To date, the effects of HA and HYAL on structural cells of the skin have been poorly characterized. Here, we examined these effects by comprehensive genome-wide gene chip analyzes followed by qPCR validation and quantitative protein analyzes.
There are a variety of chemical signals known to stimulate HA synthesis in human fibroblasts such as cytokines, decreased pH, growth factors as well as enzymatic degradation of HA (Heldin et al., 1989, Larnier et al., 1989, Laurent and Fraser, 1986). Underlying mechanisms remain unclear. In line with other findings, enzymatic degradation of HA but also HA itself was found to stimulate HA in an in vitro cell culture system. In 3H-glucosamine labeling experiments Moczar and Robert found that treatment of human skin fibroblasts with bovine testicular hyaluronidase increased the amount of newly synthesized HA in the medium (Moczar and Robert, 1993). In line with these results, our results show that HYAL increased HA amounts in conditioned supernatants of NHDF as measured by ELISA.
Interestingly, increased HA amounts were found particular in supernatants of those cells which showed high gene expression of HAS2 but no other isoforms. In various studies the HAS isoform HAS2 has been suggested to be most important for HA synthesis. HAS2 is the only HAS gene which deletion causes a lethal phenotype: HAS2 knockout mice die at embryonic day (E) 9.5 due to a failure to form HA-rich organs (Camenisch et al., 2002). This confirms the predominant role of HAS2 in the regulation of HA and reveals its important role for HA-metabolism. However, the increase of HA amount in the supernatants could either result from (i) increase in HA synthesis (ii) clearing of membrane-bound HA but also (iii) increase of HA degradation mediated by HYAL. Since HYAL activity was not investigated in our experiments, further studies are required to address this specific question.
In titration experiments we showed that HAS2 gene expression increased with decreasing concentrations of HYAL. Interestingly, HYAL at its lowest concentration (0.015 U/ml) led to the strongest induction of HAS2. Correspondingly, the amount of newly synthesized HA was the highest in cells treated with in low doses of HYAL. Furthermore, immunohistochemical analyses of human skin samples incubated with HYAL ex vivo demonstrated that low concentrations of HYAL (0.015 U/ml) led to a pronounced accumulation of HA, whereas high concentrations of HYAL (15 U/ml) reduced dermal HA levels. In similar observations Philipson et al. (Philipson et al., 1985) found that HYAL treatment at very low concentrations stimulated HA synthesis not only in cultured cells but also in isolated membrane preparations (Philipson and Schwartz, 1984) suggesting an existing feedback mechanism that enables cells to sense levels of HA that has been synthesized (Stern, 2003). The exogenously added HYAL cleaves newly synthesized HA chains as they are being extruded through pore-like structures out of the cell into the extracellular space (Prehm, 1990) leaving a message for fibroblasts that insufficient quantities of HA have been synthesized which might result in induced HA synthesis (Stern, 2004). As early as 1986 Mian postulated the existence of a multi-protein-membrane associated complex that is able to synthesize HA but also has catabolic activity (Mian, 1986a, Mian, 1986b). Two decades later Stern suggested a name for this mini-organelle – the hyaluronasome (Stern, 2003). Comparable to glycogen granules formed in muscle and liver, the hyaluronasome might respond dynamically to extracellular and intracellular events being able to regulate levels of HA deposition (Stern, 2003). An organelle in which all components are tethered together (containing HA receptors such as RHAMM and CD44 and HAS but also HYAL and HA-binding proteins) would provide the structural organization for such reactions to occur with maximum efficiency (Stern, 2003, Wahby et al., 2012). The existence of a multiplayer like the hyaluronasome could be a reason why HYAL in its lowest concentration is rather able to modulate and stimulate HA-metabolism in a positive feedback loop compared to high dose HYAL which would rather lead to a total breakdown of all available HA as demonstrated in our ELISA experiments (see Fig. 4).
