Discoidin Domain Receptor-1 Regulates Calcific Extracellular Vesicle Release in Vascular Smooth Muscle Cell Fibrocalcific Response via Transforming Growth Factor-β Signaling
- Other version(s) of this article
-
You are viewing the most recent version of this article. Previous versions:
-
Abstract
Objective—
Collagen accumulation and calcification are major determinants of atherosclerotic plaque stability. Extracellular vesicle (EV)–derived microcalcifications in the collagen-poor fibrous cap may promote plaque rupture. In this study, we hypothesize that the collagen receptor discoidin domain receptor-1 (DDR-1) regulates collagen deposition and release of calcifying EVs by vascular smooth muscle cells (SMCs) through the transforming growth factor-β (TGF-β) pathway.
Approach and Results—
SMCs from the carotid arteries of DDR-1−/− mice and wild-type littermates (n=5–10 per group) were cultured in normal or calcifying media. At days 14 and 21, SMCs were harvested and EVs isolated for analysis. Compared with wild-type, DDR-1−/− SMCs exhibited a 4-fold increase in EV release (P<0.001) with concomitantly elevated alkaline phosphatase activity (P<0.0001) as a hallmark of EV calcifying potential. The DDR-1−/− phenotype was characterized by increased mineralization (Alizarin Red S and Osteosense, P<0.001 and P=0.002, respectively) and amorphous collagen deposition (P<0.001). We further identified a novel link between DDR-1 and the TGF-β pathway previously implicated in both fibrotic and calcific responses. An increase in TGF-β1 release by DDR-1−/− SMCs in calcifying media (P<0.001) stimulated p38 phosphorylation (P=0.02) and suppressed activation of Smad3. Inhibition of either TGF-β receptor-I or phospho-p38 reversed the fibrocalcific DDR-1−/− phenotype, corroborating a causal relationship between DDR-1 and TGF-β in EV-mediated vascular calcification.
Conclusions—
DDR-1 interacts with the TGF-β pathway to restrict calcifying EV-mediated mineralization and fibrosis by SMCs. We therefore establish a novel mechanism of cell-matrix homeostasis in atherosclerotic plaque formation.
Introduction
Vascular calcification is a predictor of cardiovascular events and a major determinant of atherosclerotic plaque stability.1 Although macrocalcifications in the collagen-dense fibrous cap may promote structural stability of the plaque, microcalcifications in collagen-poor areas exert mechanical stress on the surrounding tissue, thus potentiating the risk of rupture and subsequent cardiovascular events.2 Several determinants that further influence plaque stability have been identified, among which a high number and close spatial proximity of microcalcific deposits as well as a thin collagenous fibrous cap constitute the conditions most favorable for stress-induced rupture.3 Emerging evidence suggests that vascular wall cells, including smooth muscle cells (SMCs) and macrophages, release calcifying extracellular vesicles (EVs) prone to aggregation in the extracellular matrix (ECM) and formation of calcific foci in the plaque.4–8 However, the regulation of collagen production and calcifying EV release by SMCs is incompletely understood.
See cover image
Discoidin domain receptors (DDRs)-1 and 2 are a family of 2 receptor tyrosine kinases that exhibit substrate specificity for both fibrillar and nonfibrillar collagens.9 DDR activity has been implicated in physiological processes such as cell migration,10 differentiation,11 and ECM remodeling,12 whereas dysregulated DDR function has been linked to the progression of fibrosis, arthritis, and cancer.13 DDR-1, composed of 5 membrane-bound and 2 secreted isoforms generated by alternative splicing,14 was found to play a complex role in the progression of atherosclerotic plaque formation.15 By contrast, DDR-2 does not affect SMC migration, proliferation, or ECM remodeling in vitro.16 Ahmad et al17 found that DDR-1 influences in vivo vascular calcification using DDR-1/low-density lipoprotein (LDL)-receptor double-knockout mice. Recent findings on the role of DDR-1 in fibrocalcific response and the emerging influence of EVs in vascular calcification highlight a possible impact of DDR-1 on the fibrocalcific potential of SMCs and their release of calcifying EVs. Despite extensive knowledge of the numerous pathways involved in DDR-1 downstream signaling,14 a functional link between DDR-1 signaling and its role in EV-mediated calcification as a predictor of atherosclerotic plaque stability remains to be elucidated.
Transforming growth factor-β (TGF-β) and its signaling pathways are strongly associated with SMC-mediated fibrosis18 and calcification,19 suggesting a potential regulatory role in SMC osteogenic differentiation and fibrocalcific response. A canonical pathway featuring the Smad proteins, Smad2 and Smad3, and several noncanonical pathways comprising the MAPkinases Erk1/2, JNK, and p38/MAPK are involved in TGF-β receptor activation. Both the pathways are known to cross-talk with each other resulting in synergistic or antagonistic effects on potentially desirable biological outcomes.20 In the context of atherosclerotic plaque progression, TGF-β1 promotes the osteogenic differentiation21 and calcifying potential22 of SMCs in vitro, and it is abundantly expressed in calcified human atheromata.23 The effect of TGF-β pathways on EV release and the fibrocalcific response in SMCs, however, remains unclear.
In this study, we hypothesize that DDR-1 signaling regulates EV-mediated SMC fibrocalcific responses via TGF-β pathways. Wu et al24 observed that DDR-1 depletion increases chondrogenic differentiation through enhanced expression of chondrogenic markers, such as SOX-9 and collagen II, in human adipose-derived stem cells. Moreover, the knockdown of DDR-1 in a tumor cell model reportedly stimulated the expression of TGF-β1 on the RNA and protein level.25 On the basis of these findings, we demonstrate a functional cross-talk between DDR-1 and the TGF-β pathways involving Smad3 and p38 to regulate the release of calcifying EVs by SMCs. This study introduces a novel mechanism balancing vascular fibrocalcific response in atherosclerotic plaque formation.
Materials and Methods
Materials and Methods are available in the online-only Data Supplement.
