Differential Effects of Neurectomy and Botox-induced Muscle Paralysis on Bone Phenotype and Titanium Implant Osseointegration

Jingyao Deng; David J Cohen; James Redden; Michael J McClure; Barbara D Boyan; Zvi Schwartz

doi:10.1016/j.bone.2021.116145

. Author manuscript; available in PMC: 2022 Dec 1.

Published in final edited form as: Bone. 2021 Aug 12;153:116145. doi: 10.1016/j.bone.2021.116145

Differential Effects of Neurectomy and Botox-induced Muscle Paralysis on Bone Phenotype and Titanium Implant Osseointegration

Jingyao Deng ¹, David J Cohen ¹, James Redden ¹, Michael J McClure ¹, Barbara D Boyan ^1,³, Zvi Schwartz ^1,²

PMCID: PMC8480339 NIHMSID: NIHMS1735014 PMID: 34390886

Abstract

Metabolic bone is highly innervated by both sensory and sympathetic nerves. In addition to skeletal development, neural regulation participates in local bone remodeling, which is important for successful osseointegration of titanium implants. Neurectomy is a model used to investigate the lack of neural function on bone homeostasis, but the relative impacts of direct denervation to bone or denervation-induced muscle paralysis are less well defined. To investigate this difference, we used two nerve intervention models, sciatic and femoral neurectomy (SFN) v. botox-induced muscle paralysis (BTX) and assessed the resulting femoral bone phenotype and Ti implant osseointegration. Male Sprague Dawley rats (19) were randomly divided into three groups: implant control (n=5), SFN (n=7), and BTX (n=7). Ti implants (microrough/hydrophilic [modSLA], Institut Straumann AG) were placed in the distal metaphysis of each femur on day 24 post-SFN or BTX. Bone and muscle were examined on day 28 after implant insertion. Both nerve intervention models impaired osseointegration. MicroCT and histology indicated that both models had reduced trabecular bone formation. Only BTX reduced cortical bone formation and increased cortical bone porosity. BTX resulted in more bone loss characterized by the least trabecular and cortical bone, as well as osseointegration. Osteoblasts isolated from the tibia exhibited a model-specific phenotype when they were grown on Ti substrates in vitro. Neurectomy caused more severe muscle atrophy than botox injection. These results indicate that neural regulation directly modulates bone formation and osseointegration. Muscle paralysis modulated the effects of loss of neural inputs into bone, supporting the hypothesis that mechanical loading of bone is a factor in achieving successful osseointegration. The different effects of botox and neurectomy on bone phenotype indicated that the sensory and sympathetic nerves had a role in the osseointegration process.

Keywords: osseointegration, titanium implant, bone formation, neural regulation, botox, muscle paralysis

Graphical Abstract

graphic file with name nihms-1735014-f0001.jpg

1. Introduction

Bone is a highly dynamic tissue that is continuously remodeled to adapt to mechanical loading and the surrounding environment¹. During implant insertion in bone, the surgical procedure will inevitably cause injury and a complex biological cascade of cellular events is required to achieve regeneration of the bone and integration of the implant with the surrounding bone tissue². Osseointegration is the determinant factor for implant success, which is the structural and functional connection between bone and implant surfaces³. Initial implant stability occurs via primary bone formation, but ultimately, successful osseointegration depends on optimal bone remodeling, with communication between osteoclasts and osteoblasts, resulting in balanced bone resorption and bone formation⁴.

Titanium is a popular and common implant material in dental and orthopaedic implant markets because of the high biocompatibility and corrosion resistance of its titanium dioxide surface chemistry. Our lab and other research groups have investigated surface modifications that mimic osteoclast resorption pits with a complex hierarchical structure that attract osteoprogenitor cells^5,6. Our findings indicate that surfaces with this topography stimulate bone marrow stromal cells (BMSCs) to differentiate into osteoblasts without the need for media additives such as dexamethasone or beta-glycerophosphate⁵. This change in phenotype is enhanced when the surface is hydrophilic and depends on the production of bone morphogenetic protein 2 (BMP2) and Wnt5a⁷, and signaling via α2β1 integrins⁸.

Even though titanium surfaces with microroughness and hydrophilic properties have high successful osseointegration rates in healthy patients, the success rate of osseointegration drops significantly in patients with compromised conditions, such as smoking, diabetes, and low bone density^9,10. Osteoporosis is also a risk factor for implant failure^11–13, potentially due to the imbalance between bone resorption and bone formation in bone remodeling¹⁴.

One approach is to improve osseointegration by increasing net bone formation. Research has investigated the contributions of hormonal regulation of cellular response¹⁵, such as estrogen¹⁶, 1α,25(OH)₂D₃¹⁷, and local growth factors, such as transforming growth factor-beta (TGFβ1)¹⁸, insulin-like growth factors¹⁹ and BMPs^20,21, as well as nerve-derived factors, such as neuropeptides²² and semaphorins²³. While most studies have focused on endocrine regulation, local growth factors, and surface modifications, few studies emphasized the role of neural regulation.

The importance of neural regulation of bone homeostasis has been addressed at the anatomical level by investigating the innervation of bone tissue²⁴. The autonomic nervous system has been shown to participate in skeleton development and bone remodeling. The sympathetic nervous system is involved in increasing bone resorption and decreasing bone formation; an antagonist to the β-adrenergic receptor can mitigate osteoporosis by decreasing sympathetic tone^25,26. Sympathetic nerves accompanying vasculature through Volkmann’s canals play a role in controlling blood flow, which is important for transporting nutrients and cells for bone growth and remodeling^27,28. The presence of sensory neuropeptide receptors and sympathetic receptors in osteoblasts, identified by immunocytochemistry, indicates the direct interaction between neurons and osteoblasts^24,29. Nerve-derived regulators, such as semaphorins, have been shown to regulate bone cell homeostasis¹⁹.

Class 3 semaphorins were originally identified as axon guidance proteins important for nerve development³⁰. Recently, these semaphorins have been reported to be important for regulating bone remodeling. For example, semaphorin 3A (sema3A) was reported to have dual regulation in both bone resorption and bone formation^19,31,32. This nerve-derived regulator is also produced by bone lineage cells, including BMSCs and osteoblasts, and titanium surface properties regulate its production²⁴. Exogenous sema3A increases the osteoblastic differentiation of BMSCs, and this effect is further enhanced when the cells are cultured on microrough and hydrophilic Ti implant surfaces³³. These findings suggest that neural regulation participates in local bone remodeling, making it important to investigate the contribution of neural regulation to osseointegration.

This study investigated direct neural regulation in osseointegration, emphasizing the correlation between nerve and muscle interaction. Neurectomy is a standard model used to investigate the role of neural regulation in tissue homeostasis³⁴. However, neurectomy also induces muscle paralysis at the same time, making it difficult to distinguish whether the neurectomy-induced effect is caused by direct bone denervation or by denervation-induced muscle paralysis and loss of muscle contraction. Botulinum toxin A (botox) is a neurotoxin that temporarily prevents muscle contraction by blocking the release of the neurotransmitter acetylcholine at the neuromuscular junction^35–38. Therefore, in this study, by contrasting a model that imposes both direct denervation and diminished muscle activity (neurectomy model) with a model of diminished muscle activity (botox model), the role of direct neuronal dysfunction on osseointegration will be inferred.

2. Material and Methods

2.1. Implant Preparation

Ti screw implants were designed for unicortical implantation in a rat femur and fabricated by Institut Straumann AG (Basel, Switzerland). 3.5 mm long implants with a 2.5 mm outer diameter and a 0.8 mm pitch were initially machined from a rod of grade 4 Ti. The implants had a grit blasted and acid-etched surface (SLA) that was prepared under conditions that resulted in high hydrophilicity (modSLA), as described previously^39,40. Implants were sterilized by γ-irradiation and were used immediately after removal from their packaging for all experiments.

2.2. Animals and Surgical Procedures

All animal procedures were approved by the Institutional Animal Care and Use Committee at Virginia Commonwealth University. All experiments were carried out following the National Institutes of Health guide for the care and use of laboratory animals. Animals were single-housed in an individually ventilated, solid-bottomed polysulfone cage and were kept in an AALAC accredited animal facility in indoor housing capable of temperature and humidity control within limits appropriate for the species and a 12h/12h light/dark cycle.

In the initial study, twenty-one healthy male Sprague Dawley rats (Charles River Laboratories, Wilmington, MA, USA) weighing 300–325g were randomly assigned to three groups: implant control (Control, n=5), sciatic and femoral neurectomy (SFN, n=8), and botox injection (BTX, n=8). One SFN animal and one BTX animal were withdrawn early from the study for distress according to the endpoints criteria detailed in the IACUC protocol, resulting in nineteen rats (SFN, n=7; BTX, n=7). Control L and Control R were defined as left hindlimbs and right hindlimbs in the implant control group. BTX C and BTX T were defined as contralateral hindlimbs and botox injected hindlimbs, respectively. SFN C and SFN T were defined as contralateral hindlimbs and neurectomy intervention legs.

The same experimental design was used to assess peri-implant histology. Twelve rats were divided into two cohorts (SFN: n=6; BTX: n=6). Three SFN animals and three BTX animals were withdrawn early as they met humane endpoints criteria, resulting in n=3 for each group.

For all procedures, anesthesia was induced with 5% isoflurane gas inhalation in O₂ and maintained with 2.5% isoflurane gas inhalation in O₂. Animals were recovered from anesthesia on a water-circulating warming pad before being returned to the vivarium. For all surgical procedures, 1 mg/kg of sustained-release Buprenorphine (ZooPharm, Windsor, CO, USA) was administered pre-operatively via subcutaneous injection to provide a minimum of 72 hours of postoperative analgesia.

2.2.1. Sciatic and femoral nerve neurectomy surgeries on right hindlimbs (SFN)

The hair was clipped on the right dorsum from the mid-back, down the lower hind limb to the knee, and on the ventral side from the midline, through the inguinal area, down to the knee. The skin was prepared for surgery by cleaning with chlorohexane and 70% isopropanol alternately. After palpating the ischial tuberosity, a 3 cm incision was made from that landmark parallel to the femur. The biceps femoris and tensor fascia latae were then separated to locate the sciatic nerve. A 3 cm section of the sciatic nerve was isolated by tying 5–0 nylon sutures at the proximal and distal ends and removed between the sutures. The muscles were reapproximated and the skin was closed with 5–0 nylon sutures in a running technique. Rats were then placed in a supine position, and a 2 cm skin incision was made in line with the inguinal ligament. The peritoneal fat was separated from the bundle of the femoral nerve, artery, and vein. The nerve was carefully dissected away from the femoral artery and vein, and a 5 mm segment of the nerve was removed. The tissues were then reapproximated and the skin was closed with 5–0 nylon suture in a running technique. Wound clips were placed over the sutured incisions.

2.2.2. Botox injections on right hindlimbs (BTX)

Hair was shaved from the right hindlimbs as above, and skin was prepared with 70% isopropanol before injections. Botulinum toxin type A (onabotulinumtoxin A; BOTOX®, Allergan Inc., Irvine, CA, USA [botox]) was dissolved in sterile 0.9% saline at a concentration of 10 U/ml. The right hindlimb of each rat was injected intramuscularly with a total of eight units of botox distributed as two units per syringe into the following locations: the paraspinal muscles, the quadriceps, the hamstrings, and the gastrocnemius muscles. Botox injections were administered on day 1 and day 32 to maintain muscle paralysis for the entire study.

2.2.3. Unicortical implant placements on both hindlimbs in all groups

SFN and BTX animals were aged for 24 days following their respective procedures to allow for the development of a bone phenotype. The aging time was determined by a preliminary study showing that SFN- and BTX-induced bone phenotypes were established after 24 days (data not shown). The experiment was designed to insert implants into bones that were compromised by nerve intervention in order to mimic clinical situations. ModSLA implants were placed unicortically into the distal metaphysis of both hindlimbs, as described previously³⁹. Briefly, the hindlimbs were shaved from the mid-abdomen to the foot bilaterally and sterilized with 70% isopropanol and chlorohexidine alternately. Rats were then placed in a supine position. A 3 cm incision was made on the medial aspect of the knee parallel to the femur. The rectus femoris and vastus medialis were separated in order to isolate the medial aspect of the distal femur, below the femoral notch. The periosteum was elevated, and the implant insertion site was created using a low-speed dental drill to progressively drill a defect using a series of increasing diameter drill bits (Ø1.0 mm, Ø1.6 mm, Ø2.0 mm, and Ø2.2 mm) to a depth of 3.5 mm in the distal metaphysis of the femur (Supplemental Figure 1a). The implant was then unpackaged and screwed into the defect until flush with the bone (Supplemental Figure 1b). After insertion, the implant was capped with a cover screw to prevent bone ingrowth into the internal threading of the implant. The surrounding muscle was re-approximated and sutured together using 5–0 vicryl suture using a horizontal mattress technique, and the skin was closed by 5–0 nylon suture in a running technique. 9 mm wound clips were placed over the sutured incision site. The procedure was then repeated on the contralateral limb. Animals were killed 28 days following implant surgery.

