Neuregulin-ErbB signaling promotes microglial proliferation and chemotaxis contributing to microgliosis and pain after peripheral nerve injury

Abstract

A key component in the response of the nervous system to injury is the proliferation and switch to a "proinflammatory" phenotype by microglia (microgliosis). In situations where the blood-brain barrier is intact, microglial numbers increase via the proliferation and chemotaxis of resident microglia; however, there is limited knowledge regarding the factors mediating this response. After peripheral nerve injury, a dorsal horn microgliosis develops, which directly contributes to the development of neuropathic pain. Neuregulin-1 (NRG-1) is a growth and differentiation factor with a well characterized role in neural and cardiac development. Microglia express the NRG1 receptors erbB2, 3, and 4, and NRG1 signaling via the erbB2 receptor stimulated microglial proliferation, chemotaxis, and survival, as well as interleukin-1beta release in vitro. Intrathecal treatment with NRG1 resulted in microglial proliferation within the dorsal horn, and these cells developed an activated morphology. This microglial response was associated with the development of both mechanical and cold pain-related hypersensitivity. Primary afferents express NRG1, and after spinal nerve ligation (SNL) we observed both an increase in NRG1 within the dorsal horn as well as activation of erbB2 specifically within microglia. Blockade of the erbB2 receptor or sequestration of endogenous NRG after SNL reduced the proliferation, the number of microglia with an activated morphology, and the expression of phospho-P38 by microglia. Furthermore, consequent to such changes, the mechanical pain-related hypersensitivity and cold allodynia were reduced. NRG1-erbB signaling therefore represents a novel pathway regulating the injury response of microglia.

Figure 1.

ErbB receptor expression in primary cultured microglia. a, Using RT-PCR analysis, we found that primary cultured rat microglia express the erbB2, 3, and 4 receptors. The 2% agarose gels stained with ethidium bromide and photographed under ultraviolet light are shown for the PCR products of each of the genes amplified. Pos, Positive control (E16 embryonic rat brain cDNA); MG, microglia; Neg, negative control (water). b, ErbB receptors expression is shown in microglial cells by immunohistochemistry (IB4 labels microglia green, the erbB receptors are in red). Scale bar: 20 μm. c, Western blot analysis showing a 185 kDa band for erbB2, 3, and 4 in lysates from microglial cells.

Figure 2.

NRG1 stimulates microglial survival, proliferation, and chemotaxis in vitro. Survival of microglia in vitro was assessed by incubating cultured microglia for 3 d in serum-free medium, which was supplemented with increasing doses of NRG1 (0.5–10 nm) or GM-CSF (1 nm) (as a positive control). a, b, Representative wells are shown in which microglia are identified by Iba1 immunostaining after no treatment (a) or addition of 10 nm NRG1 (b). NRG1 promoted microglial survival, and erbB2 inhibition using PD168393 (INH, 10 μm) or mAb 7.16.4 (AB, 4 μg/ml) blocked this effect. c, Quantification of microglial numbers is shown. d–f, Cultured microglia were incubated for 1 day in serum-free medium with (e) or without (d) the addition of NRG1. DAPI staining (blue) demonstrated nuclei, and TUNEL (yellow) revealed those undergoing apoptosis. Treatment with NRG1 (10 nm) significantly decreased apoptosis, an effect that was prevented when microglia were treated with the erbB inhibitor PD168393 in combination with NRG1 (quantification in f). g–i, Proliferation was assessed by incubating microglia in medium supplemented with 5% FBS for 3 d and pulse-labeling with BrdU (yellow). NRG1 treatment significantly increased the proportion of BrdU-positive microglial nuclei, an effect that was erbB2-dependent. j–l, The effects of NRG1 on microglial migration were studied using a Boyden chamber. The addition of NRG1 to the lower well of the chamber (k) increased microglial migration to the inner membrane surface compared with control (j). Inhibition of erbB2 blocked this action. I, Quantification of migrated cells. Scale bars: 100 μm. Error bars represent ± SEM (three to five independent experiments). The statistical tests used were one-way ANOVA with Bonferroni post hoc analysis for all comparisons except for the proliferation experiment in which the data were not normally distributed and ANOVA on ranks with Dunn's post hoc test was used instead. *p < 0.05, **p < 0.005. CON, Control; INH, inhibitor (PD168393); AB, neutralizing antibody (7.16.4).

