Collagen VI Deficiency Impairs Tendon Fibroblasts Mechanoresponse in Ullrich Congenital Muscular Dystrophy

Abstract

The pericellular matrix (PCM) is a specialized extracellular matrix that surrounds cells. Interactions with the PCM enable the cells to sense and respond to mechanical signals, triggering a proper adaptive response. Collagen VI is a component of muscle and tendon PCM. Mutations in collagen VI genes cause a distinctive group of inherited skeletal muscle diseases, and Ullrich congenital muscular dystrophy (UCMD) is the most severe form. In addition to muscle weakness, UCMD patients show structural and functional changes of the tendon PCM. In this study, we investigated whether PCM alterations due to collagen VI mutations affect the response of tendon fibroblasts to mechanical stimulation. By taking advantage of human tendon cultures obtained from unaffected donors and from UCMD patients, we analyzed the morphological and functional properties of cellular mechanosensors. We found that the length of the primary cilia of UCMD cells was longer than that of controls. Unlike controls, in UCMD cells, both cilia prevalence and length were not recovered after mechanical stimulation. Accordingly, under the same experimental conditions, the activation of the Hedgehog signaling pathway, which is related to cilia activity, was impaired in UCMD cells. Finally, UCMD tendon cells exposed to mechanical stimuli showed altered focal adhesions, as well as impaired activation of Akt, ERK1/2, p38MAPK, and mechanoresponsive genes downstream of YAP. By exploring the response to mechanical stimulation, for the first time, our findings uncover novel unreported mechanistic aspects of the physiopathology of UCMD-derived tendon fibroblasts and point at a role for collagen VI in the modulation of mechanotransduction in tendons.

Keywords: COL6-related myopathies; Ullrich congenital muscular dystrophy; collagen VI; contractures; focal adhesion; mechanosignaling; primary cilium; tendon fibroblasts.

Figure 1

Immunofluorescence and ultrastructural changes of collagen VI in UCMD tendon cultures. (A) Representative immunofluorescence for collagen VI (green) in TF cultures from control (CTRL) and UCMD (UCMD1-3) patients. Cells were grown to confluence and maintained 4 days in the presence of 0.25 mM L-ascorbic acid. In control cultures, collagen VI formed a dense extracellular network constituted by interconnected microfibrils (arrows). UCMD cultures displayed an altered arrangement of collagen VI networks, featuring dot-like deposits (arrowheads) in the ECM. In UCMD2 and UCMD3, the protein also accumulated in the cytoplasm (asterisks) of cells. (B) Representative image for double labeling with anti-collagen XII (upper panels, green) and collagen I (lower panels, red) in TF cultures, showing a coarse ECM arrangement (arrowheads) in UCMD cultures. In A and B, nuclei were stained with DAPI. Scale bar, 10 μm. (C) Transmission electron microscopy visualization of rotary shadowed replicas of proliferating TFs from control (CTRL), UCMD2, and UCMD3 patients, immunolabeled with anti-collagen VI and 5 nm-colloidal gold conjugated secondary antibody. In normal controls, colloidal gold particles associate with the globular domain of collagen VI tetramers, that, in turn, form microfilaments featuring “pearl necklace strands” (left upper panel, inset, white arrowheads), and complex webs. Other components of the ECM (upper right panel, white arrows), appear almost completely masked by the complex collagen VI webs. Note that in both UCMD TFs’ cultures, the few secreted collagen VI tetramers show irregular globular domains (left panels, white arrowheads): they form short microfilaments (right panels, white arrowheads), and the ECM, in general, due to the absence of collagen VI, has a smooth appearance (white arrows). Scale bar, 0.1 μm.

