Stimulation of human primary keratinocytes with PRGF followed by whole transcriptome sequencing and analysis of gene ontology clustering revealed the upregulation of several factors involved in extracellular matrix (ECM) physiology, cellular interactions and angiogenesis. One might speculate that the induction of these ECM-related factors may contribute to the observed beneficial effects of autologous thrombocyte concentrate formulations (e.g., Vivostat PRF®
) to treat chronic and hard-to-heal wounds [12
]. Based on abundance and fold-upregulation we selected seven ECM-related factors for detailed expression analyses. These factors, transforming growth factor beta-induced protein (TGFBI), fibronectin 1 (FN1), matrix metalloproteinase-9 (MMP9), transglutaminase 2 (TGM2), fermitin family member 1 (FERMT1), collagen type I alpha 1 chain (COL1A1) and collagen type XXII alpha 1 chain (COL22A1) are all associated with ECM physiology and will be discussed in the following.
Transforming growth factor beta-induced protein (TGFBI, also known as βig-H3 and keratoepithelin) is an extracellular matrix molecule secreted by many cell types including keratinocytes [29
] that regulates keratinocyte function [30
], plays a critical role in extracellular matrix interactions [32
], could increase adhesion, migration and proliferation of epithelial cells [33
], could impair the function of fibroblasts in chronic wounds [34
] and is therefore supposed to play an essential role in skin wound healing [34
]. According to the whole transcriptome analyses gene expression of TGFBI was approx. 30-fold induced in PRGF-treated keratinocytes and showed the highest abundance of the 50 most upregulated genes. Real-time PCR analyses confirmed induction of TGBI gene expression in primary keratinocytes treated with PRGF. Although the exact role of TGFBI in keratinocytes and skin wounding remains to be determined, it is likely that the observed PRGF-mediated TGFBI induction in keratinocytes contribute to the beneficial effects of the clinical use of PRF in wound treatment.
Transglutaminase-2 (TGM2 or TG2, also named tissue glutaminase) is a multifunctional protein [50
] that has cross-linking and hydrolysis activity, is enrolled in many biological processes in the human body [51
] and is implicated in the complex wound healing process [52
]. TGM2 enhanced the proteolytic resistance and strength of the collagen matrix [56
] and promotes angiogenesis and wound healing [51
]. TGM2 is also expressed by keratinocytes and it is important to differentiate TGM2 from keratinocytes transglutaminase TGM1 which is a transglutaminase mainly expressed by keratinocytes and involved in the formation of the epidermal cornified cell envelope [58
]. TGM2 is a protein cross-linking enzyme and is involved in ECM stabilization by mediating the interaction of integrins with fibronectin [59
]. Interestingly, TGM2 suppression greatly reduced MMP9 expression in a lung cancer cell line [60
] suggesting that TGM2 may influence MMP9 expression also in skin. The whole transcriptome sequencing identified TRM2 as one of the 25 most induced genes in PRGF-stimulated human primary keratinocytes. Real-time PCR analyses confirmed this PRGF-mediated induction of TGM2 in keratinocytes.
FERMT1 (fermitin family member 1 or kindlin-1) is a focal adhesion protein mainly expressed in basal keratinocytes with accumulation at cell-matrix adhesion sites that increases the integrin activity causing cell adhesion, spreading and migration [61
]. In the context of wound healing it regulates the assembly of the extracellular matrix (ECM) as well as survival, proliferation and differentiation of involved cells [63
]. A deficiency or defect of FERMT1 is associated with Kindler syndrome, a skin disease characterized by cutaneous blistering, atrophied skin, photosensitivity, progressive poikiloderma in sun-exposed areas [62
] and lethal neonatal intestinal epithelial dysfunction [64
]. Moreover, it has been shown that Kindlin-1 is essential in EGF-induced re-epithelialization in skin wound healing [65
] and regulates cutaneous stem cell proliferation [63
] that is also supposed to be important for wound healing.
