Platelet-rich fibrin matrix improves wound angiogenesis via inducing endothelial cell proliferation

Abstract

The economic, social, and public health burden of chronic ulcers and other compromised wounds is enormous and rapidly increasing with the aging population. The growth factors derived from platelets play an important role in tissue remodeling including neovascularization. Platelet-rich plasma (PRP) has been utilized and studied for the last four decades. Platelet gel and fibrin sealant, derived from PRP mixed with thrombin and calcium chloride, have been exogenously applied to tissues to promote wound healing, bone growth, hemostasis, and tissue sealing. In this study, we first characterized recovery and viability of as well as growth factor release from platelets in a novel preparation of platelet gel and fibrin matrix, namely platelet-rich fibrin matrix (PRFM). Next, the effect of PRFM application in a delayed model of ischemic wound angiogenesis was investigated. The study, for the first time, shows the kinetics of the viability of platelet-embedded fibrin matrix. A slow and steady release of growth factors from PRFM was observed. The vascular endothelial growth factor released from PRFM was primarily responsible for endothelial mitogenic response via extracellular signal-regulated protein kinase activation pathway. Finally, this preparation of PRFM effectively induced endothelial cell proliferation and improved wound angiogenesis in chronic wounds, providing evidence of probable mechanisms of action of PRFM in healing of chronic ulcers.

Conflict of interest statement

Disclosure of conflict. Cascade Medical (New Jersey, USA) provided the FIBRINET® Platelet Rich Fibrin Matrix (PRFM) system and partial funding for this study.

2011 by the Wound Healing Society.

Figures

Figure 1. Preparation of PRFM, quantification of platelet recovery and viability

A. An outline of the PRFM preparation protocol. B. Red arrow indicates a ready PRFM membrane that was used for in vitro and in vivo wound studies. C. Platelet recovery determined in the first 20 subjects enrolled. Individual % recovery data as well as average (error bar) recovery is shown. Average data shown are mean ±SD (n=20). D. Changes in platelet viability in PRFM over time. LDH release from cells was used as measure of loss of platelet viability. Data shown are % change compared to 0h or baseline sample (100% viable). Mean ±SD (n=10); **, p<0.001 compared to baseline.

Figure 2. Release of growth factors from PRFM in culture

RFM were prepared and placed in 6-well culture plates with RPMI 1640 media. Growth factors (PDGF-BB, TGF-β1, and VEGF-A) released in media was measured using ELISA. Data shown are actual growth factor levels (ng/ml) in media bathing PRFM. Data are mean ±SD (n=10 for PDGF and n=5 for VEGF and TGFβ1);**, p<0.01 compared to 0h (baseline) sample.

Figure 3. Conditioned media from PRFM induced human endothelial cell proliferation

Human microvascular endothelial cells (HMEC) were cultured in 96-well plates. The cells were serum starved and simultaneously treated with conditioned media from PRFM. The cells were cultured with 10% fetal bovine serum (FBS) in culture media served as positive control. A. Cell proliferation in the presence of PRFM conditioned media (dilution = % conditioned media of the total culture media). Data are mean ± SD (n=5); †, p<0.01 compared to positive control; *, p<0.05&**, p<0.01 compared to conditioned media untreated group. B. HMEC cells cultured in serum free conditions were treated with 25% (i.e. 1 part condition media added to 3 parts of cell culture media) PRFM conditioned media collected on days 1,3 or 7 post PRFM formation. Data are mean ± SD (n=4). †, p<0.01 compared to positive control; *, p<0.05&**, p<0.01 compared to conditioned media untreated group; C, Cell proliferation in the presence of platelet poor plasma-gel conditioned media (PPP-M, dilution = % conditioned media of the total culture media). Data are mean ± SD (n=5); †, p<0.01 compared to positive control.

Figure 4. Role of PDGF-BB and ERK activation in HMEC proliferation induced by PRFM-conditioned media

A. PRFM conditioned media was treated or not with anti-PDGF or anti-VEGF (20μg/ml)blocking antibody to sequester PDGF-BB or VEGF in PRFM conditioned media. Cells cultured in serum free conditions were treated with 25% PRFM for 48h. Data are mean ± SD (n=4). *, p<0.05 compared to PRFM treated & blocking antibody untreated group. B. PRFM-conditioned media (25%) induced rapid(30 min) ERK phosphorylation (p-ERK) in HMEC. C. Quantification of the immunoblot data (B) using densitometry. Ratio of p-ERK/ERK is shown. * p<0.05 compared to PRFM untreated group. D. ERK inhibition attenuated HMEC proliferation induced by PRFM (25%). ERK inhibition was achieved by pretreatment of cells with the ERK inhibitor UO126 (10μM) 5 min prior to PRFM exposure. Data are mean ± SD (n=5).*, p<0.05 compared to PRFM treated and inhibitor untreated group.