Figure 3.

The migration of microglial cells in response to NRG1 is a true chemotactic response. In a Boyden chamber, microglial cells migrate from an upper to a lower well through a polycarbonate filter in response to a concentration gradient of the putative chemotactic agent. A checkerboard analysis of three independent experiments assessing the chemotactic response to NRG1 performed in triplicate and normalized to the unstimulated control is shown. Microglial cells were suspended in medium alone or with 0.1, 1, or 10 nm NRG1 and then allowed to migrate for 3 h at 37°C toward different concentrations of NRG1 in the lower compartments. Highlighted in gray boxes are the results achieved by using the same concentration of NRG1 in upper and lower wells. Note that in this circumstance NRG1 increases microglial migration, indicating chemokinesis. When there is an increasing concentration gradient from the upper to the lower well (values in gray), migration is clearly enhanced, indicating a true chemotactic response. *p < 0.05 for migration across a gradient vs migration when the NRG1 concentration is the same in both wells. Numbers represent the mean ± SEM.

Figure 4.

NRG1 promotes the release of IL-1β from LPS-primed microglia but does not stimulate MHC-2 expression. a, An ELISA was used to quantify IL-1β released into the medium of cultured microglia. Neither ATP (1 mm) nor NRG1 (10 nm) alone produced IL-1β release from naive microglia; however, both significantly promoted IL-1β release from LPS (1 μg/ml)-primed microglia in a dose-dependent manner. Error bars: ± SEM. Statistical test: ANOVA on ranks. *p < 0.05, n = 3. b, Another aspect of the activation (or effector) response of microglia is that they increase expression of MHC-2. To investigate wherever NRG1 treatment leads to such a response in microglia, we incubated primary cultures in serum-free medium and treated them with LPS (1 μg/ml), NRG1 (10 nm), or medium alone (CON). Microglia were immunostained with OX-6 (which labels MHC-2), Iba1 (microglial marker), and DAPI (nuclear staining). LPS treatment led to a fivefold increase in OX-6 expression compared with control (p < 0.001), while NRG1 treatment did not elicit such a response (p = 1). Treatment with LPS also resulted in microglia adopting an amoeboid morphology, unlike NRG1. Scale bars: 100 μm. Error bars: ± SEM (three independent experiments) Statistical test: one-way ANOVA, Bonferroni post hoc analysis. **p < 0.001.

Figure 5.

Intrathecal administration of NRG1 results in dorsal horn microgliosis associated with mechanical and cold pain-related hypersensitivity. NRG1β was administered intrathecally (0.4 or 4 ng given daily for 3 d). a, b, Dorsal horn of animals treated with saline (control in a) or NRG1 (4 ng in b) 3 d after the first injection, immunostained with Iba1. Note the increase in the numbers of microglia with an activated morphology. c, Shows quantification of this response at 1 and 3 d after NRG1 injections. *p < 0.05, **p < 0.001 (comparing NRG1 doses vs control). d–l, We assessed proliferation (pulse labeling with BrdU) after NRG1 injections (Iba1 is shown in red, BrdU in green, DAPI in blue, and in the last panel merged images are shown). d–g, Dorsal horn microglia from a saline-treated animal (note that no BrdU is present). h–k, Dorsal horn microglia from a NRG1 (4 ng)-treated animal. l, Quantification of all BrdU-positive microglia in the dorsal horn of all groups at days 1 and 3 after injections were started. *p < 0.05 (comparing 4 ng vs control). Mechanical (m) and cold (n) pain-related hypersensitivity developed after NRG1 injections in a dose-dependent manner. *p < 0.05, **p < 0.001 for 4 ng vs control; ^#p < 0.05 for 0.4 ng versus control. Scale bars: (in b) a, b: 100 μm; (in k) d–k: 10 μm.