Results
DDR-1 Deficiency Increases the Release of Calcifying EVs in SMCs
Nanoparticle tracking analysis of collagenase digested, purified supernatant from SMCs after 14 and 21 days of culture showed a uniform size distribution with a mean particle size of 180 to 200±100 nm throughout all samples and time points (Figure 1A). These findings are consistent with previous data on EV size distribution.26 Compared with wild-type (WT) littermates, DDR-1−/− SMCs exhibited a significant increase in the release of EVs by 3.5-fold in normal media (P=0.0004) and 4-fold in calcifying media (P=0.003) at day 14 and a 2.5-fold increase in both media at day 21 (P<0.001; Figure 1B). Transmission electron microscopic imaging of SMCs after 21 days of culture displayed dense clusters of DDR-1−/− SMCs surrounded by abundant ECM, whereas WT cells showed sparse cell growth and little ECM production (Figure 1C). Transmission electron microscopic images demonstrated an increased number of membrane-bound vesicular bodies in proximity with DDR-1−/− compared with WT SMCs. The particles displayed via transmission electron microscopy were found to match the EV size distribution observed using nanoparticle tracking analysis (Figure IB in the online-only Data Supplement), confirming our finding of elevated EV release by DDR-1−/− SMCs. Further analysis of the calcifying potential of isolated EVs revealed an increase in alkaline phosphatase (ALP) activity in EVs from DDR-1−/− compared with WT SMCs by 1 order of magnitude in both normal and calcifying media after 14 (P<0.0001) and 21 days (P<0.01, Figure 1D). The presence of β-glycerophosphate (β-GP) in calcifying media, mimicking osteogenic conditions, further induced EV release and ALP activity in EVs from WT (46.1±23.0 versus 107.5±7.4 ng/well per mg protein, P<0.05; Figure V in the online-only Data Supplement) and DDR-1−/− SMCs (581.3±68.3 versus 878.2±155.4 ng/well per mg protein, P<0.05) significantly, thus augmenting their calcifying potential.
Figure 1. Nanoparticle tracking analysis of extracellular vesicles (EVs) released from wild-type (WT) and discoidin domain receptor-1 (DDR-1)−/− vascular smooth muscle cells (SMCs) cultured for 14 and 21 days in normal media (NM) and calcifying media supplemented with 10 mmol/L β-glycerophosphate (β-GP). A, Size distribution of EVs at days 14 and 21 shows a characteristic uniform size between 100 and 300 nm. B, Concentration of EVs released by DDR-1−/− SMCs is significantly increased compared with WT SMCs in normal and calcifying media at 14 and 21 days of culture. C, Transmission electron microscopic imaging of WT (left) and DDR-1−/− (right) SMCs after 21 days of culture. EVs in the extracellular matrix of DDR-1−/− SMCs (white arrowheads). Representative images at ×11 000 magnification; scale bar, 500 nm. D, Alkaline phosphatase (ALP) activity assay in purified EVs reveals increased calcification potential in EVs released from DDR-1−/− SMCs. Results shown as mean±SD, n=5 for each group, **P≤0.01, ***P≤0.001, ****P≤0.0001.
DDR-1−/− SMCs Propagate Calcification Through the Expression of a Procalcific Phenotype
Consistent with elevated EV calcification potential, DDR-1−/− SMCs exhibited increased ALP activity compared with WT SMCs as shown by a colorimetric staining assay after 14 and 21 days of culture (Figure 2A). An ALP activity assay performed on cell lysates concordantly showed a significant increase in the presence of cytoplasmic and cell membrane–bound ALP in DDR-1−/− SMCs in both normal and calcifying media at days 14 (P=0.0003) and 21 (P<0.01, Figure 2B). The addition of β-GP in calcifying media resulted in a linear increase in ALP activity during the 21 days of SMC culture similar to that observed in EVs, suggesting a direct relationship between ALP expression in SMCs and EV calcifying potential. Gene expression at 14 days of culture showed a concordant 6- and 9-fold upregulation of ALP mRNA in normal and calcifying media, respectively (P<0.01, Figure 2E). Further analysis of osteogenic differentiation in DDR-1−/− SMCs revealed a 56.1±38.5-fold increase in the synthetic marker osteopontin, along with a 6.6±4.3-fold increase in the osteogenic marker Msx2 (P<0.01), whereas no significant difference compared with WT was observed in related markers vimentin and Runx2 (Figure VI in the online-only Data Supplement).
Figure 2. Alkaline phosphatase (ALP) expression and calcification in wild-type (WT) and discoidin domain receptor-1 (DDR-1)−/− vascular smooth muscle cells (SMCs). A, Representative images of ALP activity staining. B, ALP activity assay in cell lysates shows increased activity of cytoplasmic and cell membrane–bound ALP. C, Representative images of Alizarin Red S staining in WT (top) and DDR-1−/− SMCs (bottom). D, Quantification of mineralization measured by Alizarin Red S absorbance at 550 nm. Mean±SD, n=5 per group, representative results of 3 independent experiments. E, ALP mRNA levels at 14 days of culture relative to WT normal media and normalized to rplp0, n=10 per group. Statistical differences labeled as follows: *P≤0.05, **P≤0.01, ***P≤0.001, ****P≤0.0001.
Compared with WT, DDR-1−/− SMCs exhibited increased Alizarin Red S–positive mineral deposition. Mineralization of DDR-1−/− cultures in calcifying media visibly progressed from days 14 to 21 with a >30-fold increase in Alizarin Red S absorbance compared with WT at day 21 (P=0.0005, Figure 2C and 2D).
Calcific mineral deposition in the ECM is regulated by a balance of inorganic phosphate and pyrophosphate, which is dependent on ALP and its antagonist ectonucleotide pyrophosphatase/phosphodiesterase-1 (Figure IA in the online-only Data Supplement).27 ALP produces phosphate for calcium phosphate complexation, whereas ectonucleotide pyrophosphatase/phosphodiesterase-1 generates pyrophosphate, an inhibitor of mineralization. In DDR-1−/− SMCs, the increase in ALP activity in EVs and ALP expression in SMCs was accompanied by a concomitant increase in ectonucleotide pyrophosphatase/phosphodiesterase-1 activity in EVs in calcifying media (Figure IB in the online-only Data Supplement) and in cell lysates in both media (Figure IC in the online-only Data Supplement).