2.3. Tissue Analysis

2.3.1. Bone phenotype and peri-implant growth analyzed by microCT

Micro-computed tomography (microCT) (SkyScan 1173, Bruker, Kontich, Belgium) was used to evaluate bone phenotypes, as well as peri-implant growth. Femurs were isolated and stored in cold 1X PBS at 4°C and scanned fresh without fixation within 24 hours of harvest in order to conduct mechanical torque testing on the same samples. Both proximal and distal ends of the femurs were scanned at a resolution of 1120×1120 pixels (isotropic voxel size of 15.82 μm) using a 1.0 mm aluminum filter, at an exposure of 250 ms, with scanning energies of 85 kV and 94 μA. 5 x-ray projections were acquired every 0.2° and averaged. A standard Feldkamp reconstruction was performed by NRecon Software (Bruker) with a beam hardening correction of 20%, and no smoothing was applied. Quantitative trabecular morphometric parameters, including bone volume percentage (BV/TV), trabecular number, trabecular thickness, and total porosity, were determined. Quantitative cortical morphometric parameters, such as BV/TV, total porosity, mean total cross sectional bone area, mean cross sectional bone perimeter, and cortical thickness were determined.

Peri-implant bone growth and bone to implant contact (BIC) were evaluated at the distal femoral metaphysis at a resolution of 1120×1120 pixels (isotropic voxel size of 15.8 μm) using 0.25 mm brass filter and at an exposure of 420 ms, 120 kV voltage and 66 μA current. After reconstruction using the same method as described above, a volume of interest (VOI) was selected to analyze BIC at different conditions (Supplemental figure 2). For total BIC, the VOI was selected from the bottom of the implant toward the apex to the highest position that cortical bone was still in contact with the implant surface (indicated by blue line in Supplemental figure 2a); for bone marrow to implant contact, the VOI was selected in the bone marrow region (indicated by the purple line in Supplemental figure 2b); for cortical bone to implant contact, the VOI was selected in the region that implant surfaces were in contact with cortical bone (indicated by the green line in Supplemental figure 2c). The VOI was shrink-wrapped, dilated 2 pixels around the implant, and subtracted from the original VOI, which enabled us to remove the implant from the calculation. The implant VOI was loaded as the region of interest. The new VOI was dilated 2 pixels away from the implant VOI and subtracted from the implant VOI. The bone volume within the new VOI was calculated and normalized to the new VOI, which was recorded as BIC.

In a separate set of analyses, a box with uniform dimensions was selected to be the VOI, covering the bone around implant (Supplemental figure 2d–f). This volume was consistent among all groups and used as total volume. After removing implants (shaded implant indicated its removal), total bone volume, bone volume in the medullary compartment, and cortical bone volume in the box were calculated and normalized to the box volume (Supplemental figure 2d–f).

2.3.2. Histology

A separate set of rats were used to assess peri-implant histology (SFN: n=3; BTX: n=3). Tissues were harvested 28 days after the implant surgery. Following harvest, samples were fixed in 10% neutral buffered formalin for a minimum of 24 hours prior to microCT imaging. Samples were then sent to be commercially processed for histological staining (Histion, Everett, WA, USA). Briefly, all fixed samples were radiographed and trimmed as needed to facilitate processing. Trimmed samples were embedded in methyl methacrylate, and one ground section was taken through the approximate center of each implant in a plane longitudinal to the implant and transaxially relative to the long axis of the bone shaft (unicortical). All sections from the four experimental groups, the SFN neurectomy leg and contralateral control, as well as the BTX treatment leg and contralateral control, were stained with Stevenel’s Blue/van Gieson stain and cover slipped. Sections were imaged by bright field light microscopy with a Zeiss AxioCam MRc5 camera and Axio Observer Z1 and analyzed by ImageJ⁴¹. Peri-implant bone growth and BIC were evaluated by histomorphometry.

Peri-implant bone growth was quantified within a consistent rectangular region of interest (ROI) that was 4.8 mm in width and 3.4 mm in length. The ROI was selected at the edge of the distal end of the bone and centered. BV was defined as the area of all bone within the ROI and normalized to the area of the ROI (TV). This was consistent for all samples and resulted in the BV/TV calculation. A series of lines 250 μm apart was drawn to measure the distance between periosteum and endosteum. These distances were averaged and graphed as cortical thickness. Additionally, the perimeter of the bone in contact with the implant surface was measured and normalized to the perimeter of the implant containing both bone marrow and cortical bone, which yielded total BIC. Evaluation of BIC was performed separately for 2 subregions, the cortical (cortical BIC) and medullary compartment (bone marrow BIC).

2.3.3. Removal torque mechanical test

After imaging with microCT, femurs were mounted in pre-designed polylactic acid (PLA) holders with polyurethane adhesive and allowed cure in the holder overnight at 4°C to eliminate unexpected movement during testing. The unicortical implant in each hindlimb was fit to a customized driver and aligned to the ElectroForce 3200 Series III test instrument axis (TA Instruments, New Castle, DE, USA) (Figure 5a). The clamp on the instrument was tightened to secure each holder so that there was no initial torque or compressive load applied before the testing. The removal torque testing protocol was similar to a previous study in our lab³⁹. Briefly, torque was applied to each implant with a rotational speed of 0.1°s⁻¹ with an axial displacement of 0.8 mm/360°. Torque vs radian graphs were generated for each implant (Figure 5b) and a bilinear model (red solid line) was fit to the experimental data (blue dash line) to calculate the maximum torque, torsional stiffness, and torsional energy by an open-source least squares spline modeling package (SLA-Shape Language Modeling version 1.14) for MATLAB (MathWork, Natick, MA, USA). The first vertical green dash line, which separated the toe region (left section) and the linear region (right region) were defined in MATLAB. The second green dash line was at the yield point to the separate elastic area (left region) and plastic area (right region). The peak of the curve was considered as maximum torque, the slope of the linear region of the curve was calculated for torsional stiffness, and the area below the linear region of the curve was defined as torsional energy (Figure 5b).

Figure 5. — The effect of sciatic and femoral neurectomy and botox injection on implant osseointegration was assessed by removal torque mechanical tests. Fresh femurs were harvested and fixed in a polylactic acid (PLA) tube with glue (a). Torque vs. radian graphs were generated for each femur and fit to a bilinear model (b) to calculate maximum torque (c, d, e), torsional stiffness (f, g, h), and torsional energy (I, j, k). Data shown are the mean of treatment leg over the contralateral leg for each group ± standard error of n=5 for control rats, n=7 for BTX rats and n=7 for SFN rats or median ± range of n=5 for Control L, Control R, n=7 for BTX C, BTX T, SFN C and SFN T. Groups not sharing a letter are significant at an α = 0.05 by one way ANOVA with Tukey post correction or Kruskal-Wallis test with Dunns’ post correction. Data comparing treatment legs with their contralateral legs designated with an asterisk are statistically different at α = 0.05 by Wilcoxon matched-paired signed rank test.

2.3.4. Muscle histology

Whole gastrocnemius and whole tibialis anterior muscles were removed from both hindlimbs from the implant control group, SFN neurectomy legs, and contralateral control, and BTX injected legs and contralateral control and weighed. Muscle histology was then produced the same as in a previous study⁴². Briefly, the gastrocnemius was fixed in 4% paraformaldehyde, dehydrated, and embedded in paraffin. Muscles were cross-sectioned ~0.5 cm from the margins. Sections (5 μm) were placed on Histobond slides (VWR, Radnor, PA, USA), deparaffinized and rehydrated, and stained with Masson’s trichrome using Weigert’s hematoxylin (Sigma-Aldrich, St. Louis, MO, USA), Biebrich scarlet-acid fuschin (Sigma-Aldrich), and aniline blue (Sigma-Aldrich). Coverslips were mounted with xylene-based mounting media and allowed to dry flat before imaging.

2.4. Response of Primary Bone Marrow Stromal Cells (rBMSCs) and Osteoblasts (rOBs)

In order to investigate the effects of denervation and botox injection on the response of bone lineage cells to modSLA in vitro, rBMSCs and rOBs were isolated from tibias due to the unavailability of femurs, which were used for other analyses. Since sciatic/femoral denervation occurred to the entire hindlimb, including femur and tibia, the effect of denervation would also be experienced in the tibia. In the botox injection rats, botox was injected into the paraspinal muscles, quadriceps, hamstrings, and gastrocnemius, which were major muscle groups involved in both femur and tibia movement. Therefore, the primary cells harvested from the tibia would be representative of primary cells harvested from the femur with respect to muscle paralysis. Tibias were isolated and stored in serum-free Dulbecco’s modified Eagle medium (DMEM; Mediatech, Manassas, VA, USA) with 150 U/ml penicillin-streptomycin (Life Technologies, Carlsbad, CA, USA). After removing soft tissues on the tibia, a Rongeur bone cutter was used to cut the top and bottom end of the bone to expose the bone marrow. rBMSCs were isolated using DMEM with 50 U/ml Pen-Strep to flush bone marrow twice from each end to a 50ml conical tube. Tubes were then spun down at 500 g for 10 minutes, and cell pellets were resuspended in ACK (ammonium-chloride-potassium) lysis buffer (Quality Biological Inc, Gaithersburg, MD, USA) to lyse red blood cells. Cells were plated in DMEM with 10% fetal bovine serum (FBS) and 100 U/ml Pen-Strep (DMEM FM) in 6-well plates for 24 hours to allow for cell attachment. Media were switched to rBMSC growth media (CCM; Lonza Biosciences, Morrisville, NC, USA) and changed every 48 hours until cells reached 90% confluence. rBMSCs were then plated at 10,000 cells/cm² on tissue culture polystyrene (TCPS) and modSLA titanium surfaces with n=6 independent cultures for each group. Media were changed 24 hours after plating and every 48 hours until day 7. On day 7, fresh media were added for 24 hours and then collected for further analysis. Cell monolayer lysates were rinsed with 1 ml 1X PBS once and lysed in 0.5 ml 0.05% Triton 100X. Sonication at 40 amplitude using a Vibra-Cell ultrasonicator (Sonics & Materials Inc., Newtown, CT, USA) was used to homogenize cell monolayer lysates.

Osteoblasts were isolated as described previously⁴³. After flushing bone marrow, tibias were digested in 0.25% Trypsin-EDTA (Life Technologies) at 37°C for 15 minutes and then the trypsinization was stopped by adding DMEM FM. Tibias were minced into small pieces approximately 1 mm*1 mm and placed in 100 mm*20 mm Petri dish with DMEM FM. Bone pieces were rinsed with sterile PBS and incubated with 260 U/ml collagenase II solution (Worthington Biochemical, Lakewood, NJ, USA) prepared in DMEM without FBS at 37°C for 2 hours to remove all remaining soft tissues. After 2 hours, bone pieces were rinsed with DMEM FM three times and then spread out in a new 100 mm*20 mm Petri dish with DMEM FM. At confluence, cells were subpassaged and plated in a 24 well plate at 10,000 cells/cm² density. The osteoblastic phenotype of isolated cells was confirmed by their ability to produce higher amounts of osteocalcin when cultured on modSLA.

Confirmed rat osteoblasts (rOBs) from each experimental group were plated on TCPS or modSLA surfaces at 10,000 cells/cm² in DMEM FM and incubated at 37°C in an atmosphere of 5% CO₂ and 100% humidity. Media were changed 24 hours after plating and every 48 hours for 7d. At 7d, cells were incubated with fresh DMEM FM for 24 hours, which was collected for further analysis. Cell layer lysates were rinsed by 1X PBS and lysed in 0.05% Triton 100X followed by sonication. The following analyses were done on the rBMSCs and rOBs. DNA quantification was conducted on cell layer lysates as described above. Immunoassays were used to measure levels of BMP2, osteopontin (OPN), osteoprotegerin (OPG), vascular endothelial growth factor 165 (VEGF165) (DuoSet ELISA, R&D Systems, Minneapolis, MN, USA) in collected media. Osteocalcin (OCN; Thermo Fisher Scientific, Waltham, MA, USA), semaphorin 3A (Sema3A; LifeSpan Biosciences, Seattle, WA, USA), and semaphorin 3C (Sema3C; MyBioSource, San Diego, CA, USA) were measured in rOBs’ collected media as well, and protein levels were normalized to DNA content.

To measure cell proliferation, confirmed rat osteoblasts from each experimental group were also plated on TCPS until 60% confluent in DMEM FM. Then cells were serum-starved in DMEM for 24 hours to ensure that proliferation was quenched. At 20 hours, Alamar blue reagent (ThermoFisher Scientific, Waltham, MA, USA) was added to the media for 4 hours, and plates were read at an excitation of 560 nm and emission of 590 nm.

2.5. Statistical Analysis

Based on previous studies^39,44,45, to detect a 30% mean difference with 20% variance and a type I error rate of 0.05, eight successful implants for each group were determined to ensure the difference of implant success with different factors (mechanical, hormonal, or pharmacological). Based on literature^38,46, comparing botox injected legs and contralateral legs, or comparing neurectomy legs and contralateral legs by paired t test to ensure the difference in bone phenotype, the effective size dz were 2.43 and 2.7 respectively. With an alpha=0.05 and power=0.8, the projected sample size needed with these effect size dz were n=4 for BTX and n=4 for SFN. Thus, N=8 rats per group were determined for the initial experiment to result in n=8 for BTX C, BTX T, SFN C, and SFN T. N=4 rats were prepared for the implant control group because identical implants were inserted into both hindlimbs. One extra animal was prepared for surgery incidents and assigned to the implant control group (N=5 rats).

A statistical power analysis was performed for sample size estimation for the second animal study, based on data from total BIC microCT from the initial experiment. Comparing BTX C to BTX T and SFN C to SFN T by Wilcoxon signed rank test, the effective size dz were 2.36 and 1.37 respectively. With an alpha=0.05 and power=0.7, the projected sample size needed with these effect size dz are n=4 for BTX and n=6 for SFN. Therefore, the sample size with n=6 for the histology study was determined.

