Dulbecco's modified Eagle's medium (DMEM) and fetal bovine serum (FBS) were obtained from Hyclone (Logan, UT, USA). Endothelial cell growth medium-2 (EGM-2) Bulletkit was purchased from Lonza (Walkersville, MD, USA). MMP-1 antibody was bought from Calbiochem (San Diego, CA, USA). Anti-phospho-STAT3 (Tyr705), anti-phospho-c-Jun N-terminal kinase (JNK), and anti-phospho-nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB) antibodies were purchased from Cell Signaling Technology (Beverly, MA, USA), and anti-β actin antibody was obtained from Santa Cruz Biotechnology (Santa Cruz, CA, USA).

Ten healthy volunteers each donated 10 mL of whole blood. The donors gave informed consent for participation in the study. For the preparation of PRP, 10 mL of blood was collected in an acid citrate dextrose (ACD-A) tube coated with anti-coagulating agent (Sigma-Aldrich, 10% (v/v)) and centrifuged at 250×g for 15 min at 37℃. The supernatant fraction was harvested and incubated for 10 min at 37℃, then centrifuged at 1,900×g for 8 min. Pellets were resuspended in Tyrode's buffer (137 mM NaCl, 2.7 mM KCl, 16 mM NaHCO₃, 5 mM MgCl₂, 3.5 mM HEPES, 2.5 mM CaCl₂, 5.5 mM glucose, 2 mg/mL BSA, pH 7.4) containing 10 U/mL heparin and 0.5 µM PGI2, and incubated for 10 min at 37℃. After centrifugation at 1,900×g for 8 min, Tyrode's buffer containing 0.02 U/mL of heparin was added to resuspend pellets and obtain PRP. Fluorescence-activated cell sorting (FACS) analysis was performed to stain P-selectin on platelets after stimulation with thrombin, and it was confirmed that 90% of the cell population expressed P-selectin. PRP containing 1.5×10⁹ platelets/mL was prepared and used for further study. PRP stimulation was performed for 5 min with 5 µM ADP, 1% Ca²⁺, 20 µg/mL collagen, or 20 U/mL thrombin.

To analyze cytokines released from PRP, a multiplex MAP human cytokine/chemokine immunoassay (Millipore, Billerica, MA, USA) was used following the manufacturer's protocol (Luminex 100™ IS Total System, XMAP Technology with MasterPlex software, Millipore, Billerica, MA, USA). Forty growth factors, inflammatory cytokines, and chemokines, including PDGF-AA, AB/BB, TGF-α, VEGF, FGF, eotaxin, granulocyte-macrophage colony-stimulating factor (GM-CSF), interferon (IFN)-α2, IFN-γ, interleukin (IL)-10, IL-12, IL-1α, IL-1β, IL-1Rα, IL-2, IL-6, interferon gamma-induced protein 10 (IP-10), monocyte chemoattractant protein (MCP)-1, Regulated on activation normal T cell expressed and secreted (RANTES), and tumor necrosis factor-α (TNF-α), were quantitatively analyzed. For the analysis of TGF-β1, 2, and 3 isoforms, multiplex TGF-β1, 2, 3 immunoassay (Millipore, Billerica, MA, USA) was used following the manufacturer's protocol.

HUVECs were kindly gifted by Dr. Y.S. Choi (CHA University, Seoul, Korea) and cultured in endothelial cell growth medium (EGM) supplemented with growth factors, 10% FBS, and 1% penicillin/streptomycin solution (10,000 IU/mL). HaCaT cells (an immortalized keratinocyte cell line) were kindly gifted from Dr H.K. Lee (Yonsei University, Seoul, Korea), and HDFs were purchased from BioBud Co. (Seungnam, Korea). HaCaT cells and HDFs were cultured in DMEM (Sciencell, USA) containing 10% fetal bovine serum (FBS), 100 U/mL penicillin, and 100 µg/mL streptomycin (GIBCO) at 37℃, 5% CO₂. HUVEC, and HDF cultures used for the study were passage 8 or less.