DDR-1 Regulates EV-Induced Fibrocalcific Response in SMCs Through TGF-β1 Downstream Signaling Pathways
Members of the TGF-β family modulate vascular fibrosis and calcification.20 In calcifying media, DDR-1−/− SMCs release elevated levels of TGF-β1 compared with WT after 21 days of culture (Figure 3A). TGF-β1 gene expression is doubled relative to WT in DDR-1−/− SMCs in both media at day 14 (Figure 3B). Western blot analysis of TGF-β canonical and noncanonical pathways after 21 days of culture showed a near-complete suppression of Smad3 phosphorylation (P<0.0001, Figure 3C) and a significant decrease in phosphorylated JNK (Figure IVB in the online-only Data Supplement) in DDR-1−/− SMCs, whereas phosphorylation of p38 was increased significantly (P=0.02, Figure 3D). The inverse regulation of canonical and noncanonical TGF-β pathways in DDR-1−/− SMCs observed in vitro was assessed in vivo through quantitative immunofluorescence of LDL-R−/− and DDR-1−/− LDL-R−/− aortic sections. The data show a significant increase in phospho-p38 (P=0.03) and a trending suppression of phospho-Smad3 in the double-knockout mice (P=0.06, Figure VIII in the online-only Data Supplement). No difference between WT and DDR-1−/− SMCs was observed in the levels of phospho-Erk1/2 (not shown), Smad2 (Figure IVC in the online-only Data Supplement), and Smad1/5 (Figure IVD in the online-only Data Supplement). The addition of the selective TGF-β receptor-I inhibitor SB431542 abrogated the changes in TGF-β1 release (P<0.0001, Figure IVA in the online-only Data Supplement) and TGF-β pathways observed in DDR-1−/− SMCs, resulting in a complete suppression of Smad3 phosphorylation in WT and p38 phosphorylation in DDR-1−/− SMCs (P<0.0001 and P=0.02, respectively; Figure 3D). These in vitro and in vivo findings support our hypothesis on a relationship between DDR-1 and TGF-β downstream signaling with manifold implications in vascular calcification.
Figure 3. Discoidin domain receptor-1 (DDR-1) knockout and transforming growth factor-β (TGF-β) signaling. A, Concentration of TGF-β1 detected by ELISA in cell media supernatant after 21 days of culture. B, TGF-β1 mRNA at 14 days of culture relative to wild-type (WT) normal media and normalized to rplp0. Representative results of 3 independent experiments in triplicates. C, Western blot analysis for Smad3 (top) and p38 (bottom) phosphorylation under native conditions (left) and with addition of 2 μmol/L SB431542 (right) after 21 days of culture. D, Quantification of Western blot band intensities normalized to GAPDH using NIH ImageJ v1.48. Mean±SD, n=10 per group, *P≤0.05, ***P≤0.001, ****P≤0.0001.
Inhibition of TGF-β Receptor Type I and p38 Mitigate the Osteogenic Potential of the DDR-1−/− Phenotype
The influence of DDR-1 on p38 and Smad3 phosphorylation as well as on TGF-β1 release by SMCs raises the question whether these pathways contribute to the osteogenic phenotype exhibited by DDR-1−/− SMCs and their release of calcifying EVs. After 21 days of culture in calcifying media, the addition of specific TGF-β receptor-I inhibitor SB431542 abrogated the previously observed 30-fold increase in calcific mineral deposition by DDR-1−/− SMCs (P<0.0001, Figure 4A and 4B). Moreover, SB431542 treatment reduced EV release by DDR-1−/− SMCs (P<0.0001, Figure 4C) and mitigated EV calcification potential demonstrated by a decrease in ALP activity (P<0.01, Figure 4D). Similarly, SB203580, a selective inhibitor of phospho-p38 MAPkinase downstream signaling, significantly suppressed EV release in DDR-1−/− SMCs in normal (P<0.0001) and calcifying media (P<0.0001, Figure 4C) along with a significant decrease in EV ALP activity in normal (P<0.001) and calcifying media (P<0.0001, Figure 4D). Thus, inhibition of TGF-β signaling by either SB431542 or SB203580 has an equal impact on major calcification end points, including EV release and EV calcifying potential (P<0.0001, Figure 4A and 4B). In DDR-1−/− SMCs, SB431542 and SB203580 consistently attenuated the activity of cytoplasmic and membrane-bound ALP (P<0.05) and ectonucleotide pyrophosphatase/phosphodiesterase-1 (P<0.01) after 14 days (Figure IIIA and IIIB in the online-only Data Supplement).
Figure 4. SB431542 and SB203580 diminish the osteogenic potential of discoidin domain receptor-1 (DDR-1)−/− vascular smooth muscle cells (SMCs). A, Alizarin Red S staining after 21 days of culture in native culture media (left), with addition of 2 μmol/L SB431542 (middle) or 10 μmol/L SB203580 (right); n=5 per group. B, Alizarin Red S absorbance of the stained wells of A, normalized to wild-type SMCs cultured in native media. C, Relative extracellular vesicle (EV) concentration, normalized to wild-type. D, Relative alkaline phosphatase (ALP) activity in EVs, normalized to wild-type. Mean±SD, n=5 per group. n.s. indicates not significant. **P≤0.01, ***P≤0.001, ****P≤0.0001.