Figure 6.

Peripheral nerve injury leads to increased phospho-erbB2 expression within dorsal horn microglia. a, Three days after L5 SNL, the expression of the phosphorylated form of the erbB2 receptor was increased in the ipsilateral dorsal horn. b–d, Phospho-erbB2 immunostaining (green) colocalized with Iba1 (microglial marker in red, b), but not with GFAP (astrocyte marker in red, c) or NeuN (neuronal marker in red, d). e, f, P-erbB2 (green, e) also colocalized with OX-42 (a marker of activated microglia; red, f). Scale bars: a, 200 μm; b–d, 50 μm; e–g, 10 μm.

Figure 7.

The time course of phospho-erbB2 expression is coincident with the development of microgliosis. The microglial response to nerve injury develops during the first week after injury (demonstrated by immunostaining with the microglial marker Iba1). a, An increase in microglial numbers is observed in the ipsilateral dorsal and ventral horn of the spinal cord of L5 SNL animals. b, c, A closer view shows that microglia change their morphology from having long processes and a small soma (b, naive animal, termed surveying microglia) to having retracted processes and a hypertrophic soma (c, 3 d after L5 SNL, termed an effector or activated morphology). Simultaneously with this microgliosis, we observed an increased expression of the p-erbB2 receptor (green) in the ipsilateral dorsal horn microglia (labeled with Iba1 in red). d–g, Naive (d), 1 d after SNL (e), 3 d after SNL (f), and 21 d after SNL (g). There was a significant increase in both total microglial number and the proportion that were p-erbB2-positive within the dorsal horn after nerve injury (quantification shown in h). *p < 0.05 (n = 3–4; one-way ANOVA on ranks, post hoc Student–Newman–Keuls method). Scale bars: a, 200 μm; b, c, 10 μm; d–g, 100 μm.

Figure 8.

NRG1 expression after L5 spinal nerve ligation. a–f, In situ hybridization images using a probe directed against the βEGF domain of NRG1 in L5 DRG of naive animals (a–c) and 3 d after SNL (d–f). Expression is highest in large-diameter DRG cells positive for NF 200 (green) but is also observed in small DRG cells positive for either IB4 (blue) or CGRP (red). Arrows provide illustrative examples of double labeled cells. There was no significant change in the proportion of DRG cells expressing EGF mRNA at 3 d after SNL. g–k, IHC images. g, h, We used the H210 antibody, which is raised against the N-terminal extracellular domain of NRG1. To confirm specificity, we tested it in Nrg1 knock-out tissue (Nfh Cre Nrg1^f1/f1 mice in which Nrg1 is ablated in sensory neurons in the late embryonic period). Note that normal NRG1 immunoreactivity is observed in DRG cells of Nrg^f1/f1 mice (i.e., control), whereas no immunoreactivity is present DRG cells of Nfh Cre Nrg1^f1/f1 mice. i–k, Immunohistochemistry confirmed that NRG1 was present in ∼60% of DRG cells (i, j) and within the dorsal horn of the spinal cord (especially within superficial laminae, k). l, NRG1 expression was also assessed using a NRG1β-specific ELISA. Lysate from dorsal horn spinal cord of naive, and 1 and 3 d after SNL animals were run (50 μg of protein for each sample). All sample values were within the range of the ELISA detection. A significant increase in NRG1 was detected at 3 d after SNL (p = 0.04, one-way ANOVA, Bonferroni post hoc test, n = 3). Scale bars: f, h, j, 50 μm; k, 200 μm.

Figure 9.