Identical implants were inserted into both right and left hindlimbs in all groups to rule out the possibility that the existence of implants constituted a complicating factor. For the in vivo study, the results of phenotypic changes in the bones, peri-implant bone growth, removal torque mechanical tests and muscle weights were graphed as treatment/control, which was defined as treatment leg normalized to contralateral leg in the same animal for control (n=5), SFN (n=7) and BTX (n=7). The analysis of normal distribution was observed by Shapiro-Wilk test. First comparison was done between contralateral legs and intervention legs by Wilcoxon matched pairs signed rank test (α=0.05) represented by asterisk (*) and then one-way ANOVA (α=0.05) with Tukey post correction represented by letter comparing differences among groups (control, SFN, BTX) in treatment/control graphs. Second comparison was conducted on data without normalization, and non-parametric analysis (Kruskal-Wallis test with Dunns’ post corrections, α=0.05) was conducted on all groups when data did not follow normal distribution. For the second animal study, both paired t test and Wilcoxon matched pairs signed rank test (α=0.05) were used to analyze the data. Values of ρ<0.05 were considered significant. To provide better evaluation of the in vivo data, results were also graphed as the median ± range when analyzed by non-parametric analysis for Control L (n=5), Control R (n=5), BTX C (n=7), BTX T (n=7), SFN C (n=7) and SFN T (n=7) or graphed as mean ± SEM when analyzed by parametric analysis (one way ANOVA with Tukey post correction) provided in the supplemental figures.

For each in vitro experiment, data presented are the mean ± SEM for n=6 independent cultures derived from 6 individual tibias per group. All in vitro experiments were repeated twice to ensure validity. One-way ANOVA with Tukey correction was used to compared differences between groups (BTX C, BTX T, SFN C, SFN T), and an unpaired t test with α=0.05 was used to compare differences between modSLA and TCPS within the same group. Values of ρ<0.05 were considered significant. All power analyses were provided by G*Power^47,48. All graphs were generated by GraphPad Prism 6.0 (GraphPad, La Jolla, CA, USA). All statistical analyses were performed in JMP statistical software (SAS Institute, Cary, NC, USA)

3. Results

3.1. Evaluation of Bone Phenotype

Results were analyzed by two methods: 1. In the same rat, the nerve intervention limb was compared to its contralateral leg to study the outcomes while taking animal variances into account. Graphs are presented as treatment/control as mentioned in the methods. 2. While comparing the treatment side to the contralateral side compensated for the inherent biological variance, it had limits because neurectomy and botox treatment could potentially lead to overloading or underloading in the contralateral legs. In this case, contralateral legs were not ideal controls. As a result, neurectomy legs, and botox legs were compared to the same side of the implant control group, which mimics the normal mechanical loading situation in healthy rats. Graphs are presented as raw data without normalization.

The trabecular and cortical bone phenotype at distal ends of the femur near the implant insertion site were analyzed by microCT and representative images shown in Figure 1a. The insertion site is shown in Figure 1b. Both the botox injection leg (BTX T) and the neurectomy leg (SFN T) had qualitatively less trabecular bone formation and higher porosity compared to their respective contralateral legs and the implant control group (Figure 1a). Quantitative analyses were presented as treatment/control, where comparisons were conducted between treatment legs to contralateral leg. Meanwhile, to minimize variances between animals, normalized outcomes between treatment sides and contralateral sides were compared in three groups. Both BTX and SFN treatment decreased trabecular bone formation, which was indicated by less BV/TV (Figure 1c), lower trabecular number (Figure 1d), and higher total porosity (Figure 1e); while trabecular thickness was not affected by BTX or SFN (Figure 1f). The severity of the decrease in trabecular bone formation was more robust in BTX rats than in the SFN rats.

Figure 1. — Characterization of the neurectomy and botox injection-induced bone phenotype at the distal femur. 2-month-old, male, Sprague Dawley rats underwent sciatic and femoral neurectomy (SFN) or botox injection (BTX) surgery. After 3 weeks, microrough/hydrophilic titanium implants (modSLA Ti) were inserted. After 28 days of osseointegration, femurs were harvested and placed in 1X PBS for microCT scanning. 3D microCT reconstructions (a) were made of the distal ends of the femurs (b). The trabecular bone phenotype was assessed as bone volume/total volume (c), trabecular number (d), total porosity (e), and trabecular thickness (f). The cortical bone phenotype was assessed as bone volume/total volume (g), total porosity (h), mean total cross sectional bone area (i), mean cross sectional bone perimeter (j), and cortical thickness (k). Data shown are the means of treatment leg/contralateral leg for each group ± standard error of n=5 for control rats, n=7 for BTX rats, and n=7 for SFN rats. Data within each group not sharing a letter are significantly different at an α = 0.05 by one-way ANOVA with Tukey post correction. Comparisons between treatment legs and contralateral legs identified with an asterisk are statistically different at α = 0.05 by Wilcoxon matched-paired signed rank test. Scale bar=1mm

The results without normalizing to contralateral legs are also shown in Supplemental Figure 3. The intervention resulted in less BV/TV, less trabecular number, and higher total porosity in BTX and SFN rats compared to the same side of hindlimbs in the implant control group (Supplemental Figure 3a–c). Trabecular thickness was not affected by BTX nor SFN (Supplemental Figure 3d).

SFN did not affect the cortical bone phenotype. There was no change in BV/TV, total porosity, the mean total cross sectional bone area and mean cross sectional bone perimeter, or cortical thickness (Figure 1g–k; Supplemental Figure 3e–i). In contrast, BTX caused reduced cortical bone indicated by less BV/TV (Figure 1g), higher total porosity (Figure 1h), less mean total cross sectional bone area (Figure 1i) and less cortical thickness (Figure 1k) compared to their contralateral legs. Compared to the implant control group, BTX also statistically decreased mean total cross sectional bone area (Figure 1i). No change in mean cross sectional bone perimeter was observed among all comparisons (Figure 1j). Supplemental Figure 3 showed that without normalization, BTX T had a significantly less mean total cross sectional bone area compared to BTX C and Control R (Supplemental Figure 3g), and did not affect BV/TV, total porosity, mean cross-sectional bone perimeter, and cortical thickness (Supplemental Figure 3e,f,h and i).

To further investigate the effect on cortical bone, the cortical bone at the mid-diaphysis was analyzed (Figure 2b). Figure 2a qualitatively showed that cortical bone from the implant control group was the thickest and BTX T had the lowest thickness. Quantitatively, botox injected legs were significantly different than their contralateral legs for all parameters (Figure 2c, and d). Additionally, BTX legs had significantly less mean total cross sectional bone area (Figure 2e, and Supplemental Figure 3l) compared to the implant control group, further demonstrating the inhibitory effect of the botox injection on cortical bone formation. There was also a reduced effect on the mean cross sectional bone perimeter compared BTX T to Control R (Supplemental Figure 3m).

SFN did not inhibit cortical bone formation except the mean total cross sectional bone area (Figure 2e). Without normalizing to the contralateral legs, similar results were also shown in Supplemental Figure 3j–n. The trabecular and cortical bone in proximal femurs were also evaluated, as shown in Supplemental Figure 4. SFN did not alter trabecular bone formation (Supplemental Figure 4a–d), and it did not affect cortical bone formation in femoral heads (Supplemental Figure 4e–i); however, the botox injection reduced trabecular bone formation, indicated by decreases in BV/TV, trabecular thickness, trabecular number, and increases in total porosity compared to the implant control group (Supplemental Figure 4a–d). Furthermore, botox injection impaired cortical bone formation, indicated by less BV/TV, higher total porosity, reduced mean total cross sectional bone area and decreased cortical thickness (Supplemental Figure 4e, f, g, and i). There was no change in the mean total cross sectional bone perimeter (Supplemental Figure 4h).

Data without normalizing to contralateral legs are shown in Supplemental Figure 5. BTX T reduced both trabecular bone formation in the femoral head compared to both BTX C and Control R, and also decreased cortical bone formation compared to Control R (Supplemental Figure 5a–i). Additionally, SFN T did not affect either trabecular bone or cortical bone (Supplemental Figure 5).

Furthermore, there was no significant difference in any of the bone phenotyperelated outcomes among Control L, BTX C, and SFN C (Supplemental Figure 3, 4 and 5).

3.2. Evaluation of Osseointegration

3.2.1. MicroCT analysis

To ensure that the treatment regimens used in the two models had an impact on the bone phenotype, one rat from the BTX group and the SFN group were euthanized to evaluate the effect on the bone before implant insertion surgery. MicroCT verified that both models resulted in reduced BV/TV of trabecular bone at distal femurs compared to the contralateral control limbs (Supplemental Figure 6).

There was a differential effect of each model on osseointegration as assessed by BIC and BV/TV. MicroCT images showed osseointegration occurred in all treatment groups, indicated by bone growth around the implant (brown tissues indicated cortical bone, and blue tissues indicated trabecular bone in bone marrow region) (Figure 3a). Both Figure 3a and Supplemental Figure 1 indicated the implant insertion is a unicortical penetration with attachment to the other side of the cortical bone, therefore, BIC and BV/TV analysis included both cortices. BTX T and SFN T rats had qualitatively less bone around the implant compared to the contralateral legs and the implant control group. There was statistically less total BIC (Figure 3b), and this effect was greatest in the BTX cohort. The reduction in total BIC was due to reduced cortical BIC. Whereas no difference was observed in the amount of bone marrow in contact with the implant surface (Figure 3c), there was a marked reduction in the amount of cortical bone (Figure 3d).

To ensure that the effect was not due to the unintended effect on the contralateral legs in BTX and SFN groups, BTX T and SFN T were also compared to Control R without normalization. BTX T had significantly less total BIC than Control R while SFN T was not different than Control R. Moreover, figure 3f showed there was no difference in total BIC among contralateral legs in the three groups. Comparison among all groups for total BIC is presented in Supplemental Figure 7a. Bone marrow BIC among all groups were not affected by nerve intervention (Supplemental Figure 7b), while cortical BIC was reduced in BTX T compared to BTX C (Supplemental Figure 7c).

Similarly, total BV/TV was reduced in both treatment groups, with the greatest effect in the BTX cohort (Figure 3g). Bone Marrow Volume/Total Volume was not statistically different (Figure 3h), but cortical BV/TV was reduced in the BTX and SFN rats (Figure 3i). Compared to Control R, BTX T caused a reduction in total BV/TV, while SFN T had no effect (Figure 3j). Furthermore, there was no difference between contralateral legs in total BV/TV in the three groups (Figure 3k). Supplemental Figure 7d showed a comparison of total BV/TV across all groups. Bone Marrow Volume/TV was lower in both BTX T and SFN T compared to Control R (Supplemental Figure 7e), and cortical BV/TV was only decreased in BTX T compared to its own contralateral legs (Supplemental Figure 7f). No significant difference was observed between contralateral legs in three groups for BIC and BV/TV (Supplemental Figure 7).

3.2.2. Histology

Stevenel’s blue and van Gieson staining provided more details of the peri-implant bone growth (Figure 4a–d). In accordance with the microCT, both models exhibited reduced bone to implant contact compared to their contralateral control limbs, and this was confirmed by their total BIC from histology (Figure 4e and j). Bone marrow to implant contact was reduced in animals treated by BTX and SFN (Figure 4f and k). Histomorphometric analysis of BIC in the cortical bone region indicated that SFN caused a statistically significant reduction (Figure 4l). BV/TV (Figure 4h and m) and cortical thickness (Figure 4i and n) were reduced to a similar extent in both models. None of the outcomes were significantly different when compared by Wilcoxon matched pair signed rank test (Figure 4o–s). Data presented in Supplemental Figure 8 demonstrated that the effects of nerve interventions on trabecular bone and cortical bone confirmed those shown by microCT, validating the reproducibility of the models.

3.2.3. Removal torque mechanical test

The implant design enabled us to assess removal torque using a screw system with a customized screw that extended from the machine and inserting to the interior of the implant (Figure 5a). The experimental data fit a bilinear fit curve (Figure 5b), from which we were able to calculate maximum torque (Figure 5c, d, and e), torsional stiffness (Figure 5f, g, and h), and torsional energy (Figure 5i, j, and k). Implants in the BTX treated limbs had reduced maximum torque when compared to the implant control group as well as to their contralateral legs (Figure 5c). Without normalization, BTX T and SFN T were not significantly different than Control R (Figure 5d). Torsional stiffness of the implants in the BTX rats was reduced compared to the contralateral limbs (Figure 5f). SFN did not affect the mechanical properties of the bone to implant bond when compared to the implant control rats, although torsional stiffness was reduced compared to their contralateral limbs (Figure 5f). BTX T impeded the torsional stiffness compared to Control R, while SFN T did not affect torsional stiffness compared to Control R (Figure 5g). No change in torsional energy was observed in either nerve intervention model (Figure 5i, and j). Additionally, no outcomes measured for mechanical properties in Control L, BTX C, and SFN C were affected by neurectomy and botox injection (Figure 5e, h, and k). Furthermore, without normalizing to contralateral legs, nerve intervention did not show an inhibitory effect on maximum torque (Supplemental Figure 9a). BTX T had the lowest torsional stiffness, but SFN T did not affect torsional stiffness (Supplemental Figure 9b). Torsional energy was not affected by either nerve intervention model (Supplemental Figure 9c).

3.3. Cell Response

Tibial osteoblasts isolated from BTX C, BTX T, SFN C, and SFN T reacted to modSLA by producing significantly more osteocalcin than when cultured on TCPS, indicating their enhanced osteoblastic phenotype on microrough and hydrophilic titanium surfaces (Supplement Figure 10).