Figure 2

PC organization and structure in control and UCMD tendon cultures. (A) Representative double immunofluorescence for Arl13b (red) and collagen VI (green) in control (CTRL) and UCMD (UCMD1-3) TFs. Arrows indicate the PC. Note the absence of collagen VI microfibrils in the proximity of the PC in all three UCMD cultures, opposite from the dense collagen VI network enveloping the PC in control cells. Nuclei were stained with DAPI. Scale bar, 5 μm. (B) Confocal imaging and 3D reconstruction of a single PC in control (CTRL) and collagen VI-deficient (UCMD) TFs, double-labeled with antibodies against collagen VI (green) and against acetylated α-tubulin (red), confirming the absence of collagen VI associated with the surface of the PC in UCMD TFs. Arrows indicate the extension of the cilium axoneme. Scale bar, 2 μm. (C) Scanning electron microscopy of normal (CTRL) and UCMD TFs, showing the presence of the PC (arrows) protruding from the cell surface. Higher magnifications on the right show the typical antenna-like structure of the PC. Scale bar of panels on the left 30 μm, and panels on the right 2 μm. (D) Representative immunofluorescence images of the PC in control (CTRL) and UCMD (UCMD1-3) TFs with an antibody against Arl13b (red), showing longer cilia in UCMD TFs. Nuclei were stained with DAPI. Scale bar, 2 μm. (E) Quantification of the PC length (upper panel) and PC prevalence (i.e., number of ciliated cells in percentage; lower panel) in cultured TFs from control (CTRL) and UCMD (UCMD1-3) patients, determined by immunofluorescence with an anti-Arl13b antibody. Measurements were performed with NIS AR 4.50 software, counting 100 cells for each culture. Significance was calculated by one-way ANOVA with Tukey’s multiple comparison test and was expressed between each UCMD and control TF under the same experimental conditions. * p < 0.05; ns, not significant; n = 3.

Figure 3

Modulation of PC and Hh signaling in control and UCMD tendon cultures under mechanical stress. (A) Representative immunofluorescence analysis of the PC in control (CTRL) and UCMD (UCMD1-3) TF cultures labeled with anti-Arl13b antibody (red), under unstrained condition (basal), after uniaxial strain (strain), and after a 3 h recovery from strain (recovery). Nuclei were stained with DAPI. Scale bar, 5 μm. (B) Quantification of PC prevalence (upper panel) and average PC length (lower panel) in control (CTRL) and UCMD TFs under the same conditions as in panel A. Statistical analysis by one-way ANOVA with Tukey’s multiple comparisons test. *, p < 0.05; **, p < 0.005; n = 3. (C) RT-qPCR analysis of GLI1 expression in control (CTRL) and UCMD TFs maintained in the above three conditions. RPLP0 served as the internal reference. Data are expressed as 2^−ΔCt × 10⁵ and presented as mean ± SD of at least three biological replicates. **, p < 0.001; ns, not statistically significant; n = 4 control cultures, each condition; n = 3 UCMD cultures, each condition. (D) Representative immunofluorescence for GLI1 in control (CTRL) and UCMD TFs maintained in the above three conditions. Scale bar, 5 μm. (E) Quantification of the nuclear relative fluorescence intensity (F.I.) of GLI1 staining in control (CTRL) and UCMD TFs, based on GLI1 immunofluorescence as in D. For each culture, values were normalized on the respective basal condition. Measurements were performed with NIS AR 4.50 software, counting 70 cells for each culture. Statistical analysis by one-way ANOVA with Tukey’s multiple comparisons test. ** p < 0.005, *** p < 0.0001; ns, not significant; n = 3.

Figure 4

Organization and activation of focal adhesions are impaired in UCMD tendon cultures during and after exposure to mechanical stress. (A,B) Magnification of confocal imaging of control (CTRL) and UCMD TFs under unstrained conditions (basal), at the end of mechanical stimulation (strain), and upon 3 h recovery from strain (recovery), following double labeling with fluorescent phalloidin (red) and with antibodies against A. talin (green) and B. phosphorylated FAK (pFAK, green). Nuclei were stained with DAPI. Scale bar, 10 µm. (C) 3D reconstruction of confocal imaging performed on control (CTRL) and UCMD TFs under unstrained conditions (basal), at the end of mechanical stimulation (strain), and upon 3 h recovery from strain (recovery), labeled with talin (green) and phalloidin (red). Nuclei were stained with DAPI. Scale bar, 5 µm. (D) Scanning electron microscopy of control (CTRL) and UCMD TF cultures under unstrained conditions (basal), at the end of mechanical stimulation (strain), and at the end of mechanical stimulation (recovery), showing an increased number of microvilli on the surface of UCMD cells in both stretched and recovery conditions. Scale bar 15 µm for low magnification, 5 µm inset. (E) Quantification of the number of microvilli per cell surface area in control and UCMD TFs in the above three conditions, based on scanning electron microscopy imaging as in D. Statistical analysis was performed by Student’s t test. **, p < 0.05; n = 10 fields, each condition.

Figure 5

Activation of signaling pathways downstream of focal adhesions is not properly restored in UCMD tendon cultures following recovery from mechanical stress. (A) Representative western blot analysis of the expression and the activation profile of talin and different protein kinases in TF cultures of control (CTRL) and UCMD (UCMD1-2) patients in the absence (−) or presence (+) of mechanical strain, as well as upon recovery for the indicated time (0, 0.5 and 3 h). Blots were incubated with antibodies against talin, phospho-Y397 FAK (P-FAK), total FAK, phospho-S473 Akt (P-Akt), total Akt, phospho-T202/Y204 ERK1/2 (P-ERK1/2), ERK2, phospho-T180/Y182 p38MAPK (P-p38MAPK), and total p38MAPK. GAPDH was used as a loading control. Molecular weight markers are reported on the right of the blots and are expressed in kDa. (B) Histograms of the relative densitometric quantifications normalized for GAPDH. ***, p < 0.001; **, p < 0.05; *, p < 0.01; n = 3, each condition. (C) Representative immunofluorescence analysis of the subcellular localization of YAP (green) in control (CTRL) and UCMD TFs under unstrained conditions (basal), at the end of mechanical stimulation (strain), and upon 3 h recovery from strain (recovery). Upon stretching, YAP nuclear staining is displayed by control, but not by UCMD cells. Nuclei were stained with DAPI. Scale bar, 5 µm. (D) Quantification of the ratio of the mean fluorescence intensity of nuclear and cytoplasmic YAP, based on immunofluorescence analysis as in C. Data are presented as mean ± SEM. Statistical analysis by one-way ANOVA with Tukey’s multiple comparisons test. **, p < 0.002; n = 30 measurement, each condition. (E) RT-qPCR analysis of CTGF and CYR61 mRNAs levels in control (CTRL) and UCMD TFs under unstrained conditions (basal), at the end of mechanical stimulation (strain), and upon 3 h recovery from strain (rec). Data are expressed as 2^−ΔCt × 10⁵ and presented as mean ± SD of at least three biological replicates. **, p < 0.005; *, p < 0.01; ns, not significant; n = 4 control cultures, each condition; n = 3 UCMD cultures, each condition.

Figure 6

Mutations of collagen VI genes result in defective adaptation to mechanical strain. In normal TFs, following exposure to mechanical strain, at the PCM, the primary cilium (PC) and specific integrins interact with collagen VI, triggering the proper formation of focal adhesions (FA) and stress fibers, as well as the activation of a plethora of molecular pathways that lead to the translocation and activation of GLI1 and YAP. As a result, the upregulation of GLI1 itself, CYR61, and CTGF is observed, with possible consequences on ECM remodeling and on the proper cellular adaptation to strain. The absence of a functional collagen VI, here observed on the right, impairs these mechanisms starting with an altered or even absent interaction between collagen VI and the PC, or with specific membrane receptors that result in a poor activation of the downstream molecular pathways. As a consequence, mechanoresponse and mechanotransduction are impaired, possibly leading to an impaired adaptation to stress. Green and red lines indicate active and impaired pathways, respectively.