Peripheral nerve trauma leads to degeneration of axons and their myelin sheaths distal to the injury site, a process called Wallerian degeneration. The degeneration of injured axons triggers a complex multicellular response involving the injured neurons, Schwann cells, and immune cells. Macrophages are the most notable immune cell population; playing a key role in degeneration and repair. Immediately after axotomy Schwann cells begin to express proinflammatory cytokines and chemokines that recruit and activate endoneurial macrophages, and later, blood-derived macrophages. Endoneural and blood-derived macrophages rapidly phagocytose disintegrated axons and myelin debris. The removal of degenerated axons and myelin debris is an important prerequisite for axonal regrowth because peripheral myelin contains axonal growth inhibitors such as myelin-associated glycoprotein [1,2,3]. Macrophages also play an active role in axonal regeneration and nerve repair. They release cytokines such as tumor necrosis factor (TNF)- α, chemokines, growth factors, and stimulate extracellular matrix remodeling to promote axonal regrowth. Furthermore, they orchestrate other cell types and polarize vasculatures that aid in Schwann cell migration [2,3,4]. On the other hand, macrophages and macrophage-derived proinflammatory cytokines like TNF- α are critically involved in maladaptive responses as neuropathic pain .
Depending on the severity and location of the injury, functional recovery is often incomplete resulting in motor, sensory, and autonomic deficits . The elucidation of signaling molecules that modulate the macrophage response, may help to develop new therapeutic strategies to improve peripheral nerve system (PNS) regeneration.
CXCR3 is chemokine (C-X-C motif) receptor 3, which is expressed by multiple leukocyte lineages, including T cells and macrophages . Three ligands, the interferon-γ-inducible chemokines CXCL9, CXCL10, and CXCL11 bind to and activate CXCR3 . CXCR3 is commonly known to regulate trafficking of T lymphocytes and natural killer cells, but there is increasing evidence, that CXCR3 also plays an important role in macrophage chemotaxis, both under physiological and pathological conditions . In different central nervous system (CNS) disease models, CXCR3-deficiency leads to reduced microglial activation accompanied by improved clinical outcomes and preserved neuronal structures [8,9,10]. In the PNS, an upregulation of the CXCR3 ligand, CXCL10 was shown in nerve injury-induced neuropathic pain and in autoimmune inflammatory polyneuropathy, both diseases with prominent macrophage involvement [11,12,13].
In the present study, we investigated the effect of CXCR3 on macrophage number, expression of the proinflammatory cytokine TNF- α and the CXCR3 agonist CXCL10, as well as on functional recovery following crush injury of the sciatic nerve with the use of CXCR3 knock out mice (CXCR3–/–) and wildtype littermates.
Materials and methods
The generation and characterization of the CXCR3–/– mice were described elsewhere . The mice used in this study were backcrossed for 12 generations to the C57BL/6J strain. The animals were housed in groups of six mice per cage under controlled illumination (light-dark cycle: 12:12 h) and stable environmental conditions (temperature: 22 ± 2°C; humidity: 55 ± 5%). All mice had free access to water and food pellets. Animals were housed in the cages for at least 1 week before the start of the experiments. Experiments were carried out in accordance with the Council Directive 2010/63 EU of the European Parliament, the Council of 22 September 2010 on the protection of animals used for scientific purposes, the guidelines of the German Animal Protection Law, and were approved by local authorities (Landesamt für Natur, Umwelt und Verbraucherschutz NRW, AZ: 8.87-51.04.20.09.353). To estimate animal numbers and group sizes, a power analysis based on previous experiments was performed. The significance level (Type 1 error) was 5%, the power (Type 2 error) was 20%, and Cohen’s effect size was 0.8. Furthermore, animal numbers were adopted from previous experiments . All efforts were made to minimize the number of animals used and their suffering.
Sciatic nerve injury
The surgical procedure was carried out following sterile precautions. For this, mice were initially anesthetized with an oxygen/isoflurane mixture (2–2.5% in 95% O2), fixed on the surgery table, and kept under a constant stream of isoflurane (1.5–2% in 95% O2) to maintain anesthesia. After shaving the fur and disinfection, a lateral skin incision was made along the length of the right femur. The right sciatic nerve was exposed and crushed distal to the sciatic notch using a fine 10.5-cm anatomical forceps for 20 s. For the sham-operated mice, the right sciatic nerve was exposed, but not crushed. The surgical site was closed using suture wound clips. All animals appeared healthy in the postoperative care room [16,17]. Crushed sciatic nerves were harvested 3, 7 and 28 days post-lesion. Left-side, untreated sciatic nerves served as controls.