Osteoblasts from BTX T legs had the least proliferative ability when grown on TCPS, which was statistically lower than BTX C (Figure 6a). The cell viability of SFN osteoblasts did not differ from contralateral limb osteoblasts, but cells from both treatment and control SFN limbs had lower viability than cells from BTX C limbs. The marked reduction in cell viability observed in BTX T osteoblasts was evident when measuring DNA content, whether the cells were cultured on TCPS or modSLA (Figure 6b). DNA content on TCPS correlated with proliferation results (Figure 6b), in which BTX T osteoblasts had the least amount of DNA. When cultured on modSLA, cells isolated from BTX C, BTX T, and SFN T showed reduced DNA when compared to TCPS (Figure 6b).

Osteoblasts from BTX T limbs exhibited an increase in osteocalcin production when grown on TCPS when compared to all other osteoblasts cultured on TCPS, and this effect was enhanced in cells cultured on modSLA (Figure 6c). Osteoblasts isolated from BTX C, BTX T, SFN C, and SFN T limbs produced higher amounts of osteocalcin on modSLA compared to TCPS (Figure 6c). Cells from BTX T produced higher OPN than BTX C on both TCPS and modSLA, and modSLA enhanced the OPN production from BTX T group when compared to TCPS (Figure 6d). Furthermore, cells from BTX T and SFN T produced a higher amount of BMP2 on modSLA than TCPS, and BTX T had the highest amount of BMP2 on both TCPS and modSLA compared to all other groups (Figure 6e). Similar results were found for VEGF 165 (Figure 6f) and OPG (Figure 6g).

Osteoblasts from BTX T limbs grown on modSLA showed increased sema3A, while osteoblasts from other treatment groups produced similar amounts of sema3A (Figure 6h). Cells from BTX T limbs produced higher levels of sema3C compared to cells from all other treatment groups and this difference was even greater on modSLA compared to TCPS (Figure 6i).

Bone marrow stromal cells isolated from tibia bone marrow from all treatment groups had less DNA (Figure 6j) on modSLA than TCPS. Moreover, osteoblastic differentiation was observed on modSLA to a comparable extent in BMSCs isolated from all experimental groups, indicated by increased production of OPN (Figure 6k), BMP2 (Figure 6l), and VEGF 165 (Figure 6m). Elevated production of OPG on modSLA was evident in cells from SFN animals, but there was no difference between treatment and control limbs (Figure 6n).

3.4. Muscle Morphology and Atrophy in Nerve Intervention Models

Muscle paralysis was qualified based on walking tests where animals were treated with either neurectomy or botox injection. Rats were allowed to walk freely on a tabletop. Treatment legs that dragged behind the animal were considered paralyzed, and both BTX T and SFN T were dragged behind while walking. Gastrocnemius and tibialis anterior muscle were chosen to analyze based on the rationale that the sciatic-femoral neurectomy and botox injections caused leg-dependent systemic effects, therefore, every muscle group in the distal region should be affected. Gastrocnemius muscles from all experimental groups were collected and stained with Masson’s trichrome to demonstrate the muscle morphology (Figure 7a). Gastrocnemius muscles from the implant control group appeared large and polygonal with a thin epimysial layer between fibers and a thick perimysium surrounding fascicles, indicating normal muscle fiber morphology. Similar morphology was also observed in muscle biopsies from BTX and SFN contralateral legs. Muscle paralysis in BTX T and SFN T appeared abnormal with thickened epimysial and perimysial layers compared to controls and small, atrophic muscle fibers. More fibrous tissue was observed in SFN legs, muscle fibers and fascicles appeared disorganized, and more nuclei surrounded the muscle fibers compared to other groups (Figure 7a).

Weight change was affected by both treatments (Figure 7b). Both models had lighter body weights compared to the implant control group and no difference between treatment groups. Muscle atrophy was observed in both models, as indicated by a significant loss in gastrocnemius weight (Figure 7c). Additionally, tibialis anterior weights were decreased in both models (Figure 7d). Without normalizing to contralateral legs, SFN reduced both gastrocnemius and tibialis anterior muscle weight compared to Control R and SFN C, while BTX did not affect muscle weight compared to Control R and BTX C (Supplemental Figure 11a, and b).

4. Discussion

For bone to be long-lasting and functioning, bone has to be innervated by sensory and sympathetic nerves. Osseointegration is a complex cascade of biological events, which needs the collaboration of various cell types, endocrine regulation, recruitment of blood vessels and nerve cells. In the present study, neural involvement and importance in osseointegration were emphasized.

Two models were used. The neurectomy model interrupted sensory nerve and sympathetic nerve innervation to both muscle and bone; therefore, motor neuron interaction with muscle was interrupted in the SFN model. In the botox injection model, only motor neuron interaction was interrupted but sensory and sympathetic nerves to the bone were still intact. Botox is also reported to mitigate pain in clinical use, inhibiting the release of pain mediators from peripheral nerve terminal, dorsal root ganglion, and spinal cord neuron⁴⁹.

Botox is effective for relieving muscle pain in muscle tone disorders, such as muscle spasm, by directly lessening the contractile activity of the muscle or indirectly via desensitization of nociceptors⁵⁰. Directly weakening muscles is through blocking the acetylcholine release from motor neurons at the neuromuscular junction. Indirect desensitization of nociceptors is also via blocking acetylcholine release because muscle hyperactivity compresses blood vessels that release agents that trigger nociceptors⁵¹. However, to the best of our knowledge, it is still unknown whether botox also relieves muscle pain directly through sensory neurons. Additionally, the specific axons present in the bone remain unclear²⁴, and no current literature demonstrates that intramuscular botox injection will affect bone sensory innervation. Therefore, the botox injection model was used in this study as a motor neuron intervention model to interrupt the motor neuron interaction and to investigate the effect of sympathetic and sensory innervation in bone. Future study needs to be performed to ensure that intramuscular muscle botox injection did not affect sensory neuron regulation of bone.

The comparison of two nerve intervention models enabled us to assess the direct role of the nervous system in modulating the bone phenotype and osseointegration. Modifying the gait in one limb will alter the gait in the contralateral limb and the mechanical instability will bring its own set of complications, including any systemic effects due to the release of factors in the treatment limb that might impact the contralateral limb^52,53. Using a separate set of animals with bilateral implants but not treated either by neurectomy or botox injection as the normal control group enabled us to eliminate this artifact. Thus, the effect of the neurectomy and treatment in the experimental limb and the contralateral limb could be compared to animals that did not have the neurectomy procedure or the treatment. In addition, each animal becomes its own control in terms of local v. systemic effects impacting osseointegration.

For this reason, the following analytical methods were used. Both limbs in all three groups were implanted, eliminating the implant surgery or the presence of the implant at time of harvest as confounding variables. By comparing the intervention leg to its contralateral leg in the same rat, the results were explored with minimum animal variations, the most clinically relevant situations, and accounting for each animal’s internal biological variables. However, when using only contralateral limbs as a control, due to the unintended loading effect from the intervention leg, this comparison can be skewed. By having a separate set of animals with implants in both hindlimbs as another control group, we were able to compare intervention models to healthy animals, and the limitation of the second comparison that intervention legs and control legs were from different animals would also be addressed. Moreover, all outcomes measured in this study did not show any difference between contralateral legs within the two models and the same side of the implant control group (Control L vs. BTX C vs. SFN C). This indicated that the nerve intervention on the right hindlimbs did not significantly disrupt the contralateral limbs in terms of the outcomes measured in this study. As a result, the focus of the discussion is mainly on the comparison between intervention legs and contralateral legs in the same animal.

Both nerve intervention models affected the bone phenotypes revealed by microCT analysis, but their effects differed. Compared to neurectomy, botox injection with only motor neuron interruption resulted in more bone loss with less trabecular bone formation. Cortical bone was also reduced, and the cortical bone porosity was increased, indicating that botox caused more bone loss overall, with unique effects on the cortical bone.

Both neurectomy and botox impaired osseointegration. Most of the reduction in osseointegration occurred in the cortical bone area, indicated by less BIC but no change was observed in the bone marrow area. This differs from our previous studies^14,39, which showed that in animals with ovariectomy-induced osteoporosis, the reduction of osseointegration was mainly found in the reduction of BIC in the trabecular bone. Botox-induced muscle paralysis resulted in less osseointegration than SFN, which correlated with the greater reduction in bone observed in the botox model. Moreover, peri-implant bone growth was the lowest in the BTX rats.

Histomorphometric analysis of peri-implant bone growth supported the microCT data and showed that total bone to implant contact was reduced in both nerve intervention models. Cortical bone to implant contact was not affected by botox; instead, botox injection decreased the bone marrow to implant contact. In contrast, neurectomy affected both cortical bone to implant contact and bone marrow to implant contact. Combining both microCT and histology results, botox injection retarded bone formation. MicroCT indicated that there was a less mature bone formation in the cortical bone area in comparison to neurectomy, and histology confirmed that there was still uncalcified osteoid. In contrast, neurectomy resulted in a faster bone formation process compared to botox injections. Thus, the cortical bone area consisted mainly of mineralized bone, supporting the microCT analysis. Despite their differences in presentation, both nerve intervention models impaired bone marrow to implant contact, suggesting that they had the potential to decrease the BIC in the bone marrow area at a later time point. However, the small sample size used for histology, limits our ability to draw a definitive conclusion. Future studies should take into account the study complications and use a larger sample size.

Mechanical testing confirmed that neural intervention reduced osseointegration. By measuring removal torque, we were able to assess the maximum force to cause implant failure; torsional stiffness, which is the linear region or elastic region of the torque-radian graph indicating the recovery ability of bone under applied torque and ability to return the applied force; and torsional energy, which is the energy absorbed by the bone before any permanent damage⁵⁴. Botox injection decreased torsional stiffness and maximum torque, which can be caused by either unloading for the course of osseointegration or the thinning of the cortices before implantation. In contrast, neurectomy did not affect these mechanical properties of the bone to implant bonds, although it did result in lower torsional stiffness compared to the contralateral legs. These mechanical results correlated with the microCT and histomorphometric findings, indicating that botox had the most deleterious effects on osseointegration.

Our in vitro studies using primary osteoblasts and bone marrow stromal cells isolated from bone and bone marrow of the affected tibias and their contralateral controls, provided further information on the specific roles that neural regulation plays in osseointegration. However, due to the limitation of femur availability for primary cell isolation, rOBs and rBMSCs were isolated from tibias. Examining cells from tibias instead of femurs was not the most representative situation to evaluate the cells responsible for osseointegration. However, as mentioned above, neurectomy was conducted at the proximal end of femurs, which caused systemic denervation to the entire hindlimbs. Botox was injected into the muscles involved in both femur and tibia movement. Therefore, the tibia was consequently affected by both nerve interventions, so it is reasonable to infer that cells were affected similarly to the situation in the femur.

We used a cell culture system developed in our laboratory to assess the response of osteoblast lineage cells to clinically used Ti implants with a microtextured, hydrophilic surface^3,5,55,56. Neurectomy and botox injection impaired the proliferative capacity compared to cells isolated from botox contralateral legs. We did not investigate whether the rate of proliferation was affected; thus, the lower numbers of cells at harvest may also have reflected a smaller osteoblast population at the outset of the culture.

The effect of botox and neurectomy did not impact the enhanced osteoblast phenotype seen when BMSCs are cultured on modSLA substrates compared to TCPS. However, osteoblasts isolated from bone chips did exhibit treatment-specific differences in response. Botox caused a marked increase in the production of osteocalcin, BMP2, VEGF 165, osteoprotegerin, osteopontin, and sema3C compared to cells from the contralateral tibias and this was even greater when cells were cultured on modSLA. When cells were cultured on modSLA, cells from botox injected legs exhibited increased production of sema3A as well.

These results refute the hypothesis that nerve intervention impaired osseointegration through impeding cells’ ability to differentiate into osteoblasts on microrough and hydrophilic titanium surfaces. Osteoblasts isolated from denervated and botox injected legs produced more osteopontin when growth on modSLA compared to osteoblasts from their contralateral control legs. Moreover, osteoblasts from botox treated limbs exhibited the most robust response to modSLA.

Osteoblasts have been reported to be able to produce neuronal factors such as semaphorins that favor and regulate nerve fiber ingrowth and bone homeostasis^19,30,57,58. Sema3A and sema3C are class 3 semaphorins, which were originally identified as axon guidance proteins. Sema3A has been shown by our lab to be produced by BMSCs, and its production is sensitive to titanium surface characteristics³³. In addition, sema3A increased osteogenic differentiation of BMSCs on the microtextured titanium surfaces³³ and increased bone formation in vivo⁵⁹. Here we showed that sema3C is also produced by osteoblasts on the modSLA surface. These results suggest that nerve intervention may regulate bone formation and osseointegration by affecting the production of neuronal factors.

Sema3A has been reported to have osteoprotective roles in bone remodeling³¹ and sema3C is a well-known axon attractant⁶⁰. Since nerve intervention did not change surface-induced osteoblastic differentiation of BMSCs on modSLA but it did impact osseointegration in vivo, it is more likely that the effect of denervation is to cause downregulation of neuronal factors that are important for bone remodeling. Osteoblasts isolated from all groups except BTX T lost the ability to secrete higher amounts of sema3A and sema3C on modSLA. It is possible that botox treatment results in more recruitment of nerve fibers into the bone system and attempt to mitigate the reduction of bone formation and osseointegration.

