Abstract
Chronic repetitive forces on the spinal column promote the development of degenerative spinal disease. Yet the mechanisms linking such macroscale mechanical forces to tissue hypertrophy remain unknown. Here we show that fibrotic regions in human ligamentum flavum naturally exposed to high stress display elevated Rho-associated kinase (ROCK) signalling and an increased density of myofibroblasts expressing smooth muscle actin α. The myofibroblasts were localized in regions of elevated stiffness and microstress, such accumulation was ROCK dependent, and ROCK inhibition partially reduced the stress-driven transcriptional responses. Our findings support the further investigation of ROCK inhibitors for the treatment of degenerative spinal disease.
This is a preview of subscription content, access via your institution
Access options
Access Nature and 54 other Nature Portfolio journals
Get Nature+, our best-value online-access subscription
27,99 € / 30 days
cancel any time
Subscribe to this journal
Receive 12 digital issues and online access to articles
118,99 € per year
only 9,92 € per issue
Buy this article
- Purchase on SpringerLink
- Instant access to full article PDF
Prices may be subject to local taxes which are calculated during checkout
Similar content being viewed by others
Data availability
The data supporting the results in this study are available within the paper and its Supplementary Information. De-identified patient data may be made available on request from the corresponding author, subject to approval from the Institutional Review Board of Massachusetts General Hospital. Source data for the figures are provided with this paper.
Code availability
Code used to implement finite-element models of lumbar spine/pelvis is available on request. Details of the implemented model have been previously published in ref. 45. The MATLAB code used to analyse the AFM data is also available on request. The code implemented standard statistical methods, as described in Methods, and the R code used to analyse gene expression is also available on request. The code implemented standard methods using open-source packages, as described in Methods.
References
Kushchayev, S. V. et al. ABCs of the degenerative spine. Insights Imaging 9, 253–274 (2018).
Tschumperlin, D. J., Ligresti, G., Hilscher, M. B. & Shah, V. H. Mechanosensing and fibrosis. J. Clin. Invest. 128, 74–84 (2018).
Ravindra, V. M. et al. Degenerative lumbar spine disease: estimating global incidence and worldwide volume. Glob. Spine J. 8, 784–794 (2018).
Austevoll, I. M. et al. Decompression with or without fusion in degenerative lumbar spondylolisthesis. N. Engl. J. Med. 385, 526–538 (2021).
Yoshiiwa, T. et al. Analysis of the relationship between ligamentum flavum thickening and lumbar segmental instability, disc degeneration, and facet joint osteoarthritis in lumbar spinal stenosis. Asian Spine J. 10, 1132–1140 (2016).
Kim, Y. U. et al. The role of the ligamentum flavum area as a morphological parameter of lumbar central spinal stenosis. Pain Physician 20, E419–E424 (2017).
Benzel, E. C. & American Association of Neurological Surgeons. Biomechanics of Spine Stabilization (Thieme, 2001).
Jang, S. Y. et al. Radiographic parameters of segmental instability in lumbar spine using kinetic MRI. J. Korean Neurosurg. Soc. 45, 24–31 (2009).
Lee, C. K. Accelerated degeneration of the segment adjacent to a lumbar fusion. Spine 13, 375–377 (1988).
Huang, Y. P. et al. Preserving posterior complex can prevent adjacent segment disease following posterior lumbar interbody fusion surgeries: a finite element analysis. PLoS ONE 11, e0166452 (2016).
Hur, J. W. et al. Myofibroblast in the ligamentum flavum hypertrophic activity. Eur. Spine J. 26, 2021–2030 (2017).
Hayashi, F. et al. Myofibroblasts are increased in the dorsal layer of the hypertrophic ligamentum flavum in lumbar spinal canal stenosis. Spine J. 22, 697–704 (2022).
Kendall, R. T. & Feghali-Bostwick, C. A. Fibroblasts in fibrosis: novel roles and mediators. Front. Pharm. 5, 123 (2014).
Lederer, D. J. & Martinez, F. J. Idiopathic pulmonary fibrosis. N. Engl. J. Med. 378, 1811–1823 (2018).
Plikus, M. V. et al. Fibroblasts: origins, definitions, and functions in health and disease. Cell 184, 3852–3872 (2021).
Zent, J. & Guo, L. W. Signaling mechanisms of myofibroblastic activation: outside-in and inside-out. Cell. Physiol. Biochem. 49, 848–868 (2018).
