In this study we show that constitutive and collagen I-induced NHLF migration through collagen IV is mediated by both DDR2 and DDR1. Additionally, silencing of the DDR2-associated kinases JAK2 and ERK1/2 with specific siRNAs inhibited migration of NHLFs through both collagen I-containing and collagen IV-containing matrices, strengthening the hypothesis that DDR2 signaling is important for human fibroblasts to enable them to recognize and degrade collagen IV. Furthermore, we have also shown that only DDR2 appears to be required for NHLF migration through collagen I.
We have previously shown that collagen I induces DDR1 and MMP-10 expression through the activation of DDR2 in primary human lung fibroblasts suggesting a key role for both DDR1 and DDR2 in fibroblast function in situations associated with excessive matrix deposition, such as fibrosis and wound healing. Knockdown experiments using specific siRNAs also confirmed an important role for both JAK2 and ERK1/2 in both constitutive and collagen I-induced DDR1 expression. Importantly, basal protein expression levels of DDR1 as well as constitutive phosphorylation of JAK2 and ERK1/2 were reduced in NHLFs in the presence of DDR2-specific siRNA, suggesting a link between DDR2 activation, JAK2 and ERK1/2 phosphorylation, and DDR1 expression .
The interaction of fibroblasts with the ECM and their subsequent migration into regions of injury and remodeling are major factors that contribute to wound healing and fibrosis . DDRs have been shown to play an important role in cell adhesion, migration, proliferation and ECM remodeling by controlling the expression and activity of MMPs [1, 27]. For example, Olaso et al. have demonstrated that skin fibroblasts from DDR2 knockout mice present impaired proliferation and migration through a reconstituted basement membrane concomitantly with the expression of MMP-2 . Nevertheless, the functional significance of DDR activation in human lung fibroblasts has not been extensively characterized.
Together with MMP-10, type IV collagenases such as MMP-2 and MMP-9 are also able to degrade collagen IV and have been implicated in cell migration . We have previously shown that collagen I induces the expression of MMP-10 and MMP-2, but not MMP-9 in primary human lung fibroblasts. Furthermore, our finding that DDR2-mediated collagen I induction of MMP-10 expression was also JAK2 and ERK1/2 dependent in NHLFs highlights the importance of these pathways in collagen I-induced expression of ECM degrading metalloproteinases . MMP-10 has been shown to play a major role in tissue remodeling as it is not only able to degrade collagen III and IV, gelatin, proteoglycans and elastin [21, 22], but is also responsible for the activation of other MMPs such as MMP-1, MMP-8, and MMP-9 . In the present study we have shown that, together with MMP-10, collagen I-induced MMP-2 mRNA and protein expression are DDR2 but not DDR1 dependent. MMP-2 is not only able to degrade a wide range of ECM components such as elastin, fibronectin and most collagens but is also involved in the processing of growth factors and cytokines such as tumor necrosis factor and interleukin-1β, into their biologically active forms [30, 31]. Experiments with MMP-10-specific and MMP-2-specific siRNA showed that collagen I-induced NHLF transmigration through collagen IV-coated inserts is MMP-2 but not MMP-10 dependent in NHLFs. MMP-2 has been shown to function as an autocrine regulator of proliferation and migration in human keratinocytes , and human airway smooth muscle cells . Several studies suggest that while matrices rich in fibrillar collagens maintain fibroblasts in an active state, basement membrane proteins slow collagen production and receptor expression [33, 34]. Thus, degradation of the basement membrane by fibroblast-derived MMPs such as MMP-2 is likely to induce continued interstitial collagen production in fibroblasts.
The finding that the expression of DDR1 and MMP-2 in NHLFs is DDR2 dependent led us to hypothesize that constitutive and collagen I-induced activation of DDR2 could increase the ability of NHLFs to recognize and degrade basement membrane collagen IV, thereby facilitating fibroblast migration through connective tissue. While we cannot formally rule out the possibility that in our experimental system collagen I coating impacts cellular adhesion and thereby modifies migration in a DDR-independent manner, our data strongly suggest a DDR2-mediated mechanism.
DDR2 has been reported to be only activated by fibril-forming collagens and collagen × however DDR1 can recognize a wide range of collagen types, including both fibril-forming collagen I and network-forming collagen IV, the main component of basal lamina [26, 35]. DDR2 is much more abundantly expressed in fibroblasts than DDR1 and constitutive expression of DDR1 in NHLFs where DDR2 has been silenced does not seem to be sufficient for NHLFs to migrate through collagen I. Furthermore, our results on NHLFs transfected with DDR1-specific siRNA suggest that primarily DDR2 is responsible for recognizing collagen I in NHLFs, further strengthening the critical role of DDR2 in fibroblast migration. In contrast, constitutive and collagen I-induced DDR1 expression seems to be involved in fibroblast migration through collagen IV confirming previous reports showing a key role of DDR1 in fibroblast migration . Interestingly, the silencing of DDR2, which does not recognize collagen IV , and the DDR2-associated kinases JAK2, and ERK1/2 also abrogated constitutive and collagen I-induced migration of NHLFs through collagen IV suggesting that the role of DDR2 in NHLF migration through collagen IV-coated inserts is likely to be due to constitutive and collagen I-induced DDR2 activation, and DDR2-dependent signal transduction and gene expression. Interestingly, DDR2-specific, but not DDR1-specific, siRNA also inhibited NHLF proliferation suggesting a wider role for DDR2 in fibroblasts function.