Collagen/glycosaminoglycan-based matrices for controlling skin cell responses

Introduction

Impaired wound healing and tissue regeneration in an increasing number of elderly, multimorbid patients lead to detrimental effects on the quality of life and impose a severe challenge to the health care system. Wound healing of adult, acute wounds normally progress through three tempero-spatially organized, overlapping phases: (i) hemostasis and inflammation, (ii) proliferation and (iii) remodeling and maturation (Schultz et al. 2011). The process is complex and involves a variety of different cell types including platelets, macrophages, fibroblasts, epithelial and endothelial cells. During hemostasis, platelets aggregate and a provisional fibrin matrix is established. Platelets release cytokines and growth factors into the wound, leading to the recruitment and activation of inflammatory cells, i.e. neutrophils and macrophages, which release further cytokines and growth factors. The latter recruit and activate fibroblasts and keratinocytes initiating the proliferative phase. Here, fibroblasts proliferate and produce ECM molecules like collagen and proteoglycans, forming granulation tissue, which replaces the provisional fibrin matrix. Keratinocytes proliferate and migrate over this granulation tissue closing the wound. At the same time, endothelial cells are activated by vascular endothelial growth factor (VEGF) forming new capillaries. Finally, granulation tissue matures by replacing the initially synthesized collagen type III by well-ordered collagen type I leading to a gradually increasing tensile strength of the tissue. Chronic wounds, on the other hand, are characterized by prolonged inflammatory, proliferative, or remodeling phases, leading to tissue fibrosis and to non-healing ulcers (Schultz et al. 2011).

These processes are strongly influenced by local and invading cells and their microenvironment, e.g. the extracellular matrix (ECM). Here, collagen and glycosaminoglycans, the latter as part of proteoglycans, are of special significance, either by acting as ligands for other ECM macromolecules and cellular integrins or by retaining water and interacting with biological mediator proteins (Salbach et al. 2012; Walimbe and Pantich 2020). Both thereby are responsible for regulating tissue homeostasis, cell migration, and cellular signaling in every phase of wound healing (Mathew-Steiner et al. 2021; Sodhi and Panitch 2020). Moreover, since collagen and GAG are biocompatible, low immunogenic, biodegradable and resorbable, they are considered ideal candidates for biomimetic biomaterials, tissue-engineered scaffolds, and wound dressings (Mathew-Steiner et al. 2021; Sodhi and Panitch 2020). The great potential of these ECM components and combinations thereof in wound healing and tissue regeneration has been reviewed previously, e.g. by Salbach et al. (2012), Köwitsch et al. (2018) and more recently by Walimbe and Pantich (2020), Sodhi and Pantich (2020) and Mathew-Steiner et al. (2021). The research focus in recapitulating the ECM for promoting these processes is, next to bone, cartilage and neuronal regeneration, the regeneration of skin. For the latter, electrospun nanofibrous scaffolds composed of hyaluronan (HA) or chondroitin sulfate (CS) in combination with gelatin or thermoplastic polymers as well as porous collagen matrices with covalently cross-linked HA, CS or heparan sulfate have been used (Sodhi and Panitch 2020; Walimbe and Panitch 2020). These scaffolds were shown to promote the attachment and proliferation of skin fibroblasts and to improve epidermal and dermal regeneration in animal wound healing models.