At the predetermined time points, mice were deeply anesthetized with isoflurane (3% in 95% O2) and intracardially perfused with 4% paraformaldehyde in 0.1 M phosphate-buffered solution. Ipsilateral (injured and sham, respectively) and contralateral (control) sciatic nerves were harvested and immediately placed in phosphate-buffered saline buffered 4% paraformaldehyde (pH 7.4; Sigma-Aldrich, Darmstadt, Hesse, Germany) for 3 h at 4°C. Sciatic nerve samples were then embedded in Tissue Tek (Sakura Finetek, Staufen im Breisgau, Baden-Württemberg, Germany), snap-frozen in liquid nitrogen-cooled methyl butane (Sigma-Aldrich) and stored at –80°C. Ten-micrometer thick transversal sections were prepared from the cryoblock at the site of the crush injury and stored at –80°C until used. Fluorescence immunohistochemistry on frozen sections was performed for 16–18 h at 4°C using the rat anti-CD68 antibody (Serotec, Puchheim, Bavaria, Germany) (dilution 1:250), followed by incubation for 45 min with the secondary species-specific antibody (dilution 1:250) conjugated with Alexa Fluor 594. Sections were mounted and counterstained with 4′,6′-diamidino-2-phenylindole (DAPI) (Sigma-Aldrich).
Immunofluorescence-stained sections were examined using a Nikon Eclipse E 800 microscope (Nikon, Chiyoda, Tokio, Japan) and images were acquired at 40x with a fluorescence camera (SPOT camera system, Visitron, Göttingen, Lower Saxony, Germany) attached to a computer (Apple, Cupertino, Silicon Valley, California, USA) for imaging processing using the software programs Spot (SPOT Imaging) and analySIS Image Processing (EMSIS). Analysis 3.1 was also used to determine the endoneurial area under study. Three whole cross-sections of the injury site were examined per animal. Only CD68+ signals, which could be clearly attributed to a cell core (counterstaining with DAPI) were assessed as macrophages. CD68+ cells were counted from captured images and the nerve tissue area was calculated by the software. The sum of immunoreactive cells was graphed and analyzed.
Quantitative real-time PCR
Total RNA was isolated and purified from aliquots of homogenized nerve tissue of the injury site using TRIzol reagent (Sigma-Aldrich). The samples with TRIzol were roughly shaken and homogenized using the Precellys homogenizer system (Bertin Technologies, Montigny-le-Bretonneux, France). RNA quantity was determined using a NanoDrop 1000 (PEQLAB, Erlangen, Bavaria, Germany). Up to 3 μg of total RNA was reverse-transcribed into cDNA using SuperScript III Reverse Transcriptase (Invitrogen, Carlsbad, California, USA). Real-time quantitative PCR assays were performed on a StepOnePlus real-time PCR (RT-PCR) System (Applied Biosystems, Foster City, California, USA) using TaqMan Gene Expression Assays (Applied Biosystems) for TNF- α and CXCL10. Samples were simultaneously analyzed for GAPDH mRNA as internal control. Each sample was assayed in duplicate and normalized to the internal control, and data were determined as copies of mRNA/GAPDH (means ± SEM).
Test of sciatic nerve function/inverted mesh grid grip test
Sciatic nerve function was tested according to a modified version of the inverted mesh grid grip test published by Kondziela . The mice were placed on the center of a 43 cm2 of wire mesh consisting of 12 mm2 of 1 mm diameter wire. It was surrounded by a 4-cm thick deep wooden beading which prevented mice from climbing to the converse side. Once a mouse was positioned, the grid was turned upside down and elevated in order to force the mouse to grip the wire to avoid falling. Mice were scored as followed: five points = no gripping with the injured hind limb, 2.5 points = the foot of the injured limb was placed on the grip, but strength was not enough to grip, 0 points = mouse grips with the injured limb to the mesh and hold itself tightly. Each mouse was tested 10 times at each time point and mean values were calculated. Between repeated measurements, mice were returned to the home cage to allow recovery of muscular strength.