Gibb, A. A., Lazaropoulos, M. P. & Elrod, J. W. Myofibroblasts and fibrosis: mitochondrial and metabolic control of cellular differentiation. Circ. Res. 127, 427–447 (2020).
Klingberg, F., Hinz, B. & White, E. S. The myofibroblast matrix: implications for tissue repair and fibrosis. J. Pathol. 229, 298–309 (2013).
Frangogiannis, N. Transforming growth factor-β in tissue fibrosis. J. Exp. Med. 217, e20190103 (2020).
Shi, J. et al. Distinct roles for ROCK1 and ROCK2 in the regulation of cell detachment. Cell Death Dis. 4, e483 (2013).
Knipe, R. S., Tager, A. M. & Liao, J. K. The Rho kinases: critical mediators of multiple profibrotic processes and rational targets for new therapies for pulmonary fibrosis. Pharm. Rev. 67, 103–117 (2015).
Rieder, F. ROCKing the field of intestinal fibrosis or between a ROCK and a hard place? Gastroenterology 153, 895–897 (2017).
Gomes, R. N., Manuel, F. & Nascimento, D. S. The bright side of fibroblasts: molecular signature and regenerative cues in major organs. npj Regen. Med. 6, 43 (2021).
Martinez-Vidal, L. et al. Causal contributors to tissue stiffness and clinical relevance in urology. Commun. Biol. 4, 1011 (2021).
Nia, H. T. et al. Solid stress and elastic energy as measures of tumour mechanopathology. Nat. Biomed. Eng. 1, 0004 (2016).
Hilibrand, A. S. & Robbins, M. Adjacent segment degeneration and adjacent segment disease: the consequences of spinal fusion? Spine J. 4, 190s–194s (2004).
Park, P., Garton, H. J., Gala, V. C., Hoff, J. T. & McGillicuddy, J. E. Adjacent segment disease after lumbar or lumbosacral fusion: review of the literature. Spine 29, 1938–1944 (2004).
Sun, C., Zhang, H., Wang, X. & Liu, X. Ligamentum flavum fibrosis and hypertrophy: molecular pathways, cellular mechanisms, and future directions. FASEB J. 34, 9854–9868 (2020).
Malakoutian, M. et al. Do in vivo kinematic studies provide insight into adjacent segment degeneration? A qualitative systematic literature review. Eur. Spine J. 24, 1865–1881 (2015).
McMains, M. C. et al. A biomechanical analysis of lateral interbody construct and supplemental fixation in adjacent-segment disease of the lumbar spine. World Neurosurg. 128, e694–e699 (2019).
Lindsey, D. P., Kiapour, A., Yerby, S. A. & Goel, V. K. Sacroiliac joint fusion minimally affects adjacent lumbar segment motion: a finite element study. Int. J. Spine Surg. 9, 64 (2015).
Srinivas, G. R., Kumar, M. N. & Deb, A. Adjacent disc stress following floating lumbar spine fusion: a finite element study. Asian Spine J. 11, 538–547 (2017).
Ohtori, S. et al. Change of lumbar ligamentum flavum after indirect decompression using anterior lumbar interbody fusion. Asian Spine J. 11, 105–112 (2017).
Limthongkul, W. et al. Indirect decompression effect to central canal and ligamentum flavum after extreme lateral lumbar interbody fusion and oblique lumbar interbody fusion. Spine 45, E1077–e1084 (2020).
Wang, T. et al. Cyclic mechanical stimulation rescues Achilles tendon from degeneration in a bioreactor system. J. Orthop. Res. 33, 1888–1896 (2015).
Hayashi, K. et al. Fibroblast growth factor 9 is upregulated upon intervertebral mechanical stress-induced ligamentum flavum hypertrophy in a rabbit model. Spine 44, E1172–E1180 (2019).
Hayashi, K. et al. Mechanical stress induces elastic fibre disruption and cartilage matrix increase in ligamentum flavum. Sci. Rep. 7, 13092 (2017).
Joannes, A. et al. FGF9 and FGF18 in idiopathic pulmonary fibrosis promote survival and migration and inhibit myofibroblast differentiation of human lung fibroblasts in vitro. Am. J. Physiol. Lung Cell. Mol. Physiol. 310, L615–L629 (2016).