There is a dynamic feedback signaling between HYAL and HAS regulating the net deposition of HA and HA fragments (Heldin et al., 2013, Takahashi et al., 2005, West et al., 1985). Out of a variety of cells, dermal fibroblasts are known to synthesize the largest amounts of HA as compared to other cells of the human organism (Li et al., 2007). In line with this observation, in our study NHDF had a higher basal HA production in contrast to epidermal keratinocytes.
The role of HA and HYAL during wound repair is only poorly described. The healing of cutaneous wounds is a complex biological process that can be divided into different phases that overlap in time and space: hemostasis, inflammation, proliferation, and tissue remodeling (Fronza et al., 2014). During early phases of wound repair, HA increases quickly in the wound bed (Chen and Abatangelo, 1999). Following peak levels, HA becomes degraded by increasing levels of wound-produced HYAL (Dunphy and Udupa, 1955). HA fragments then orchestrate specific size-dependent functions (Stern et al., 2006). Extensive literature describes that application of exogenous HA can improve wound healing (Averbeck et al., 2010, Aya and Stern, 2014, Byl et al., 1993, King et al., 1991). In the wound healing analyzes presented here, application of HA induced a significant increase in wound closure. Interestingly, scratch closure occurred as fast in the presence of HYAL. In line with these results, Fronza et al. found that not only HA but also HYAL can accelerate wound closure. In contrast to our in vitro based assay using human primary cells their group used an in vivo full-thickness excisional model in Wistar rats (Fronza et al., 2014). As a HA degrading enzyme HYAL may contribute to the balance between synthesis and deposition of HA and may therefore play a potential role as a healing promoting agent for cutaneous injuries (Fronza et al., 2014). Decreased wound healing with age is attributed in part to compromised HA metabolism and decreased ability to process HA (Meyer and Stern, 1994, Stern and Maibach, 2008). In the aged rat skin, studies have found abundance of HMW-HA, perhaps reflecting an inability to generate lower-molecular-size fragments (Reed et al., 2013). The lack to generate such small fragments would compromise the wound healing process (Aya and Stern, 2014). Voorhees and Fisher found that the injection of HA-fillers stimulates localized proliferation of fibroblasts in the human skin (Quan et al., 2013, Wang et al., 2007). These fibroblasts showed a stretched appearance, and expressed high levels of type I procollagen thereby restoring dermal matrix components that are lost in photodamaged skin (Fisher et al., 2008). When HYAL is added to the wound scratches it might break cross-links in HA which is being extruded in the medium so it behaves like native HA. Possibly, increased concentration of HA fragments resulting from HYAL activity might be important in the wounding process as they stimulate capacity of fibroblast for functional activation. Particularly low molecular weight HA has been suggested to contribute to wound healing (West et al., 1985). Therefore, we also investigated the effects of medium-sized HA on wound closure. Surprisingly, medium-sized fragments did not shorten the closure time of the scratch compared to medium control. As other fragment sizes were not investigated in our study, this could be addressed in future studies.
Wohlrab et al. investigated the influence of adjuvant HYAL on wound healing in a placebo-controlled, double-blinded clinical trial. Regarding target parameters like transepidermal water loss, hemovascular perfusion, and complete macroscopic epithelization of the wound his group found no evidence that HYAL retards wound healing (Wohlrab et al., 2012).
To conclude, HYAL is a bioactive enzyme that exerts multiple effects on the HA-metabolism as well as on the structural cells of the skin. Our study provides direct evidence that especially low doses of HYAL significantly induce HAS and as well as the synthesis and concentration of HA whereas high-dose-HYAL leads to a downmodulation of HA in dermal fibroblasts. Thus, low-dose-HYAL may be beneficial in the rejuvenation of aged skin as it stimulates dermal fibroblasts to increase HA amount. In addition, our study points toward an important role of HYAL in wound healing as HYAL accelerates wound closure in an in vitro wound scratch model of dermal fibroblasts. Future studies are required to further fully elucidate the underlying molecular pathways of HYAL and HA action in the skin.