Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Jul 5.
Published in final edited form as: Osteoarthritis Cartilage. 2022 Jan 10;30(4):586–595. doi: 10.1016/j.joca.2021.12.008

Measures of cardiovascular function suggest autonomic nervous system dysregulation after surgical induction of joint injury in the male Lewis rat

TD Yeater †, J Zubcevic ‡, KD Allen †,§,*
PMCID: PMC9255271  NIHMSID: NIHMS1815665  PMID: 35017058

SUMMARY

Objectives:

Functional changes in the autonomic nervous system may help explain variability in the progression of knee osteoarthritis (OA). Thus, the objective of this study was to evaluate autonomic nervous system shifts, measured via heart rate response variables, in rat knee joint injury and OA models.

Methods:

Cardiovascular characteristics were measured at baseline and bi-weekly for 8 weeks after skin incision, medial collateral ligament transection (MCLT), or MCLT+medial meniscus transection (MCLT+MMT). Heart rate was also assessed during a mild stressor (elevated maze). At endpoint, cardiovascular responses to mechanical knee stimuli were evaluated, as well as responses to 1-phenylbiguanide, a 5HT3A receptor agonist with reported ability to stimulate vagal responses.

Results:

During low activity, a slower heart rate occurred in MCLT (299 ± 10 bpm) and MCLT+MMT (310 ± 10 bpm) animals compared to controls (325 ± 10 bpm). Furthermore, patellar ligament mechanical stimuli produced an immediate decrease in heart rate and blood pressure in all groups. Finally, a larger drop in heart rate was observed in MCLT (252 ± 40 bpm) and MCLT+MMT (263 ± 49 bpm) following administration of 1-phenylbiguanide compared to skin incision (168 ± 45 bpm).

Conclusions:

Acute mechanical stimulation of the patellar ligament produced drops in heart rate, suggesting a possible joint-brain connection that modulates autonomic responses. With both joint injury, cardiac vagal activation was altered in response to pharmacological stimulation, with chronic longitudinal heart rate reduction. These data provide some preliminary evidence of potential functional shifts in autonomic nervous system function in models of joint injury and OA.

Keywords: Osteoarthritis, Heart rate variability, Autonomic nervous system, Vagal tone

Introduction

Osteoarthritis (OA) is a highly prevalent musculoskeletal disease characterized by local, low-grade joint inflammation that modulates local neuroplasticity. Including decreases in sensory and sympathetic innervation in highly inflamed areas of the synovium1. Although local neural and inflammatory changes in OA are commonly studied, OA also contributes to changes beyond the joint, like lowered thresholds of dorsal horn neurons2. Dysfunction of the autonomic nervous system has been implicated in the progression of OA3,4. Because measures of autonomic nervous system function are not commonly applied in OA, this work begins with a brief review of autonomic nervous system physiology and pathophysiology. Then, autonomic nervous system function is assessed in a rat model of OA.

Autonomic nervous system and vagal tone

The autonomic nervous system is often referred to in the context of involuntary actions, like heart rate, respiration, and digestion. The two branches of the autonomic nervous system, the parasympathetic and sympathetic nervous systems, often function in opposition to maintain homeostatic balance of body functions. The autonomic nervous system acts as a regulator of immune system homeostasis. Specifically, the vagus nerve controls peripheral inflammation via a bidirectional mechanism. Vagal afferent fibers sense cytokines in the periphery, such as IL-1β and TNFα5 while efferent fibers release acetylcholine, having both direct and indirect anti-inflammatory actions. The serotonin receptor, 5-HT3A is abundantly expressed on the vagus nerve6, and 1-phenylbiguanide, a 5-HT3A agonist, has been shown to be selective for vagal nerve activation, corresponding to bradycardic and hypo-tensive events7. Thus, blood pressure and heart rate responses act as indirect measures for vagus nerve and autonomic function.

Heart rate variability as a surrogate measure of the autonomic nervous system

One advantage of using heart rate variability (HRV) as a surrogate measure for autonomic function is the ability to obtain measurements longitudinally and in different environments. Moreover, the relationship of HRV measurements to autonomic function has been thoroughly characterized by experimental studies8. HRV is easily taken in clinical populations, and these assessments have been shown to have clinical relevance9.

HRV is measured in temporal and frequency domains. One common temporal variable is the root mean square of the successive differences (RMSSD) in neighboring heart rate inter-beat intervals. RMSSD is often used to characterize the parasympathetic nervous system, where increased RMSSD may correspond to increased vagal efferent drive10. Spectral analysis of HRV provides frequency domain parameters, used to describe changes in each branch of the autonomic nervous system8,9. Total power in the low frequency (LF) band is often attributed to the sympathetic nervous system while power in the high frequency (HF) band represents the parasympathetic nervous system8.

Heart rate itself is indicative of autonomic function. The parasympathetic nervous system slows down heart rate, while the sympathetic nervous system increases heart rate. Thus, slower heart rates represent an increased ratio of parasympathetic to sympathetic influence. However, this measure does not provide specific information about each branch of the autonomic nervous system, only balance between branches.

Autonomic nervous system dysregulation in inflammatory diseases and a potential role in knee osteoarthritis

In other inflammatory and chronic diseases, such as rheumatoid arthritis, autonomic dysregulation is implicated in initiating chronic systemic inflammation11. Consistent with this, α7 nicotinic acetylcholine receptor knock-out mice have more joint damage and increased levels of pro-inflammatory cytokines compared to wildtype controls12. Stimulation of the cholinergic anti-inflammatory pathway via electrical vagal nerve stimulation improves signs and symptoms of rheumatoid arthritis13. However, inflammation in rheumatoid arthritis is characterized by cyclic acute flares, while OA is characterized by chronic, low grade inflammation14,15. Nonetheless, autonomic dysregulation may present in diseases with low-grade inflammation, as autonomic dysfunction has been detected with obesity16 and in non-inflammatory diseases like fibromyalgia17.