Statistical analysis of macrophage counts, RT-PCR, and inverted Mesh Grid Grip Test was performed using two-way analysis of variance (ANOVA) followed by an appropriate post-hoc test. The significance level was set to P ≤ 0.05. Results are presented as average ± SEM.
Decreased numbers of macrophages after sciatic nerve crush-injury in CXCR3–/– mice
To assess whether macrophage recruitment in Wallerian degeneration is mediated by CXCR3, we determined macrophage numbers in wildtype and CXCR3–/– mice at different time points following sciatic nerve crush injury. In wildtype mice, macrophage numbers of the ipsilateral sciatic nerve continuously increased until day 28 post-injury (4-fold at day 3, 11-fold at day 7, and 17-fold at day 28, respectively). In contrast, macrophages count of CXCR3–/– mice increased until day 3 similar to the wildtype, but tended to be lower at day 7 and were clearly reduced at day 28 compared to wildtype mice. In both genotypes, no changes in macrophage counts were seen in unaffected contralateral or sham-treated sciatic nerves (Fig. 1) (P = 0.009, two-way ANOVA, Bonferroni post-hoc test; CXCR3–/– crush: day 3: n = 4, day 7: n = 3 and day 28: n = 6; wildtype crush: day 3: n = 4, day 7: n = 4, day 28: n = 6; CXCR3–/– contralateral: day 3, day 7, and day 28: n = 3 at each time point; wildtype contralateral: day 3, day 7, and day 28: n = 3 at each time point, CXCR3–/– sham: n = 3; wildtype sham: n = 3).
The expression of the proinflammatory cytokine tumor necrosis factor- α and the chemokine CXCL10 is reduced in CXCR3–/– mice following sciatic nerve crush
We next examined whether CXCR3 has an impact on the expression levels of the proinflammatory cytokine TNF- α and the CXCR3-agonist CXCL10. mRNA expression levels were analyzed in wildtype and CXCR3–/– mice at days 3 and 7 following sciatic nerve crush at the site of the injury and 5 mm distal. Unaffected contralateral sciatic nerves served as the control. Analysis showed a clear increase of mRNA expression levels of TNF- α in wildtype mice following sciatic nerve crush at both time points compared to control nerves, which was most pronounced at day 7. In contrast, TNF- α expression was clearly impaired in CXCR3–/– mice with a significant reduction of expression levels compared to wildtype mice 7 days after crush injury (day 7, injury site: P = 0.0032; day 7, 5 mm distal injury site: P < 0.0001, two-way ANOVA, Tukey post-hoc test, n = 6 for each subgroup) (Fig. 2a).
Three days following sciatic nerve crush, a clear increase of CXCL10 mRNA expression was observed both in wildtype and CXCR3–/– mice. Seven days after injury, the expression of CXCL10 mRNA further increased in wildtype mice, whereas in CXCR3–/– mice the expression stagnated and was significantly lower than in wildtype mice at this time point at both sites (injury site: P = 0.042, 5 mm distal: P = 0.036, two-way ANOVA, Tukey post-hoc test, n = 6 for each subgroup) (Fig. 2b).
Functional recovery following crush lesion of the sciatic nerve is accelerated in CXCR3–/– mice
To evaluate whether deletion of CXCR3 also affects functional recovery after sciatic nerve crush, we performed the inverted mesh grid grip test in wildtype and CXCR3–/– mice. The grip strength test is a widely-used non-invasive method designed to evaluate mouse limb strength. Mice of both genotypes displayed a complete loss of sciatic nerve function in the first days following sciatic nerve crush (score 5). Sciatic functional scores gradually improved to 0 as repair occurred. Comparing both genotypes revealed that CXCR3–/– mice displayed an earlier and faster regeneration of sciatic functional scores as wildtype mice (day 12: P < 0.001, day 14: P < 0.001, day 18: P < 0.05, day 21: P < 0.05, two-way ANOVA, Bonferroni post-hoc test, n = 5 for each subgroup) (Fig. 3).
