Atte S. A. Eskelinen ,Petri Tanska,Cristina Florea, Gustavo A. Orozco, Petro Julkunen, Alan J. Grodzinsky, Rami K. Korhonen
Abstract
Post-traumatic osteoarthritis (PTOA) is associated with cartilage degradation, ultimately leading to disability and decrease of quality of life. Two key mechanisms have been suggested to occur in PTOA: tissue inflammation and abnormal biomechanical loading. Both mechanisms have been suggested to result in loss of cartilage proteoglycans, the source of tissue fixed charge density (FCD). In order to predict the simultaneous effect of these degrading mechanisms on FCD content, a computational model has been developed. We simulated spatial and temporal changes of FCD content in injured cartilage using a novel finite element model that incorporates (1) diffusion of the pro-inflammatory cytokine interleukin-1 into tissue, and (2) the effect of excessive levels of shear strain near chondral defects during physiologically relevant loading. Cytokine-induced biochemical cartilage explant degradation occurs near the sides, top, and lesion, consistent with the literature. In turn, biomechanically-driven FCD loss is predicted near the lesion, in accordance with experimental findings: regions near lesions showed significantly more FCD depletion compared to regions away from lesions (p<0.01). Combined biochemical and biomechanical degradation is found near the free surfaces and especially near the lesion, and the corresponding bulk FCD loss agrees with experiments. We suggest that the presence of lesions plays a role in cytokine diffusion-driven degradation, and also predisposes cartilage for further biomechanical degradation. Models considering both these cartilage degradation pathways concomitantly are promising in silico tools for predicting disease progression, recognizing lesions at high risk, simulating treatments, and ultimately optimizing treatments to postpone the development of PTOA.
Introduction
Acute joint insult can debilitate the functioning of articular cartilage and lead to pain, joint stiffness and disability in patients. Moreover, trauma such as anterior cruciate ligament (ACL) rupture [1,2] may be associated with cartilage injury and increased susceptibility to cartilage degeneration which can culminate in post-traumatic osteoarthritis (PTOA) [3–5]. Two major mechanisms have been suggested to play a role in PTOA progression. The first mechanism involves joint inflammation [3,6,7] and the resulting diffusion of pro-inflammatory cytokines from the synovial fluid into cartilage [8–11], leading to proteolysis of cartilage matrix. The second mechanism involves biomechanical factors [12] including induction of chondral defects and aberrant loading characteristics in the knee joint milieu [13–15], leading to elevated shear strains near chondral lesions [16–18]. While these mechanisms have been recognized for years, current healthcare approaches lack effective tools to identify patients with increased risk of PTOA development. Prevention or even postponing the onset of the disease and surgical interventions would be desirable. Thus, we urgently need novel tools to predict disease progression.
Materials and methods
Experiments
Previous experimental research has assessed cytokine-mediated and biomechanically-driven cartilage degeneration [36,40,49,52,57,58]. These studies involved evaluation of (1) loss of sulphated glycosaminoglycans (GAGs) to culture medium via the dimethylmethylene blue (DMMB) assay, or (2) loss of FCD via decrease in optical density obtained with digital densitometry (DD) of Safranin-O stained sections [59,60]. Both methods are indicative of PG matrix damage. In the current study, we quantified average FCD loss near and further away from cartilage lesions based on the data of Orozco et al. [49] to (1) characterize how the presence of lesions affects early matrix damage and PG release, and (2) to compare biomechanically-induced FCD loss with our computational estimates. Due to lack of data in the literature, we did not focus on localized FCD loss in injured samples cultured with cytokines. See S1 Supplementary Material (Subsections S1.1–1.4) for additional details on the previously reported experimental data used for the present modeling analyses.
Discussion
Summary
In this study, we have developed a novel mechanobiological FE model capable to predict cartilage degradation via both biochemically and biomechanically-driven mechanisms. We provided spatio-temporal estimations of matrix damage, which are in accordance with our experimental findings (S1 Supplementary Material Subsection S2.1, S2 Fig) and the current literature reports. With the chosen concentration of pro-inflammatory cytokine IL-1 (1 ng/ml, representing moderate inflammation), the ECM was degraded more dramatically over larger areas (near free surfaces and below the lesion) compared to the effect of mechanical loading alone after injury. The shear strain-modulated biomechanical degeneration that occurred only near the lesion was rapid, especially during 1–3 days following injury.
Acknowledgments
The authors appreciate the support of the University of Eastern Finland and the Massachusetts Institute of Technology to conduct this study. CSC–IT center for Science Ltd., Finland, is acknowledged for providing the modeling software. Patrick Grahn, D.Sc. (Tech.), from COMSOL Multiphysics Support is acknowledged for technical support.
Citation: Eskelinen ASA, Tanska P, Florea C, Orozco GA, Julkunen P, Grodzinsky AJ, et al. (2020) Mechanobiological model for simulation of injured cartilage degradation via pro-inflammatory cytokines and mechanical stimulus. PLoS Comput Biol 16(6): e1007998. https://doi.org/10.1371/journal.pcbi.1007998
Editor: Alison Marsden, Stanford University, UNITED STATES
Received: February 3, 2020; Accepted: May 28, 2020; Published: June 25, 2020
Copyright: © 2020 Eskelinen et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All the underlying data are available in Fairdata Research Data Storage Service
(https://etsin.fairdata.fi/dataset/43c5e4f1-3c3e-4ea2-b63a-38aca480b77c). DOI: https://doi.org/10.23729/0fec1760-d320-44f3-baae-608aee509dc5
Funding: This work was supported by the Doctoral Programme in Science, Technology and Computing (SCITECO) of the University of Eastern Finland (http://www.uef.fi/en/web/dpsciteco; ASAE); the Finnish Cultural Foundation (https://skr.fi/en; grant number 00191044; PT); the Maire Lisko Foundation (PT); the Academy of Finland (https://www.aka.fi/en; grant numbers 286526 [RKK], 322423 [PJ], 324529 [RKK]); the Sigrid Jusélius Foundation (https://sigridjuselius.fi/en/; RKK); the Instrumentarium Science Foundation (ASAE); the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement nos 702586, 713645 (https://ec.europa.eu/programmes/horizon2020/en/h2020-section/marie-sklodowska-curie-actions; RKK). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