While the vagus nerve does not innervate the knee, an anatomical and functional link exists between peripheral sensory drive and autonomic nervous system dysregulation. First, pain-related information is transmitted as nociceptive signals to the dorsal root ganglia, then the dorsal horn of the spinal cord. Neurons from lamina I of the spinal dorsal horn, including nociceptors, project to the nucleus tractus solitarius (NTS), which acts as a homeostatic integration system of afferent signals18. Electrical stimulation of the NTS inhibits nociceptive signal transmission19, and nociceptive feedback attenuates the parasympathetic nervous system by damping vagal activity20. However, these nociceptive-autonomic circuits have not been explored in the context of chronic knee OA.

Due to a lack of evidence of parasympathetic innervation of cartilage, actions of the parasympathetic nervous system on local inflammation and OA pathophysiology via the vagus nerve may be indirect. For example, splenic sympathetic fibers can be activated by vagal efferents21. This way, vagal input controls systemic inflammation through activation of peripheral sympathetic nerves. Our collaborator's work has shown bone marrow can participate in peripheral inflammation via a similar mechanism20, which may relate to bone marrow changes in OA22. Emerging evidence demonstrates cholinergic innervation in subchondral bone of OA patients23. Despite this lack of parasympathetic innervation to the joint, chondrocytes express receptors for both parasympathetic and sympathetic neurotransmitters (acetylcholine and norepinephrine, respectively). Stimulation of these receptors modulate chondrocyte metabolism, inflammatory response, and cartilage degradation24,25. The dysregulation of the non-neuronal cholinergic system in the joint suggests an important functional role of the autonomic nervous system and may explain a possible feedback mechanism that could lead to autonomic cardiovascular changes.

Using insights from other joint diseases, we hypothesized autonomic dysregulation would occur in a surgical model of knee OA. This was first assessed by examining longitudinal HRV in a rat knee OA model. Because stress can heighten autonomic response26,27, our hypothesis was also tested in an elevated plus test, a behavioral test of anxiety28,29. Finally, heart rate and blood pressure were assessed during nociceptive mechanical stimulation on the knee, as well as a chemical vagal nerve stimulation via a systemic agonist injection. These measures of cardiovascular function provide the first insight into functional autonomic nervous system changes in a rodent OA model.

Materials and methods

Experimental Design

To assess our hypothesis, experiments consisted of biweekly radio-telemetry assessments of HRV. At endpoint, HRV was collected in an elevated plus maze (9 weeks post-op), and measures of heart rate and blood pressure were assessed in response to chemical vagal nerve stimulation (9–11 weeks post-op). These experiments were conducted on three cohorts of Lewis rats with surgical destabilization of the knee via either medial collateral ligament (MCL) transection (MCLT) alone (as an intermediate control with the joint capsule opened) or MCLT plus medial meniscus transection (MCLT+MMT). Skin incision was used as sham control, and animals were randomized into groups using a random number generator. Target number of animals per group was n = 9 (skin incision and MCLT) and n = 16 (MCLT+MMT). With a repeated measure design, these numbers allow for a detectable effect size (f) of 22% at a power of 0.8. HRV was assessed longitudinally; however, due to technical limitations of telemetry probes, data at some time points are absent (see Methods). Thus, the sample size at each time point and for different tests varies. In addition, four animals were removed from the study as a humane endpoint due to telemeter implant surgical complications. Finally, due to surgical error (meniscus not cut), four MCLT+MMT animals were moved to the MCLT group. The decision to reclassify these animals as MCLT was entirely based on histological data, looking at every slide through the knee joint (every 100 μm); data on heart rate did not affect this reclassification. In the end, final numbers for groups are: longitudinal HRV (n = 7–9 skin incision, n = 8–9 MCLT, and n = 8–9 MCLT+MMT); elevated plus maze (n = 7 for skin incision, n = 8 MCLT, and n = 7 MCLT+MMT); and endpoint autonomic response (n = 6 skin incision, n = 9 MCLT, and n = 6 MCLT+MMT). Supplemental Figure 1 is provided as a visual representation of experimental time points and data collected.

Animals

Male Lewis rats (approximately 350 g and 12 weeks old) were obtained from Charles River (Wilmington, MA, USA) and acclimated for at least 5 days at University of Florida (UF) animal care facilities. Animals were housed two per cage in a lighting controlled environment (12-h light/dark cycle) with access to food and water ad libitum. Standard bedding and housing were used. All procedures were approved by the UF Institutional Animal Care and Use Committee.

Telemeter placement surgery

Telemeters were implanted as previously described30. Briefly, rats were anesthetized with isoflurane (1.5–2.0% in oxygen) and the abdomen of the rat was aseptically prepared. A 2–3 cm incision was made on the abdomen's midline, followed by blunt dissection to the descending aorta. Using suture, the vessel was temporarily occluded and a 29G needle was used to create an opening for the radio-telemeter catheter (TSE Stellar Telemetery, Chesterfield, MO). The catheter was inserted and secured with tissue adhesive. The radio-telemeter was sutured to the muscle wall (3.0 silk), and the muscle was closed with absorbable suture (3.0). Skin was closed with wound clips, which were removed after 10 days. Buprenorphine was administered (0.05 mg/kg) twice daily for 2 days to manage post-operative pain. Rats recovered at least 10 days before joint surgery.