Bone loss is associated with muscle atrophy or muscle inactivity, either caused by disuse, neuromuscular pathology, or post-traumatic condition³⁷. In this study, both neurectomy and botox injection models introduced muscle paralysis, but the major difference between these models is that neurectomy results in lost trophic signaling and sensation. The difficulty of distinguishing whether the effect on osseointegration is from muscle paralysis or nervous system interruption is a common limitation for nerve intervention models. We used the botox injection model as a muscle paralysis model to help tease out the direct neuronal effect on osseointegration; botox induced muscle paralysis by blocking the neuromuscular junction while innervation remained intact. More, sensory and sympathetic nerves were left intact following treatments. Whether the direct neuronal effect on osseointegration in this study is through the induction of mechanical unloading or the direct neuronal effect on bone was not evident. To resolve this, muscle atrophy and morphology in both models were examined to demonstrate the correlation between muscle atrophy and osseointegration. Muscle weights and histological assessment of collagen content using aniline blue staining are commonly used confirmation indicators of atrophy. Gastrocnemius and tibialis anterior muscles, which were at the distal regions of implant insertion sites, were chosen to be examined to delineate between changes caused by the implant insertion surgery and the nerve intervention. Gastrocnemius and tibialis anterior muscle mass were significantly reduced in both models, and surprisingly, neurectomy resulted in less muscle mass than botox injection. Additionally, muscle morphology qualitatively showed that neurectomy legs had the smallest muscle fiber diameters and the most collagenous matrix.

The abnormal muscle morphology in both models confirmed the effects on muscle atrophy, and the difference in muscle mass and morphology between neurectomy legs and botox injection legs indicated that denervation caused more muscle damage than botox injection. The observation that SFN rats with more muscle atrophy showed relatively higher trabecular bone formation and better osseointegration than BTX rats implied the prevention of direct neural regulation (both nerve sensation and sympathetic innervation) might be able to modify muscle paralysis-induced bone loss and regulate impaired osseointegration. Neurectomy with more muscle atrophy than caused by botox injection did not worsen bone formation and osseointegration, which indicated the nervous system played a role in muscle paralysis induced bone change and osseointegration.

Even though the endocrine regulation of bone homeostasis is well studied and established, the participation of the nervous system in local bone remodeling is a burgeoning field, especially after the discovery of neural innervation and presence in bone. Bone is highly innervated by sensory nerves and sympathetic nerves. Loss of nerve sensation to the bone could impair bone mass accrual⁶¹. Capsaicin-induced sensory denervation also caused a trabecular bone reduction in the proximal tibia⁶², decreased the quality of bone around the implant by reducing the bonding force of the implant to the bone, and therefore, impaired osseointegration in rats⁶³. The effect of sympathetic tone on bone formation is not well understood and showed controversial results in different studies^27,64–66. Sympathetic neurons cultured with osteoblasts in a transwell co-culture system with no cell-to-cell contact promoted osteoblastic differentiation through BMP signaling⁶⁴. Other studies demonstrated that sympathetic nerve activity through activation of β2-adrenergic receptors present in osteoblasts could exaggerate bone resorption⁶⁷. Additionally, β-agonist treatment can increase osteoclast differentiation, which results in the upregulation of bone resorption⁶⁸. In our neurectomy model, the sympathetic denervation may be responsible for the higher bone formation and osseointegration compared to the botox model with sympathetic tone present, and this statement will be further investigated in the future by measuring β2-adrenergic receptors expression from primarily isolated osteoblast among all groups and the regulation of sympathetic neurons on osteoblasts response to microrough and hydrophilic titanium surfaces in vitro.

This study enabled the evaluation of direct neural regulation in osseointegration. Clinically, demand for implants in patients with nerve injuries, such as severe brain trauma; thus, translation of our outcomes to the clinical situation may be limited. However, the revelation of the importance of neural regulation in osseointegration can widen the application of neuronal factors on improving osseointegration in compromised patients, such as those with osteoporosis. Future research is needed to investigate the mechanisms that underlie the effect, such as whether the effect is on bone formation, bone resorption, or bone remodeling as well as the neural signaling pathways that are impacted.

5. Conclusion

Both nerve intervention models caused significant bone reduction and negatively impact the osseointegration of titanium implants. Selected motor neuron blockage by botox injection led to more reduction in bone formation and caused less osseointegration but less effect on muscle atrophy and morphology. The conflict observed from effects on bone phenotype and osseointegration and effects on muscle in both nerve intervention models emphasize the involvement of neural regulation in osseointegration.

Supplementary Material

Supplemental Figure 1. Unicortical implant surgery. Naïve bone after drilling a hole but before implant insertion (a); naïve bone after implant inserted (b). Brown tissue indicated cortical bone and blue tissue indicated trabecular bone.

NIHMS1735014-supplement-1.tif^{(6.9MB, tif)}

Supplemental Figure 2. Demonstration of three different ways of analyzing bone to implant contact and bone volume/total volume. Total bone to implant contact (a) was defined by the area having bone tissue contacting with implant surfaces normalized to the area covered by the blue line, which included the cortical bone region and bone marrow area. Bone marrow to implant contact (b) was defined by the area having bone tissue contacting implant surfaces normalized to the area covered by the purple line, which covered the bone marrow area. Cortical bone to implant contact (c) was defined as the area having bone tissue contacting implant surfaces normalized to the area covered by the green line, which covered the cortical bone region. After removing the implant using the CtAn program, total bone volume/total volume (d) was defined as the volume of cortical bone and trabecular bone in the uniform shaded box normalized to the volume of the shaded box. Bone marrow volume/total volume (e) was defined as the volume of trabecular bone in the bone marrow area in the shaded box normalized to the volume of the shaded box. Cortical bone volume/total volume (f) was defined as the volume of cortical bone in the shaded box normalized to the volume of the shaded box. Scale bar=1mm

NIHMS1735014-supplement-2.tif^{(7.9MB, tif)}

Supplemental Figure 3. Characterization of the neurectomy and botox injection induced bone phenotype change at the distal femur. Femurs were harvested and placed in 1X PBS for microCT scanning. The trabecular bone phenotype at the metaphysis of the distal femur was assessed as bone volume/total volume (a), trabecular number (b), total porosity (c), and trabecular thickness (d). The cortical bone phenotype at the metaphysis of the distal femur was assessed as bone volume/total volume (e), total porosity (f), mean total cross-sectional bone area (g), mean cross-sectional bone perimeter (h), and cortical thickness (i). The cortical bone phenotype at the mid-diaphysis was assessed as BV/TV (j), total porosity (k), mean total cross sectional bone area (l), mean cross sectional bone perimeter (m), and cortical thickness (n). Data shown are the median ± range of n=5 for Control L, and Control R, n=7 for BTX C, BTX T, and n=7 for SFN C, SFN T. Data within each group not sharing a letter are significantly different at an α = 0.05 by Kruskal-Wallis test with Dunns’ post correction.

NIHMS1735014-supplement-3.tif^{(4.4MB, tif)}

Supplemental Figure 4. Characterization of the neurectomy and botox injection induced bone phenotype change at the femoral head. Femurs were harvested and placed in 1X PBS for microCT scanning. 3D microCT reconstructions were made of the femoral heads. The trabecular bone phenotype was assessed as bone volume/total volume (a), trabecular thickness (b), trabecular number (c), and total porosity (d). The cortical bone phenotype was assessed as bone volume/total volume (e), total porosity (f), mean total cross-sectional bone area (g), mean cross-sectional bone perimeter (h), and cortical thickness (i). Data shown are the means of treatment leg/contralateral leg for each group ± standard error of n=5 for control rats, n=7 for BTX rats, and n=7 for SFN rats. Data within each group not sharing a letter are significantly different at an α = 0.05 by one-way ANOVA with Tukey post correction. Comparisons between treatment legs and control legs identified with an asterix are statistically different at α = 0.05 by Wilcoxon matched-paired signed rank test.

NIHMS1735014-supplement-4.tif^{(4.3MB, tif)}

Supplemental Figure 5. Characterization of the neurectomy and botox injection induced bone phenotype change at the femoral head. Femurs were harvested and placed in 1X PBS for microCT scanning. 3D microCT reconstructions were made of the femoral heads. The trabecular bone phenotype was assessed as bone volume/total volume (a), trabecular thickness (b), trabecular number (c), and total porosity (d). The cortical bone phenotype was assessed as bone volume/total volume (e), total porosity (f), mean total cross sectional bone area (g), mean cross sectional bone perimeter (h), and cortical thickness (i). Data shown are the median ± range of n=5 for Control L, and Control R, n=7 for BTX rats, and n=7 for SFN rats. Data within each group not sharing a letter are significantly different at an α = 0.05 by Kruskal-Wallis test with Dunns’ post correction.

NIHMS1735014-supplement-5.tif^{(4.7MB, tif)}

Supplemental Figure 6. Characterization of the bone phenotype change at the distal end of the femur three weeks after neurectomy surgery and botox injection. (n=1 for each group.)

NIHMS1735014-supplement-6.tif^{(2.9MB, tif)}

Supplemental Figure 7. The effect of sciatic and femoral neurectomy and botox injection on implant osseointegration assessed by microCT. Distal femurs were harvested and placed in 1X PBS. Assessments included: total bone to implant contact (a), bone marrow to implant contact (b), and cortical bone to implant contact (c), as well as total bone volume/total volume (d), bone marrow volume/total volume (e), and cortical bone volume/total volume (f). Data shown are the median ± range of n=5 for Control L, and Control R, n=7 for BTX C, BTX T and n=7 for SFN C, SFN T. Data not sharing a letter are significantly different at an α = 0.05 by Kruskal-Wallis test with Dunns’ post correction.

NIHMS1735014-supplement-7.tif^{(3.4MB, tif)}

Supplemental Figure 8. Characterization of the bone phenotype at the distal end of the femur by microCT for the repeat sciatic and femoral neurectomy and botox injection experiments. Trabecular bones were quantified by microCT reconstruction as bone volume/total volume (a), trabecular thickness (b), trabecular number (c), and total porosity (d). Cortical bones were quantified by microCT reconstruction as bone volume/total volume (e), total porosity (f), mean total cross sectional bone area (g), mean cross sectional bone perimeter (h), and cortical thickness (i). Data shown are the median of treatment leg over the contralateral leg for each group ± range of n=3 for BTX rats and n=3 for SFN rats. Groups not sharing a letter are significantly different at an α = 0.05 by one tailed paired t-test.

NIHMS1735014-supplement-8.tif^{(4.1MB, tif)}

Supplemental figure 9. The effect of sciatic and femoral neurectomy and botox injection on implant osseointegration was assessed by removal torque mechanical tests with maximum torque (a), torsional stiffness (b), and torsional energy (c). Data shown are the median ± standard error of median ± range of n=5 for Control L, and Control R, n=7 for BTX C, BTX T, and n=7 for SFN C, SFN T. Groups not sharing a letter are significant at an α = 0.05 by Kruskal-Wallis test with Dunns’ post correction.

NIHMS1735014-supplement-9.tif^{(2.7MB, tif)}

Supplemental Figure 10. Phenotypic characterization of rat osteoblasts isolated from tibial cortical bone. Primary osteoblasts were isolated from tibial bone from each experimental group and cultured separately on TCPS and modSLA. On day 7, cells cultured on TCPS and modSLA were treated with fresh DMEM FM. Media were collected 24 hours later and assayed for osteocalcin content. Data shown are the ratio of osteocalcin in conditioned media from modSLA cultures compared to conditioned from TCPS cultures and are presented as mean ± standard error of n=6 independent cultures for each group. Groups not sharing a letter are significantly different at α=0.05 by one-way ANOVA with Tukey post correction. Data indicated by a hashtag are significantly different from its TCPS control by unpaired t-test.

NIHMS1735014-supplement-10.tif^{(2.6MB, tif)}

Supplemental Figure 11. Characterization of the gastrocnemius and tibialis anterior muscles from control, BTX, and SFN rats: gastrocnemius weight (a) and tibialis anterior weight (b). Data shown are the median ± range of n=5 for Control L, and Control R, n=7 for BTX C, BTX T, and n=7 for SFN C, SFN T. Groups not sharing a letter are significantly different at an α = 0.05 by Kruskal-Wallis test with Dunns’ post correction.

NIHMS1735014-supplement-11.tif^{(2.4MB, tif)}

Highlights.

Botox-induced muscle paralysis causes more bone loss than neurectomy.
Direct neural involvement in the osseointegration of titanium implants is emphasized.
Nerve intervention may regulate osseointegration by affecting neuronal factor production.