DDR-1 Exerts a Negative Feedback on Quantitative Collagen Synthesis in Response to Extracellular Collagen
DDR-1 acts as a sensor for type I collagen in SMCs, activating a negative feedback loop on collagen synthesis in human SMCs.12 In DDR-1−/− SMCs, lack of collagen feedback from fibrillar collagen in the ECM not only resulted in increased collagen production (P<0.001, Figure 5B) but also led to defined changes in the 3-dimensional structure of collagen fibers. The fluorescently labeled collagen-binding protein CNA3528 detected a dense network of amorphous collagen fibers devoid of spatial organization in cultures of DDR-1−/− SMCs after 21 days, whereas collagen produced by WT SMCs was found to be less dense (P<0.001), more fibrillar, and restricted to the immediate vicinity of the SMCs (Figure 5A). The addition of β-GP in calcifying media increased calcification in DDR−/− SMCs compared with WT as detected by a fluorescent calcium tracer29 (P=0.002). Addition of SB431542 reduced collagen synthesis (P<0.01) and calcific deposition in calcifying media (P<0.01) in DDR-1−/− SMCs. Thus, TGF-β receptor inhibition re-established WT conditions in DDR-1−/− SMCs (Figure 5A and 5B).
Figure 5. Discoidin domain receptor-1 (DDR-1) acts as a feedback regulator on vascular fibrocalcific response. A, Fluorescent labeling of collagen by CNA probe (green) and microcalcific deposits by osteosense (red) produced by wild-type (WT) and DDR-1−/− vascular smooth muscle cells in native media (top) and with addition of 2 μmol/L SB431542 (bottom) after 21 days of culture. Bar represents 60 μm. Representative images of 4 independent experiments. B, % positive area for collagen (green) and microcalcifications (red). Mean±SD, n=5 per group, **P≤0.01, ***P≤0.001. C, Top: Picrosirius Red staining of low-density lipoprotein receptor (LDL-R)−/− and DDR-1−/− LDL-R−/− aortic arches under polarized microscope at ×4 magnification, scale bar, 200μm. Bottom: Immunohistochemical staining for transforming growth factor-β 1 at ×2 magnification. Images correspond to Picrosirius Red–stained areas of B. Scale bar, 100 μm; n=6 for each group. For each n, 2 sections were stained for evaluation. D, Quantification of relative frequency of thin, loose, and thick, densely packed collagen fibers represented by green and orange birefringence using ImageJ v1.48; n=6 per group. E, Density-dependent color scanning electron microscopy (DDC-SEM) of plaque areas from LDL-R−/− and DDR-1−/− LDL-R−/− aortic arches, with a human atheroma for comparison. Mineral is presented in orange and extracellular matrix is presented in green. Magnification ×10 000, scale bar, 1μm; n=6 per group.
DDR-1 Deficiency Triggers TGF-β–Mediated Calcific EV Release and Fibrotic Response In Vivo
On the basis of these findings demonstrating a fibrocalcific phenotype of DDR-1−/− SMCs through an interaction with TGFβ signaling in vitro, we investigated the interactions between SMCs, DDR-1, TGF-β1, and atherosclerotic plaque formation in vivo. The aortic arches of LDL-receptor knockout (LDL-R−/−) and DDR-1−/− LDL-R−/− mice on a 24-week high-fat diet were analyzed for the presence and distribution of calcific EVs, collagen, and TGF-β1 expression.
Picrosirius Red staining revealed the spatial arrangement and packing density of collagen fibers under polarized light by enhancing natural birefringence. Although green colors indicate thin, poorly packed collagen, increased thickness, and packing density causes a shift in birefringence to the reddish-orange spectrum.30 DDR-1−/− LDL-R−/− aortic arches exhibited a network of thick, densely packed collagen fibers that were nearly undetectable in LDL-R−/− tissue sections (Figure 5C). For quantitative analysis, Picrosirius Red–stained collagen was categorized into thin and loose (green), medium-sized (yellow) and thick, and densely packed fibers (orange-red). Mature, medium-sized fibers were expressed constitutively in the vessel wall of both groups regardless of fibrotic response (P=0.89, not shown), and they were therefore excluded from analysis. Quantification of thin fibers (green) relative to thick bundles (red-orange) revealed a shift toward thick, densely packed collagen (45.53±19.78% versus 54.47±19.78% thin fibers, P=0.01) in DDR-1−/− LDL-R−/−, whereas in LDL-R−/−, thin fibers were found to be more abundant (81.90 versus 18.10±9.60% thick fibers, Figure 5D).
Consistent with in vitro findings, immunohistochemistry for TGF-β1 detected positive areas in sections from DDR-1−/− LDL-R−/− mice in proximity to areas positive for thick, densely packed collagen. In LDL-R−/− aortic sections, however, TGF-β1 staining was scarce and rarely found in the vascular interstitium (Figure 5C), supporting our in vitro hypothesis on a functional connection between DDR-1 and the TGF-β pathway in the regulation of ECM homeostasis. Similarly, staining for ALP in DDR-1−/− LDL-R−/− sections showed signal in proximity to areas of plaque formation, whereas in the LDL-R−/− control, ALP expression was barely detectable (Figure IX in the online-only Data Supplement).
Analysis of calcific EVs and EV aggregates using density-dependent color scanning electron microscopy, a novel method identifying early stages of calcification in the form of calcifying vesicular structures,7 showed a multitude of isolated and aggregating vesicular particles of equal density as adjacent calcific deposits in plaque areas from DDR-1−/− LDL-R−/− mice, bearing strong morphological resemblance to calcifications found in human atheromata (Figure 5E). By contrast, plaques from LDL-R−/− mice presented as mostly fibrotic with only few vesicular structures and calcific deposits of nodular appearance, revealing a pattern of calcific EV release consistent with nanoparticle tracking analysis and transmission electron microscopic analyses.