Further, a few selected collagen- and GAG-based biomaterials have been commercialized as cellular dermal substitutes for enabling full-thickness wound healing (Philandrianos et al. 2012). In most cases, these are combinations of collagen type I with other collagen types, CS and/or chitosan and elastin, but also a dermal regeneration template made of esterified HA is available. Collagen has been implicated in promoting wound healing through the following mechanisms: (i) as a decoy e.g. for matrix metalloproteinases, (ii) as a substrate for the migration of cells relevant in the wound healing process, and (iii) as promoter of a proangiogenic, anti-inflammatory environment (Mathew-Steiner et al. 2021). HA is next to its regulatory role of cellular functions and inflammation, considered beneficial for the hydration of tissue during healing leading to an increased availability of nutrients and removal of waste products (Sodhi and Panitch 2020). However, heparin, oversulfated CS or chemically sulfated hyaluronan (sHA) have been rather marginally investigated so far for improving skin regeneration. Since the degree of sulfation and the sulfate group distribution along the polymer chain are crucial for interacting with e.g. growth factors and given their reported strong anti-inflammatory properties (Salbach et al. 2012; Scharnweber et al. 2015), these sulfated GAG are promising candidates for developing cell instructive biomaterials to control tissue regeneration in particular for health-compromised patients with non-healing chronic wounds. GAG-based biomaterials in form of coacervates, hydrogels, scaffolds and coatings are considered as promising for growth factor/cytokine sequestration and delivery in therapeutic applications as reviewed by Hachim et al. (2019). Here, most investigations were done with heparin but the better suitability of other GAG types in regulated and sustained delivery for certain applications is suggested. However, native sulfated GAG are rather heterogeneous in terms of their disaccharide composition, chain length and sulfation degree/pattern leading to a restricted utility for biomedical applications (Salbach et al. 2012; Scharnweber et al. 2015). Thus, chemically functionalized HA and CS derivatives have been implemented as mimetic GAG structures, offering a more defined and regular molecule structure, good biotechnological or, in case of CS, animal source-based availability, and low immunogenicity (Scharnweber et al. 2015). To investigate the impact of these GAG derivatives and to mimic the natural ECM, collagen(coll)/GAG-based coatings were derived as artificial extracellular matrices (aECM) by in vitro fibrillogenesis of coll type I in the presence of GAG at 37 °C. For matrices containing high-sulfated sHA (sHA3), strong immunomodulatory effects were demonstrated impairing inflammatory macrophage functions (Franz et al. 2013; Kajahn et al. 2012). This concerns the reduced differentiation of monocytes into pro-inflammatory M1 macrophages as well as the decreased pathogen uptake and inflammatory cytokine release of already differentiated M1 macrophages. In both cases, an increased secretion of the immunoregulatory interleukine-10 was observed indicating a polarization towards pro-regenerative M2 macrophages. The latter are considered to be key for inflammatory resolution and improved wound healing. This anti-inflammatory effect of solute sHA3 was further characterized by Jouy et al. (2017) using a global quantitative proteomics approach, which was combined with targeted analysis of key proteins. This revealed the involvement of antioxidant induction in attenuating inflammatory signaling pathways (Jouy et al. 2017). Thus, a beneficial effect for accelerated healing responses by sulfated GAG (sGAG)-containing biomaterials has been proposed, in particular for chronic wound situations (Franz et al. 2013). Along these lines, Lohmann et al. (2017) showed that hydrogels composed of end-functionalized star-shaped polyethylene glycol and derivatives of heparin effectively scavenged certain inflammatory cytokines in wound fluids from chronic human venous leg ulcers. Further, when used as wound dressings, these gels reduced inflammation and increased granulation tissue formation, vascularization and wound closure in an in vivo mouse model of delayed wound healing outperforming a standard-of-care product (Lohmann et al. 2017). The immunoregulatory capacity of sHA3 was addressed further in a recent in vivo study with the same mouse model by Hauck et al. (2021) using acrylated HA (HA-AC)/coll-based hydrogels as wound dressings, designed for delivery of sHA3 into wounds. These hydrogels e.g. reduced inflammation, increased pro-regenerative macrophage activation, enhanced new tissue formation and wound closure, thereby improving defective tissue repair.

Regarding the regulation of fibroblast functions, coll/GAG-based coatings demonstrated to be growth promoting substrates for human dermal fibroblasts (dFb) by increasing initial adhesion and proliferation in a sulfation-dependent manner (van der Smissen et al. 2011). Here, the “proliferative phenotype” was in particular promoted by sHA3 and high-sulfated chondroitin sulfate (sCS3). In contrast, the aECM did not promote α-smooth muscle actin (aSMA) expression and differentiation to myofibroblasts. This study suggested that an accelerated initiation of the wound healing process can be expected by these matrices due to increased numbers of dFb. Further, quantitative proteomics analysis revealed an altered expression of ECM-related proteins by human dFb in response to aECM containing sHA3 (Müller et al. 2012). In line with findings of van der Smissen et al. (2011), it was demonstrated that the presence of sHA3 altered ECM remodelling with downregulation of the coll degrading enzymes cathepsin K, and matrix metalloproteinases-2 and -14. In addition, the expression of several ECM proteins like decorin and coll type I was reduced.