6. Acknowledgements

Institut Straumann AG (Basel, Switzerland) provided the microrough and hydrophilic titanium implants and implant surfaces, and support for this study. We would also like to thank Lucas Olson, Tri Nguyen, and Dane Neilson for their help with muscle histology. Additional support was provided by the National Institute of Arthritis and Musculoskeletal and Skin Disease of the National Institute of Health under Award Numbers R01AR052102 and R01AR072500: The graphical abstract is created with BioRender.com. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The study is in partial fulfillment of the Ph.D. degree for Jingyao Deng.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

1.Mach DB, Rogers SD, Sabino MC, et al. Origins of skeletal pain: Sensory and sympathetic innervation of the mouse femur. Neuroscience. 2002;113(1):155–166. doi: 10.1016/S0306-4522(02)00165-3 [DOI] [PubMed] [Google Scholar]
2.Lee JWY, Bance ML. Physiology of osseointegration. Otolaryngol Clin North Am. 2019. doi: 10.1016/j.otc.2018.11.004 [DOI] [PubMed] [Google Scholar]
3.Boyan BD, Lotz EM, Schwartz Z. Roughness and hydrophilicity as osteogenic biomimetic surface properties. Tissue Eng - Part A. 2017;23(23–24):1479–1489. doi: 10.1089/ten.tea.2017.0048 [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Sims NA, Gooi JH. Bone remodeling: Multiple cellular interactions required for coupling of bone formation and resorption. Semin Cell Dev Biol. 2008;19(5):444–451. doi: 10.1016/j.semcdb.2008.07.016 [DOI] [PubMed] [Google Scholar]
5.Boyan BD, Cheng A, Olivares-Navarrete R, Schwartz Z. Implant surface design regulates mesenchymal stem cell differentiation and maturation. Adv Dent Res. 2016;28(1):10–17. doi: 10.1177/0022034515624444 [DOI] [PMC free article] [PubMed] [Google Scholar]
6.Schwartz Z, Lotz EM, Berger MB, Boyan BD. 7.22 The effect of substrate microtopography on osteointegration of titanium implants ☆. Compr Biomater II. 2017;7:429–443. doi: 10.1016/b978-0-12-803581-8.10218-8 [DOI] [Google Scholar]
7.Olivares-Navarrete R, Hyzy SL, Park JH, et al. Mediation of osteogenic differentiation of human mesenchymal stem cells on titanium surfaces by a Wnt-integrin feedback loop. Biomaterials. 2011;32(27):6399–6411. doi: 10.1016/j.biomaterials.2011.05.036 [DOI] [PMC free article] [PubMed] [Google Scholar]
8.Lai M, Hermann CD, Cheng A, et al. Role of α2β1 integrins in mediating cell shape on microtextured titanium surfaces. J Biomed Mater Res - Part A. 2015;103(2):564–573. doi: 10.1002/jbm.a.35185 [DOI] [PMC free article] [PubMed] [Google Scholar]
9.Buser D, von Arx T, ten Bruggenkate C, Weingart D. Basic surgical principles with ITI implants. Clin Oral Implants Res. 2000;11 Suppl 1:59–68. doi: 10.1034/j.1600-0501.2000.011S1059.x [DOI] [PubMed] [Google Scholar]
10.Klokkevold PR, Han MSTJ. How do smoking, diabetes and periodontitis affect outcomes of implant treatment? Br Dent J. 2007;203(6):333–333. doi: 10.1038/bdj.2007.841 [DOI] [PubMed] [Google Scholar]
11.Gaetti-Jardim EC, Santiago-Junior JF, Goiato MC, Pellizer EP, Magro-Filho O, Jardim EG. Dental implants in patients with osteoporosis: A clinical reality? J Craniofac Surg. 2011;22(3):1111–1113. doi: 10.1097/SCS.0b013e3182108ec9 [DOI] [PubMed] [Google Scholar]
12.Mohammad AR, Hooper DA, Vermilyea SG, Mariotti A, Preshaw PM. An investigation of the relationship between systemic bone density and clinical periodontal status in post-menopausal Asian-American women. Int Dent J. 2003;53(3):121–125. doi: 10.1111/j.1875-595X.2003.tb00735.x [DOI] [PubMed] [Google Scholar]
13.Moedano DE, Irigoyen ME, Borges-Yáñez A, Flores-Sánchez I, Rotter RC. Osteoporosis, the risk of vertebral fracture, and periodontal disease in an elderly group in Mexico City. Gerodontology. 2011;28(1):19–27. doi: 10.1111/j.1741-2358.2009.00342.x [DOI] [PubMed] [Google Scholar]
14.Lotz EM, Cohen DJ, Schwartz Z, Boyan BD. Titanium implant surface properties enhance osseointegration in ovariectomy induced osteoporotic rats without pharmacologic intervention. Clin Oral Implants Res. 2020;31(4):374–387. doi: 10.1111/clr.13575 [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Hao J, Zhang Y, Jing D, et al. Mechanobiology of mesenchymal stem cells: Perspective into mechanical induction of MSC fate. Acta Biomater. 2015;20:1–9. doi: 10.1016/j.actbio.2015.04.008 [DOI] [PubMed] [Google Scholar]
16.Berger MB, Cohen DJ, Olivares-Navarrete R, et al. Human osteoblasts exhibit sexual dimorphism in their response to estrogen on microstructured titanium surfaces. Biol Sex Differ. 2018;9(1):1–10. doi: 10.1186/s13293-018-0190-x [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Bonewald LF, Kester MB, Schwartz Z, et al. Effects of combining transforming growth factor β and 1,25- dihydroxyvitamin D3 on differentiation of a human osteosarcoma (MG-63). J Biol Chem. 1992;267(13):8943–8949. [PubMed] [Google Scholar]
18.Boyan BD, Batzer R, Kieswetter K, et al. Titanium surface roughness alters responsiveness of MG63 osteoblast- like cells to 1α,25-(OH)2D3. J Biomed Mater Res. 1998;39(1):77–85. doi: [DOI] [PubMed] [Google Scholar]
19.Kim BJ, Koh JM. Coupling factors involved in preserving bone balance. Cell Mol Life Sci. 2019;76(7):1243–1253. doi: 10.1007/s00018-018-2981-y [DOI] [PMC free article] [PubMed] [Google Scholar]
20.Bishop GB, Einhorn TA. Current and future clinical applications of bone morphogenetic proteins in orthopaedic trauma surgery. Int Orthop. 2007;31(6):721–727. doi: 10.1007/s00264-007-0424-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
21.Berger MB, Bosh KB, Jacobs TW, Joshua Cohen D, Schwartz Z, Boyan BD. Growth factors produced by bone marrow stromal cells on nanoroughened titanium–aluminum–vanadium surfaces program distal MSCs into osteoblasts via BMP2 signaling. J Orthop Res. 2020;(September):1–13. doi: 10.1002/jor.24869 [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Huang S, Li Z, Liu Y, et al. Neural regulation of bone remodeling: Identifying novel neural molecules and pathways between brain and bone. J Cell Physiol. 2019;234(5):5466–5477. doi: 10.1002/jcp.26502 [DOI] [PubMed] [Google Scholar]
23.Negishi-Koga T, Takayanagi H. Bone cell communication factors and Semaphorins. Bonekey Rep. 2012;1(September):183. doi: 10.1038/bonekey.2012.183 [DOI] [PMC free article] [PubMed] [Google Scholar]
24.Brazill JM, Beeve AT, Craft CS, Ivanusic JJ, Scheller EL. Nerves in Bone: Evolving Concepts in Pain and Anabolism. J Bone Miner Res. 2019;34(8):1393–1406. doi: 10.1002/jbmr.3822 [DOI] [PMC free article] [PubMed] [Google Scholar]
25.Ji-Ye H, Xin-Feng Z, Lei-Sheng J. Autonomic control of bone formation: Its clinical relevance. Handb Clin Neurol. 2013;117:161–171. doi: 10.1016/B978-0-444-53491-0.00014-6 [DOI] [PubMed] [Google Scholar]
26.He JY, Jiang LS, Dai LY. The roles of the sympathetic nervous system in osteoporotic diseases: A review of experimental and clinical studies. Ageing Res Rev. 2011;10(2):253–263. doi: 10.1016/j.arr.2011.01.002 [DOI] [PubMed] [Google Scholar]
27.Elefteriou F, Campbell P, Ma Y. Control of bone remodeling by the peripheral sympathetic nervous system. Calcif Tissue Int. 2014;94(1):140–151. doi: 10.1007/s00223-013-9752-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
28.Zochodne DW. Local blood flow in peripheral nerves and their ganglia: Resurrecting key ideas around its measurement and significance. Muscle and Nerve. 2018;57(6):884–895. doi: 10.1002/mus.26031 [DOI] [PubMed] [Google Scholar]
29.Elefteriou F Neuronal signaling and the regulation of bone remodeling. Cell Mol Life Sci. 2005;62(19–20):2339–2349. doi: 10.1007/s00018-005-5175-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
30.Kang S, Kumanogoh A. Semaphorins in bone development, homeostasis, and disease. Semin Cell Dev Biol. 2013;24(3):163–171. doi: 10.1016/j.semcdb.2012.09.008 [DOI] [PubMed] [Google Scholar]
31.Xu R. Semaphorin 3A a new player in bone remodeling. Cell Adhes Migr. 2014;8(1):5–10. doi: 10.4161/cam.27752 [DOI] [PMC free article] [PubMed] [Google Scholar]
32.Li Z, Hao J, Duan X, et al. The role of semaphorin 3A in bone remodeling. Front Cell Neurosci. 2017;11(February):1–8. doi: 10.3389/fncel.2017.00040 [DOI] [PMC free article] [PubMed] [Google Scholar]
33.Lotz EM, Berger MB, Boyan BD, Schwartz Z. Regulation of mesenchymal stem cell differentiation on microstructured titanium surfaces by semaphorin 3A. Bone. 2020;134(January):115260. doi: 10.1016/j.bone.2020.115260 [DOI] [PMC free article] [PubMed] [Google Scholar]
34.Ma X, Lv J, Sun X, et al. Naringin ameliorates bone loss induced by sciatic neurectomy and increases Semaphorin 3A expression in denervated bone. Sci Rep. 2016;6(122):1–14. doi: 10.1038/srep24562 [DOI] [PMC free article] [PubMed] [Google Scholar]
35.Thomsen JS, Christensen LL, Vegger JB, Nyengaard JR, Brüel A. Loss of bone strength is dependent on skeletal site in disuse osteoporosis in rats. Calcif Tissue Int. 2012;90(4):294–306. doi: 10.1007/s00223-012-9576-7 [DOI] [PubMed] [Google Scholar]
36.Bettis T, Kim BJ, Hamrick MW. Impact of muscle atrophy on bone metabolism and bone strength: implications for muscle-bone crosstalk with aging and disuse. Osteoporos Int. 2018;29(8):1713–1720. doi: 10.1007/s00198-018-4570-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
37.Gatti V, Ghobryal B, Gelbs MJ, et al. Botox-induced muscle paralysis alters intracortical porosity and osteocyte lacunar density in skeletally mature rats. J Orthop Res. 2019;37(5):1153–1163. doi: 10.1002/jor.24276 [DOI] [PubMed] [Google Scholar]
38.Warner SE, Sanford DA, Becker BA, Bain SD, Srinivasan S, Gross TS. Botox induced muscle paralysis rapidly degrades bone. Bone. 2006;38(2):257–264. doi: 10.1016/j.bone.2005.08.009 [DOI] [PMC free article] [PubMed] [Google Scholar]
39.Lotz EM, Cohen DJ, Ellis RA, Wayne JS, Schwartz Z, Boyan BD. Ibandronate treatment before and after implant insertion impairs osseointegration in aged rats with ovariectomy induced osteoporosis. JBMR Plus. 2019;3(7):e10184. doi: 10.1002/jbm4.10184 [DOI] [PMC free article] [PubMed] [Google Scholar]
40.Lotz EM, Lohmann CH, Boyan BD, Schwartz Z. Bisphosphonates inhibit surface-mediated osteogenesis. J Biomed Mater Res - Part A. 2020;108(8):1774–1786. doi: 10.1002/jbm.a.36944 [DOI] [PMC free article] [PubMed] [Google Scholar]
41.Schneider CA, Rasband WS, Eliceiri KW. NIH Image to ImageJ: 25 years of image analysis. Nat Methods. 2012;9(7):671–675. doi: 10.1038/nmeth.2089 [DOI] [PMC free article] [PubMed] [Google Scholar]
42.McClure MJ, Cohen DJ, Ramey AN, et al. Decellularized muscle supports new muscle fibers and improves function following volumetric injury. Tissue Eng - Part A. 2018;24(15–16):1228–1241. doi: 10.1089/ten.tea.2017.0386 [DOI] [PubMed] [Google Scholar]
43.Bakker AD, Klein-Nulend J. Osteoblast isolation from murine calvaria and long bones. Methods Mol Biol. 2012;816:19–29. doi: 10.1007/978-1-61779-415-5_2 [DOI] [PubMed] [Google Scholar]
44.McMillan J, Kinney RC, Ranly DM, Fatehi-Sedeh S, Schwartz Z, Boyan BD. Osteoinductivity of demineralized bone matrix in immunocompromised mice and rats is decreased by ovariectomy and restored by estrogen replacement. Bone. 2007;40(1):111–121. doi: 10.1016/j.bone.2006.07.022 [DOI] [PubMed] [Google Scholar]
45.Olivares-Navarrete R, Raines AL, Hyzy SL, et al. Osteoblast maturation and new bone formation in response to titanium implant surface features are reduced with age. J Bone Miner Res. 2012;27(8):1773–1783. doi: 10.1002/jbmr.1628 [DOI] [PMC free article] [PubMed] [Google Scholar]
46.Nakamura H, Morita S, Kumei Y, et al. Effects of neurectomy and tenotomy on the bone mineral density and strength of tibiae. Acta Astronaut. 2001;49(3–10):179–190. doi: 10.1016/S0094-5765(01)00097-2 [DOI] [PubMed] [Google Scholar]
47.Erdfelder E, FAul F, Buchner A, Lang AG. Statistical power analyses using G*Power 3.1: Tests for correlation and regression analyses. Behav Res Methods. 2009;41(4):1149–1160. doi: 10.3758/BRM.41.4.1149 [DOI] [PubMed] [Google Scholar]
48.Faul F, Erdfelder E, Lang AG, Buchner A. G*Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behav Res Methods. 2007;39(2):175–191. doi: 10.3758/BF03193146 [DOI] [PubMed] [Google Scholar]
49.Park JH, Park HJ. Botulinum toxin for the treatment of neuropathic pain. Toxins (Basel). 2017;9(9). doi: 10.3390/toxins9090260 [DOI] [PMC free article] [PubMed] [Google Scholar]
50.Intiso D Therapeutic use of botulinum toxin in neurorehabilitation. J Toxicol. 2012;2012(June):1–18. doi: 10.1155/2012/802893 [DOI] [PMC free article] [PubMed] [Google Scholar]
51.Mense S Neurobiological basis for the use of botulinum toxin in pain therapy. J Neurol Suppl. 2004;251(1):1–7. doi: 10.1016/j.toxicon.2012.07.027 [DOI] [PubMed] [Google Scholar]
52.Einhorn TA; Simon G; Devlin VJ; Warman J; Sidhu SP; Vigorita VJ. The osteogenic response to distant skeletal injury. J Bone Jt Surg. 1990;72(9):1374–1378. [PubMed] [Google Scholar]
53.Marshall TS, Schwartz Z, Swain LD, Amir D, Sela J, Gross U, Muller-mai BDB C. Matrix vesicle enzyme activity in endosteal bone following implantation of bonding and non.-bonding implant materials. 1991:112–120. doi:99123188140001101 [DOI] [PubMed] [Google Scholar]
54.Turner CH. Bone Strength : Current Concepts. 2006;446:429–446. doi: 10.1196/annals.1346.039 [DOI] [PubMed] [Google Scholar]
55.Olivares-Navarrete R, Hyzy SL, Pan Q, et al. Osteoblast maturation on microtextured titanium involves paracrine regulation of bone morphogenetic protein signaling. J Biomed Mater Res - Part A. 2015;103(5):1721–1731. doi: 10.1002/jbm.a.35308 [DOI] [PMC free article] [PubMed] [Google Scholar]
56.Lotz EM, Olivares-navarrete R, Berner S, Boyan BD, Schwartz Z. Osteogenic response of human MSCs and osteoblasts to hydrophilic and hydrophobic nanostructured titanium implant surfaces. 2016:3137–3148. doi: 10.1002/jbm.a.35852 [DOI] [PubMed] [Google Scholar]
57.Chen X, Wang Z, Duan N, Zhu G, Schwarz EM, Xie C. Osteoblast–osteoclast interactions. Connect Tissue Res. 2018;59(2):99–107. doi: 10.1080/03008207.2017.1290085 [DOI] [PMC free article] [PubMed] [Google Scholar]
58.Kim B-J, Koh J-M. Coupling factors involved in preserving bone balance. Cell Mol Life Sci. 2018;(0123456789). doi: 10.1007/s00018-018-2981-y [DOI] [PMC free article] [PubMed] [Google Scholar]
59.Fukuda T, Takeda S, Xu R, et al. Sema3A regulates bone-mass accrual through sensory innervations. Nature. 2013;497(7450):490–493. doi: 10.1038/nature12115 [DOI] [PubMed] [Google Scholar]
60.Hao J, Yu J. Semaphorin 3C and its receptors in cancer and cancer stem-like cells. Biomedicines. 2018;6(2):42. doi: 10.3390/biomedicines6020042 [DOI] [PMC free article] [PubMed] [Google Scholar]
61.Fukuda T, Takeda S, Xu R, et al. Sema3A regulates bone-mass accrual through sensory innervations. Nature. 2013;497(7450):490–493. doi: 10.1038/nature12115 [DOI] [PubMed] [Google Scholar]
62.Zhang ZK, Guo X, Lao J, Qin YX. Effect of capsaicin-sensitive sensory neurons on bone architecture and mechanical properties in the rat hindlimb suspension model. J Orthop Transl. 2017;10:12–17. doi: 10.1016/j.jot.2017.03.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
63.Huang B, Ye J, Zeng X, Gong P. Effects of capsaicin-induced sensory denervation on early implant osseointegration in adult rats. R Soc Open Sci. 2019;6(1):0–9. doi: 10.1098/rsos.181082 [DOI] [PMC free article] [PubMed] [Google Scholar]
64.He JY, Zheng XF, Jiang SD, Chen XD, Jiang LS. Sympathetic neuron can promote osteoblast differentiation through BMP signaling pathway. Cell Signal. 2013;25(6):1372–1378. doi: 10.1016/j.cellsig.2013.02.016 [DOI] [PubMed] [Google Scholar]
65.Yao Q, Zeng Y, Feng Y, Wu H, Liang H, Gong P. Chemical sympathectomy impairs peri-implant osseointegration in mice: Role of the sympathetic nervous system in osseointegration. Int J Oral Maxillofac Implant. 2019;34(1):91–98. doi: 10.11607/jomi.7086 [DOI] [PubMed] [Google Scholar]
66.Corr A, Smith J, Baldock P. Neuronal control of bone remodeling. Toxicol Pathol. 2017;45(7):894–903. doi: 10.1177/0192623317738708 [DOI] [PubMed] [Google Scholar]
67.Cizza G, Primma S, Csako G. Depression as a risk factor for osteoporosis. Trends Endocrinol Metab. 2009;20(8):367–373. doi: 10.1016/j.tem.2009.05.003 [DOI] [PMC free article] [PubMed] [Google Scholar]
68.Bonnet N, Benhamou CL, Brunet-Imbault B, et al. Severe bone alterations under β2 agonist treatments: Bone mass, microarchitecture and strength analyses in female rats. Bone. 2005;37(5):622–633. doi: 10.1016/j.bone.2005.07.012 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