Restoration after peripheral nerve injury is often incomplete resulting in poor clinical outcomes. The complex mechanisms of nerve degeneration and regeneration are still not fully understood. We showed in the present study that deletion of CXCR3 leads to a diminished increase of macrophage counts, a diminished expression of CXCL10 and TNF- α and an accelerated functional recovery following sciatic nerve crush.
Macrophages are key players in the first steps of Wallerian degeneration that perform a primary role in debris removal and cytokine production, whereas in later stages they play a regeneration promoting role [2,18]. From day 2 on, and peaking at day 14, macrophage numbers increase considerably, then exhibit a continuous decline. Nevertheless, elevated macrophage counts are detectable for weeks and months after injury [3,15]. Our data reveal that the increase of macrophage numbers in the early phase following sciatic nerve crush is unaffected in CXCR3–/– mice (day 3), but in the later course, starting at day 7 and most obviously at day 28, the increase in macrophage numbers stagnates and is reduced in comparison to wildtype mice.
The reduced density of macrophages in CXCR3–/– mice were preceded by a decreased TNF- α and CXCL10 mRNA expression. TNF- α expression is elevated during the first hours to 3 days following peripheral nerve injury, and then starts to decline by day 7. This rapid injury-induced release of TNF- α is essential for the infiltration of neutrophils and proinflammatory macrophages into the distal stump of the lesioned sciatic nerve [2,19]. The reduced TNF- α expression in CXCR3–/– mice (days 3 and 7), may, therefore, be causally linked to the reduced macrophage counts starting at day 7.
To our knowledge, there is no present study investigating a possible contribution of the CXCR3 agonist CXCL10 in Wallerian degeneration. Our data show that CXCL10 mRNA expression is upregulated following crush lesion of the sciatic nerve. This is in line with the prominent role of the CXCL10/C-X-C chemokine receptor axis in leukocyte recruitment and activation in traumatic optic nerve injury and its involvement in peripheral nerve trauma-induced neuropathic pain [12,20]. In CXCR3–/– mice, CXCl10 was significantly diminished at day 7 suggesting that deletion of CXCR3 leads to negative feedback on CXCL10 production. However, the exact pathophysiological mechanism remains to be explored.
In addition to reduced inflammatory activity, a significantly accelerated functional recovery of sciatic motor nerve function was found in CXCR3–/– mice compared to wildtype mice. Simultaneously to the reduced inflammatory activity, improvement of functional scores started at day 7/day 12 suggesting a causal link.
In summary, our data suggest, that CXCR3 does not affect the initial inflammatory response in Wallerian degeneration, but that CXCR3 maintains elevated macrophage counts and the activity of inflammatory mediators and therefore has a negative effect on functional recovery.
The effect of CXCR3 deficiency in our sciatic nerve crush model is in line with the role of CXCR3 in several CNS diseases. In animal models of prion disease and herpes simplex virus encephalitis, CXCR3 deficiency leads to reduced microglial recruitment, reduced expression of inflammatory cytokines, including CXCL10, and improved clinical course [8,9]. In optic nerve crush and axotomy of the perforant path, knock out of CXCR3 leads to a reduction of microglial recruitment and preserves neuronal structures [10,20].
Several classes of small molecule CXCR3 antagonists have been developed in the last few years . AMG487, a potent orally bioavailable CXCR3 antagonist, partially prevents neuronal loss and improves functional outcomes in the optic nerve crush model . Future studies are needed to investigate whether pharmacological CXCR3 blockers exert a positive effect on the outcome of peripheral mechanical nerve lesions.
We thank Katlynn Carter and Bernd Evert for critically reading the manuscript.
We thank Anna Laura Potthoff for illustrations.
Conflicts of interest
There are no conflicts of interest.