HUVECs were plated on a 1% gelatin-coated plate at a density of 40,000 cells/35-mm dish HaCaT cells were plated at a density of 50,000 cells/35-mm dish. The following day, cells were starved with serum-free medium for 24 hr. The culture medium was then replaced with fresh serum-free medium, and 200 µL of PRP was added (10% v/v). After incubation for 24 hr, cells were washed with PBS 3 times, trypsinized, and counted using a hemocytometer.

Cell migration was assessed by scratch wound assay [18]. Briefly, HaCaT cells were plated at a density of 80,000 cells/35-mm dish, and HDFs were plated at a density of 50,000 cells/35-mm dish. The following day, culture medium was replaced with fresh serum-free DMEM, and scratch wounds were created in a cell monolayer using a sterile pipette tip. After washing away suspended cells, the culture medium was replaced with fresh serum-free medium and 200 µL of PRP was added (10% v/v). After 24 hr, the wound width was estimated using a light microscope (Nikon, Japan) equipped with a digital camera, and 4 measurements of wound width were made at randomly chosen points. The extent of wound closure was presented as the percentage by which the original scratch width had decreased at each measured time point.

Two mL of conditioned media was concentrated to 500 µL by Vivaspin 500 (Satorius Stedim Biotech, Goettingen, Germany), and then 10 µL was loaded onto a 10% SDS-polyacrylamide gel containing 1 mg/mL gelatin. The gel was run in Tris/glycine buffer for 4 hr, and then incubated in 2.5% Triton X-100 solution for 15 min twice to remove SDS. To detect gelatinase activity, the gel was incubated in reaction buffer containing 50 mM Tris-HCl (pH 7.4), 10 mM CaCl₂, and 0.05% Brij 35 overnight at 37℃. The gelatinolytic activity was shown by staining with 0.1% (w/v) Coomassie Brilliant Blue R-250, 10% (v/v) glacial acetic acid, and 30% (v/v) methanol, followed by destaining with 10% (v/v) acetic acid and 30% (v/v) methanol.

Proteins were extracted with urea lysis buffer (4 M urea, 0.5% SDS, 0.5% Nonidet P-40, 100 mM Tris (pH 7.4) and 5 mM EDTA). Cell lysates were centrifuged, and supernatants containing soluble proteins were collected and subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). Separated proteins were then transferred onto 0.45-µm PVDF membranes. Blots were blocked with TBST (10 mM Tris-base, 100 mM NaCl, and 0.01% Tween 20) containing 5% bovine serum albumin for 1 hr, and then probed with corresponding antibodies. Membranes were incubated with HRP-coupled appropriate secondary antibodies for 1 hr, and immunoreactive signals were detected by using the ECL detection kit (GE healthcare).

For cell proliferation and cell migration assays, each experiment was repeated at least three times. The experimental data were analyzed by a one-way analysis of variance (ANOVA) or student t test. Statistical analysis was performed using Microsoft Exel Program. Data were presented as mean±SD, and P<0.05 was considered statistically significant.