NIHMS1735014-supplement-1.tif^{(6.9MB, tif)}

NIHMS1735014-supplement-2.tif^{(7.9MB, tif)}

NIHMS1735014-supplement-3.tif^{(4.4MB, tif)}

NIHMS1735014-supplement-4.tif^{(4.3MB, tif)}

NIHMS1735014-supplement-5.tif^{(4.7MB, tif)}

Supplemental Figure 6. Characterization of the bone phenotype change at the distal end of the femur three weeks after neurectomy surgery and botox injection. (n=1 for each group.)

NIHMS1735014-supplement-6.tif^{(2.9MB, tif)}

NIHMS1735014-supplement-7.tif^{(3.4MB, tif)}

NIHMS1735014-supplement-8.tif^{(4.1MB, tif)}

NIHMS1735014-supplement-9.tif^{(2.7MB, tif)}

NIHMS1735014-supplement-10.tif^{(2.6MB, tif)}

NIHMS1735014-supplement-11.tif^{(2.4MB, tif)}

[R1] 1.Mach DB, Rogers SD, Sabino MC, et al. Origins of skeletal pain: Sensory and sympathetic innervation of the mouse femur. Neuroscience. 2002;113(1):155–166. doi: 10.1016/S0306-4522(02)00165-3 [DOI] [PubMed] [Google Scholar]

[R2] 2.Lee JWY, Bance ML. Physiology of osseointegration. Otolaryngol Clin North Am. 2019. doi: 10.1016/j.otc.2018.11.004 [DOI] [PubMed] [Google Scholar]

[R3] 3.Boyan BD, Lotz EM, Schwartz Z. Roughness and hydrophilicity as osteogenic biomimetic surface properties. Tissue Eng - Part A. 2017;23(23–24):1479–1489. doi: 10.1089/ten.tea.2017.0048 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R4] 4.Sims NA, Gooi JH. Bone remodeling: Multiple cellular interactions required for coupling of bone formation and resorption. Semin Cell Dev Biol. 2008;19(5):444–451. doi: 10.1016/j.semcdb.2008.07.016 [DOI] [PubMed] [Google Scholar]

[R5] 5.Boyan BD, Cheng A, Olivares-Navarrete R, Schwartz Z. Implant surface design regulates mesenchymal stem cell differentiation and maturation. Adv Dent Res. 2016;28(1):10–17. doi: 10.1177/0022034515624444 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R6] 6.Schwartz Z, Lotz EM, Berger MB, Boyan BD. 7.22 The effect of substrate microtopography on osteointegration of titanium implants ☆. Compr Biomater II. 2017;7:429–443. doi: 10.1016/b978-0-12-803581-8.10218-8 [DOI] [Google Scholar]

[R7] 7.Olivares-Navarrete R, Hyzy SL, Park JH, et al. Mediation of osteogenic differentiation of human mesenchymal stem cells on titanium surfaces by a Wnt-integrin feedback loop. Biomaterials. 2011;32(27):6399–6411. doi: 10.1016/j.biomaterials.2011.05.036 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R8] 8.Lai M, Hermann CD, Cheng A, et al. Role of α2β1 integrins in mediating cell shape on microtextured titanium surfaces. J Biomed Mater Res - Part A. 2015;103(2):564–573. doi: 10.1002/jbm.a.35185 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R9] 9.Buser D, von Arx T, ten Bruggenkate C, Weingart D. Basic surgical principles with ITI implants. Clin Oral Implants Res. 2000;11 Suppl 1:59–68. doi: 10.1034/j.1600-0501.2000.011S1059.x [DOI] [PubMed] [Google Scholar]

[R10] 10.Klokkevold PR, Han MSTJ. How do smoking, diabetes and periodontitis affect outcomes of implant treatment? Br Dent J. 2007;203(6):333–333. doi: 10.1038/bdj.2007.841 [DOI] [PubMed] [Google Scholar]

[R11] 11.Gaetti-Jardim EC, Santiago-Junior JF, Goiato MC, Pellizer EP, Magro-Filho O, Jardim EG. Dental implants in patients with osteoporosis: A clinical reality? J Craniofac Surg. 2011;22(3):1111–1113. doi: 10.1097/SCS.0b013e3182108ec9 [DOI] [PubMed] [Google Scholar]

[R12] 12.Mohammad AR, Hooper DA, Vermilyea SG, Mariotti A, Preshaw PM. An investigation of the relationship between systemic bone density and clinical periodontal status in post-menopausal Asian-American women. Int Dent J. 2003;53(3):121–125. doi: 10.1111/j.1875-595X.2003.tb00735.x [DOI] [PubMed] [Google Scholar]

[R13] 13.Moedano DE, Irigoyen ME, Borges-Yáñez A, Flores-Sánchez I, Rotter RC. Osteoporosis, the risk of vertebral fracture, and periodontal disease in an elderly group in Mexico City. Gerodontology. 2011;28(1):19–27. doi: 10.1111/j.1741-2358.2009.00342.x [DOI] [PubMed] [Google Scholar]

[R14] 14.Lotz EM, Cohen DJ, Schwartz Z, Boyan BD. Titanium implant surface properties enhance osseointegration in ovariectomy induced osteoporotic rats without pharmacologic intervention. Clin Oral Implants Res. 2020;31(4):374–387. doi: 10.1111/clr.13575 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R15] 15.Hao J, Zhang Y, Jing D, et al. Mechanobiology of mesenchymal stem cells: Perspective into mechanical induction of MSC fate. Acta Biomater. 2015;20:1–9. doi: 10.1016/j.actbio.2015.04.008 [DOI] [PubMed] [Google Scholar]

[R16] 16.Berger MB, Cohen DJ, Olivares-Navarrete R, et al. Human osteoblasts exhibit sexual dimorphism in their response to estrogen on microstructured titanium surfaces. Biol Sex Differ. 2018;9(1):1–10. doi: 10.1186/s13293-018-0190-x [DOI] [PMC free article] [PubMed] [Google Scholar]

[R17] 17.Bonewald LF, Kester MB, Schwartz Z, et al. Effects of combining transforming growth factor β and 1,25- dihydroxyvitamin D3 on differentiation of a human osteosarcoma (MG-63). J Biol Chem. 1992;267(13):8943–8949. [PubMed] [Google Scholar]

[R18] 18.Boyan BD, Batzer R, Kieswetter K, et al. Titanium surface roughness alters responsiveness of MG63 osteoblast- like cells to 1α,25-(OH)2D3. J Biomed Mater Res. 1998;39(1):77–85. doi: [DOI] [PubMed] [Google Scholar]

[R19] 19.Kim BJ, Koh JM. Coupling factors involved in preserving bone balance. Cell Mol Life Sci. 2019;76(7):1243–1253. doi: 10.1007/s00018-018-2981-y [DOI] [PMC free article] [PubMed] [Google Scholar]

[R20] 20.Bishop GB, Einhorn TA. Current and future clinical applications of bone morphogenetic proteins in orthopaedic trauma surgery. Int Orthop. 2007;31(6):721–727. doi: 10.1007/s00264-007-0424-8 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R21] 21.Berger MB, Bosh KB, Jacobs TW, Joshua Cohen D, Schwartz Z, Boyan BD. Growth factors produced by bone marrow stromal cells on nanoroughened titanium–aluminum–vanadium surfaces program distal MSCs into osteoblasts via BMP2 signaling. J Orthop Res. 2020;(September):1–13. doi: 10.1002/jor.24869 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R22] 22.Huang S, Li Z, Liu Y, et al. Neural regulation of bone remodeling: Identifying novel neural molecules and pathways between brain and bone. J Cell Physiol. 2019;234(5):5466–5477. doi: 10.1002/jcp.26502 [DOI] [PubMed] [Google Scholar]

[R23] 23.Negishi-Koga T, Takayanagi H. Bone cell communication factors and Semaphorins. Bonekey Rep. 2012;1(September):183. doi: 10.1038/bonekey.2012.183 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R24] 24.Brazill JM, Beeve AT, Craft CS, Ivanusic JJ, Scheller EL. Nerves in Bone: Evolving Concepts in Pain and Anabolism. J Bone Miner Res. 2019;34(8):1393–1406. doi: 10.1002/jbmr.3822 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R25] 25.Ji-Ye H, Xin-Feng Z, Lei-Sheng J. Autonomic control of bone formation: Its clinical relevance. Handb Clin Neurol. 2013;117:161–171. doi: 10.1016/B978-0-444-53491-0.00014-6 [DOI] [PubMed] [Google Scholar]