Discussion
These findings provide a novel perspective on EV-mediated vascular fibrocalcific response, a crucial determinant of atherosclerotic plaque stability. DDR-1−/− SMCs exhibited a predominantly osteogenic and synthetic phenotype, concomitant with an increase in all measured end points of SMC-mediated fibrocalcific response both in vitro and in vivo. These included calcifying EV release4–6 and ALP activity29 as facilitators of mineralization and collagen synthesis as a hallmark of fibrosis, along with enhanced expression of synthetic and osteogenic markers. Microcalcifications in collagen-poor areas of the plaque have proven detrimental to its structural stability, exerting mechanical stress on surrounding tissue, and increasing the risk of rupture.2,3 We recently demonstrated that collagen acts as a scaffold for EV accumulation and formation of macro- or microcalcifications8; however, the cellular mechanisms regulating the release of calcifying EVs in response to collagen are unknown. This study introduces DDR-1 as a novel molecular regulator controlling fibrocalcific remodeling of the ECM in atherosclerotic plaques through a previously unknown interaction with TGF-β pathways, thus associating TGF-β1 with EV release in SMCs. DDR-1 was found to restrict TGF-β1 release and subsequent p38 phosphorylation in SMCs, thereby limiting EV-mediated calcification and collagen matrix production. Furthermore, this study complements previous work on a DDR-1–activated negative feedback loop on collagen synthesis and remodeling.12,15 DDR-1 acts as a sensor for ECM collagen and, in response, limits collagen synthesis and calcific EV release by SMCs, promoting ECM homeostasis. Our findings of enhanced fibrocalcification in vitro, increased fibrotic remodeling, ALP expression and an abundance of calcific EVs in the DDR-1−/− model in vivo support this conclusion.
An explanation for the differences between this study and the study by Ahmad et al17 on SMC-mediated calcification is given by the use of different calcifying conditions. Although Ahmad et al17 induced calcification in vitro using a high-phosphate medium prone to spontaneous, ALP-independent mineral formation,31 this study used media supplemented with β-GP, a specific ALP substrate with low tendency for cell-independent Ca2+ complexation.32 The differences between currently established in vitro calcification models may reflect different pathologies,33 the high-phosphate media mimicking the pathogenesis of medial calcification in chronic kidney disease,34 whereas β-GP may promote osteogenic changes by stimulating ALP activity,35 a common mechanism of atherogenic calcification. Indeed, culturing WT and DDR-1−/− SMCs in the high-phosphate media used by Ahmad et al17 showed a remarkable inversion of the DDR-1−/− phenotype in the characteristic end points of fibrocalcific response including EV release (P=0.02, Figure VIIB in the online-only Data Supplement), ALP activity (P<0.05, Figure VIIB in the online-only Data Supplement), and TGF-β1 release (P=0.0002, Figure VIID in the online-only Data Supplement) but not including TGF-β1 gene expression (Figure VIIC in the online-only Data Supplement). In our in vivo model, the abundant presence of accumulating, electron-dense vesicular particles observed using density-dependent color scanning electron microscopy combined with increased expression of ALP in DDR-1−/− LDL-R−/− vasculature corroborates the role of DDR-1 in early plaque calcification. Future studies are required to elucidate the contribution of stage of plaque maturation as well as varying conditions in the in vivo DDR-1−/− model on the differential effects of DDR-1 in fibrocalcific responses observed in our 2 studies.
The effects of TGF-β1 on atherosclerotic plaque formation are pleiotropic and presumably pathway dependent.36 Our study introduces a novel link between DDR-1 and the TGF-β pathway in conjunction with the release of calcifying EVs, establishing a potent regulatory circuit involving DDR-1 activation, calcified plaque formation, and subsequent destabilization. Our findings indicate that DDR-1 cross-talk with TGF-β1 signaling is independent of bone morphogenetic protein and osteogenic regulator Runx2,37 whereas influencing the expression of Msx2, a Runx2-independent promoter of cytokine-mediated osteogenic vascular calcification interacting with the Wnt pathway38 (Figure VI in the online-only Data Supplement). Under calcifying conditions, DDR-1−/− SMCs released increased TGF-β1 in the ECM, concomitant with elevated phospho-p38 and suppressed phospho-Smad3 associated with the fibrocalcific DDR-1−/− phenotype in vitro and in vivo. These findings complement previous work on the positive effect of phospho-p38 on ALP expression and activity.39 The reversal of enhanced mineral deposition, calcific EV release, and ALP activity in DDR-1−/− SMCs by SB203580 further suggests a causal relationship between p38 phosphorylation and EV-induced fibrocalcific response. We hereby add new evidence to the controversial claims in current literature, either confirming14,40 or refuting41 an interaction between DDR-1 and p38 signaling. The near-complete suppression of phospho-Smad3 concomitant with enhanced mineralization observed in DDR-1−/− models both in vitro and in vivo supports the findings obtained by Shimokado et al42 using Smad3−/− SMCs. TGF-β1 itself is identified as an atherogenic agent, its expression and release related to advanced fibrosis and calcific deposition in human atheromata.24,43 Moreover, the observed increase in expression and release of TGF-β1 by the DDR-1−/− model in vitro and in vivo concurs with a previous study on DDR-1 knockdown in a human cancer cell model.25 Interestingly, the suppression of p38 phosphorylation and TGF-β1 release in DDR-1−/− SMCs by SB431542 suggests a link between TGF-β receptor activation and the p38 pathway that has been implicated in aortic valve calcification.44 In our DDR-1−/− LDL-R−/− mouse model, we observed a strong association of TGF-β1–positive areas with Picrosirius Red–positive fibrotic regions and abundantly present calcific EVs forming compact calcifications similar to those found in human atheromata. This evidence purports a role of TGF-β pathways crucial to the fibrocalcific DDR-1−/− phenotype both in vitro and in vivo.
Collectively, we conclude that DDR-1 provides a novel connection between vascular SMC-induced fibrosis and EV-mediated calcification in early-stage atherosclerotic plaque formation through the adverse regulation of 2 TGF-β signaling pathways. By restricting TGF-β1 release in SMCs, DDR-1 suppresses phosphorylation of proatherogenic p38 and increases phospho-Smad3, resulting in attenuated fibrosis and calcifying EV release (Figure 6). Although further studies are required to elucidate the detailed mechanisms of DDR-1 signaling, the novel role of DDR-1 in maintaining fibrocalcific homeostasis in SMCs using the TGF-β pathway provides a new perspective on the pathogenesis of plaque formation. DDR-1 hereby exerts a TGF-β1–mediated negative feedback on both fibrotic and calcific responses, preventing early-onset ECM remodeling in the process of plaque formation. The interaction between DDR-1, the TGF-β pathway, and the release of calcifying EVs further provides a range of potential implications in congenital and acquired disorders associated with pathological vascular calcification. Our findings introduce a heretofore-undisclosed mechanism regulating vascular fibrocalcific response and ECM remodeling as major determinants of atherosclerotic plaque stability.
