Stem Cells Transl Med, 1:142–149.

51.

Rosenberger

, Schwab

, Mirakaj

, Masekowsky

, Mager

, Morote-Garcia

, Unertl

and Eltzschig

. (2009). Hypoxia-inducible factor-dependent induction of netrin-1 dampens inflammation caused by hypoxia. Nat Immunol, 10:195–202.

52.

Santa Maria

, Redmond

, McInnes

, Atlas

and Ghassemifar

. (2011). Tympanic membrane wound healing in rats assessed by transcriptome profiling. Laryngoscope, 121:2199–2213.

53.

Dursun

, Dogru

, Gungor

, Cincik

, Poyrazoglu

and Ozdemir

. (2008). Comparison of paper-patch, fat, and perichondrium myringoplasty in repair of small tympanic membrane perforations. Otolaryngol Head Neck Surg, 138:353–356.

54.

Saliba

, Alzahrani

, Zhu

and Chemtob

. (2014). Growth factors expression in hyaluronic acid fat graft myringoplasty. Laryngoscope, 124:E224–E230.

55.

Madlener

, Parks

and Werner

. (1998). Matrix metalloproteinases (MMPs) and their physiological inhibitors (TIMPs) are differentially expressed during excisional skin wound repair. Exp Cell Res, 242:201–210.

56.

Legrand

, Gilles

, Zahm

, Polette

, Buisson

, Kaplan

, Birembaut

and Tournier

. (1999). Airway epithelial cell migration dynamics. MMP-9 role in cell-extracellular matrix remodeling. J Cell Biol, 146:517–529.

57.

Makela

, Larjava

, Pirila

, Maisi

, Salo

, Sorsa

and Uitto

. (1999). Matrix metalloproteinase 2 (gelatinase A) is related to migration of keratinocytes. Exp Cell Res, 251:67–78.

58.

Mohan

, Chintala

, Jung

, Villar

, McCabe

, Russo

, Lee

, McCarthy

, Wollenberg

, et al. (2002). Matrix metalloproteinase gelatinase B (MMP-9) coordinates and effects epithelial regeneration. J Biol Chem, 277:2065–2072.

59.

Werner

, Breeden

, Hubner

, Greenhalgh

and Longaker

. (1994). Induction of keratinocyte growth factor expression is reduced and delayed during wound healing in the genetically diabetic mouse. J Invest Dermatol, 103:469–473.

60.

Wilgus

, Matthies

, Radek

, Dovi

, Burns

, Shankar

and DiPietro

. (2005). Novel function for vascular endothelial growth factor receptor-1 on epidermal keratinocytes. Am J Pathol, 167:1257–1266.

61.

Brown

, Yeo

, Berse

, Yeo

, Senger

, Dvorak

and van de Water

. (1992). Expression of vascular permeability factor (vascular endothelial growth factor) by epidermal keratinocytes during wound healing. J Exp Med, 176:1375–1379.

62.

Detmar

, Brown

, Berse

, Jackman

, Elicker

, Dvorak

and Claffey

. (1997). Hypoxia regulates the expression of vascular permeability factor/vascular endothelial growth factor (VPF/VEGF) and its receptors in human skin. J Invest Dermatol, 108:263–268.

63.

Sen

, Khanna

, Babior

, Hunt

, Ellison

and Roy

. (2002). Oxidant-induced vascular endothelial growth factor expression in human keratinocytes and cutaneous wound healing. J Biol Chem, 277:33284–33290.

64.

, Higgins

, Lu

, Samarakoon

, Wilkins-Port

, Ye

, Higgins

, Staiano-Coico

and Higgins

. (2008). SERPINE1 (PAI-1) is a prominent member of the early G0—> G1 transition “wound repair” transcriptome in p53 mutant human keratinocytes. J Invest Dermatol, 128:749–753.

65.

Keski-Oja

and Koli

. (1992). Enhanced production of plasminogen activator activity in human and murine keratinocytes by transforming growth factor-beta 1. J Invest Dermatol, 99:193–200.

66.

Simone

, Longmate

, Law

and Higgins

. (2015). Targeted inhibition of PAI-1 activity impairs epithelial migration and wound closure following cutaneous injury. Adv Wound Care (New Rochelle), 4:321–328.

67.

Simone

, Higgins

, Czekay

, Law

, Higgins

, Archambeault

, Kutz

and Higgins

. (2014). SERPINE1: a molecular switch in the proliferation-migration dichotomy in wound-“activated” keratinocytes. Adv Wound Care (New Rochelle), 3:281–290.

68.

Providence

, Higgins

, Mullen

, Battista

, Samarakoon

, Higgins

, Wilkins-Port

and Higgins

. (2008). SERPINE1 (PAI-1) is deposited into keratinocyte migration “trails” and required for optimal monolayer wound repair. Arch Dermatol Res, 300:303–310.

69.

Grossman

, Krueger

, Yourish

, Granelli-Piperno

, Murphy

, May

, Kupper

, Sehgal

and Gottlieb

. (1989). Interleukin 6 is expressed in high levels in psoriatic skin and stimulates proliferation of cultured human keratinocytes. Proc Natl Acad Sci U S A, 86:6367–6371.

