Elastolytic activity of cysteine cathepsins K, S, and V promotes vascular calcification

Abstract

Elastin plays an important role in maintaining blood vessel integrity. Proteolytic degradation of elastin in the vascular system promotes the development of atherosclerosis, including blood vessel calcification. Cysteine cathepsins have been implicated in this process, however, their role in disease progression and associated complications remains unclear. Here, we showed that the degradation of vascular elastin by cathepsins (Cat) K, S, and V directly stimulates the mineralization of elastin and that mineralized insoluble elastin fibers were ~25-30% more resistant to CatK, S, and V degradation when compared to native elastin. Energy dispersive X-ray spectroscopy investigations showed that insoluble elastin predigested by CatK, S, or V displayed an elemental percentage in calcium and phosphate up to 8-fold higher when compared to non-digested elastin. Cathepsin-generated elastin peptides increased the calcification of MOVAS-1 cells acting through the ERK1/2 pathway by 34-36%. We made similar observations when cathepsin-generated elastin peptides were added to ex vivo mouse aorta rings. Altogether, our data suggest that CatK-, S-, and V-mediated elastolysis directly accelerates the mineralization of the vascular matrix by the generation of nucleation points in the elastin matrix and indirectly by elastin-derived peptides stimulating the calcification by vascular smooth muscle cells. Both processes inversely protect against further extracellular matrix degradation.

Figure 1

Digestion of non-calcified and calcified elastin by CatK, S, and V. Native or calcified bovine neck elastin (1 mg) was incubated with 2 µM CatK, S, or V in 100 mM sodium acetate pH 5.5, 2.5 mM DTT, 2.5 mM EDTA for 18 h at 37 °C. (a) Non-calcified and (b) calcified elastin were analyzed by SEM after digestion. (c) Remaining elastin weight was measured after digestion and expressed as a percentage of the control. (d) Desmosine release in the digestion supernatant was quantified by ELISA. Digestion supernatants from (e) non-calcified and (f) calcified elastin were loaded on a C-18 column and eluted by HPLC. (g) The area under the chromatograms were quantified with Empower software (Waters). Data from non-calcified elastin (n = 6) were compared to the corresponding calcified samples (n = 6) by one-way Mann-Whitney U test. (*p < 0.025, **p < 0.01).

Figure 2

Mineralization of CatK-, S-, and V-digested elastin fibers. Bovine neck elastin was digested by CatK, S, and V prior to mineralization. (a) Samples were analyzed by Back Scattered (BS) SEM. EDS analysis confirmed the mineral deposits on calcified elastin fibers highlighting (b) calcium, (c) phosphate and (d) the elemental content was expressed as percentage. (e) Intact and digested mineralized elastin was decalcified with 0.6 N HCl and acid extracts were analyzed for calcium and phosphate content (µmol/g of elastin). Pairwise comparison of Calcium/Phosphate levels between control elastin (n = 5) and digested samples (n = 5 for each set) was carried out by Mann-Whitney U test (*p < 0.01). (f) The amount of desmosine detected in the digestion supernatant was correlated with the amount of calcium determined in the corresponding sample. The correlation between the two variables was analyzed by the Spearman method.

Figure 3

Characterization of the mineral phase of calcified elastin. (a) Raman spectra obtained from cortical bone sample (black) and calcified elastin (red) mineral fractions. The peaks corresponding the phosphate anion vibrational frequencies are indicated with arrows. (b) Powder X-ray diffraction patterns of cortical bone (black) and calcified elastin (red) mineral fractions. The diffractions patterns were plotted with the hydroxyapatite reference pattern (ICDD 9-432) obtained from PDF-2 2012 database.

Figure 4

Effects of cathepsin-generated elastin peptides on vascular smooth muscle cell calcification. MOVAS-1 cells were incubated in low phosphate (LP) complete DMEM (1 mM Pi) or high phosphate (HP) complete DMEM (2 mM Pi) in the presence or absence of a tryptic BSA digest, synthetic VGVAPG peptide or CatK-, S-, V or MMP-12-digested elastin (each 10 µg/mL) for 21 days. (a) Cell layers were stained with 2% Alizarin-Red pH 4.2 (calcium staining), images were recorded at 20x magnification and (b) the intensity of Alizarin-Red-positive staining was quantified with NIS-Elements software (Nikon), data are represented as mean ± SD. (c) Cytochemical staining of active alkaline phosphatase was carried out with an alkaline phosphatase detection kit (Sigma-Aldrich) and (d) staining intensity was quantified with NIS-Elements software (Nikon), data are represented as mean ± SD. (e) Cell layers were decalcified with 0.6 N HCl for 24 h and solubilized calcium was quantified by cresolphthalein assay (Randox) and normalized to protein concentration. (f) Lysates from 1 h, 12 h and 24 h-old cell culture were loaded (30 µg total protein) and separated onto 12% acrylamide gel, transferred onto a nitrocellulose membrane and probed with anti-phospho-ERK1/2 (p-ERK1/2) Membranes were then stripped and probed with anti-ERK1/2 antibodies, then stripped again and probed with anti-β-actin antibody (all three panels are showing the same membrane). The intensity of the immunoreactive bands was quantified by ImageJ software (NIH software) and normalized to the β-actin corresponding signal. Comparison between control HP (n = 4) and peptides treated HP samples (n = 4 for each set) was carried out with the one-way Mann-Whitney U test (n.s.: non-significant, *p < 0.05).

Figure 5

Effect of cathepsin-generated elastin peptides on the calcification of aortic tissue. Aortas from 3-month-old C57BL6 mice were incubated in low phosphate (LP) complete αMEM (1 mM Pi) or in high phosphate (HP) αMEM (2 mM Pi) in the presence or absence of tryptic BSA peptides or CatK-, S-, or V-digested elastin for 12 days at 37 °C. (a) Aortas were washed with PBS, fixed with 10% buffered formalin for 24 h and stained with 2% Alizarin-Red pH 4.2 and images were recorded at low magnification. (b) 4 mm fragments from descending aortas were embedded in paraffin and cut into 5 µm sections and stained with Alizarin-Red. Images were recorded at 100x magnification. (c) Alizarin-Red intensity from 10 sections for each condition was quantified with NIS-Elements software (Nikon); data are represented as mean ± SD. (d) 1 mm aorta sections were decalcified with 0.6 N HCl for 24 h and solubilized calcium was quantified by a cresolphthalein assay (Randox) and normalized to protein concentration. Comparison between control HP (n = 10) and peptides treated HP samples (n = 10 for each set) was carried out with the one-way Mann-Whitney U test (n.s.: non-significant, *p < 0.01).

Figure 6

Effects of cysteine cathepsin and MMPs inhibition on aortic ring calcification. 4-mm-long aortic rings were incubated in low phosphate (LP) complete αMEM (1 mM Pi) or in high phosphate (HP) αMEM (2 mM Pi) in the presence or absence of E64 or GM6001 (10 µM) or lactose (50 mM) for 12 days at 37 °C. (a) 5 µm sections were stained with hematoxylin and eosin (H&E), Masson’s Trichrome (TC) and Alizarin-Red. Images were recorded at 100x magnification. (b) Alizarin-Red intensity was quantified with NIS-Elements software (Nikon), data are represented as mean ± SD. Comparison between control HP (n = 20) and peptides treated HP samples (n = 20 for each set) was carried out with the one-way Mann-Whitney U test (*p < 0.01). (c) The amount of desmosine in the culture medium was determined by ELISA.