Mead, T. J. ADAMTS6: emerging roles in cardiovascular, musculoskeletal and cancer biology. Front. Mol. Biosci. 9, 1023511 (2022).
Hou, J. et al. TNF-α-induced NF-κB activation promotes myofibroblast differentiation of LR-MSCs and exacerbates bleomycin-induced pulmonary fibrosis. J. Cell. Physiol. 233, 2409–2419 (2018).
Weiss, J. A., Gardiner, J. C. & Bonifasi-Lista, C. Ligament material behavior is nonlinear, viscoelastic and rate-independent under shear loading. J. Biomech. 35, 943–950 (2002).
Komeili, A., Rasoulian, A., Moghaddam, F., El-Rich, M. & Li, L. P. The importance of intervertebral disc material model on the prediction of mechanical function of the cervical spine. BMC Musculoskelet. Disord. 22, 324 (2021).
Kiapour, A. et al. Biomechanical analysis of stand-alone lumbar interbody cages versus 360° constructs: an in vitro and finite element investigation. J. Neurosurg. Spine 36, 928–936 (2021).
Joukar, A. et al. Biomechanics of the sacroiliac joint: surgical treatments. Int. J. Spine Surg. 14, 355–367 (2020).
Kiapour, A. et al. Biomechanical analysis of stand-alone lumbar interbody cages versus 360° constructs: an in vitro and finite element investigation. J. Neurosurg. Spine 36, 928–936 (2022).
Mihara, A. et al. Tensile test of human lumbar ligamentum flavum: age-related changes of stiffness. Appl. Sci. 11, 3337 (2021).
Kirby, M. C., Sikoryn, T. A., Hukins, D. W. & Aspden, R. M. Structure and mechanical properties of the longitudinal ligaments and ligamentum flavum of the spine. J. Biomed. Eng. 11, 192–196 (1989).
Patwardhan, A. G., Havey, R. M., Meade, K. P., Lee, B. & Dunlap, B. A follower load increases the load-carrying capacity of the lumbar spine in compression. Spine 24, 1003–1009 (1999).
Remus, R., Selkmann, S., Lipphaus, A., Neumann, M. & Bender, B. Muscle-driven forward dynamic active hybrid model of the lumbosacral spine: combined FEM and multibody simulation. Front. Bioeng. Biotechnol. 11, 1223007 (2023).
Dobin, A. et al. STAR: ultrafast universal RNA-seq aligner. Bioinformatics 29, 15–21 (2013).
Frankish, A. et al. GENCODE reference annotation for the human and mouse genomes. Nucleic Acids Res. 47, D766–D773 (2019).
Love, M. I., Huber, W. & Anders, S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 15, 550 (2014).
Zhou, Y. et al. Metascape provides a biologist-oriented resource for the analysis of systems-level datasets. Nat. Commun. 10, 1523 (2019).
Dimitriadis, E. K., Horkay, F., Maresca, J., Kachar, B. & Chadwick, R. S. Determination of elastic moduli of thin layers of soft material using the atomic force microscope. Biophys. J. 82, 2798–2810 (2002).
Acknowledgements
We acknowledge the master craftsmanship of J. (R.) McConnell and the MGH Biomedical Engineering Model Shop for assistance in fabrication of the loading device used in this study. All images were created by authors of this manuscript. We thank D. Glazer for critical review of the manuscript. North America Spine Society (NASS) Basic Science Research Grant provided funding for the study. M.H. received salary support from an NIH R25 Research Grant.
Author information
Authors and Affiliations
Contributions
G.M.S., M.H., M.S.S. and J.H.S. conceptualized the project. M.S.S., M.H., L.R. and B.D.C. performed wet lab investigation. G.P.N., J.-V.C.C., J.H.S., G.M.S., M.S.S. and M.H. performed patient sample collection. E.M. conducted RNA-seq analysis. A.K. performed finite-element modelling. G.N. and M.A.S. conducted clinical data collection. J.B., I.D.C., E.E., R.B., S.S. and B.D.C. provided technical support. G.M.S., M.H. and M.S.S. acquired funding. G.M.S., A.J.G., H.T.N., L.F.B., J.H.S. and B.D.C. supervised the project. M.S.S. and M.H. analysed data. M.S.S. and M.H. wrote the original manuscript draft. G.M.S., A.J.G., H.T.N. and L.F.B. reviewed and edited the manuscript.
Corresponding author
Ethics declarations
Competing interests
A provisional patent application related to this work (63/322,621; G.M.S. and M.H.) was filed on 22 March 2022. The other authors declare no competing interests.
Peer review
Peer review information
Nature Biomedical Engineering thanks Mazda Farshad, Chiseung Lee, Christopher McCulloch and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Peer reviewer reports are available.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Extended data
Extended Data Fig. 1 Patterns of motion and stress on LF remain consistent across multiple FEMs derived from three additional patients (one female and two male) when compared to the previously analyzed model.
a, Hybrid loading ranging from 0 to 30 degrees of flexion from L1 to the sacrum was applied to all four models. Resulting L3/4 segmental motion and b, L3/4 LF stress were quantified and compared. c, Representative error (in standard deviation) of L3/4 segmental motion was calculated and plotted. d, Representative error (in standard deviation) of L3/4 LF stress was calculated and plotted. e, Data from the four models derived from the four different patients was combined on singular plots to demonstrate the precision of the calculations of L3/4 segmental motion and of f, L3/4 LF stress despite the varying geometries.
Extended Data Fig. 2 Ligamentum flavum (LF) measurement on MRI.
On a patient’s T2 weighted spine MRI, the desired vertebral level was identified on the sagittal scan (A) and used to identify the ligamentum flavum at the desired level on the axial scan (B). The axial slice of the MRI with the thickest section of the ligamentum flavum was identified. Using Visage, the ligamentum flavum was circled (B, yellow outline), the area was calculated and recorded, and a screenshot was saved for future reference.
Extended Data Fig. 3 Globally, there is progressive, stepwise loss of elastin from non-DSD to DSD to ASD and locally, thorough examination of tissue reveals small regions of torn elastin and loosened elastin present in all clinical conditions.
A, LF from non-DSD, DSD, and ASD patients was stained for elastin by EVG. B, Percent elastin was quantified by using a standard color thresholding protocol for all images; elastin was significantly reduced from non-DSD (n = 3) levels in both DSD (n = 3, p = 0.020) and ASD (n = 3, p = 0.008) conditions. C, Representative images (elastin stain, top; schematic, bottom) demonstrating normal matrix (i), elastin tears (ii), and loosened matrix (iii), which were seen across non-DSD, DSD, and ASD conditions. Quantitative data are shown as mean ± sem, all tests are two-sided, unpaired t-tests.
Extended Data Fig. 4 By trichrome staining, ratio of blue to red increases stepwise from non-DSD to DSD to ASD.
A, Representative images of LF from non-DSD, DSD, and ASD patients stained with trichrome staining. B, Representative images of the color thresholding protocol for quantification. Percent blue and red was quantified using a standard color thresholding protocol for all images; red pixels matching a set range of values were selected (top, selected pixels highlighted red). The remainder of pixels were considered blue and the images were converted to binary images for quantification (bottom, white=red and black=blue). C, The ratio of blue to red significantly increases from non-DSD (n = 2) to DSD (n = 3, p = 0.022) and increases again from DSD to ASD (n = 3). Quantitative data are shown as mean ± sem, all tests are two-sided, unpaired t-tests.
Extended Data Fig. 5 CD45 cells decrease stepwise from non-DSD to DSD to ASD.
A, Representative images of immunohistochemistry stains for CD45 (brown) with nuclear counterstain (purple) in non-DSD, DSD, and ASD LF. B, Qualitative schematic demonstrating the observation that CD45 cells, when present, seemed to cluster together in regions of abnormal ECM. C, The most CD45 cells were seen in non-DSD samples (n = 3), fewer were seen in DSD samples (n = 3), and significantly fewest were seen in ASD samples (n = 3, p = 0.0254 from non-DSD, p = 0.0196 from DSD). Quantitative data are shown as mean ± sem, all tests are two-sided, unpaired t-tests.
Extended Data Fig. 6 qPCR and western blots of LF samples by disease group.
A, while overall qPCR analysis of LF from non-DSD, DSD, and ASD patients was highly variable, Fibronectin increased stepwise between non-DSD, DSD, and ASD groups. CYP also significantly decreased from non-DSD levels. B, As quantified by western blots on LF samples, similar levels of latent TGFβ were present in all disease groups but the amount of released, active TGFβ was highest in non-DSD samples and lowest in ASD samples. Quantitative data are shown as mean ± sem.
Extended Data Fig. 7 Effects of TGFβRI inhibitors, Rho/ROCK inhibitors, and stretch on primary LF cells.
A, Incubation with TGFβ markedly increases SMAD2 nuclear localization, but this effect is abolished by adding the TGFβ inhibitor SB431542 (n = 4). B, in primary LF cells growing on collagen coated FlexCell plates without stretch, incubation for 7 hours with Ripasudil reduced the number of cells with assembled SMAα stress actin fibers. The number of cells positive for SMAα increased after 6 hours of stretch, but this effect was protected against by Ripasudil (n = 3). Quantitative data are shown as mean ± sem.
Extended Data Fig. 8 FEM functions as both a descriptive and predictive tool: Values of LF stress calculated using FEM for clinical conditions associated with higher SMAα cell densities predicted optimal stresses at which to apply cyclic strain to LF fascicles in the bioreactor to increase SMAα cells.
A, The range of LF degrees of hypertrophy seen among all patients, as measured by LF area on axial cross section on MRI. B, FEMs with predicted stress at maximum spine flexion: 50 kPa when fused across the LF, 206 kPa with no fusion, 271 kPa when fused a vertebral level below the LF. C, LF areas plotted in (A), color-coded in accordance with FEM-predicted force experienced by the measured LF. D, Fold increase in SMAα cells compared to paired, unpulled fascicles (n = 24 pairs), binned by stress in kPa. Below, FEM models are aligned by maximum stress experienced by LF at maximum spine flexion.
Extended Data Fig. 9 Progressively increased doses of rock inhibitors causes progressingly increased cytoskeletal actin disassembly in LF primary cells; at low concentrations, ripasudil rock inhibition uncouples the LF cell’s sensing and alignment in the direction of experienced stretch and results in fewer SMAα cells in conditions with and without 2-dimensional stretch.
A, LF primary cells treated with increasing doses of rock inhibitors stained for actin (top) or SMAα (bottom). Insets contain schematics emphasizing morphology observed with each level of treatment: full-length fibers in untreated cells (i,iv) dissolve into fragments (ii), then into sparse granules on the periphery of the cell cytoplasm (iii,v). B, LF primary cells were assigned conditions of ‘no stretch’ or ‘stretch’ and then treated with vehicle control or ripasudil. After stretch regimen, cells were stained for actin and SMAα (i-iv; insets contain representative schematics of the actin fibers directions of cells) and the percentage of SMAα cells was quantified (v). Percent SMAα cells increased with stretch, and decreased with ripasudil treatment, in both stretched and unstretched conditions. C, Immunofluorescence staining for SMAα and DAPI in paired, unstretched and stretched LF fascicles, with quantification of n = 12 triply paired fascicles. Images and graph in c have been reproduced from main manuscript Figs. 4g and 5f, respectively, for efficient comparison to results in b,i-v. Quantitative data are shown as mean ± sem.
Extended Data Fig. 10 LF fascicle Young’s modulus significantly decreases after 24 hour incubation with Ripasudil, and decreases significantly more than when incubated with DMSO.
A, Timeline describing experimental protocol for LF collection from the operating room through beginning incubation with DMSO or ripasudil, referred to in graphs as ‘0 h’, and competition of the incubation period, referred to in graphs as ‘24 h’. B,C, Schematics demonstrating the hypothesized components resisting tension and thus contributing to LF fascicle bulk modulus - ECM fibers and cellular cytoskeletal tension, the latter of which would be disrupted with Ripasudil treatment. D, In 6 sets of quadruply-paired fascicles from 5 patients, the Young’s modulus of fascicles treated with Ripasudil decreased significantly (p = 0.0012) while the Young’s modulus of fascicles treated with DMSO decreased non-significantly (p = 0.0523). E, Fascicles treated with Ripasudil had a significantly greater negative change in modulus (YMend - YMstart) than fascicles treated with DMSO (p = 0.0467). Tests are paired, one-tailed t-tests.
Supplementary information
Supplementary Information
Supplementary figures and tables.
Data
Source data for the supplementary figures.
Data
Uncropped western blots for Supplementary Fig. 4.
Source data
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Hadzipasic, M., Sten, M.S., Massaad, E. et al. ROCK-dependent mechanotransduction of macroscale forces drives fibrosis in degenerative spinal disease. Nat. Biomed. Eng (2025). https://doi.org/10.1038/s41551-025-01396-7
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41551-025-01396-7