Further, it was demonstrated that sHA3 alone and as part of a coll-based matrix has a strong effect on transforming growth factor β1 (TGF-β1)-driven dFb differentiation to myofibroblasts (van der Smissen et al. 2013).

TGF-β is an important multifunctional growth factor with pleiotropic effects on different cell types in all phases of wound healing as reviewed by Pakyari et al. (2013) and Lichtman et al. (2016). It’s immediate release after injury is key in recruiting macrophages and fibroblasts to the wound (Lichtman et al. 2016; Pakyari et al. 2013). Further, TGF-β enhances the migration of keratinocytes facilitating reepithelization (Lichtman et al. 2016; Pakyari et al. 2013) and is involved in wound contraction enabling wound closure (Pakyari et al. 2013). Being expressed at high levels in the wound microenvironment after initiation of epithelialization (Lichtman et al. 2016), TGF-β promotes fibroblast to myofibroblast differentiation and induces fibroblasts to synthesize extracellular matrix components like fibronectin as well as collagen types I and III (Pakyari et al. 2013). However, excessive TGF-β1 levels due to pathological overexpression during wound healing lead to an increased accumulation of ECM proteins clinically manifesting in fibrotic skin disorders, like hypertrophic scarring and keloids (Lichtman et al. 2016; Pakyari et al. 2013). Thus, reducing TGF-β1 activities in these circumstances may improve wound healing and scarring outcome (Lichtman et al. 2016).

The presence of sHA3 significantly reduced the bioactivity of TGF-β1 by impaired downstream signaling in dFb (van der Smissen et al. 2013). This manifested in a strongly reduced Smad2/3 translocation to the nucleus which finally led to a significantly reduced induction of TGF-β1-stimulated aSMA, coll type I and ED-A fibronectin expression. The study suggests that sHA3-effects were due to occupying the TGF-β-receptor I (TβR-I) binding site as shown by in silico docking experiments. In a following study by Koehler et al. (2017) the underlying mechanism of sHA3 action on TGF-β-signaling was further addressed experimentally and computationally at the receptor level, demonstrating that sHA3 interferes with TGF-β-binding to both receptors, i.e. TβR-I and -II (Koehler et al. 2017). However, the order of binding events played a major role indicating the formation of an inactive signaling complex upon TβR-I binding following sHA3 interacting with a preformed TβR-II/TGF-β1 complex. In consequence, TβR-I and, furthermore, Smad2 phosphorylation were diminished in a human foreskin fibroblast cell line (Hs 27) in the presence of sHA3.

In summary, aECM containing sGAG might act as growth promoting substrates enhancing fibroblast adhesion and proliferation in a sulfate-dependent manner, while at the same time reducing ECM synthesis. Thus, these aECM might increase the number of potential wound healing fibroblasts for initiation of wound closure. Further, it was proposed that sHA either as a pharmaceutical agent or as component of biomaterials might be a promising option, e.g. in locally interfering with TGF-β1-driven wound scarring or skin fibrosis. Hence, for wound healing applications, aECM with a moderate amount of sulfate groups, e.g. coll/sHA1 was suggested as the matrix of choice in initializing granulation tissue formation and de novo synthesis of matrix, since, when applying these matrices, dFb proliferation was found to be increased but ECM synthesis was not reduced (van der Smissen et al. 2011). However, at this point further investigations have been warranted translating the impact of these aECM on the speed and quality of wound closure, e.g. in more complex in vitro culture models and in animal studies.

In view of the above finding, this review presents recent investigations addressing the following open questions:

1. Are sHA-containing matrices also a beneficial growth environment for other skin-related cells, e.g. keratinocytes and melanocytes?

2. Do sHA-containing matrices promote epithelial differentiation of human mesenchymal stem cells (hMSC)?

3. How do aECM perform in a more complex environment, e.g. 3D spheroid cell culture and ex vivo organ culture models?

4. Are photo-crosslinked HA-AC/coll-based hydrogels containing acrylated sHA applicable as delivery systems preserving the integrity of relevant mediators in the wound milieu and release active compounds over longer time periods?