PRP is considered to be a rich source of cytokines (growth factors/chemokines), thus, we characterized growth factor profiles after stimulation with agonists. One milliliter of resting PRP containing 1.5×10⁹ platelets was stimulated with 5 µM ADP, 20 µg/mL collagen, 1% Ca²⁺, or 20 U/mL thrombin for 5 min. After centrifugation at 12,000×g for 30 min, supernatants were collected, and growth factor analysis was performed using a multiplex human cytokine/chemokine immunoassay system (Millipore Co, Billerica, MA, USA). Our study showed that thrombin acted as a powerful stimulator for the release of PDGF-AB/BB (>10,000 pg/mL), PDGF-AA (>10,000 pg/mL), and bFGF (370 pg/mL). However, the release of VEGF was negligible, and neither EGF nor IGF was detected following thrombin stimulation. When PRP was stimulated with ADP, collagen, or Ca²⁺, the levels of released PDGF-AA, AB/BB, and bFGF were relatively lower than in thrombin-stimulated PRP supernatants (Fig. 1A). This study indicates that the activation of 2 thrombin receptors, protease activated receptor type 1 (PAR-1) and PAR-4 on the platelet surface [19] occurred within 5 min. However, PRP activation was slower in response to ADP, Ca²⁺, and collagen, indicating differences in signaling dynamics between the platelet membrane receptors for these agonists. In an analysis of chemokine profiles, thrombin again acted as a strong stimulator for chemokine release by PRP. RANTES, GRO (cotton rat growth-regulated protein), and soluble CD40 ligand (sCD40L) were the 3 most abundantly expressed chemokines, with levels ranging between 2,000 pg/mL and 5,000 pg/mL. As expected, the highest level of RANTES release was observed by thrombin stimulation (3,228 pg/mL). This effect was not specific to thrombin, since unstimulated PRP, as well as PRP stimulated with ADP, Ca²⁺, or collagen, also released RANTES to concentrations ≥2,500 pg/mL. RANTES in α-granules of platelets is known to be released during the acute stages of inflammation and is a potent chemoattractant for T cells, monocytes, natural killer cells, basophils, and eosinophils [20]. It is notable that unstimulated PRP releases a high level of RANTES, as this indicates that the RANTES release from PRP is less dependent on agonist stimulation than other cytokines/chemokines. In contrast, the release of sCD40L (3,418 pg/mL) and GRO (3,895 pg/mL) from PRP was specific to thrombin stimulation (Fig. 1B). Platelets are a major source of sCD40L, and binding of sCD40L to CD40 receptor initiates inflammatory processes; thus, the pharmaceutical inhibition of sCD40L is being vigorously pursued as a potential avenue for the prevention of cardiovascular disease [21]. GRO, another chemokine stored in platelet α-granules, is a member of the CXC chemokine family, which also includes IL-4, platelet factor 4 (PF4), and macrophage inflammatory protein 2 (MIP-2). GRO is expressed in macrophages, endothelial cells, and fibroblasts, but the functions of GRO are not well understood at present, although increased serum levels of GRO were observed in women with preeclampsia [22]. In our analysis of TGF-β isoforms, thrombin was found to stimulate the release of TGF-β1 (2,435 pg/mL) and TGF-β2 (103.5 pg/mL). However, neither TGF-β3 nor TGF-α was released from PRP stimulated with thrombin, ADP, Ca²⁺, or collagen. PRP was mildly stimulated to release TGF-β1 (100 pg/mL) by ADP, Ca²⁺, and collagen (Fig. 1C).

The proliferation of skin cells is an active process in wound healing, in which coordinated upregulations are seen for PDGF, TGF-β, VEGF, bFGF, keratinocyte growth factor (KGF), and hepatocyte growth factor (HGF) [23]. On the basis of the results of our previous experiments, which showed a large release of growth factors and chemokines from both unstimulated and agonist-stimulated PRP, we further investigated the effects of PRP on cell proliferation. When HUVECs were treated with PRP for 24 hr, cell proliferation was increased 7-fold compared to serum-free controls (P<0.05, N=3). Surprisingly, there were no significant differences in cell proliferation rates between unstimulated PRP, and PRP stimulated with thrombin, ADP, or collagen. However, treatment of HUVECs with Ca²⁺ stimulated PRP significantly induced HUVEC proliferation. Ca²⁺ treatment without PRP treatment did not have any effect on HUVEC proliferation (data not shown). For HaCaT cells, we only tested the effect of unstimulated PRP on cell proliferation. The HaCaT cell proliferation rate was increased 5-fold by PRP compared to serum-free controls (P<0.05, N=4) (Fig. 2). This result is encouraging, because PRP itself may be safe to apply for therapeutics, while PRP agonists could cause adverse non-target effects. For example, treatment with bovine thrombin as a potent platelet activator can cause the development of antibodies against thrombin, which lead to severe postoperative bleeding following bypass graft surgery to treat thrombosis in animal studies [24].

