Neurostimulation of the cholinergic anti-inflammatory pathway ameliorates disease in rat collagen-induced arthritis

Abstract

Introduction: The inflammatory reflex is a physiological mechanism through which the nervous system maintains immunologic homeostasis by modulating innate and adaptive immunity. We postulated that the reflex might be harnessed therapeutically to reduce pathological levels of inflammation in rheumatoid arthritis by activating its prototypical efferent arm, termed the cholinergic anti-inflammatory pathway. To explore this, we determined whether electrical neurostimulation of the cholinergic anti-inflammatory pathway reduced disease severity in the collagen-induced arthritis model.

Methods: Rats implanted with vagus nerve cuff electrodes had collagen-induced arthritis induced and were followed for 15 days. Animals underwent active or sham electrical stimulation once daily from day 9 through the conclusion of the study. Joint swelling, histology, and levels of cytokines and bone metabolism mediators were assessed.

Results: Compared with sham treatment, active neurostimulation of the cholinergic anti-inflammatory pathway resulted in a 52% reduction in ankle diameter (p = 0.02), a 57% reduction in ankle diameter (area under curve; p = 0.02) and 46% reduction overall histological arthritis score (p = 0.01) with significant improvements in inflammation, pannus formation, cartilage destruction, and bone erosion (p = 0.02), accompanied by numerical reductions in systemic cytokine levels, not reaching statistical significance. Bone erosion improvement was associated with a decrease in serum levels of receptor activator of NF-κB ligand (RANKL) from 132±13 to 6±2 pg/mL (mean±SEM, p = 0.01).

Conclusions: The severity of collagen-induced arthritis is reduced by neurostimulation of the cholinergic anti-inflammatory pathway delivered using an implanted electrical vagus nerve stimulation cuff electrode, and supports the rationale for testing this approach in human inflammatory disorders.

Conflict of interest statement

Competing Interests: YAL, MF, AC and RZ are employees of SetPoint Medical Corporation. PPT is a consultant to and has received research grants from SetPoint Medical. PPT has become an employee of GlaxoSmithKline (GSK) after completion of the study. GSK has obtained an equity interest in Setpoint Medical Corporation. FAK, AB, and MJV have no competing interests to declare. This does not alter the authors' adherence to all PLOS ONE policies on sharing data and materials.

Figures

Figure 1. NCAP Delivery System.

(A) Schematic drawing of the percutaneous vagus nerve cuff electrode (B) Surgical implantation of the lead in the neck of a supine animal is shown. The slotted belt (asterisk) is wound around the entire carotid sheath (small arrow) containing the vagus nerve, bringing the electrode pair in close apposition to and in proper orientation on the nerve, allowing effective induction of vagus depolarization. The distal portion of the lead (large arrow) is tunneled subcutaneously to the back, and externalized between the scapulae. (C) Following surgical healing, the distal end of the lead is externalized and protected under a jacket with a Velcro closure. The distal lead is intermittently connected using alligator clips to an external pulse generator for daily active or sham NCAP delivery. (D) The waveform of one cycle of the charge-balanced biphasic pulse delivered by the pulse generator is illustrated schematically. Rats were stimulated once daily for 60 seconds from study day 9 to 15 with a pulse waveform amplitude (PA) of 3 mA, pulsewidth (PW) of 200 microseconds, pulse frequency of 10 Hz, and a 50 microsecond inter-pulse interval (IPI).

Figure 2. NCAP Reduces Clinical Signs of Joint Inflammation.

Rats were placed into four treatment groups, three of which received vagus nerve lead implants, with a fourth group of unimplanted controls. The three implanted groups had CIA induction or sham CIA induction, and active versus sham electrical neurostimulation on study days 9–15 as outlined in Table 1 (Control/No Implant n = 4; Control/NCAP n = 4; CIA/NCAP n = 9; CIA/Sham NCAP n = 12). Ankle diameter over time is shown as mean+SE (A), and AUC of ankle diameter over Day 9–16 is shown as mean+SE (B), *p≤0.05 ANOVA versus CIA/Sham NCAP.

Figure 3. NCAP Reduces Histological Measures of Joint Damage.

(A–B) Ankle joints were harvested on study day 16, stained with Toluidine Blue, and scored on a scale of 0-5 for inflammation, pannus formation, cartilage damage and bone resorption (A), with a composite summated score of 0–20 (B). Data are shown as mean+SE score. *p≤0.05 t-test versus CIA/Sham NCAP. (C–D) Representative 50X photomicrographs of ankle joints are shown which have the approximate mean summed score as that of the entire treatment group. Ankle from CIA/Sham NCAP group (C) demonstrates marked inflammation and synovitis (S) and mild cartilage damage (large arrow) and bone resorption (small arrow). Ankle from CIA/NCAP group (D) demonstrates mild inflammation and synovitis (S) and minimal cartilage damage (large arrow) and minimal bone resorption (small arrow).

Figure 4. NCAP Effect on Circulating Cytokines.

Study day 16 serum was assayed for IL-1α, IL-1β, IL-2, IL-6, IFN-γ and TNF. Data are shown as mean+SE level, t-test p = NS for all individual cytokine comparisons between CIA/NCAP and CIA/Sham NCAP.

Figure 5. NCAP Effect on Mediators of Bone Metabolism.

Study day 16 serum was assayed for RANKL (A), and OPG (B), and the ratio of OPG/RANKL group means calculated (C). Data are shown as mean+SE level, *p≤0.05, t-test versus CIA/Sham NCAP. Osteocalcin (D), P1NP (E), TRAP-5b (F) and CTX-1 (G) data are shown as mean+SE level, *p≤0.05 t-test versus CIA/Sham NCAP.