The role of chemokines in Guillain-Barré syndrome

Abstract

Introduction: Chemokines and their receptors are important mediators of inflammation. Guillain-Barré syndrome (GBS) is the most common cause of acute paralysis worldwide. Despite current treatments, outcomes are suboptimal. Specific chemokine receptor antagonists have the potential to be efficacious against pathogenic leukocyte trafficking in GBS.

Methods: A 36-year literature review was performed to summarize available data on chemokine expression in GBS and its representative animal model, experimental autoimmune neuritis (EAN).

Results: Although there were a few observational human and animal studies demonstrating chemokine ligand/receptor expression in GBS and EAN, in vitro and in vivo functional studies using gene knockouts, neutralizing antibodies, or small molecular antagonists were limited. CCL2-CCR2, CCL5-CCR5, and CXCL10-CXCR3 have been most strongly implicated in EAN and GBS pathogenesis, providing targets for molecular blockade.

Conclusions: Preclinical human in vitro and in vivo EAN studies are needed to evaluate the potential efficacy of chemokine signaling inhibition in GBS.

Keywords: Guillain-Barré syndrome; chemokine; chemokine receptor; experimental autoimmune neuritis; therapy.

FIGURE 1

Chemokine ligand and receptor expression and possible roles in GBS. Based on human observational studies and data obtained from EAN models, it is hypothesized that CCR2⁺ monocytes cross the BNB via interaction with Schwann-cell–secreted CCL2 expressed by the BNB endothelium. These monocytes differentiate into endoneurial macrophages that migrate to CCL2-expressing Schwann cells to induce demyelination. CCR1/CCR5⁺ macrophages may migrate to axons expressing CCL5 to induce axonal degeneration. CXCR3⁺ T cells may migrate across the BNB via interactions with BNB endothelium CXCL10 and directly contribute to nerve injury via cytokine secretion or expression of pro-inflammatory chemokines that further attract hematogenous monocytes and lymphocytes, enhancing the inflammatory process. The relationship between chemokines expressed within the endoneurium of peripheral nerves and nerve roots and chemokines expressed in CSF is undetermined.

FIGURE 2

Effect of CXCR3 antagonism on the behavioral manifestations of sm-EAN. The severity of EAN was assessed in age-matched female SJL/J mice using the 6-point neuromuscular severity scale (NMSS) after disease induction with purified BPNM. Maximum disease severity is expected between days 26–32 PI in this model. After disease onset, equally affected mice were treated with 5 mg/kg AMG 487 (CXCR3 antagonist) in 20% cyclodextrin, 400 mg/kg human intravenous immunoglobulin (hIVIg), or vehicle (20% cyclodextrin) via intraperitoneal injection once daily for 5 days from day 13 PI (black arrow). AMG 487 dramatically reversed disease progression to near baseline levels after 2 days, which persisted for the next 15 days. NMSS scale: 0—no weakness; 1—tail weakness; 2—mild-to-moderate fore- or hind-limb weakness; 3—severe fore- or hind-limb weakness; 4—mild-to-moderate tetraparesis; 5—severe tetraparesis. N = 3 per experimental group. Error bars indicate standard errors of the mean.

FIGURE 3

Effect of CXCR3 antagonism on sm-EAN motor nerve electrophysiology. Bilateral sciatic and dorsal caudal tail nerve electrophysiological studies were performed on anesthetized mice 30 days after EAN induction. Mice were treated with AMG 487 (CXCR3 antagonist), vehicle (negative control, 20% cyclodextran), and human intravenous immunoglobulin (hIVIg; positive control) on days 13–17. Distal compound motor action potential (dCMAP) amplitudes, conduction velocities, and total dCMAP waveform duration (d-Duration) were recorded. No statistically significant differences were observed in mean dCMAP amplitudes after AMG 487 treatment with either nerve (A, B), whereas mean conduction velocities were higher in both nerves, relative to vehicle and hIVIg (C, D). Total mean d-Durations were lower after AMG 487 treatment in both nerves (E, F). ^*P <0.05 relative to vehicle; ^#P <0.05 relative to hIVIg; N.S., not significant relative to vehicle (A–D, F) or hIVIg (E). N = 6 nerves per experimental group. Error bars indicate standard errors of the mean.