Joint surgeries

Rats were anesthetized with isoflurane (1.5–2% in oxygen) and the knee was aseptically prepared. For all surgeries, a 1–2 cm incision was made along the stifle. Muscle was bluntly dissected to expose the MCL. After visualization of the ligament, the surgeon was informed of the surgery. If a sham procedure, muscle and skin were re-approximated using absorbable suture (4.0). If an MCLT procedure, the MCL was transected. For MCLT+MMT procedure, the MCL was transected and the medial meniscus was radially transected at its central aspect31. Muscle and skin were then closed as previously described. Rats were again administered buprenorphine (0.05 mg/kg) twice daily for 2 days to manage post-operative pain.

Recordings of cardiovascular function

As discussed above, autonomic balance can be indirectly examined through heart rate frequency bands8,9. Here, blood pressure recordings were used to determine relative power of frequency bands. Before collecting data, fidelity of recordings was confirmed by inspection of the blood pressure trace for pulse pressure waveform. If no pulse pressure was observed, signal fidelity was deemed unreliable and not collected. Data were recorded once per hour for 24 h (5 min, 500 Hz sampling rate). Using MATLAB, systolic peaks in blood pressure were identified. Inter-beat intervals were calculated and imported into Kubios HRV Standard (version 3.2.0) for time and frequency domain analyses. Data underwent equidistant re-sampling at 10 Hz and detrending via smoothing priors method with a smoothing parameter of 1,000. For fast-Fourier transform, window width was set to 512 points with 50% overlap. The LF and HF bands were defined as 0.27–0.75 Hz and 0.75–3.3 Hz, respectively32,33. Data were corrected for artifacts using Kubios, and trials were excluded if more than 5.0% of the trial required artifact correction, as recommended by manufacturer34. Data are presented separately for high activity (night) and low activity (day) periods. Based on accelerometer data from the telemeter, the low activity period was identified as 12pm–2pm and high activity was identified as 10pm-12am (see Fig. 3 of results).

Fig. 3.

Fig. 3

Open in a new tab

Movement data collected via radiotelemetry accelerometer. (A) 24-h recording over pre-surgical time point for all animals. (B) Longitudinal low-activity (day) movement data and (C) longitudinal high-activity (night) movement data. Error bars represent mean ± 95% confidence interval and main effect statistics are indicated on the plot.

Elevated plus maze

Stress is known to exaggerate autonomic nervous system responses26,27. Therefore, blood pressure recordings were collected while animals were in an elevated plus maze (a behavioral test of anxiety28,29). Baseline recordings were taken for 5 min prior to placing the animal in the maze and for 5 min while exploring the maze. Inter-beat intervals were calculated and imported into Kubios (version 3.2.0) and both time and frequency domain parameters were analyzed. Data are presented and analyzed as a change from baseline to maze exploration.

Endpoint autonomic response

Rats were anesthetized with isoflurane (1.5–2.0% in oxygen). Body temperature was maintained at 37°C. Fur was shaved from surgical regions and a 2 cm incision was made over the inguinal region of the left hind limb. The femoral artery, vein, and nerve were exposed via blunt dissection and gently separated. Polyethylene tubing was inserted and secured to the femoral artery for blood pressure measurements and to the femoral vein for drug administration. Electrocardiogram and blood pressure were recorded continuously using Grass Technologies P511 AC Amplifier, and a CED Power1401 mk II data acquisition interface. Signals were collected using CED Spike II (version 8) software.

While under isoflurane anesthesia, an 18-inch silk suture (4.0) was threaded behind the patellar ligament then placed over a pulley to apply tensile force to the ligament (Supplemental Fig. 2). The knee was placed in a custom holder to bend the joint to 90°. Force of increasing magnitude (200, 500, and 1,000 g) was applied for 20 s using a laboratory weight set. Multiple forces were chosen to test whether a force–response relationship exists. Five minutes were allowed for recovery between stimuli.

For vagal activation, 1-phenylbiguanide (Sigma–Aldrich) was systemically administered. To assess changes to the known dose–response of phenylbiguanide to heart rate and blood pressure drops, multiple concentrations were administered (1, 10, and 100 μg/kg)35. Volume of administration ranged from 100 to 150 μL. There was a recovery period of 10 min between chemical stimuli. For all procedures, surgeon was blinded to group assignments.

Joint histopathology

Knee joints were collected then fixed in 10% neutral buffered formalin for 2 days. Following fixation, knee joints were decalcified in 10% formic acid for 2 weeks and then vacuum infiltrated in paraffin wax. Sections (10 μm) were acquired every 100 μm, with one frontal section collected from the loading region of each joint stained with safranin O and fast green to confirm cartilage degradation. Joints were then graded by three observers based on the OARSI score36. Observers were blinded with respect to group. In addition to OARSI scores, select histological parameters (cartilage area, bone area, cell density in synovial intima, synovial width, entropy of cells in synovial intima, and eccentricity of cell shape in synovial intima) were analyzed using open source software developed by our lab37,38.