3.6. Collagens COL1A1 and COL22A1
Collagens are the most abundant proteins in the ECM. Collagen type I alpha 1 chain (COL1A1) is involved in the formation of type I collagen fibers [66
] and is therefore involved in the profibrotic gene induction program [68
]. Although mainly expressed by fibroblasts COL1A1 is also expressed in keratinocytes as confirmed by our study. Collagen type XXII alpha 1 chain (COL22A1) is also a member of the ECM and has been reported to act as a cell adhesion ligand for skin keratinocytes and fibroblasts [69
]. It is speculated that COL22A1 is involved in the formation of myofibroblasts from fibroblasts and epithelial cells [70
]. These cells have an important role in ECM matrix production and fibrosis [70
]. Moreover, myofibroblasts can accelerate the wound healing process by contracting the edges of the wound [71
]. Thus, COL22A1 may play at least an indirect role in supporting wound healing. As shown here, the expression of both, COL1A1 and COL22A1, are induced in human keratinocytes by PRGF and Vivostat PRF®
Our PCR analyses confirmed induction of all investigated ECM-related factors in human primary keratinocytes treated with PRGF in vitro. A time kinetic analysis revealed highest PRGF-induced expression of the ECM-related factors after 12–48 h. After 4 h PRGF treatment of the keratinocytes, none of the investigated genes was significantly upregulated. This is in line with the PRGF-mediated induction of the antimicrobial peptides hBD-2 and hBD-3 in keratinocytes, which showed no induction after 4 h and 12 h [22
]. Interestingly, IL-6 was strongly induced in keratinocytes by PRGF already after 4 h and the PRGF-mediated hBD-2 induction was–at least in part–dependent on activation of the IL-6 pathway [22
]. To evaluate whether IL-6 signaling is also involved in the observed PRGF-induced expression of the ECM-related factors, we incubated the keratinocytes with the IL-6-receptor blocking antibody tocilizumab. In contrast to hBD-2, none of the PRGF-induced ECM-related genes was significantly influenced by tocilizumab (data no shown) suggesting that IL-6 signaling is not involved.
It is known that PRGF preparations contain several growth factors. We could recently demonstrate that the PRGF-mediated induction of various antimicrobial peptides was depended on the activation of the epidermal growth factor receptor [22
]. Thus, we speculated that the EGFR may be involved in the observed PRGF-mediated induction of analyzed ECM-related factors. Indeed, our data revealed a significant influence of the EGFR signaling on the PRGF-induced ECM-related factors. Blockade of the EGFR by the monoclonal antibody cetuximab revealed a significantly decreased PRGF-mediated induction of FN1, MMP9, TGM2, FERMT1 and COL22A1. The EGFR-dependent MMP9 induction by PRGF is in concordance with a previous study reporting that the induction of MMP-9 by TNF-alpha in keratinocytes was dependent on the activity of the EGFR [72
]. Similarly, EGFR activation was reported to activate TGM2 expression [73
]. Our study also revealed that the induction of FERMT1 by PRGF required the EGFR. Interestingly, a recent study reported that FERMT1 itself had a positive influence on EGFR level in keratinocytes by direct interaction with the EGFR which protected EGFR from lysosomal degradation [62
]. Thus, it seems that FERMT1 and EGFR interact with each other via a direct positive feedback loop. In contrast to FN1, MMP9, TGM2, FERMT1 and COL22A1, our results revealed that the PRGF-mediated induction of TGFBI and COL1A1 was significantly enhanced by blocking the EGFR suggesting that EGFR activation inhibited the PRGF-induced expression of TGFBI and COL1A1. The observed differential effects of EGFR blockade on the PRGF-induced expression of the ECM-related genes are in line with a previous study reporting differential effects of EGFR inhibition on the skin barrier and skin inflammation. On the one hand blockade of EGFR signaling resulted in the down-regulation of antimicrobial peptides and tight junction genes, on the other hand skin-associated chemokines were induced thus explaining cutaneous inflammation after anti-EGFR therapy [74
]. Together, our data indicate that the influence of PRGF on the expression of ECM-related factors strongly depends on the EGFR. This is in line with the known importance of EGFR signaling in the context of the ECM [75
]. It remains to be shown whether EGFR-ligands present in the PRGF directly activate the EGFR or if a PRGF-mediated release of EGFR ligands influence expression of the ECM-related factors in an auto-/paracrine manner. Either way, our data supports the hypothesis that the PRGF-mediated cutaneous EGFR activation is associated with the beneficial effects associated with PRGF/PRF-related therapies in the context of wound healing.
Our study also revealed a positive influence of PRGF and PRF in an ex vivo setting using skin explants and in an in vivo setting with experimentally generated wounds. Interestingly, expression of TGM2, COL1A1, COL1A22 were not induced in the ex vivo experiments but significantly (in the case of COL1A1 with p = 0.0507 almost significantly) induced in the in vivo situation. A possible explanation for these observed differences may be related to the fact that the influence of PRGF on the ECM-related factors was investigated in the ex vivo setting after approximately one day incubation time. In contrast, the expression of the ECM-related factors in the in vivo setting was analyzed only 5 days after the last PRF treatment. Another explanation for the different results may be related to potential differences in the composition of PRGF and Vivostat PRF®. However, both formulations are based on concentrated platelets. Therefore, we expect that the main effector molecules are present in both formulations. In this regard it would be interesting to perform a detailed analysis of the major active factors present in both formulations.
Taken together, our study identifies the ECM organization as a major target of the effects elicited by PRGF and Vivostat PRF® treatment. ECM remodeling and an intact ECM is important for restoration of the skin barrier after wounding. Thus, our data highlight the PRGF-mediated induction of ECM-related factors as underlying effect that may contribute to the beneficial effects of thrombocytes-derived factors in wound healing. Clearly, future studies are needed to further investigate the influence of thrombocytes lysates on the wound healing process and to decipher the underlying mechanisms.