70.

Yoshizaki

, Nishimoto

, Matsumoto

, Tagoh

, Taga

, Deguchi

, Kuritani

, Hirano

, Hashimoto

, et al. (1990). Interleukin 6 and expression of its receptor on epidermal keratinocytes. Cytokine, 2:381–387.

71.

Gallucci

, Simeonova

, Matheson

, Kommineni

, Guriel

, Sugawara

and Luster

. (2000). Impaired cutaneous wound healing in interleukin-6-deficient and immunosuppressed mice. FASEB J, 14:2525–2531.

72.

Scott

, Arnott

, Robinson

, Moore

, Thompson

, Marshall

and Balkwill

. (2004). TNF-alpha regulates epithelial expression of MMP-9 and integrin alphavbeta6 during tumour promotion. A role for TNF-alpha in keratinocyte migration?. Oncogene, 23:6954–6966.

73.

Song

, Ju

, Goh

, Kwon

, Choi

and Park

. (2011). Suppression of TNF-alpha-induced MMP-9 expression by a cell-permeable superoxide dismutase in keratinocytes. BMB Rep, 44:462–467.

74.

Shipley

, Pittelkow

, Wille

Jr , Scott

and Moses

. (1986). Reversible inhibition of normal human prokeratinocyte proliferation by type beta transforming growth factor-growth inhibitor in serum-free medium. Cancer Res, 46(Pt 2):2068–2071.

75.

Hashimoto

. (2000). Regulation of keratinocyte function by growth factors. J Dermatol Sci, 24(Suppl 1):S46–S50.

76.

Secker

, Shortt

, Sampson

, Schwarz

, Schultz

and Daniels

. (2008). TGFbeta stimulated re-epithelialisation is regulated by CTGF and Ras/MEK/ERK signalling. Exp Cell Res, 314:131–142.

77.

Finch

and Rubin

. (2004). Keratinocyte growth factor/fibroblast growth factor 7, a homeostatic factor with therapeutic potential for epithelial protection and repair. Adv Cancer Res, 91:69–136.

78.

Shirakata

(2010). Regulation of epidermal keratinocytes by growth factors. J Dermatol Sci, 59:73–80.

79.

Sogabe

, Abe

, Yokoyama

and Ishikawa

. (2006). Basic fibroblast growth factor stimulates human keratinocyte motility by Rac activation. Wound Repair Regen, 14:457–462.

80.

Werner

. (1998). Keratinocyte growth factor: a unique player in epithelial repair processes. Cytokine Growth Factor Rev, 9:153–165.

81.

Finch

, Rubin

, Miki

, Ron

and Aaronson

. (1989). Human KGF is FGF-related with properties of a paracrine effector of epithelial cell growth. Science, 245:752–755.

82.

Tsuboi

, Sato

, Kurita

, Ron

, Rubin

and Ogawa

. (1993). Keratinocyte growth factor (FGF-7) stimulates migration and plasminogen activator activity of normal human keratinocytes. J Invest Dermatol, 101:49–53.

83.

Coffey

Jr , Derynck

, Wilcox

, Bringman

, Goustin

, Moses

and Pittelkow

. (1987). Production and auto-induction of transforming growth factor-alpha in human keratinocytes. Nature, 328:817–820.

84.

Dorfleutner

, Hintermann

, Tarui

, Takada

and Ruf

. (2004). Cross-talk of integrin alpha3beta1 and tissue factor in cell migration. Mol Biol Cell, 15:4416–4425.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.02 MB

0.53 MB

0.00 MB

Paracrine Activity from Adipose-Derived Stem Cells on In Vitro Wound Healing in Human Tympanic Membrane Keratinocytes

Abstract

Introduction

Materials and Methods

Culture of primary hTM keratinocytes and HaCaT cells

Culture of adipose-derived stem cell

Collection of ADSC-conditioned medium

Total RNA isolation and complementary DNA synthesis

Real-time polymerase chain reaction of human wound healing array

Gene ontology analysis of hypoxia-induced gene expression changes in cultured ADSC

Total protein assay

Quantification of specific protein levels

MTS assay for proliferation

Scratch test assay for migration

Statistics

Results

Effect of CMADSC on keratinocyte proliferation

Effect of CMADSC on keratinocyte migration

Molecular identification of paracrine activity

PCR array for wound healing gene expression in ADSCs and hTM keratinocytes

Gene expression

Hypoxia in ADSCs

Gene expression regulated by hypoxia

Gene ontology analysis

Protein concentration in conditioned media

Discussion

ADSCs in TM wound healing and keratinocytes

Response to normoxic and hypoxic CMADSC

Identity of the paracrine activity from ADSCs

Conclusion

Footnotes

Acknowledgments

Author Disclosure Statement

References

Supplementary Material

Effect of CM_ADSC on keratinocyte proliferation

Effect of CM_ADSC on keratinocyte migration

Response to normoxic and hypoxic CM_ADSC