Statistical analysis

For longitudinal measurements, a linear mixed effects model was conducted using week, group, and the interaction as fixed effects and animal identifier as a random effect. To account for any pre-surgical variation, baseline data were included as a covariate in the model. For missing longitudinal baseline data (n = 1, skin incision), a value was imputed via baseline group average. If indicated, least squared means were compared using Tukey's HSD adjusted p-values. Separate models were run for high and low activity periods. For endpoint analysis of blood pressure and heart rate, data were subtracted from no stimulation value; then, a linear mixed effects model was conducted using stimulus, group and the interaction as fixed effects and animal as a random effect. For data in the elevated plus maze and histological grading, data were analyzed with a one-way ANOVA and Tukey HSD, if group main effect was significant. For missing baseline data in the elevated maze (n = 2, skin incision and n = 1 MCLT+MMT), a value was imputed via baseline group average.

All data are presented as mean ± 95% confidence interval (CI), with p-values considered significant if below 0.05. The assumptions of our statistical models were checked via visual inspection of QQ Plot by a biostatistician and homogeneity of variance was checked via Levene's Test.

Results

Joint histopathology

Fig. 1 shows representative images of the medial compartment of operated limbs in each group. Consistent with previous findings38, successful MCLT+MMT surgeries caused full depth cartilage lesions, subchondral bone changes, joint capsule thickening, and osteophytes by 8 weeks post-operation. MCLT+MMT animals exhibited OARSI scores of 4.6 ± 0.5 (95% CI), while all skin incision and MCLT surgeries resulted in OARSI scores of 0. Quantitative analysis revealed differences between groups for cartilage area, bone area, entropy of cells in synovial intima, and eccentricity of cell shape in the synovial intima (p's < 0.05). Synovial width also appeared to increase with MCLT alone and MCLT+MMT, but note, these results are not statistically significant (p = 0.06). Quantitative histological data and associated CIs are presented in Fig. 2 and Table I, respectively.

Fig. 1.

Fig. 1

Open in a new tab

Example histological images of (A) skin incision, (B) MCLT and (C) MCLT+MMT knee joints. The Scale bar represents 500 microns.

Fig. 2.

Fig. 2

Open in a new tab

Quantitative histological data for (A) cartilage area, (B) bone area, (C) cell density in the synovial intima, (D) synovial width, (E) entropy of cells in synovial intima, and (F) eccentricity of cell shape in the synovial intima. Error bars represent mean ± 95% confidence interval. Asterisks indicate p < 0.05.

Table I.

Results of quantitative histological grading. Values represent mean ± 95% confidence interval

Cartilage Area (mm2) Bone Area (mm2) Cell Density in the
Synovial Intima (px/pix)		 Synovial Width
(mm)		 Entropy of Cells in the
Synovial Intima		 Eccentricity of Cell Shape
in the Synovial Intima
Skin Incision 0.14 + 0.01 0.36 + 0.05 0.026 + 0.007 1.51 + 0.31 0.072 + 0.024 0.89 + 0.01
MCLT Alone 0.15 + 0.01 0.41 + 0.05 0.025 + 0.007 1.78 + 0.29 0.121 + 0.023 0.89 + 0.01
MCLT+MMT 0.10 + 0.01 0.48 + 0.05 0.029 + 0.007 2.04 + 0.31 0.209 + 0.024 0.88 + 0.01
Open in a new tab

Relative movement

Fig. 3 displays average animal activity over the baseline recording period. Using this plot, we selected 10pm-12am as the three consecutive most active times and 12pm–2pm as the three consecutive least active times. Values for longitudinal variables were averaged across these time periods for each time point, such that each animal contributed one value per time point. Fig. 3(B) and (C) shows activity levels over time. No statistical differences were seen between groups.

Time domain heart rate variability

Heart rate, shown in Fig. 4(A) and (B), varied between groups during the low activity period (p < 0.05, main effect) with MCLT (299 ± 10 bpm) and MCLT+MMT (310 ± 10 bpm) having lower heart rate than skin incision (325 ± 10 bpm) (p = 0.002, and 0.039, respectively). During high activity, differences in heart rate were not observed between groups. No differences between groups were observed for RMSSD [Fig. 4(C) and (D)].

Fig. 4.

Fig. 4

Open in a new tab

Time domain results from longitudinal collection of cardiovascular parameters via radiotelemetry. Heart rate is shown for (A) low-activity (day) and (B) high-activity (night) periods. Similarly, temporal domain heart rate variability (RMSSD) is shown for (C) low-activity (day) and (D) high-activity (night) periods. Error bars represent mean ± 95% confidence interval and p-values of interest are indicated on plots.

Frequency domain heart rate variability

The HF component of power spectra has been previously used as a measure of parasympathetic tone. During high activity, MCLT+MMT animals had an estimated mean HF power of 8.79 ± 1.77 ms2, compared to skin incision control (11.10 ± 1.79 ms2) and MCLT alone (11.34 ± 1.74). While the CIs of control and MCLT+MMT only marginally overlap, statistical differences were not observed (Fig. 5(B), p = 0.099 group main effect). Differences in HF power were not observed during low activity [Fig. 5(A)]. For low activity LF power, no differences between groups were observed [Fig. 5(C)]. However, during high activity, skin incision control animals had an estimated mean LF power of 2.99 ± 0.74 ms2 compared to MCLT alone (3.88 ± 0.69 ms2) and MCLT+MMT (2.70 ± 0.69 ms2). Again, while the CIs between MCLT and MCLT+MMT marginally overlap, no statistical differences were seen (p = 0.051, group main effect and 0.091 interaction).

Fig. 5.

Fig. 5

Open in a new tab

Frequency domain results from longitudinal collection of cardiovascular parameters via radiotelemetry. High frequency power (parasympathetic drive) is shown for (A) low-activity (day) and (B) high-activity (night) periods. Similarly, low frequency power (sympathetic drive) is shown for (C) low-activity (day) and (D) high-activity (night) periods. Finally, autonomic balance (LF/HF ratio) is shown for (E) low-activity (day) and (F) high-activity (night) periods. Error bars represent mean ± 95% confidence interval and p-values of interest are indicated on plots.

