Granzyme B mediates both direct and indirect cleavage of extracellular matrix in skin after chronic low-dose ultraviolet light irradiation

Extracellular matrix (ECM) degradation is a hallmark of many chronic inflammatory diseases that can lead to a loss of function, aging, and disease progression. Ultraviolet light (UV) irradiation from the sun is widely considered as the major cause of visible human skin aging, causing increased inflammation and enhanced ECM degradation. Granzyme B (GzmB), a serine protease that is expressed by a variety of cells, accumulates in the extracellular milieu during chronic inflammation and cleaves a number of ECM proteins. We hypothesized that GzmB contributes to ECM degradation in the skin after UV irradiation through both direct cleavage of ECM proteins and indirectly through the induction of other proteinases. Wild-type and GzmB-knockout mice were repeatedly exposed to minimal erythemal doses of solar-simulated UV irradiation for 20 weeks. GzmB expression was significantly increased in wild-type treated skin compared to nonirradiated controls, colocalizing to keratinocytes and to an increased mast cell population. GzmB deficiency significantly protected against the formation of wrinkles and the loss of dermal collagen density, which was related to the cleavage of decorin, an abundant proteoglycan involved in collagen fibrillogenesis and integrity. GzmB also cleaved fibronectin, and GzmB-mediated fibronectin fragments increased the expression of collagen-degrading matrix metalloproteinase-1 (MMP-1) in fibroblasts. Collectively, these findings indicate a significant role for GzmB in ECM degradation that may have implications in many age-related chronic inflammatory diseases.

Increase in epidermal and dermal thickness after UV irradiation. Representative H&E stained sections of dorsal skin collected from wild-type non-irradiated control (WT-C), wild-type UV-irradiated (WT-UVR) and GzmB-KO UV-irradiated (KO-UVR) mice after 20 weeks. There was a significant increase in epidermal and dermal thickness after UV irradiation in both genotypes, leading to an overall increase in skin thickness (mean ± SEM; p<0.05 Tukey's multiple comparison).

Fig S2:
Quantification of immune cells in control and UV-irradiation mouse skin. Dorsal skin sections were immuno-stained for neutrophil elastase (neutrophils), CD68 (macrophages) and CD3 (T cells) and cells were either counted or the intensity of staining was measured via the number of positive pixels detected (above a set threshold) and normalized to area. Results are expressed at a percentage of wild-type non-irradiated control (WT-C) (mean ± SEM; *p<0.05, Tukey's multiple comparison). Scale bars = 60 µm. Figure S3: GzmB-mediated FN fragments increase MMP-3 expression and release from fibroblasts Figure S3: GzmB-mediated FN fragments increase MMP-3 expression and release from fibroblasts. Primary fibroblasts were added to GzmB-mediated FN-fragments and MMP-3 release was assayed in the supernatants after 20h by western blot. GAPDH was probed from cell lysates collected from the same wells as loading controls. Results are expressed as a percentage of intact fibronectin control (mean ± SEM from quadruplicate wells, *p<0.05 t-test).

Tissue collection and processing
At 7 or 20 weeks, mice were euthanized by carbon dioxide inhalation and equivalent areas of dorsal skin were excised. An approximate 1x1cm piece of skin was fixed in 10% buffered formalin for 24 h and then processed to paraffin embedding for histological and immunohistochemical analyses. Skin samples were also flash frozen in liquid nitrogen and stored at -80 o C for subsequent multi-photon microscopy, protein extraction and western blot.

Histology and immunohistochemistry
Five micron cross-sections of skin were deparaffinized and stained with hematoxylin and eosin (H&E) for evaluation of morphology. Measurement of skin thickness was completed by scanning slides with an Aperio Scanscope and using the associated ImageScope software (Aperio Technologies Inc. v11.2.0.780) to take measurements across the entire section at multiple sites and averaged for each mouse. Collagen was observed in picrosirius red-stained sections using 100% polarized light and pictures were taken at a fixed exposure. Quantification of collagen was performed as previously described (Hiebert et al. 2013). Toluidine Blue (TBO) at pH 2.0 was used to stain mast cells.
Mast cells, neutrophils and CD68 macrophages were manually counted and normalized to the length of tissue counted. CD3 T cells, MMP-1 and decorin staining was quantified using Aperio ImageScope. The number of color pixels (above a set threshold) were counted and expressed as the number of positive pixels per unit area. For all quantifications, at least three different areas were analyzed and averaged per mouse/section. Fibroblast detachment assay 3T3 mouse fibroblasts were plated to wells of a 96-well tissue culture plate (4 x 10 4 cells/well) in DMEM containing 10% FBS and 1% P/S (Invitrogen), and allowed to adhere overnight. Cells were washed with PBS and incubated in serum free media (SFM) containing the indicated concentrations of GzmB (Beryllium, Seattle, WA) ± GzmB specific inhibitor Compound 20 (50 µM) (Centre for Drug Research and Development, Vancouver, BC) (Willoughby et al. 2002). Cells were incubated at 37°C for 7 h, after which images of the cells in the plate were captured under phase contrast using an Olympus microscope camera. In some experiments, the supernatant was collected, centrifuged to remove cellular debris and western blot was performed for fibronectin as previously described. Cells plated in parallel wells were lysed and probed for GAPDH on western blot as a loading control to show equivalent cell plating. For quantification of remaining adherent cells, MTS assays were performed after washing once with PBS to remove non-adherent cells, and results expressed as a percentage of untreated control.

Decorin cleavage assay
Decorin cleavage assay was performed as previously described (Hiebert et al. 2011).

Collagen degradation assay
Formation of collagen fibrils and decorin coating was based upon a previously described protocol, with modifications (Geng et al. 2006). For each sample, Type 1 collagen fibrils were formed in eppendorf tubes that had been blocked with 1% BSA and thoroughly washed. Briefly, 3.5 µL PureCol Collagen (Bovine Collagen, 3 mg/mL, Advanced BioMatrix, San Diego, CA) was mixed with 21.5 µL 10 mM HCl and 25 µL 50 mM Tris buffer (pH 7.5) and incubated for 2 h at 37°C, after which fibrils were recovered as a pellet by centrifugation. To coat the fibrils with decorin, fibril pellets were re-suspended in 33 µL Tris buffer, 15 µL recombinant decorin (3 µg, R&D Systems) and 2 µL 10 mM ZnSO 4 to facilitate binding. After incubating overnight at 37°C, the fibrils were recovered by centrifugation and then re-suspended in Tris buffer ± 100 nM GzmB ± 50 µM Compound 20 for 8 h. Fibrils were again recovered by centrifugation. MMP-1 proenzyme (100 ng/sample) (Calbiochem, Millipore, San Deigo, CA) was activated by trypsin (50 nM) in MMP buffer (50 mM Tris, 0.5 M NaCl, 5 mM CaCl 2 , pH 7.5) for l hour, after which trypsin activity was inhibited with soybean trypsin inhibitor. Fibrils were either re-suspended with activated MMP-1 or MMP buffer alone for 16 h at 37 °C. Samples were separated on 10 % SDS-PAGE gels and proteins were visualized in the gel with SimplyBlue Safe Stain (Invitrogen).
Intact collagen was quantified via Image J.
Bernstein EF, Underhill CB, Hahn PJ, Brown DB, Uitto J (1996) Chronic sun exposure alters both the content and distribution of dermal glycosaminoglycans. Br J Dermatol. 135,