Epidermal cell responses on coll/GAG-based matrices

2D matrices

In contrast to fetal wounds with HA as the major GAG, CS is reported to be prevalent in the newly formed ECM of adult wounds (Melrose 2016). Animal derived CS mainly composed of CS-A and -C is described for the treatment of burns and chronic wounds (Cuadra et al. 2012; Salbach et al. 2012) and, in form of hydrogel preparations, demonstrated to be beneficial in wound healing of murine and porcine epidermis (Kirker et al. 2002, 2004). However, the biological effects of sCS3 on keratinocyte proliferation, migration as well as the expression of selected proteins remained unidentified.

Solute CS variants sulfation-dependently decelerated the migration of an immortalized human keratinocyte cell line (HaCaT) in a scratch assay with sCS3 leading to the strongest reduction (Corsuto et al. 2018). Further, when included in coll-based aECM, sCS3 lead to stagnating cell numbers during the cultivation period, while low sulfated CS slightly induced cellular proliferation as indicated by increased DNA content after 72 h culture time. However, HaCaT cultured on sCS3-containing coatings produced increased amounts of solute, active TGF-β1 (Figure 1), which could be translated into biomaterials able to decrease epidermal hyperproliferation in chronic wounds.

Figure 1:

Amount of active TGF-β1 released into the medium by HaCaT cells grown on aECM containing low- and high-sulfated CS of animal (CS, sCS3) and semi-synthetic (CSb, sCS3b) sources as quantified by Sandwich ELISA.

Results are shown in relation to the respective cell numbers determined by lactate dehydrogenase assay. Values are the mean ± SD of three different experiments with significant differences between respective treatments according to Student’s t-test (*p < 0.05, **p < 0.01, ***p < 0.001). From Corsuto et al. (2018); reprinted with permission from Carbohydr Polym.

Currently performed autologous transplantation treatments of non-healing wounds via autologous split-skin grafts and cell-suspensions involve invasive skin harvesting, which might result in scarring, while at the same time yielding suboptimal cell numbers and quality as reviewed by Kanapathy et al. (2017). Thus, non-invasive procedures for deriving human keratinocytes and melanocytes from the outer root sheath (ORS) of hair follicles are a promising alternative for generating a pigmented, autologous epidermal graft. To achieve this goal, beneficial cultivation conditions for both cell types are essential. Schneider et al. (2019) demonstrated that aECM coatings on tissue culture polystyrene (TCPS) can be used to improve the cultivation conditions for human keratinocytes (HUKORS) and melanocytes (HUMORS) from ORS (Schneider et al. 2019). While coll/sHA4 (high-sulfated HA with four sulfate groups/disaccharide unit) enhanced the proliferative phenotype of HUKORS, it slightly but significantly decreased cellular growth of HUMORS compared to TCPS. On the other hand, sHA-containing matrices (coll/sHA1 and coll/sHA4) promoted the expression of melanotic marker genes and the melanin content in HUMORS in a sulfation-dependent manner (Figure 2). In all cases, sHA4-containing matrices displayed the strongest effect with the melanin content on coll/sHA4 significantly increased compared to coll and coll/HA.

Figure 2:

Gene expression of the melanocyte marker paired box 3 gene (A) and melanin content/volume of HUMORS (B) on PS and aECM matrices upon 5 days of cultivation.

For gene expression analysis values correspond to n = 10; depicted as mean ± SD; *p ≤ 0.05 and **p ≤ 0.01. For quantification of melanin content values correspond to n = 9; depicted as mean ± SD; *p ≤ 0.05 and **p ≤ 0.01. Statistically significant differences were assessed by one-way ANOVA between experimental groups. HPRT1: housekeeping gene hypoxanthine phosphor-ribosyltransferase 1. From Schneider et al. (2019); reprinted with permission from J Biomed Mater Res A.