1. Barrette B, Hébert MA, Filali M, Lafortune K, Vallières N, Gowing G, et al. Requirement of myeloid cells for axon regeneration. J Neurosci. 2008; 28:9363–9376
2. Chen P, Piao X, Bonaldo P. Role of macrophages
in Wallerian degeneration and axonal regeneration after peripheral nerve injury. Acta Neuropathol. 2015; 130:605–618
3. Gaudet AD, Popovich PG, Ramer MS. Wallerian degeneration: gaining perspective on inflammatory events after peripheral nerve injury. J Neuroinflammation. 2011; 8:110
4. Cattin AL, Burden JJ, Van Emmenis L, Mackenzie FE, Hoving JJ, Garcia Calavia N, et al. Macrophage-induced blood vessels guide Schwann cell-mediated regeneration of peripheral nerves. Cell. 2015; 162:1127–1139
5. Leung L, Cahill CM. TNF-α and neuropathic pain – a review. J Neuroinflammation. 2010; 7:1–11
6. Torraca V, Cui C, Boland R, Bebelman JP, van der Sar AM, Smit MJ, et al. The CXCR3
-CXCL11 signaling axis mediates macrophage recruitment and dissemination of mycobacterial infection. Dis Model Mech. 2015; 8:253–269
7. Müller M, Carter S, Hofer MJ, Campbell IL. Review: the chemokine receptor CXCR3
and its ligands CXCL9, CXCL10
and CXCL11 in neuroimmunity – a tale of conflict and conundrum. Neuropathol Appl Neurobiol. 2010; 36:368–387
8. Zimmermann J, Hafezi W, Dockhorn A, Lorentzen EU, Krauthausen M, Getts DR, et al. Enhanced viral clearance and reduced leukocyte infiltration in experimental herpes encephalitis after intranasal infection of CXCR3
-deficient mice. J Neurovirol. 2017; 23:394–403
9. Riemer C, Schultz J, Burwinkel M, Schwarz A, Mok SW, Gültner S, et al. Accelerated prion replication in, but prolonged survival times of, prion-infected CXCR3
-/- mice. J Virol. 2008; 82:12464–12471
10. Rappert A, Bechmann I, Pivneva T, Mahlo J, Biber K, Nolte C, et al. CXCR3
-dependent microglial recruitment is essential for dendrite loss after brain lesion. J Neurosci. 2004; 24:8500–8509
11. Chiang S, Ubogu EE. The role of chemokines in Guillain-Barré syndrome. Muscle Nerve. 2013; 48:320–330
12. Fu ES, Zhang YP, Sagen J, Candiotti KA, Morton PD, Liebl DJ, et al. Transgenic inhibition of glial NF-kappa B reduces pain behavior and inflammation after peripheral nerve injury. Pain. 2010; 148:509–518
13. Kieseier BC, Tani M, Mahad D, Oka N, Ho T, Woodroofe N, et al. Chemokines and chemokine receptors in inflammatory demyelinating neuropathies: a central role for IP-10. Brain. 2002; 125:823–834
14. Hancock WW, Lu B, Gao W, Csizmadia V, Faia K, King JA, et al. Requirement of the chemokine receptor CXCR3
for acute allograft rejection. J Exp Med. 2000; 192:1515–1520
15. Mueller M, Leonhard C, Wacker K, Ringelstein EB, Okabe M, Hickey WF, Kiefer R. Macrophage response to peripheral nerve injury: the quantitative contribution of resident and hematogenous macrophages
. Lab Invest. 2003; 83:175–185
16. Elfar JC, Jacobson JA, Puzas JE, Rosier RN, Zuscik MJ. Erythropoietin accelerates functional recovery after peripheral nerve injury. J Bone Joint Surg Am. 2008; 90:1644–1653
17. Lee H, Baek J, Min H, Cho IH, Yu SW, Lee SJ. Toll-like receptor 3 contributes to Wallerian degeneration after peripheral nerve injury. Neuroimmunomodulation. 2016; 23:209–216
18. Bosse F. Extrinsic cellular and molecular mediators of peripheral axonal regeneration. Cell Tissue Res. 2012; 349:5–14
19. Nadeau S, Filali M, Zhang J, Kerr BJ, Rivest S, Soulet D, et al. Functional recovery after peripheral nerve injury is dependent on the pro-inflammatory cytokines IL-1β and TNF: implications for neuropathic pain. J Neurosci. 2011; 31:12533–12542
20. Ha Y, Liu H, Zhu S, Yi P, Liu W, Nathanson J, et al. Critical role of the CXCL10
/C-X-C chemokine receptor 3 axis in promoting leukocyte recruitment and neuronal injury during traumatic optic neuropathy induced by optic nerve crush. Am J Pathol. 2017; 187:352–365
21. Andrews SP, Cox RJ. Small molecule CXCR3
antagonists. J Med Chem. 2016; 59:2894–2917