Mast Cell/Proteinase Activated Receptor 2 (PAR2) Mediated Interactions in the Pathogenesis of Discogenic Back Pain

Abstract

Mast cells (MCs) are present in the painful degenerate human intervertebral disc (IVD) and are associated with disease pathogenesis. MCs release granules containing enzymatic and inflammatory factors in response to stimulants or allergens. The serine protease, tryptase, is unique to MCs and its activation of the G-protein coupled receptor, Protease Activated Receptor 2 (PAR2), induces inflammation and degradation in osteoarthritic cartilage. Our previously published work has demonstrated increased levels of MC marker tryptase in IVD samples from discogenic back pain patients compared to healthy control IVD samples including expression of chemotactic agents that may facilitate MC migration into the IVD. To further elucidate MCs' role in the IVD and mechanisms underlying its effects, we investigated whether (1) human IVD cells can promote MC migration, (2) MC tryptase can mediate up-regulation of inflammatory/catabolic process in human IVD cells and tissue, and (3) the potential of PAR2 antagonist to function as a therapeutic drug in in vitro human and ex vivo bovine pilot models of disease. MC migration was quantitatively assessed using conditioned media from primary human IVD cells and MC migration examined through Matrigel. Exposure to soluble IVD factors significantly enhanced MC migration, suggesting IVD cells can recruit MCs. We also demonstrated significant upregulation of MC chemokine SCF and angiogenic factor VEGFA gene expression in human IVD cells in vitro in response to recombinant human tryptase, suggesting tryptase can enhance recruitment of MCs and promotion of angiogenesis into the usually avascular IVD. Furthermore, tryptase can degrade proteoglycans in IVD tissue as demonstrated by significant increases in glycosaminoglycans released into surrounding media. This can create a catabolic microenvironment compromising structural integrity and facilitating vascular migration usually inhibited by the anti-angiogenic IVD matrix. Finally, as a "proof of concept" study, we examined the therapeutic potential of PAR2 antagonist (PAR2A) on human IVD cells and bovine organ culture IVD model. While preliminary data shows promise and points toward structural restoration of the bovine IVD including down-regulation of VEGFA, effects of PAR2 antagonist on human IVD cells differ between gender and donors suggesting that further validation is required with larger cohorts of human specimens.

Keywords: PAR2; discogenic back pain; intervertebral disc; mast cell; tryptase.

FIGURE 1

(A) Macroscopic IVD schematic with zoomed in proposed mechanistic models of (B) MC invasion into the intradiscal environment in acute onset and chronic state as visualized through (1) Acute injury or matrix perturbation provides physical access of MC progenitors into usually immune-privileged intra-discal microenvironment acutely and (2) in the chronic state access can occur through vascularization, (3) facilitated by the MC chemoattractant SCF (KITLG). (C) Biochemical pathways of immune modulation in DBP as follows; (4) MCs in the IVD respond to mechanical load and nutrient deprivation in the harsh microenvironment by releasing tryptase (Wiet et al., 2017¹; Zhang et al., 2012²; Fowlkes et al., 2013³). Tryptase can activate (5) PAR2 directly (Molino et al., 1997⁴) as well as (5’) matrix enzymes, matrix metalloproteinases 3 and 13 (MMPs) (Magarinos et al., 2013⁵). (6) PAR2 activation by tryptase, abated by PAR2A (Chen et al., 2011⁶) in the present study, leads to downstream factors related to catabolism, (6’) such as MMPs (Lee et al., 2010⁸), which can degrade the proteoglycan aggrecan (Magarinos et al., 2013⁵) creating an increasingly permissive environment for inflammation and stimulating increased activation of endogenous pro-MMP (Vo et al., 2013⁹). (7) PAR2’s activation also leads to downstream factors related to inflammation, neo-vascularization/neo-innervation, and MC chemotaxis contributing to a synergistic positive feedback loop, whereas PAR2’s activation upregulates its own expression through IL-1β, ad general inflammation, perpetuating a chronic degenerate state (Ritchie et al., 2007⁷). Black Arrows represent labeling or directionality, Blue arrows represent forward mechanism steps, Green arrows represent feedback interactions. (C) Blunt arrow ends represent effects coming from, pointed ends represent effects going to, and bidirectional pointed ends represent co-regulatory interactions.

FIGURE 2

(A) Representative images of (calcein) stained fluorescent MCs migrating in response to 0% FBS BMC (negative control), 0% FBS ACCM (cell control), 0% FBS IVDCM, and 10% FBS BMC (positive control). (B) Mast cell migration in response to IVDCM (N = 3), ACCM (N = 3) and 0% FBS (negative control) (N = 6) normalized to 10% (positive control = 100% migrated cells) (N = 6; ^∗p ≤ 0.05). Red scale bar = 500 μm (4×).

FIGURE 3

(A) Average fold changes in qRT-PCR gene expression in human NP cells (N = 4) in 3D agarose constructs treated with rhTryptase and normalized to untreated controls (^∗p < 0.05). (B) GAG release into media from freeze-thawed human NP explants from cadaveric (autopsy) IVD specimens (N = 8; ^∗p < 0.05).

FIGURE 4

(A) Representative images of immunohistochemical (IHC) PAR2 staining in IVD tissue from all regions; NP, AF and CEP including relevant positive and negative lung tissue controls. (B) PAR2 expression presented as % of cells expressing PAR2 as assessed by IHC and confirmed with western blot using the same PAR2 primary antibodies. Red scale bar = 100 μm, Black scale bar = 50 μm. Arrows point to same cells at both magnifications.

FIGURE 5

(A) Representative Alcian blue (aggrecan) and picrosirius red (collagen) staining of bovine caudal IVDs after 96 h of culture. Discs were either left intact (control) or injured and cultured with MCCM with or with a PAR2A. (B) Representative cell viability image for Injury groups treated with MCCM+PAR2A. (C) Average fold change in qRT-PCR gene expression in AF region of bovine ex vivo samples (^∗p < 0.05). Red scale bar = 1000 μm, Black scale bar = 50 μm.

FIGURE 6

(A) Histogram of average fold change in qRT-PCR gene expression in human NP cells (N = 7) treated with MCCM and MCCM+PAR2A normalized to basal untreated controls. (B) Histograms of fold changes in qRT-PCR gene expression in human NP cells for individual donors normalized to basal control where left (M1, M2, M3) are males and right (F1, F2, F3, F4) are females.