Figure 6. Ways of discoidin domain receptor-1 (DDR-1) interaction with transforming growth factor-β (TGF-β) signaling through direct suppression of TGF-β1 (1), phosphorylation of p38 (2), or activation of Smad3 (3), leading to downregulation of calcifying extracellular vesicle (EV) release and EV-bound alkaline phosphatase (ALP) by vascular smooth muscle cells. Inverse regulation of Smad3 and p38 activation through possible downstream interaction (4).
Nonstandard Abbreviations and Acronyms
| β-GP | β-glycerophosphate |
| DDR-1 | discoidin domain receptor-1 |
| DDR-1−/− | DDR-1 knockout |
| ECM | extracellular matrix |
| EV | extracellular vesicle |
| LDL-R−/− | low-density lipoprotein receptor knockout |
| SMC | vascular smooth muscle cell |
| WT | wild-type |
| TGF-β | transforming growth factor-β |
Acknowledgments
We thank Eugenia Shvartz and Jung Choi for their excellent technical assistance.
Sources of Funding
This work was funded by
Disclosures
None.
Footnotes
References
- 1.
Maldonado N, Kelly-Arnold A, Vengrenyuk Y, Laudier D, Fallon JT, Virmani R, Cardoso L, Weinbaum S . A mechanistic analysis of the role of microcalcifications in atherosclerotic plaque stability: potential implications for plaque rupture.Am J Physiol Heart Circ Physiol. 2012; 303:H619–H628. doi: 10.1152/ajpheart.00036.2012.CrossrefMedlineGoogle Scholar - 2.
Vengrenyuk Y, Carlier S, Xanthos S, Cardoso L, Ganatos P, Virmani R, Einav S, Gilchrist L, Weinbaum S . A hypothesis for vulnerable plaque rupture due to stress-induced debonding around cellular microcalcifications in thin fibrous caps.Proc Natl Acad Sci U S A. 2006; 103:14678–14683. doi: 10.1073/pnas.0606310103.CrossrefGoogle Scholar - 3.
Kelly-Arnold A, Maldonado N, Laudier D, Aikawa E, Cardoso L, Weinbaum S . Revised microcalcification hypothesis for fibrous cap rupture in human coronary arteries.Proc Natl Acad Sci U S A. 2013; 110:10741–10746. doi: 10.1073/pnas.1308814110.CrossrefGoogle Scholar - 4.
New SE, Aikawa E . Role of extracellular vesicles in de novo mineralization: an additional novel mechanism of cardiovascular calcification.Arterioscler Thromb Vasc Biol. 2013; 33:1753–1758. doi: 10.1161/ATVBAHA.112.300128.LinkGoogle Scholar - 5.
Kapustin AN, Chatrou ML, Drozdov I, . Vascular smooth muscle cell calcification is mediated by regulated exosome secretion.Circ Res. 2015; 116:1312–1323. doi: 10.1161/CIRCRESAHA.116.305012.LinkGoogle Scholar - 6.
Kapustin AN, Davies JD, Reynolds JL, McNair R, Jones GT, Sidibe A, Schurgers LJ, Skepper JN, Proudfoot D, Mayr M, Shanahan CM . Calcium regulates key components of vascular smooth muscle cell-derived matrix vesicles to enhance mineralization.Circ Res. 2011; 109:e1–12. doi: 10.1161/CIRCRESAHA.110.238808.LinkGoogle Scholar - 7.
Bertazzo S, Gentleman E, Cloyd KL, Chester AH, Yacoub MH, Stevens MM . Nano-analytical electron microscopy reveals fundamental insights into human cardiovascular tissue calcification.Nat Mater. 2013; 12:576–583. doi: 10.1038/nmat3627.CrossrefMedlineGoogle Scholar - 8.
Hutcheson JD, Goettsch C, Bertazzo S, Maldonado N, Ruiz JL, Goh W, Yabusaki K, Faits TS, Bouten CV, Franck G, Quillard T, Libby P, Aikawa M, Weinbaum S, Aikawa E . Genesis and growth of extracellular vesicle-derived microcalcification in atherosclerotic plaques.Nat Mater. 2016; doi: 10.1038/nmat4519.CrossrefGoogle Scholar - 9.
Vogel W, Gish GD, Alves F, Pawson T . The discoidin domain receptor tyrosine kinases are activated by collagen.Mol cell. 1997; 1:13–23.CrossrefMedlineGoogle Scholar - 10.
Hou G, Vogel WF, Bendeck MP . Tyrosine kinase activity of discoidin domain receptor 1 is necessary for smooth muscle cell migration and matrix metalloproteinase expression.Circ Res. 2002; 90:1147–1149.LinkGoogle Scholar - 11.
Yeh YC, Lin HH, Tang MJ . A tale of two collagen receptors, integrin β1 and discoidin domain receptor 1, in epithelial cell differentiation.Am J Physiol Cell Physiol. 2012; 303:C1207–C1217. doi: 10.1152/ajpcell.00253.2012.CrossrefMedlineGoogle Scholar - 12.
Ferri N, Carragher NO, Raines EW . Role of discoidin domain receptors 1 and 2 in human smooth muscle cell-mediated collagen remodeling: potential implications in atherosclerosis and lymphangioleiomyomatosis.Am J Pathol. 2004; 164:1575–1585. doi: 10.1016/S0002-9440(10)63716-9.CrossrefMedlineGoogle Scholar - 13.
Leitinger B . Discoidin domain receptor functions in physiological and pathological conditions.Int Rev Cell Mol Biol. 2014; 310:39–87. doi: 10.1016/B978-0-12-800180-6.00002-5.CrossrefMedlineGoogle Scholar - 14.