As cell motility and migration play pivotal roles in skin wound healing, we investigated the effect of PRP on cell migration by scratch wound assay. When HaCaT cells and HDFs were treated with PRP for 24 hr, cell migration was increased by 60% in HaCaT cells (P<0.05, N=4) and 35% in HDFs (P<0.05, N=3) (Fig. 3). As PDGF-AA is a key chemoattractant in the process of tissue repair, we consider that PDGF-AA released from PRP may have played a role in inducing the migration of HDFs in this experiment [25]. However, an effect of PDGF-AA on the migration of HaCaT cells can be ruled out, since HaCaT cells do not express PDGF receptors on the cell surface [26]. Instead, TGF-β1 released from PRP is more likely to have been responsible for the migration of HaCaT cells via epithelial-to-mesenchymal transformation (EMT) [27].

Analysis of MMP expression has shown that MMP-1, -2, -3, -9, -10, and -14 are prevalently expressed during the skin wound healing process [28]. HaCaT cells were plated at a density of 300,000 cells/60-mm dish, and HDFs were plated at a density of 200,000 cells/60-mm dish in DMEM containing 10% FBS and 1% P/S. The following day, cells were starved with serum-free medium for 24 hr. Culture media were then replaced with fresh serum-free medium, and cells were incubated with 200 µL PRP (10% v/v) for 48 hr. Culture media were concentrated, MMP-2 and -9 were analyzed by gelatin zymography, and MMP-1 was analyzed by Western blotting. Our study showed that unstimulated PRP contained detectable levels of active MMP-9 and MMP-2. In HaCaT cells, treatment with PRP induced MMP-9 activity but did not have any effect on MMP-2 activity. In addition, MMP-1 expression was observed in unstimulated PRP, and treatment with PRP induced MMP-1 expression in HaCaT cells. However, PRP treatment did not have any significant effect on MMP-1, MMP-2, or MMP-9 activity in HDFs. Our study suggests that the effects of PRP on the regulation of MMP expression differ between HaCaT cells and HDFs (Fig. 4).

To investigate the regulation by PRP of transcription factors capable of transactivation at MMP promoters, we analyzed the activation of STAT-3, activator protein-1 (AP-1) and NF-κB by Western blotting using phospho-specific antibodies [29]. HaCaT cells were plated at a density of 300,000 cells/35-mm dish in DMEM containing 10% FBS and 1% P/S. The following day, cells were starved with serum-free medium for 48 hr. The culture medium was then replaced with fresh serum-free medium, and 200 µL PRP (10%, v/v) was added for 0, 5, 15, 30, 60, or 180 min. Following PRP incubations, cell lysates were prepared, and 40 µg protein was analyzed by Western blotting with anti-phospho-STAT-3, phospho-JNK and phospho-NF-κB antibodies. We found that PRP treatment induced the activation of STAT-3 by phosphorylation of tyr-705 from 5-30 min. A STAT-3 binding site is located in the promoter region of MMP-1 [29]. We found that PRP treatment induced the phosphorylation of JNK at 30 min, which may be responsible for MMP-1 and MMP-9 induction, as AP-1 binding sites are found in the promoter regions of MMP-1, -2, and -9. The activation of NF-κB by PRP was initiated at 5 min and sustained till 60 min. As NF-κB binding sites are also located in the promoter regions of MMP-1 and MMP-9, we assume that the activation of NF-κB may contribute to the induction of MMP-1 and MMP-9 expression (Fig. 5). Studies using specific inhibitors for these transcription factors may be able to further elucidate the mechanisms of MMP regulation by PRP treatment.