Maze data

Stress is known to exacerbate autonomic responses26. For this reason, HRV parameters were measured during an elevated plus maze test (9 weeks after OA induction). While in the maze, heart rate increased in all groups, with no differences between groups [Fig. 6(A)]. Similarly, RMSSD, HF power, and LF power did not change with introduction to the maze [Fig. 6(B)-(D) ].

Fig. 6.

Fig. 6

Open in a new tab

Cardiovascular responses to a mild stressor (elevated plus maze). Temporal domain measures include the (A) heart rate and (B) time domain heart rate variability (RMSSD). Frequency domain measures include the (C) high frequency power (parasympathetic drive) and (D) low frequency power (sympathetic drive).

Endpoint data

With increasing doses of the vagal activator 1-phenylbiguanide, dose-dependent drops in heart rate were observed (p < 0.01, dose main effect). For the largest dose of phenylbiguanide, skin incision animals had mean heart rate decrease of 168 ± 45 bpm, compared to MCLT animals (252 ± 40 bpm) and MCLT+MMT (263 ± 49 bpm). Again, while there minimal overlap in these CIs, these results were not significant in the linear mixed effect model (Fig. 7(A), p = 0.07, interaction). Phenylbiguanide also caused dose-dependent drops in blood pressure (p < 0.01, dose main effect, Fig. 6(B)), with no differences between groups. For mechanical stimulation of the patellar ligament, pressure and heart rate dropped with the application of a weight on the ligament [Fig. 7(C) and (D)]. However, events were similar in magnitude for all weights and no differences were observed between groups.

Fig. 7.

Fig. 7

Open in a new tab

Endpoint cardiovascular responses to pharmacological and mechanical stimulation. (A) Heart rate and (B) blood pressure drops in response to 1-phenylbiguanide to stimulate vagus nerve. (C) Heart rate and (D) blood pressure response to mechanical stimulation of the knee. Error bars represent mean ± 95% confidence interval and p-values of interest are indicated on plots.

Discussion

Courties et al. recently proposed a working model for the role of the autonomic nervous system in OA pathogenesis3. Our study expands on these ideas to quantify shifts in autonomic nervous system function through cardiovascular parameters. Particularly, MCLT+MMT animals had an estimated mean HF power (parasympathetic) of 8.79 ± 1.77 ms2 compared to skin incision control at 11.10 ± 1.79 ms2; however, these results were not statistically significant (p = 0.099 group main effect). Moreover, mechanical stimulation of the patellar ligament caused immediate reductions in cardiovascular parameters, demonstrating a potential connection between the knee and its extra-articular musculature and the autonomic function. Finally, chemical activation of the vagus (at 100 μg/kg) showed altered cardiovascular responses in MCLT+MMT animals; however, this difference was not statistically significant (p = 0.07, interaction). Taken together, our data provides preliminary, though non-conclusive, evidence of functional autonomic nervous system changes in a rodent ligament injury and an OA model, indicating potential physiologic relationships between knee injury and cardiovascular dysfunction.

Mechanical stimulation of the patellar ligament resulted in drops in heart rate and blood pressure in all groups. These data suggest a possible direct neural connection between joint mechanosensitive nociceptive fibers, the joint's surrounding musculature, and cardiac vagal efferent fibers. This possible crosstalk between joint-level signaling and vagal efferent fibers, along with increased joint nociceptor firing with OA progression39, may chronically modulate the autonomic nervous system in OA. However, responses may be attributed to loading on the patellar ligament, quadriceps muscle, or synovium rather than bone and cartilage loading per se. This warrants further study with additional loading regimes, which may delineate group differences by including loading more relevant to OA pathogenesis, such as applying mechanical rotation or compression.

In our study, joint injury was induced via two methods, MCLT and MCLT+MMT. Although MCLT animals do not develop cartilage damage within 8 weeks, we have previously described pathological bone and synovial remodeling with MCLT alone38. In this study, bone remodeling was not found in MCLT animals, likely due to sample size and timepoint differences relative to prior work. However, entropy of cells in synovial intima showed a difference between all groups with increasing entropy due to increasing severity of joint injury. Synovial inflammation response may influence autonomic cardiovascular changes rather than bone and cartilage damage per se. Thus, further exploration of autonomic changes due to different knee traumas may help provide explanations for the variability related to the type of injury and rate of joint remodeling.

During low activity, heart rate decreased in ligament injury and combined ligament and meniscal injury groups. While the cause of changes in resting heart rate were not directly tested, it may be due to autonomic imbalance. Cardiac vagal efferent fibers work to slow heart rate; thus, a decrease in heart rate could be caused by increased cardiac vagal tone and possible vagal sensitivity. Our data also suggests sensitization of joint injury groups to cardiac vagal activation via a pharmacological stimulant of vagus nerve afferent fibers (1-phenylbiguanide). Possibly, vagal afferents are more sensitive with ligament injury or OA due to persistent stimulation via inflammatory mediators in the periphery. Sympathetic afferent input is altered due to persistent inflammation40; however, this has not been shown for vagal afferent input in OA.

The discrepancy between potential longitudinal HRV and endpoint autonomic measures may be due to differences in autonomic tone and reactivity. MCLT+MMT animals may develop a chronic decrease in vagal tone but experience sensitized vagal reactivity to stimuli; however, these measures were inconclusive. Additionally, while HRV is a physiological parameter to indicate disease, we should be careful to not over-interpret results. For example, vagus nerve electrophysiology does not necessarily correlate with HRV41. This is a clear limitation of our study; however, our data do motivate future work to directly measure vagus nerve electrophysiology via non-survival surgeries.