[R26] 26.He JY, Jiang LS, Dai LY. The roles of the sympathetic nervous system in osteoporotic diseases: A review of experimental and clinical studies. Ageing Res Rev. 2011;10(2):253–263. doi: 10.1016/j.arr.2011.01.002 [DOI] [PubMed] [Google Scholar]

[R27] 27.Elefteriou F, Campbell P, Ma Y. Control of bone remodeling by the peripheral sympathetic nervous system. Calcif Tissue Int. 2014;94(1):140–151. doi: 10.1007/s00223-013-9752-4 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R28] 28.Zochodne DW. Local blood flow in peripheral nerves and their ganglia: Resurrecting key ideas around its measurement and significance. Muscle and Nerve. 2018;57(6):884–895. doi: 10.1002/mus.26031 [DOI] [PubMed] [Google Scholar]

[R29] 29.Elefteriou F Neuronal signaling and the regulation of bone remodeling. Cell Mol Life Sci. 2005;62(19–20):2339–2349. doi: 10.1007/s00018-005-5175-3 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R30] 30.Kang S, Kumanogoh A. Semaphorins in bone development, homeostasis, and disease. Semin Cell Dev Biol. 2013;24(3):163–171. doi: 10.1016/j.semcdb.2012.09.008 [DOI] [PubMed] [Google Scholar]

[R31] 31.Xu R. Semaphorin 3A a new player in bone remodeling. Cell Adhes Migr. 2014;8(1):5–10. doi: 10.4161/cam.27752 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R32] 32.Li Z, Hao J, Duan X, et al. The role of semaphorin 3A in bone remodeling. Front Cell Neurosci. 2017;11(February):1–8. doi: 10.3389/fncel.2017.00040 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R33] 33.Lotz EM, Berger MB, Boyan BD, Schwartz Z. Regulation of mesenchymal stem cell differentiation on microstructured titanium surfaces by semaphorin 3A. Bone. 2020;134(January):115260. doi: 10.1016/j.bone.2020.115260 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R34] 34.Ma X, Lv J, Sun X, et al. Naringin ameliorates bone loss induced by sciatic neurectomy and increases Semaphorin 3A expression in denervated bone. Sci Rep. 2016;6(122):1–14. doi: 10.1038/srep24562 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R35] 35.Thomsen JS, Christensen LL, Vegger JB, Nyengaard JR, Brüel A. Loss of bone strength is dependent on skeletal site in disuse osteoporosis in rats. Calcif Tissue Int. 2012;90(4):294–306. doi: 10.1007/s00223-012-9576-7 [DOI] [PubMed] [Google Scholar]

[R36] 36.Bettis T, Kim BJ, Hamrick MW. Impact of muscle atrophy on bone metabolism and bone strength: implications for muscle-bone crosstalk with aging and disuse. Osteoporos Int. 2018;29(8):1713–1720. doi: 10.1007/s00198-018-4570-1 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R37] 37.Gatti V, Ghobryal B, Gelbs MJ, et al. Botox-induced muscle paralysis alters intracortical porosity and osteocyte lacunar density in skeletally mature rats. J Orthop Res. 2019;37(5):1153–1163. doi: 10.1002/jor.24276 [DOI] [PubMed] [Google Scholar]

[R38] 38.Warner SE, Sanford DA, Becker BA, Bain SD, Srinivasan S, Gross TS. Botox induced muscle paralysis rapidly degrades bone. Bone. 2006;38(2):257–264. doi: 10.1016/j.bone.2005.08.009 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R39] 39.Lotz EM, Cohen DJ, Ellis RA, Wayne JS, Schwartz Z, Boyan BD. Ibandronate treatment before and after implant insertion impairs osseointegration in aged rats with ovariectomy induced osteoporosis. JBMR Plus. 2019;3(7):e10184. doi: 10.1002/jbm4.10184 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R40] 40.Lotz EM, Lohmann CH, Boyan BD, Schwartz Z. Bisphosphonates inhibit surface-mediated osteogenesis. J Biomed Mater Res - Part A. 2020;108(8):1774–1786. doi: 10.1002/jbm.a.36944 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R41] 41.Schneider CA, Rasband WS, Eliceiri KW. NIH Image to ImageJ: 25 years of image analysis. Nat Methods. 2012;9(7):671–675. doi: 10.1038/nmeth.2089 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R42] 42.McClure MJ, Cohen DJ, Ramey AN, et al. Decellularized muscle supports new muscle fibers and improves function following volumetric injury. Tissue Eng - Part A. 2018;24(15–16):1228–1241. doi: 10.1089/ten.tea.2017.0386 [DOI] [PubMed] [Google Scholar]

[R43] 43.Bakker AD, Klein-Nulend J. Osteoblast isolation from murine calvaria and long bones. Methods Mol Biol. 2012;816:19–29. doi: 10.1007/978-1-61779-415-5_2 [DOI] [PubMed] [Google Scholar]

[R44] 44.McMillan J, Kinney RC, Ranly DM, Fatehi-Sedeh S, Schwartz Z, Boyan BD. Osteoinductivity of demineralized bone matrix in immunocompromised mice and rats is decreased by ovariectomy and restored by estrogen replacement. Bone. 2007;40(1):111–121. doi: 10.1016/j.bone.2006.07.022 [DOI] [PubMed] [Google Scholar]

[R45] 45.Olivares-Navarrete R, Raines AL, Hyzy SL, et al. Osteoblast maturation and new bone formation in response to titanium implant surface features are reduced with age. J Bone Miner Res. 2012;27(8):1773–1783. doi: 10.1002/jbmr.1628 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R46] 46.Nakamura H, Morita S, Kumei Y, et al. Effects of neurectomy and tenotomy on the bone mineral density and strength of tibiae. Acta Astronaut. 2001;49(3–10):179–190. doi: 10.1016/S0094-5765(01)00097-2 [DOI] [PubMed] [Google Scholar]

[R47] 47.Erdfelder E, FAul F, Buchner A, Lang AG. Statistical power analyses using G*Power 3.1: Tests for correlation and regression analyses. Behav Res Methods. 2009;41(4):1149–1160. doi: 10.3758/BRM.41.4.1149 [DOI] [PubMed] [Google Scholar]

[R48] 48.Faul F, Erdfelder E, Lang AG, Buchner A. G*Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behav Res Methods. 2007;39(2):175–191. doi: 10.3758/BF03193146 [DOI] [PubMed] [Google Scholar]

[R49] 49.Park JH, Park HJ. Botulinum toxin for the treatment of neuropathic pain. Toxins (Basel). 2017;9(9). doi: 10.3390/toxins9090260 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R50] 50.Intiso D Therapeutic use of botulinum toxin in neurorehabilitation. J Toxicol. 2012;2012(June):1–18. doi: 10.1155/2012/802893 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R51] 51.Mense S Neurobiological basis for the use of botulinum toxin in pain therapy. J Neurol Suppl. 2004;251(1):1–7. doi: 10.1016/j.toxicon.2012.07.027 [DOI] [PubMed] [Google Scholar]

[R52] 52.Einhorn TA; Simon G; Devlin VJ; Warman J; Sidhu SP; Vigorita VJ. The osteogenic response to distant skeletal injury. J Bone Jt Surg. 1990;72(9):1374–1378. [PubMed] [Google Scholar]

[R53] 53.Marshall TS, Schwartz Z, Swain LD, Amir D, Sela J, Gross U, Muller-mai BDB C. Matrix vesicle enzyme activity in endosteal bone following implantation of bonding and non.-bonding implant materials. 1991:112–120. doi:99123188140001101 [DOI] [PubMed] [Google Scholar]

[R54] 54.Turner CH. Bone Strength : Current Concepts. 2006;446:429–446. doi: 10.1196/annals.1346.039 [DOI] [PubMed] [Google Scholar]

[R55] 55.Olivares-Navarrete R, Hyzy SL, Pan Q, et al. Osteoblast maturation on microtextured titanium involves paracrine regulation of bone morphogenetic protein signaling. J Biomed Mater Res - Part A. 2015;103(5):1721–1731. doi: 10.1002/jbm.a.35308 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R56] 56.Lotz EM, Olivares-navarrete R, Berner S, Boyan BD, Schwartz Z. Osteogenic response of human MSCs and osteoblasts to hydrophilic and hydrophobic nanostructured titanium implant surfaces. 2016:3137–3148. doi: 10.1002/jbm.a.35852 [DOI] [PubMed] [Google Scholar]

[R57] 57.Chen X, Wang Z, Duan N, Zhu G, Schwarz EM, Xie C. Osteoblast–osteoclast interactions. Connect Tissue Res. 2018;59(2):99–107. doi: 10.1080/03008207.2017.1290085 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R58] 58.Kim B-J, Koh J-M. Coupling factors involved in preserving bone balance. Cell Mol Life Sci. 2018;(0123456789). doi: 10.1007/s00018-018-2981-y [DOI] [PMC free article] [PubMed] [Google Scholar]

[R59] 59.Fukuda T, Takeda S, Xu R, et al. Sema3A regulates bone-mass accrual through sensory innervations. Nature. 2013;497(7450):490–493. doi: 10.1038/nature12115 [DOI] [PubMed] [Google Scholar]

[R60] 60.Hao J, Yu J. Semaphorin 3C and its receptors in cancer and cancer stem-like cells. Biomedicines. 2018;6(2):42. doi: 10.3390/biomedicines6020042 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R61] 61.Fukuda T, Takeda S, Xu R, et al. Sema3A regulates bone-mass accrual through sensory innervations. Nature. 2013;497(7450):490–493. doi: 10.1038/nature12115 [DOI] [PubMed] [Google Scholar]

[R62] 62.Zhang ZK, Guo X, Lao J, Qin YX. Effect of capsaicin-sensitive sensory neurons on bone architecture and mechanical properties in the rat hindlimb suspension model. J Orthop Transl. 2017;10:12–17. doi: 10.1016/j.jot.2017.03.001 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R63] 63.Huang B, Ye J, Zeng X, Gong P. Effects of capsaicin-induced sensory denervation on early implant osseointegration in adult rats. R Soc Open Sci. 2019;6(1):0–9. doi: 10.1098/rsos.181082 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R64] 64.He JY, Zheng XF, Jiang SD, Chen XD, Jiang LS. Sympathetic neuron can promote osteoblast differentiation through BMP signaling pathway. Cell Signal. 2013;25(6):1372–1378. doi: 10.1016/j.cellsig.2013.02.016 [DOI] [PubMed] [Google Scholar]

[R65] 65.Yao Q, Zeng Y, Feng Y, Wu H, Liang H, Gong P. Chemical sympathectomy impairs peri-implant osseointegration in mice: Role of the sympathetic nervous system in osseointegration. Int J Oral Maxillofac Implant. 2019;34(1):91–98. doi: 10.11607/jomi.7086 [DOI] [PubMed] [Google Scholar]

[R66] 66.Corr A, Smith J, Baldock P. Neuronal control of bone remodeling. Toxicol Pathol. 2017;45(7):894–903. doi: 10.1177/0192623317738708 [DOI] [PubMed] [Google Scholar]

[R67] 67.Cizza G, Primma S, Csako G. Depression as a risk factor for osteoporosis. Trends Endocrinol Metab. 2009;20(8):367–373. doi: 10.1016/j.tem.2009.05.003 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R68] 68.Bonnet N, Benhamou CL, Brunet-Imbault B, et al. Severe bone alterations under β2 agonist treatments: Bone mass, microarchitecture and strength analyses in female rats. Bone. 2005;37(5):622–633. doi: 10.1016/j.bone.2005.07.012 [DOI] [PubMed] [Google Scholar]

PERMALINK

Differential Effects of Neurectomy and Botox-induced Muscle Paralysis on Bone Phenotype and Titanium Implant Osseointegration

Jingyao Deng

David J Cohen

James Redden

Michael J McClure

Barbara D Boyan

Zvi Schwartz

Abstract

Graphical Abstract

1. Introduction

2. Material and Methods

2.1. Implant Preparation

2.2. Animals and Surgical Procedures

2.2.1. Sciatic and femoral nerve neurectomy surgeries on right hindlimbs (SFN)

2.2.2. Botox injections on right hindlimbs (BTX)

2.2.3. Unicortical implant placements on both hindlimbs in all groups

2.3. Tissue Analysis

2.3.1. Bone phenotype and peri-implant growth analyzed by microCT

2.3.2. Histology

2.3.3. Removal torque mechanical test

Figure 5.

2.3.4. Muscle histology

2.4. Response of Primary Bone Marrow Stromal Cells (rBMSCs) and Osteoblasts (rOBs)

2.5. Statistical Analysis

3. Results

3.1. Evaluation of Bone Phenotype

Figure 1.

Figure 2.

3.2. Evaluation of Osseointegration

3.2.1. MicroCT analysis

Figure 3.

3.2.2. Histology

Figure 4.

3.2.3. Removal torque mechanical test

3.3. Cell Response

Figure 6.

3.4. Muscle Morphology and Atrophy in Nerve Intervention Models

Figure 7.

4. Discussion

5. Conclusion

Supplementary Material

Highlights.

6. Acknowledgements

Footnotes

References

Associated Data

Supplementary Materials

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases