(D-138) Axonal damage is induced after very slow manual tensile loading of neuron collagen constructs that simulate the spinal facet capsule ligament

Friday, October 13, 2023

3:30 PM – 4:30 PM PDT

Location: Exhibit Hall - Row D - Poster # 138

Presenting Author(s)

EN

Ella M. Nevo (she/her/hers)

undergraduate researcher
University of Pennsylvania
Philadelphia, Pennsylvania, United States

Co-Author(s)

CW

Chang Wang

graduate student
University of Pennsylvania, United States
PG

Prabesh Ghimire

Research Specialist
University of Pennsylvania, United States

Last Author(s)

BW

Beth Winkelstein, PhD

Professor
University of Pennsylvania
Philadelphia, Pennsylvania, United States

Introduction:: Neck pain is a leading cause of disability [1]. The spinal facet capsular ligament (FCL) has been shown to have the potential to generate pain since it has neurons embedded in its collagen network and can undergo loading and motions that can exceed its normal physiological range either during trauma or with slower onset joint or spinal degeneration [2-4]. Ligaments, including the FCL, are known to exhibit rate-dependent mechanical responses [5,6]. Similarly, neurons have varied responses to stretch depending on the rate it is applied, with slow stretch resulting in growth [7], and faster stretch inducing injury and reduced viability [8]. Our lab has developed a neuronal collagen construct (NCC) system serving as an in vitro model of the FCL to investigate relationships between biomechanics and neuronal responses; although both fast tensile loading (120%/s) and slower strain rates (1%/s, 7%/s) have been imposed [9,10], the effect of slow tensile stretch to failure on neuronal responses has not been investigated. This study used that NCC system to investigate failure responses, local strains and neuronal injury in a custom-built manual stretch device.

Materials and Methods:: NCCs (n=8) were fabricated from rat tail collagen I (2mg/mL) with dorsal root ganglia harvested from E18 Sprague-Dawley rats [9,10]. NCCs were cut into strips (21mm x 8mm) and labeled with a grid (Figure 1). Gels were attached via grips to a stretching device equipped with two manual linear stages submerged in a 1X PBS bath (Figure 1). NCCs were stretched until failure incrementally (5 micrometer turns per increment). At each increment, gels were imaged to capture marker and stage locations (Figure 1). A subset of stretched NCCs (n=3) and an unloaded control gel (n=1) were placed in neurobasal media for 24hrs before being fixed for immunolabeling to evaluate axon health and identify nuclei. Gels were blocked and incubated in mouse anti-β-III tubulin (1:200) and goat anti-mouse 488 (1:500) and DAPI (1:1000).

At the unloaded reference position and each stretch increment, the position of the mobile stage was calculated and marker locations were digitized. Maximum principal strains (MPS) on the gel were estimated using LS-DYNA [9-11]. Average gel distraction and MPS were calculated across all gels for each increment up to the last increment before failure was observed, which was taken as the failure image. Axon morphology was evaluated by two blinded graders according to: no damage (0), mild damage (1), or moderate damage (2) [12,13] (Figure 2). For gels with more than one ratable image in each element (n=3; 1 control), the average axon rating and strain at failure were plotted against each other and fit by linear regression.

Results, Conclusions, and Discussions:: NCCs were manually loaded at 0.105±0.004mm/s, corresponding to a strain rate of 0.2±0.023%/s. Gels (n=8) failed at 8.80±1.78mm and sustained average MPS of 22.1±5.4% at failure. For the subset of gels in which neuron health was assessed (n=4, from 34 images), the majority of images in stretched gels had an average axonal rating of 1 or higher (27 images) and only a few images had an average rating of 0.5 (n=6 images) or 0 (n=1 image). Overall, failed gels exhibited an average rating of 1.104±0.415 and the control gel exhibited an average rating of 0.833±0.516. Although the axon rating was higher after stretch than control, that difference is not significant. Axonal rating and NCC MPS at failure are positively correlated with a strong linear fit (R²=0.914) (Figure 2).

Although stretch was applied incrementally in this study, the displacements at failure are consistent with those reported in previous studies at higher strain rates of 1%/s (~4.5-8.3mm) and an even faster rate of 127%/s (8.7mm) [9,11]. In those studies, both force and mean MPS at failure vary with loading rate [9,11]. Since the manual device does not measure force, additional studies should be performed at this very slow rate using a system with a force transducer to fully define the biomechanical environment. A previous study using the same NCC system found that axons undergo the same degree of damage for non-painful and painful stretches applied at 0.5mm/s, with the majority of axons being rated a 2 [13]. In our study, although strains are similar to the painful stretches in that earlier study axon damage is less severe, with the majority having a rating of 1 or higher. Although damage and strain are related, additional studies are needed with more spatiotemporal resolution to probe relationships between local mechanics and neuron health. In addition, these results reflect findings from a very small sample size. Nevertheless, although slow stretch of axons has been reported to induce growth [7], our findings suggest it may also be injurious. Future work needs to also investigate effects on specific neuronal signaling pathways.

Acknowledgements (Optional): :

References (Optional): : [1] Jaumard et al. J Biomech Eng. 133(7):071010, 2011; [2] Zhang et al. Ann Biomed Eng. 45:2813–25, 2017; [3] Ita et al. JOSPT. 47(7):450-61, 2017; [4] Bass et al. Spine J. 32(1):E7-E13, 2007; [5] Noyes et al. JBJB. 56(2):236-53, 1974; [6] Stephen et al. J Mech Behav Biomed Mater. 10:216-26, 2012; [7] Pfsiter et al. Neurosurgery. 60(1):137-42, 2007; [8] Cullen & LaPlaca. J Neurotrauma. 23(9):1304-19, 2006; [9] Nuethong et al. SB3C Mtg. 2022; [10] Zhang et al. J Biomech Eng. 138(2):021013, 2016; [11] Ita & Winkelstein. J Biomech Eng. 141(9):0910131, 2019; [12] Kallakuri et al. Eur Spine J. 17(4):556-63, 2008; [13] Villasmil et al. BMES Mtg. 2017.