Finally, the MCLT+MMT model is commonly performed in the Lewis rat, recommended based on noted strain differences42. However, after completion of this study, additional literature was uncovered regarding the hypothalamic-pituitary-adrenal (HPA)-axis dysfunction in Lewis rats, including decreased corticosterone release, altered stress response, and impaired response to inflammatory insult43,44. The autonomic nervous system interacts with the HPA-axis to regulate immune system functions. Due to these shifts in Lewis rats, our measures of autonomic impairment related to OA may be blunted. Future work should investigate other strains to evaluate if these responses are greater in other strains, and include corticosterone measurements to evaluate HPA-axis function.

In conclusion, the combination of our data provides preliminary quantitative evidence of functional changes in the autonomic nervous system related to knee ligament injury and OA in the rat, including evidence of crosstalk between mechanosensitive joint afferent fibers and cardiac vagal efferent fibers. Moreover, our measures of cardiovascular function suggest the vagus nerve may become sensitized in surgical models of ligament injury and OA, which may relate to changes in heart rate parameters. Combined, our data indicate the brain-joint axis may become altered with joint injury and OA pathophysiology and help explain physiologic links between knee OA and cardiovascular dysfunction. While these data are inconclusive and clear limitations exist regarding animal strain and mechanical stimuli used for the knee, the data support the potential for autonomic dysfunction with chronic joint injury and the need for further study into the brain-joint axis shifts with OA progression.

Supplementary Material

Supplemental Figures

Acknowledgements

The authors gratefully thank Dr. Terrie Vasilopoulos for her biostatistical support and Dr. David Baekey for his continued training and support for this project. We also thank Sarika Karri, Taylor Harrell, Allison Kennedy, and Markia Bowe for their assistance in data collection, data analysis, and histological sectioning and staining. We are grateful to Dr. Aaron Mickle for allowing us to use his lab space and equipment. We also thank Dr. Brittany Partain, Savannah Dewberry, Kiara Chan and Dr. Yash Shah for assistance during animal surgeries. We thank Tolulope Ajayi, Carlos Cruz, Jacob Griffith, and Jessica Aldrich and the reviewers of this manuscript for providing insightful feedback instrumental in the improvement of the presentation of our work. Finally, we thank Pavel Sul, Matthew Becker, and Rachel Hybart for their emotional support during long hours of data collection, analysis, and manuscript drafting.

Funding sources

This study was supported by the National Institute of Arthritis and Musculoskeletal and Skin Diseases of the National Institutes of Health (NIAMS/NIH) under award number R01AR071431, by a fellowship from the Nanoscale Institute for Medical and Engineering Technology at the University of Florida, and by a graduate student fellowship from the Herbert Wertheim College of Engineering and J. Crayton Pruitt Family Department of Biomedical Engineering at the UF. This study was also supported by a NIAMS/NIH Ruth L Kirschstein National Research Service Award (TDY), award number F31AR077996. Beyond providing funds, these sources did not participate in data collection, analysis, interpretation or decision to submit this publication.

Footnotes

Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.joca.2021.12.008.

Conflict of interest

The authors declare that they have no conflicts of interest.