Hence, coll/sHA4 provided a promising aECM by promoting a proliferative phenotype in HUKORS and the onset of melanogenesis in HUMORS, which is beneficial for improving the generation of epidermal grafts. In contrast, fully differentiated human epidermal melanocytes were less sensitive to the matrix composition. Promoting a proliferative phenotype in HUKORS was also contrary to the above-mentioned findings by Corsuto et al. (2018) with immortalized HaCaT cells. This discrepancy was interpreted as an extended ability of HUKORS in reacting to environmental stimuli and possible deviating effects of the sCS3 versus the sHA4 derivative. Currently, binding of sHA to the CD44 receptor followed by CD44 activation and sHA-mediated concentration of biological mediators from the cultivation medium were suggested to be relevant for the effect on HUKORS and HUMORS (Corsuto et al. 2018). The involvement of CD44 in mediating sHA effects was likewise reported by Jouy et al. (2017) demonstrating the rapid internalization of sHA by CD44 and scavenger receptors in fully differentiated inflammatory macrophages, contributing to the anti-inflammatory activity of sHA. However, the exact mechanism of sHA action on HUKORS and HUMORS was not investigated further and needs to be addressed in future studies.

3D matrices

In order to develop novel biomimetic materials including bioactive wound dressings, the ECM components coll and GAG-derivatives were included in biomaterials via different preparation routes, thereby creating cell instructive microenvironments. Against this background, Bhowmick et al. (2017a) fabricated nanofibrous scaffolds by electrospinning gelatin-GAG blend solutions containing CS, HA and/or sHA3. While, in contrast to findings of van der Smissen et al. (2011), no significant difference in adhesion could be detected between scaffold types, cell proliferation was significantly enhanced for HaCaT cells on scaffolds containing sGAG depending on sulfation degree and concentration (Bhowmick et al. 2017a). The highest cell numbers were found on scaffolds containing 1% sHA3 and 1.25% CS (Figure 3). Whereas these findings are again in contrast to those of Corsuto et al. (2018) demonstrating a decelerated HaCaT proliferation on matrices containing sCS3, they are in line with findings of Schneider et al. (2019) showing a sulfation-dependent increase in proliferation of HUKORS on coll/GAG coatings. One reason for these contradictory results could be again possible deviating effects of high-sulfated sCS versus sHA derivatives. Another one could be differences in GAG composition and concentration between coatings and scaffold. Stimulating the proliferation of keratinocytes is particularly beneficial for early stages of wound healing accelerating the epidermal regeneration process (Metcalfe and Ferguson 2007).

Figure 3:

Scanning electron microcopy image of nanofibrous electrospun gelatin-GAG blended scaffold containing 1% sHA and 1.25% CS (A) and HaCaT keratinocyte proliferation on scaffolds as determined by DNA quantification (B).

10% gelatin scaffolds served as controls. Statistically significant differences were determined by two-way ANOVA between experimental groups and Bonferroni Post-hoc analysis with p values *p < 0.05, **p < 0.01, ***p < 0.001. From Bhowmick et al. (2017a); reprinted with permission from Mater Sci Eng C.

In another study, HA-AC/coll-based hydrogels supplemented with acrylated sHA (sHA1-AC) were established by photo-crosslinking and investigated for binding and release of HB-EGF (Thönes et al. 2019). This growth factor is a ligand for epidermal growth factor receptor (EGFR) signaling, required for the successful closure of skin wounds due to activating keratinocyte migration and proliferation (Johnson and Wang 2013; Shirakata et al. 2005) as well as dermal fibroblasts to produce further mediator proteins (Puccinelli et al. 2010).

It was found that pure HA-AC gels and those supplemented with sHA1-AC release promigratory concentrations of HB-EGF for at least 72 h after immobilization of the growth factor. Importantly, the biological activity of HB-EGF, e.g. the stimulation of HaCaT migration, released from sHA1-AC-containing gels was significantly increased, since it was as effective as 2.5–8-fold higher growth factor concentrations released from pure HA-AC gels. Obviously, HB-EGF was protected from degradation to a certain extent due to sequestration by sHA1-AC.

The effect of sHA1-AC-containing hydrogels on epithelial keratinocyte migration and proliferation was further assessed in a porcine skin organ culture model indicating enhanced wound closure by HB-EGF-loaded gels, as measured by epidermal tip formation, compared to untreated controls, solute HB-EGF or unloaded gels (Figure 4). Thus, HB-EGF released form these hydrogels (1–10 ng/mL) was significantly more efficient than a higher concentration of solute HB-EGF (400 ng/wound) added directly to the wounds.

Figure 4:

HB-EGF released from sHA1-AC-containing hydrogels promoted wound closure by epidermal tip formation in porcine ear skin.

Gels were partially loaded with 200 ng HB-EGF/gel and subjected for 48 h to experimental wounds. Instead, 400 ng solute HB-EGF was added directly. Epidermal tip length was investigated by Masson-Goldner Trichrome staining of paraffin sections. Bars: 500 µm (whole wounds); 100 µm (insets). From Thönes et al. (2019); reprinted with permission from Acta Biomater.

This suggests that reversible immobilization of HB-EGF by sHA1-AC-containing gels could overcome the ineffectiveness of directly applied growth factor solutions to non-healing and burn wounds improving its local bioavailability in skin, e.g. by preventing rapid degradation.

Hence, these sHA1-AC-containing hydrogels demonstrated a suitability for the delivery of biologically active HB-EGF promoting keratinocyte effects, which is in line with previous findings using heparin as the delivery component and reporting enhancing keratinocyte and corneal epithelial cell functions and accelerated wound re-epithelization in normal and diabetic mice (Johnson and Wang 2013, 2015; Princz and Sheardown 2012).

Further, effective sequestration and release of biologically active growth factor by sGAG-containing HA-AC/coll-based hydrogels was also shown for VEGF (Rother et al. 2020). However, whether these findings translate into effective wound dressings promoting an improved healing response in injured skin needs to be further assessed in vivo.

Epithelial differentiation of hMSC on coll/GAG-based matrices

In a further study by Bhowmick et al. (2017b), the impact of the previously described GAG-containing electrospun nanofibrous scaffolds on epithelial differentiation of hMSC was investigated while contact co-cultivating with HaCaT keratinocytes. Whereas co-culturing on 1% sHA3 and 1.25% CS-containing scaffolds led to cellular fusion of the majority of hMSC with HaCaT, the non-fused hMSC displayed an increase of epithelial markers, e.g. keratin-14, as determined by qRT-PRC and immunohistochemistry (Figure 5), indicating epithelial transdifferentiation of hMSC (Bhowmick et al. 2017b). This is in line with a previous study demonstrating that keratinocyte proximity and contact plays a major role in determining mesenchymal stem cell fate (Sivamani et al. 2011). In contrast, scaffolds containing 1% non-sulfated HA and 1.25% CS did not increase the expression of epithelial markers to the same extent. These results suggest that co-cultivating hMSC and keratinocytes on sGAG-containing nanofibrous scaffolds might synergistically promote the utilization of hMSC for improved wound healing by cellular fusion, indicated in promoting hMSC transdifferentiation (Dörnen et al. 2020), and to a certain extent by enhancing direct epithelial transdifferentiation of a hMSC subpopulation.

Figure 5:

Direct cocultivation of hMSC with HaCaT keratinocytes on nanofibrous electrospun gelatin-GAG blended scaffold led to cellular fusion (Q2 from FACS analysis) and triggered epithelial transdifferentiation of a hMSC subpopulation (Q3 from FACS analysis).

(A) FACS analysis of hMSC (green) and HaCaT (blue) labeled with the cell trackers CTB and CTG, respectively, on scaffolds containing 1% sHA and 1.25% CS. (B) Epithelial marker expression analysis of keratin 14 (K14) and ΔNp63α for non-fused hMSCs (Q3, FACS analysis) as determined by quantitative RT-PCR on scaffolds with different GAG content. hMSC only cultured on gelatin scaffolds served as control. Significant difference to respective treatment: **p < 0.01 as determined by two-way ANOVA with Bonferroni Post-hoc analysis. From Bhowmick et al. (2017b); reprinted with permission from J Mater Sci Mater Med.