Figure 5. ERK knock-down inhibited PRFM-conditioned media induced proliferation of HMEC

Cells were transfected with siRNA to knockdown ERK 1 and 2. Control cells were transfected with scrambled (control) siRNA. A. ERK immunoblot showing effective knock down of ERK following siRNA tranfection. B–C. Quantification of the immunoblot densitometry data of ERK1 (B) and ERK2 (C) shown in A. Ratio of ERK/GAPDH is shown. GAPDH was used as a housekeeping protein standardization of sample loading. Data are mean ± SD (n=3);* p<0.05 compared to control siRNA treated group. D. ERK knockdown attenuated PRFM (25%) induced proliferation of HMEC. HMEC were transfected with ERK 1 & 2 siRNA to knockdown ERK levels in cells. Control cells were transected with scrambled (control) siRNA. ERK siRNA and control siRNA transfected cells were activated with 25% PRFM as described earlier. Results are mean ± S.D. *, p<0.05 compared to scrambled (control) siRNA treated group.

Figure 6. Treatment of porcine ischemic wound with PRFM

A. Four Ischemic wounds (solid circles) were developed on the back of pigs as described in Methods. Six additional non-ischemic wounds were developed (open circles). Arrows indicate direction of blood flow in flaps containing ischemic wounds. The upper left and lower right ischemic wounds were treated with PRFM at the time of wounding. B. Representative digital images of excisional wounds treated or not with PRFM on days 0 and 4 post-wounding.

Figure 7. Ischemic wound histology following treatment with PRFM membranes

Wound-edge tissues were collected on 14 d post-wounding. Formalin-fixed paraffin-embedded biopsy tissues were sectioned (5μm) and stained using A. Masson’s Trichrome staining. This staining results in blue-black nuclei; blue collagen and cytoplasm. Epidermal cells appear reddish. The arrows indicate the start and the end of wound. Scale bar = 500 μm. Right panels are the zoom of boxed area of images shown in left panel. Scale bar = 50 μm. B. Picro-sirius red staining. The red staining is the birefringence of collagen fibers, which is largely due to co-aligned molecules of Type I collagen. Bar graph show quantitation of the collagen fibers in PRFM-treated or untreated wounds. Scale bar = 50 μm. Data are mean ± SD (n=3); * p<0.05 compared to untreated wounds.

Figure 8. Increased vascularization of PRFM-treated ischemic wounds

Wound-edge tissues were collected 14d post-wounding. Formalin fixed paraffin embedded biopsy tissues were sectioned (5 μm) and stained using von Willebrand Factor (brown) and counterstained with hematoxylin (blue). Compared to non-treated ischemic wounds, the treated wounds showed increased vascular endothelial cell staining. Arrows indicate vWF positive vessels. Right panel. Zoom of the boxed area shown in corresponding images to the left. Scale bar = 50 μm. Bar graph show quantitation of the blood vessels in PRFM-treated or untreated wounds. Data are mean ± SD (n=3); * p<0.05 compared to untreated wounds.

Figure 9. Increased number of Ki67 positive cells in PRFM treated porcine wounds

Wound-edge tissues were collected 14 d postwounding. Formalin fixed paraffin embedded biopsy tissues were sectioned (5 μm) and stained using Ki67 (brown), a marker of proliferating cells. The sections were counterstained with hematoxylin (blue). Compared to non-treated ischemic wounds, the treated wounds showed increased number of proliferating cells. Arrows indicate the Ki67 positive proliferating cells. Lower panel. Scale bar = 50 μm. Bar graph show quantitation of the Ki67 positive cells in PRFM treated or untreated wounds. Data are mean ± SD (n=3); * p<0.05 compared to untreated wounds.

Figure 10. Co-localization of Ki67 and vWF positive cells in PRFM treated porcine wounds

Wound-edge tissues were collected 14 d post wounding. Formalin fixed paraffin embedded biopsy tissues were sectioned (5 μm) and stained using Ki67 (red) and vWF (green). Compared to non-treated ischemic wounds, the treated wounds showed increased number of VWF positive proliferating cells. Arrows indicate Ki67 and vWF positive blood vessels. Top two panels. Scale bar = 200 μm. Bottom panel. Scale bar = 20 μm.

Figure 11. Increased CD31 expression in LCM captured granulation tissue from PFRM treated wounds

Wound-edge tissues were collected 14 d post wounding. LCM capture was performed on H&E stained OCT embedded tissue A, visualization of section; B, demarcation (in blue) of region of interest; C, post-dissection and catapulting. Scale bar, 150 μm; D, Real-time PCR was performed on LCM captured granulation tissue from wounds treated or not with PFRM to quantify CD31 expression normalized against PPIA levels. Data are mean ± SD (n=4); *, p<0.05.

Figure 12. Increased blood flow in PFRM treated wounds

Doppler images at 14 d post wounding of the back of swine with ischemic flaps treated or not with PRFM. Bar graph presents the quantitative data from the Doppler images. Data are mean ± SD (n=3); *, p