References

  • 1.Eitner A, Pester J, Nietzsche S, Hofmann GO, Schaible H-G. The innervation of synovium of human osteoarthritic joints in comparison with normal rat and sheep synovium. Osteoarthritis Cartilage 2013;21(9):1383–91, 10.1016/j.joca.2013.06.018. [DOI] [PubMed] [Google Scholar]
  • 2.Woolf CJ. Central sensitization: implications for the diagnosis and treatment of pain. Pain 2011;152(Suppl 3), 10.1016/j.pain.2010.09.030. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Courties A, Sellam J, Berenbaum F. Role of the autonomic nervous system in osteoarthritis. Best Pract Res Clin Rheumatol 2017;31(5):661–75, 10.1016/j.berh.2018.04.001. [DOI] [PubMed] [Google Scholar]
  • 4.Berenbaum F, Meng QJ. The brain-joint axis in osteoarthritis: nerves, circadian clocks and beyond. Nat Rev Rheumatol 2016;12(9):508–16, 10.1038/nrrheum.2016.93. [DOI] [PubMed] [Google Scholar]
  • 5.Zanos TP, Silverman HA, Levy T, Tsaava T, Battinelli E, Lorraine PW, et al. Identification of cytokine-specific sensory neural signals by decoding murine vagus nerve activity. Proc Natl Acad Sci U S A 2018;115(21):E4843–52, 10.1073/pnas.1719083115. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Bornstein JC, Damon DH, Keating D, Browning KN. Role of central vagal 5-HT 3 receptors in gastrointestinal physiology and pathophysiology. Front Neurosci 2015;9:413, 10.3389/fnins.2015.00413, www.frontiersin.org. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Uneyama H, Niijima A, Tanaka T, Torii K. Receptor subtype specific activation of the rat gastric vagal afferent fibers to serotonin. Life Sci 2002;72:415–23, 10.1016/S0024-3205(02)02271-3. [DOI] [PubMed] [Google Scholar]
  • 8.Berntson GG, Thomas Bigger J, Eckberg DL, Grossman P, Kaufmann PG, Malik M, et al. Heart rate variability: origins methods, and interpretive caveats. Psychophysiology 1997;34(6):623–48, 10.1111/j.1469-8986.1997.tb02140.x. [DOI] [PubMed] [Google Scholar]
  • 9.Task Force of the European Society of Cardiology the North American Society of Pacing Electrophysiology. Heart rate variability Standards of measurement, physiological interpretation, and clinical use. Circulation 1996;93(5):1043–65. Heart Rate Variability : Standards of Measurement , Physiological Interpretation , and Clinical Use. [PubMed] [Google Scholar]
  • 10.Shaffer F, Ginsberg JP. An overview of heart rate variability metrics and norms. Front Public Health 2017;5, 10.3389/fpubh.2017.00258. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Koopman FA, Tang MW, Vermeij J, de Hair MJ, Choi IY, Vervoordeldonk MJ, et al. Autonomic dysfunction precedes development of rheumatoid arthritis: a prospective cohort study. EBIOM 2016;6:231–7, 10.1016/j.ebiom.2016.02.029. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.van Maanen MA, Stoof SP, LaRosa GJ, Vervoordeldonk MJ, Tak PP. Role of the cholinergic nervous system in rheumatoid arthritis: aggravation of arthritis in nicotinic acetylcholine receptor alpha7 subunit gene knockout mice. Dis 2010;69: 1717–23, 10.1136/ard.2009.118554. [DOI] [PubMed] [Google Scholar]
  • 13.Koopman FA, Chavan SS, Miljko S, Grazio S, Sokolovic S, Schuurman PR, et al. Vagus nerve stimulation inhibits cytokine production and attenuates disease severity in rheumatoid arthritis. Proc Natl Acad Sci 2016;113(29):8284–9, 10.1073/pnas.1605635113. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Goldring MB, Otero M. Inflammation in osteoarthritis. Curr Opin Rheumatol 2011;23(5):471–8, 10.1097/BOR.0b013e328349c2b1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Houard X, Goldring MB, Berenbaum F. Homeostatic mechanisms in articular cartilage and role of inflammation in osteoarthritis. Curr Rheumatol Rep 2013;15(11):375, 10.1007/s11926-013-0375-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Spitoni GF, Ottaviani C, Petta AM, Zingaretti P, Aragona M, Sarnicola A, et al. Obesity is associated with lack of inhibitory control and impaired heart rate variability reactivity and recovery in response to food stimuli. Int J Psychophysiol 2017;116:77–84, 10.1016/j.ijpsycho.2017.04.001. [DOI] [PubMed] [Google Scholar]
  • 17.Vincent A, McAllister SJ, Singer W, Toussaint LL, Sletten DM, Whipple MO, et al. A report of the autonomic symptom profile in patients with fibromyalgia. J Clin Rheumatol 2014;20(2): 106–8, 10.1097/RHU.0b013e3182a225dd. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Cortelli P, Giannini G, Favoni V, Cevoli S, Pierangeli G. Nociception and autonomic nervous system. Neurol Sci 2013;34(Suppl 1):S41–6, 10.1007/s10072-013-1391-z. [DOI] [PubMed] [Google Scholar]
  • 19.Huan-Ji D, Shi-Yi Z. Involvement of solitary tract nucleus in control of nociceptive transmission in cat spinal cord neurons. Pain 1990;40(3):323–31, 10.1016/0304-3959(90)91129-7. [DOI] [PubMed] [Google Scholar]
  • 20.Santisteban MM, Zubcevic J, Baekey DM, Raizada MK. Dysfunctional brain-bone marrow communication: a paradigm shift in the pathophysiology of hypertension. Curr Hypertens Rep 2013;15(4):377–89, 10.1007/s11906-013-0361-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Rosas-Ballina M, Ochani M, Parrish WR, Ochani K, Harris YT, Huston JM, et al. Splenic nerve is required for cholinergic antiinflammatory pathway control of TNF in endotoxemia. Proc Natl Acad Sci 2008;105(31):11008–13, 10.1073/pnas.0803237105. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Witt KL, Vilensky JA. The anatomy of osteoarthritic joint pain. Clin Anat 2014;27(3):451–4, 10.1002/ca.22120. [DOI] [PubMed] [Google Scholar]
  • 23.Courties A, Do A, Leite S, Senay S, Pigenet A, Belle M, et al. Activating the cholinergic system a novel opportunity for treating osteoarthritis. Osteoarthritis Cartilage 2019;27:S38, 10.1016/j.joca.2019.02.054. [DOI] [Google Scholar]