Dermal cell responses on coll/GAG-based matrices

3D matrices

In their attempt to develop novel biomimetic wound dressings, by using the ECM components coll and GAG derivatives, Bhowmick et al. (2017a) not only studied the response of the keratinocyte but also that of Hs27 fibroblasts and hMSC on their electrospun nanofibrous scaffolds blending gelatin-GAG solutions containing CS, HA and/or sHA3. While in contrast to findings of van der Smissen et al. (2011), no significant difference in adhesion could be detected between scaffold types, cell proliferation was significantly enhanced for Hs27 fibroblasts on scaffolds containing sGAG depending on sulfation degree and concentration, as was the case for HaCaT and hMSC (Bhowmick et al. 2017a). Again, the highest cell numbers were detected on scaffolds containing 1% sHA3 and 1.25% CS. Findings on proliferation are in line with those of van der Smissen et al. (2011) showing a sulfation-dependent increase in proliferation of human dFb on coll/GAG coatings. One reason for deviating results regarding adhesion could be related to differences of the material surface architectures, in GAG composition and concentration between coatings and scaffolds as well as differences between primary dFb and the Hs27 cell line. Stimulating (migration and) proliferation of fibroblasts and keratinocytes is particularly beneficial for early stages of wound healing, promoting the healing process (Metcalfe and Ferguson 2007). Hence, these findings on both, keratinocytes and fibroblasts, suggest that sGAG-containing electrospun nanofibrous scaffolds might be beneficial as part of skin tissue engineered constructs for deep, non-healing wounds by stimulating cellular proliferation and thereby accelerating epidermal–dermal regeneration processes.

Besides investigating the impact of HB-EGF released from HA-AC/coll-based hydrogels with and without sHA1-AC on keratinocyte response, Thönes et al. (2019) also studied the effect on dFb. HB-EGF induced hepatocyte growth factor (HGF) gene expression in human dFb when added directly and after release from both hydrogel types (Figure 6). It was again of note that the biological activity of HB-EGF released from sHA1-AC-containing gels was significantly higher, since it was as effective as 2.5–8-fold higher growth factor concentrations released from pure HA-AC gels. HB-EGF exerts its effects by activating EGFR signaling (Heo et al. 2018). Hence, the influence of HB-EGF released from hydrogels on EGFR signaling was further studied in porcine dFb. The induction of EGFR signaling was detected for HB-EGF released from HA-AC- or sHA1-AC-containing hydrogels over 72 h via increased protein kinase B/Akt-phosphorylation as determined by Western Blot analysis. This finding is a further proof for the bioactivity of the released growth factor for at least 72 h. Since HGF, induced via EGFR signaling, is known to stimulate keratinocyte migration and differentiation during wound healing via it’s paracrine activity (Seeger and Paller 2015), an enhanced expression in dFb might also contribute to epithelial regeneration and thus enhanced wound healing.

Figure 6:

HB-EGF release from HA-AC/sHA1-AC hydrogels and growth factor induced HGF expression in human dFb.

(A) Quantification of growth factor release per 24 h by ELISA. Significant difference to respective treatment: n = 3, ***p < 0.005 as determined by two-way ANOVA with Bonferroni post-hoc test. (B) Fold induction of HGF mRNA expression after direct application (stock) as well as after release from HA-AC- and HA-AC/sHA1-AC-containing hydrogels (24, 48 and 72 h) and incubation with dFb for 48 h. HGF expression in untreated dFb was set to 1 (dashed line). n = 4; *p < 0.05, **p < 0.01, ****p < 0.0001 as determined with unpaired t-test versus control. From Thönes et al. (2019); reprinted with permission from Acta Biomater.

Conclusion

The presented studies concerning coll/GAG-based matrices for controlling skin cell responses demonstrated that 2D- and 3D-matrices containing sGAG are growth regulating substrates not only for fibroblasts, but also for keratinocytes. In addition, they promote the melanotic phenotype of melanocytes from the ORS of hair follicles suggesting an accelerated initiation of the wound healing process and opening up the opportunity for generating a pigmented, autologous epidermal graft. Further, aECM, together with co-culturing keratinocytes, might promote the utilization of hMSC for wound healing approaches by cellular fusion and by enhancing epithelial transdifferentiation of a hMSC subpopulation. Finally, it was demonstrated that HA-AC/coll-based hydrogels are applicable as delivery systems for HB-EGF preserving its integrity in the wound milieu and thus release active growth factor over longer time periods. Together with findings demonstrating the immunomodulatory potential of sGAG, these studies substantiate the hypothesis that biomimetic materials containing sHA bare a great potential for the development of smart materials to improve disturbed wound healing by the material itself or releasable bioactive proteins. Potential constraints are the limited in vivo proof of an enhanced wound healing and the presently pending binding/release and biological activity studies of further biological mediator proteins relevant for the wound healing process. Further, the here presented in vitro and ex vivo studies are not performed with cells/tissues from health compromised patients or animals, which needs to be addressed in further studies to fully reveal the benefit of the presented aECM.