Dynamic Load Model Systems of Tendon Inflammation and Mechanobiology

Abstract

Dynamic loading is a shared feature of tendon tissue homeostasis and pathology. Tendon cells have the inherent ability to sense mechanical loads that initiate molecular-level mechanotransduction pathways. While mature tendons require physiological mechanical loading in order to maintain and fine tune their extracellular matrix architecture, pathological loading initiates an inflammatory-mediated tissue repair pathway that may ultimately result in extracellular matrix dysregulation and tendon degeneration. The exact loading and inflammatory mechanisms involved in tendon healing and pathology is unclear although a precise understanding is imperative to improving therapeutic outcomes of tendon pathologies. Thus, various model systems have been designed to help elucidate the underlying mechanisms of tendon mechanobiology via mimicry of the in vivo tendon architecture and biomechanics. Recent development of model systems has focused on identifying mechanoresponses to various mechanical loading platforms. Less effort has been placed on identifying inflammatory pathways involved in tendon pathology etiology, though inflammation has been implicated in the onset of such chronic injuries. The focus of this work is to highlight the latest discoveries in tendon mechanobiology platforms and specifically identify the gaps for future work. An interdisciplinary approach is necessary to reveal the complex molecular interplay that leads to tendon pathologies and will ultimately identify potential regenerative therapeutic targets.

Keywords: dynamic loading; extracellular matix; inflammation; mechanotransduction; tendinopathy; tendon; tendon pathology.

FIGURE 1

Features of the healthy and pathological tendon fascicle. Created with Biorender.com. Features of the clinically pathological tendon fascicle (a hierarchical subunit of whole tendon) include lipid deposition, angiogenesis, nerve ingrowth, adipocyte and chondrocyte-like TPC differentiation (Kannus and Józsa, 1991; Józsa and Kannus, 1997; Zhang and Wang, 2010, 2014; Agarwal et al., 2017), increases in proteoglycan content (Fu, Chan and Rolf, 2007; Samiric et al., 2009; Attia et al., 2014), collagen type III RNA upregulation/deposition (Liu et al., 1995; Samiric et al., 2009), and irregularities in collagen alignment/crosslinking at previous microscopic to macroscopic rupture sites (Kannus and Józsa, 1991; Józsa and Kannus, 1997; Järvinen et al., 2004). Overall, the pathological tendon is mechanically weaker than the healthy tendon. TPC, tendon progenitor cell; Scx, Scleraxis; Tnmd, Tenomodulin; Col 1, collagen type I; Col 3, collagen type III; MMP, matrix metalloproteinases; ADAMT, disintegrin and metalloproteinase with thrombospondin motifs; IL-1β, interleukin-1β; TGF-β, transforming growth factor β.

FIGURE 2

Schematic of proposed tenocyte molecular response to physiological and hyper-physiological mechanical loading thresholds. Created with Biorender.com. (A) Molecular cascade following physiological loading resulting in both integrin-mediated pathways and postulated stretch-activated ion channel pathways to induce a TGF-β-/Smad-2/3 activation and thus transcription of various extracellular matrix related genes and regulatory enzymes. (B) Hyper-physiological loading leads to macro-scale ruptures that induce inflammatory cascades through greater activation of TGF-β pathways and IL-1β that lead to paracrine signaling and ultimately matrix degradation and inflammatory cell localization at the injury site. Scx, Scleraxis; Tnmd, Tenomodulin; Col 1, collagen type I; Col 3, collagen type III; MMP, matrix metalloproteinases; ADAMT, disintegrin and metalloproteinase with thrombospondin motifs; IL-1β, interleukin-1β; TGF-β, transforming growth factor β.

FIGURE 3

Tendon healing estimated timeline and features associated with a healthy or pathological state following injury. Created with Biorender.com. Initial injury due to overloading induces an inflammatory state that is characterized by matrix degradation, cell apoptosis, and immune cell localization (monocytes) that transition to a pro-inflammatory (M1) macrophage for removal of cell and ECM debris. The proliferative or repair phase (estimated 1–6 weeks) is characterized by injury site repair via increased cellularity, growth factors, inflammatory cytokines, and eventually the inflammation resolving M2 macrophage transition. The final remodeling stage can take months to years, where pathological features may persist with sustained inflammatory cytokines (though not immune cells) and degeneration of the ECM. In either the healthy or pathological case, a persistent scar-like tissue remains present (Fu et al., 2010; Lipman et al., 2018; Alim et al., 2020; Gracey et al., 2020). MMPs, matrix metalloproteinases; VEGF, vascular endothelial growth factor; TGF-β, transforming growth factor β; IL-1β, interleukin-1β; TNF-α, tumor necrosis factor α.

FIGURE 4

Representative estimations of tendon loading schemes. Created with Biorender.com. (A) Pull-to-failure test results in a stress-strain curve when normalized to tendon cross-sectional area (Wang, 2006; Wang et al., 2012). The toe region is due to natural collagen fiber crimping, the linear region is equated to physiological stress values, and the deformation region is where macro-scale ruptures start to occur (Louis-Ugbo et al., 2004). (B) Fatigue test indicates similar regions that occur due to repetitive physiological loading, where “creep” occurs due to macro-scale ruptures eventually leading to complete rupture of the tendon after a number of cycles (usually less than 50) (Thornton et al., 2002).

FIGURE 5

Bioreactor schematics for in vitro and ex vivo systems with static or dynamic loading and high-throughput options. Created with Biorender.com. (A) Example ex vivo systems are commonly developed in-house with whole tendon clamping systems that can be uniaxially stretched under set strains or loads (Wang et al., 2015; Tohidnezhad et al., 2020; Pedaprolu and Szczesny, 2021). (B) Tenocyte cell culture monolayers (2D in vitro systems) under static load, or no load, and dynamic loads have the greatest high-throughput capabilities. Dynamic loads on flexible membranes may be produced by mechanical actuators or with vacuum pressure, as illustrated and modified from Flexcell International’s Tension System, Burlington, NC (Arnoczky et al., 2002; Fleischhacker et al., 2020; Gaut et al., 2020; Kubo et al., 2020). (C) Cell-seeded 3D constructs can be gel-like, as shown in the side view (modified from Flexcell’s 3D Tissue Train System), or like whole-tissue if decellularized constructs are used. These 3D systems are generally bound by the dimensions of the material or well in which they are contained, yet there is flexibility in load types (tensile, shear, etc.) that can be analyzed (Zhang et al., 2015; Patel et al., 2017; Wang and Thien, 2018; Sawadkar et al., 2020; Pentzold and Wildemann, 2022).