  • 24.Lorenz J, Schäfer N, Bauer R, Jenei-Lanzl Z, Springorum RH, Grässel S. Norepinephrine modulates osteoarthritic chondrocyte metabolism and inflammatory responses. Osteoarthritis Cartilage 2016;24(2):325–34, 10.1016/J.JOCA.2015.08.007. [DOI] [PubMed] [Google Scholar]
  • 25.Courties A, Do A, Leite S, Legris M, Sudre L, Pigenet A, et al. The role of the non-neuronal cholinergic system in inflammation and degradation processes in osteoarthritis. Arthritis Rheumatol 2020;72(12):2072–82, 10.1002/ART.41429. [DOI] [PubMed] [Google Scholar]
  • 26.Esler M. Mental stress and human cardiovascular disease. Neurosci Biobehav Rev 2017;74:269–76, 10.1016/j.neubiorev.2016.10.011. [DOI] [PubMed] [Google Scholar]
  • 27.Pulopulos MM, Vanderhasselt MA, De Raedt R. Association between changes in heart rate variability during the anticipation of a stressful situation and the stress-induced cortisol response. Psychoneuroendocrinology 2018;94:63–71, 10.1016/j.psyneuen.2018.05.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Rodgers RJ, Dalvi A. Anxiety, defence and the elevated plusmaze. In: Neuroscience and Biobehavioral Reviews 1997;vol. 21:801–10, 10.1016/S0149-7634(96)00058-9. [DOI] [PubMed] [Google Scholar]
  • 29.Zubcevic J, Watkins J, Perez PD, Colon-Perez LM, Long MT, Febo M, et al. MEMRI reveals altered activity in brain regions associated with anxiety, locomotion, and cardiovascular reactivity on the elevated plus maze in the WKY vs SHR rats. Brain Imaging Behav 2018;12(5):1318–31, 10.1007/s11682-017-9798-4. [DOI] [PubMed] [Google Scholar]
  • 30.Huetteman DA, Bogie H. Direct blood pressure monitoring in laboratory rodents via implantable radio telemetry. In: DiPetrillo K, Ed. Cardiovascular Genomics: Methods and Protocols. Totowa, NJ: Humana Press; 2009:57–73, 10.1007/978-1-60761-247-6_4. [DOI] [PubMed] [Google Scholar]
  • 31.Janusz MJ, Bendele AM, Brown KK, Taiwo YO, Hsieh L, Heitmeyer SA. Induction of osteoarthritis in the rat by surgical tear of the meniscus: inhibition of joint damage by a matrix metalloproteinase inhibitor. Osteoarthritis Cartilage 2002;10(10):785–91, 10.1053/joca.2002.0823. [DOI] [PubMed] [Google Scholar]
  • 32.Waki H, Katahira K, Polson JW, Kasparov S, Murphy D, Paton JFR. Automation of analysis of cardiovascular autonomic function from chronic measurements of arterial pressure in conscious rats. Exp Physiol 2006;91(1):201–13, 10.1113/expphysiol.2005.031716. [DOI] [PubMed] [Google Scholar]
  • 33.Porter K, Ahlgren J, Stanley J, Hayward LF. Modulation of heart rate variability during severe hemorrhage at different rates in conscious rats. Auton Neurosci Basic Clin 2009;150(1–2): 53–61, 10.1016/j.autneu.2009.04.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Tarvainen MP, Lipponen JA, Niskanen J-P, Ranta-aho PO. Kubios HRV User's Guide 2021. [DOI] [PubMed] [Google Scholar]
  • 35.Dutta A, Deshpande SB. Cardio-respiratory reflexes evoked by phenylbiguanide in rats involve vagal afferents which are not sensitive to capsaicin. Acta Physiol 2010;200(1):87–95, 10.1111/j.1748-1716.2010.02105.x. [DOI] [PubMed] [Google Scholar]
  • 36.Pritzker KPH, Gay S, Jimenez SA, Ostergaard K, Pelletier J-P, Revell PA, et al. Osteoarthritis cartilage histopathology: grading and staging. Osteoarthritis Cartilage 2006;14(1): 13–29, 10.1016/j.joca.2005.07.014. [DOI] [PubMed] [Google Scholar]
  • 37.Kloefkorn HE, Jacobs BY, Xie DF, Allen KD. A graphic user interface for the evaluation of knee osteoarthritis (GEKO): an open-source tool for histological grading. Osteoarthritis Cartilage 2019;27(1):114–7, 10.1016/j.joca.2018.09.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Kloefkorn HE, Allen KD, Crayton Pruitt J. Quantitative histological grading methods to assess subchondral bone and synovium changes subsequent to medial meniscus transection in the rat HHS Public Access. Connect Tissue Res 2017;58(4): 373–85, 10.1080/03008207.2016.1251425. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Schuelert N, McDougall JJ. Grading of monosodium iodoacetate-induced osteoarthritis reveals a concentration-dependent sensitization of nociceptors in the knee joint of the rat. Neurosci Lett 2009;465(2):184–8, 10.1016/j.neulet.2009.08.063. [DOI] [PubMed] [Google Scholar]
  • 40.Pongratz G, Straub RH. The sympathetic nervous response in inflammation. Arthritis Res Ther 2014;16(1), 10.1186/s13075-014-0504-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Marmerstein JT, McCallum GA, Durand DM. Direct measurement of vagal tone in rats does not show correlation to HRV. Sci Rep 2021;11(1):1210, 10.1038/s41598-020-79808-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Gerwin N, Bendele AM, Glasson S, Carlson CS. The OARSI histopathology initiative - recommendations for histological assessments of osteoarthritis in the rat. Osteoarthritis Cartilage 2010;18(Suppl 3), 10.1016/j.joca.2010.05.030. [DOI] [PubMed] [Google Scholar]
  • 43.Cohen H, Zohar J, Gidron Y, Matar M, Belkind D, Loewenthal U, et al. Blunted HPA Axis response to stress influences susceptibility to posttraumatic stress response in rats. Biol Psychiatr 2006;59(12):1208–18, 10.1016/j.biopsych.2005.12.003. [DOI] [PubMed] [Google Scholar]
  • 44.Sternberg EM, Hill JM, Chrousos GP, Kamilaris T, Listwak SJ, Gold PW, et al. Inflammatory Mediator-Induced Hypothalamic-Pituitary-Adrenal axis Activation Is Defective in Streptococcal Cell Wall Arthritis-Susceptible Lewis Rats. Proc Natl Acad Sci U S A 1989;86(April):2374–8. April. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

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

Supplementary Materials

Supplemental Figures
NIHMS1815665-supplement-Supplemental_Figures.docx (146.2KB, docx)

RESOURCES