Fu HL, Valiathan RR, Arkwright R, Sohail A, Mihai C, Kumarasiri M, Mahasenan KV, Mobashery S, Huang P, Agarwal G, Fridman R . Discoidin domain receptors: unique receptor tyrosine kinases in collagen-mediated signaling.J Biol Chem. 2013; 288:7430–7437. doi: 10.1074/jbc.R112.444158.CrossrefMedlineGoogle Scholar - 15.
Franco C, Hou G, Ahmad PJ, Fu EY, Koh L, Vogel WF, Bendeck MP . Discoidin domain receptor 1 (ddr1) deletion decreases atherosclerosis by accelerating matrix accumulation and reducing inflammation in low-density lipoprotein receptor-deficient mice.Circ Res. 2008; 102:1202–1211. doi: 10.1161/CIRCRESAHA.107.170662.LinkGoogle Scholar - 16.
Hou G, Wang D, Bendeck MP . Deletion of discoidin domain receptor 2 does not affect smooth muscle cell adhesion, migration, or proliferation in response to type I collagen.Cardiovasc Pathol. 2012; 21:214–218. doi: 10.1016/j.carpath.2011.07.006.CrossrefMedlineGoogle Scholar - 17.
Ahmad PJ, Trcka D, Xue S, Franco C, Speer MY, Giachelli CM, Bendeck MP . Discoidin domain receptor-1 deficiency attenuates atherosclerotic calcification and smooth muscle cell-mediated mineralization.Am J Pathol. 2009; 175:2686–2696. doi: 10.2353/ajpath.2009.080734.CrossrefMedlineGoogle Scholar - 18.
Ruiz-Ortega M, Rodríguez-Vita J, Sanchez-Lopez E, Carvajal G, Egido J . TGF-beta signaling in vascular fibrosis.Cardiovasc Res. 2007; 74:196–206. doi: 10.1016/j.cardiores.2007.02.008.CrossrefMedlineGoogle Scholar - 19.
Pardali E, Ten Dijke P . TGFβ signaling and cardiovascular diseases.Int J Biol Sci. 2012; 8:195–213. doi: 10.7150/ijbs.3805.CrossrefMedlineGoogle Scholar - 20.
Guo X, Wang XF . Signaling cross-talk between TGF-beta/BMP and other pathways.Cell Res. 2009; 19:71–88. doi: 10.1038/cr.2008.302.CrossrefMedlineGoogle Scholar - 21.
Simionescu A, Philips K, Vyavahare N . Elastin-derived peptides and TGF-beta1 induce osteogenic responses in smooth muscle cells.Biochem Biophys Res Commun. 2005; 334:524–532. doi: 10.1016/j.bbrc.2005.06.119.CrossrefMedlineGoogle Scholar - 22.
Watson KE, Boström K, Ravindranath R, Lam T, Norton B, Demer LL . TGF-beta 1 and 25-hydroxycholesterol stimulate osteoblast-like vascular cells to calcify.J Clin Invest. 1994; 93:2106–2113. doi: 10.1172/JCI117205.CrossrefMedlineGoogle Scholar - 23.
Jeziorska M . Transforming growth factor-betas and cd105 expression in calcification and bone formation in human atherosclerotic lesions.Z Kardiol. 2001; 90(suppl 3):23–26.CrossrefMedlineGoogle Scholar - 24.
Wu SC, Hsiao HF, Ho ML, Hung YL, Chang JK, Wang GJ, Wang CZ . Suppression of discoidin domain receptor 1 expression enhances the chondrogenesis of adipose-derived stem cells.Am J Physiol Cell Physiol. 2015; 308:C685–C696. doi: 10.1152/ajpcell.00398.2014.CrossrefMedlineGoogle Scholar - 25.
Rudra-Ganguly N, Lowe C, Mattie M, Chang MS, Satpayev D, Verlinsky A, An Z, Hu L, Yang P, Challita-Eid P, Stover DR, Pereira DS . Discoidin domain receptor 1 contributes to tumorigenesis through modulation of TGFBI expression.PLoS One. 2014; 9:e111515. doi: 10.1371/journal.pone.0111515.CrossrefMedlineGoogle Scholar - 26.
György B, Szabó TG, Pásztói M, Pál Z, Misják P, Aradi B, László V, Pállinger E, Pap E, Kittel A, Nagy G, Falus A, Buzás EI . Membrane vesicles, current state-of-the-art: emerging role of extracellular vesicles.Cell Mol Life Sci. 2011; 68:2667–2688. doi: 10.1007/s00018-011-0689-3.CrossrefMedlineGoogle Scholar - 27.
Orimo H, Shimada T . The role of tissue-nonspecific alkaline phosphatase in the phosphate-induced activation of alkaline phosphatase and mineralization in SaOS-2 human osteoblast-like cells.Mol Cell Biochem. 2008; 315:51–60. doi: 10.1007/s11010-008-9788-3.CrossrefMedlineGoogle Scholar - 28.
Aper SJ, van Spreeuwel AC, van Turnhout MC, van der Linden AJ, Pieters PA, van der Zon NL, de la Rambelje SL, Bouten CV, Merkx M . Colorful protein-based fluorescent probes for collagen imaging.PLoS One. 2014; 9:e114983. doi: 10.1371/journal.pone.0114983.CrossrefMedlineGoogle Scholar - 29.
Aikawa E, Nahrendorf M, Figueiredo JL, Swirski FK, Shtatland T, Kohler RH, Jaffer FA, Aikawa M, Weissleder R . Osteogenesis associates with inflammation in early-stage atherosclerosis evaluated by molecular imaging in vivo.Circulation. 2007; 116:2841–2850. doi: 10.1161/CIRCULATIONAHA.107.732867.LinkGoogle Scholar - 30.
Velidandla S, Gaikwad P, Ealla KK, Bhorgonde KD, Hunsingi P, Kumar A . Histochemical analysis of polarizing colors of collagen using picrosirius red staining in oral submucous fibrosis.J Int Oral Health: JIOH. 2014; 6:33–38.MedlineGoogle Scholar - 31.
Prosdocimo DA, Wyler SC, Romani AM, O’Neill WC, Dubyak GR . Regulation of vascular smooth muscle cell calcification by extracellular pyrophosphate homeostasis: synergistic modulation by cyclic AMP and hyperphosphatemia.Am J Physiol Cell Physiol. 2010; 298:C702–C713. doi: 10.1152/ajpcell.00419.2009.CrossrefMedlineGoogle Scholar - 32.
Shioi A, Nishizawa Y, Jono S, Koyama H, Hosoi M, Morii H . Beta-glycerophosphate accelerates calcification in cultured bovine vascular smooth muscle cells.Arterioscler. Thromb Vasc Biol. 1995; 15:2003–2009.LinkGoogle Scholar - 33.
Hutcheson JD, Maldonado N, Aikawa E . Small entities with large impact: microcalcifications and atherosclerotic plaque vulnerability.Curr Opin Lipidol. 2014; 25:327–332. doi: 10.1097/MOL.0000000000000105.CrossrefMedlineGoogle Scholar - 34.
Jono S, McKee MD, Murry CE, Shioi A, Nishizawa Y, Mori K, Morii H, Giachelli CM . Phosphate regulation of vascular smooth muscle cell calcification.Circ Res. 2000; 87:E10–E17.LinkGoogle Scholar - 35.
Bottagisio M, Lovati AB, Lopa S, Moretti M . Osteogenic differentiation of human and ovine bone marrow stromal cells in response to β-glycerophosphate and monosodium phosphate.Cell Reprogram. 2015; 17:235–242. doi: 10.1089/cell.2014.0105.CrossrefMedlineGoogle Scholar - 36.
Redondo S, Santos-Gallego CG, Tejerina T . TGF-beta1: a novel target for cardiovascular pharmacology.Cytokine Growth Factor Rev. 2007; 18:279–286. doi: 10.1016/j.cytogfr.2007.04.005.CrossrefMedlineGoogle Scholar - 37.
Bae JS, Gutierrez S, Narla R, Pratap J, Devados R, van Wijnen AJ, Stein JL, Stein GS, Lian JB, Javed A . Reconstitution of Runx2/Cbfa1-null cells identifies a requirement for BMP2 signaling through a Runx2 functional domain during osteoblast differentiation.J Cell Biochem. 2007; 100:434–449. doi: 10.1002/jcb.21039.CrossrefMedlineGoogle Scholar - 38.
Shao JS, Aly ZA, Lai CF, Cheng SL, Cai J, Huang E, Behrmann A, Towler DA . Vascular Bmp Msx2 Wnt signaling and oxidative stress in arterial calcification.Ann N Y Acad Sci. 2007; 1117:40–50. doi: 10.1196/annals.1402.075.CrossrefMedlineGoogle Scholar - 39.
Sowa H, Kaji H, Yamaguchi T, Sugimoto T, Chihara K . Activations of ERK1/2 and JNK by transforming growth factor beta negatively regulate Smad3-induced alkaline phosphatase activity and mineralization in mouse osteoblastic cells.J Biol Chem. 2002; 277:36024–36031. doi: 10.1074/jbc.M206030200.CrossrefMedlineGoogle Scholar - 40.
Matsuyama W, Wang L, Farrar WL, Faure M, Yoshimura T . Activation of discoidin domain receptor 1 isoform b with collagen up-regulates chemokine production in human macrophages: role of p38 mitogen-activated protein kinase and NF-kappa B.J Immunol. 2004; 172:2332–2340.CrossrefMedlineGoogle Scholar - 41.
Lu KK, Trcka D, Bendeck MP . Collagen stimulates discoidin domain receptor 1-mediated migration of smooth muscle cells through Src.Cardiovasc Pathol. 2011; 20:71–76. doi: 10.1016/j.carpath.2009.12.006.CrossrefMedlineGoogle Scholar - 42.
Shimokado A, Sun Y, Nakanishi M, Sato F, Oikawa K, Akasaka T, Muragaki Y . Smad3 plays an inhibitory role in phosphate-induced vascular smooth muscle cell calcification.Exp Mol Pathol. 2014; 97:458–464. doi: 10.1016/j.yexmp.2014.10.005.CrossrefMedlineGoogle Scholar - 43.
Panutsopulos D, Papalambros E, Sigala F, Zafiropoulos A, Arvanitis DL, Spandidos DA . Protein and mRNA expression levels of VEGF-A and TGF-beta1 in different types of human coronary atherosclerotic lesions.Int J Mol Med. 2005; 15:603–610.MedlineGoogle Scholar - 44.
Hutcheson JD, Ryzhova LM, Setola V, Merryman WD . 5-HT(2B) antagonism arrests non-canonical TGF-β1-induced valvular myofibroblast differentiation.J Mol Cell Cardiol. 2012; 53:707–714. doi: 10.1016/j.yjmcc.2012.08.012.CrossrefMedlineGoogle Scholar
Significance
Vascular fibrosis and calcification determine atherosclerotic plaque stability. Microcalcifications formed through the accumulation of calcifying extracellular vesicles (EVs) destabilize the plaque and contribute to the risk of plaque rupture. The interaction of collagen synthesis and calcifying EV release by vascular smooth muscle cells driving fibrocalcific response in plaque formation is unknown. We identify discoidin domain receptor-1 as a novel molecular switch restricting fibrosis and EV-mediated calcification of the extracellular matrix. By affecting the release of transforming growth factor-β1 and several transforming growth factor-β signaling pathways in vascular smooth muscle cells, discoidin domain receptor-1 provides negative feedback on collagen synthesis, the release of calcifying EVs and the expression of alkaline phosphatase, diminishing EV calcifying potential. Discoidin domain receptor-1 activation thus shuts off excess collagen production and calcific deposition, maintaining homeostasis of the extracellular matrix. Our study introduces the discoidin domain receptor-1–transforming growth factor-β axis as the missing link between fibrosis and EV-mediated calcification in atherosclerotic plaque formation.
Submit a Response to This Article