ErbB2 receptor blockade or sequestration of endogenous NRG inhibits the development of microgliosis after spinal nerve ligation. The effect of erbB2 receptor blockers (PD168393, 5 μg/d, i.t.; or mAb 9 or 7.16.4, 5 μg, i.t.) or a NRG-sequestering molecule (HBD-S-H4, 3 μg, i.t.) was determined at 3 d after SNL. a–e, ErbB2 receptor blockade (b and e, low- and high-power photomicrographs, respectively) resulted in a reduction in both p-erbB2 (green) expression and in the number of microglia in the dorsal horn (Iba1 immunostaining, red) compared with vehicle (a and d, low- and high-power photomicrographs, respectively). Quantification (c) demonstrates that SNL results in an increased number of microglia with activated morphology compared with sham surgery animals. This increase was significantly attenuated after treatment with erbB2 inhibitor, erbB2 receptor-blocking antibodies, or the NRG1 antagonist (HBD-S-H4) with respect to control. f–j, SNL results in increased numbers of microglia expressing phospho-p38. The number of microglia (red) expressing p-p38 MAPK (green) after SNL was significantly reduced after erbB2 receptor blockade or sequestration of NRG using HBD-S-H4. h, Quantification. Error bars represent ± SEM (n = 3–4 per group). SH, Sham; VH, vehicle; INH, inhibitor (PD168393); IgG2a, nonimmune IgG (control); ANT, antagonist (HBD-S-H4). Statistical test: t test (vehicle vs inhibitor and control vs antagonist) or one-way ANOVA, Bonferroni post hoc test (IgG2a vs erbB2 receptor-blocking antibodies).**p < 0.005. Scale bars: a, b, f, g, 100 μm; d, e, i, j, 50 μm.

Figure 10.

Microglial proliferation after spinal nerve ligation is significantly reduced after erbB2 receptor inhibition. a–h, Three days after L5 SNL, a proliferative response is seen in microglia within the dorsal horn, as shown here by labeling newly dividing cells with BrdU (yellow) and microglia with Iba1 (red). DAPI is shown in blue to delineate nuclei. The BrdU-labeled nuclei are almost exclusively within microglial cells (as seen in the merge image). Blocking the erbB2 receptor with PD168393 (5 μg/d, i.t.) or with the blocking antibodies (mAb 9 or 7.16.4, 5 μg, i.t.) resulted in a significant reduction in microglial proliferation (quantification in i). Scale bars: 100 μm. Error bars represent ± SEM (n = 3–4 per group). SH, Sham; VH, vehicle; INH, inhibitor (PD168393); IgG2a, nonimmune IgG (control). Statistical test: one-way ANOVA, Bonferroni post hoc test. **p < 0.005.

Figure 11.

ErbB2 receptor inhibition or sequestration of endogenous NRG reduced mechanical pain hypersensitivity and cold allodynia after spinal nerve ligation. Animals underwent L5 SNL and received a continuous intrathecal infusion of the erbB2 inhibitor (PD168393, 1.25–10 μg/d) or vehicle for 14 d. a, Mechanical hypersensitivity developed at day 1 after SNL in both groups and after day 2 was significantly attenuated in animals receiving the erbB2 inhibitor in a dose-dependent fashion. *p < 0.05, **p < 0.001, for 10 μg/d vs vehicle; ^#p < 0.05 for 5 μg/d vs vehicle; n = 7–11/group. b, Once the pump had emptied, the withdrawal thresholds of the group receiving the erbB2 inhibitor (5 μg/d) returned to those of control. c, Delayed treatment with the erbB2 inhibitor (5 μg/d) from day 3 onward (by which time microgliosis is well established) was not effective at reversing mechanical pain-related hypersensitivity (p = 0.75) d, Cold allodynia was also significantly reduced in a dose-dependent fashion by inhibiting the erbB2 receptor. **p < 0.001 for 10 μg/d vs vehicle; ^#p < 0.05 for 5 μg/d vs vehicle, n = 7–11/group. e, f, Intrathecal administration of HBD-S-H4 (3 μg) at days 0 and 4 after SNL (shown by arrows) significantly reduced mechanical (p < 0.001) (e) and cold (p < 0.05) (f) pain-related hypersensitivity. Error bars represent ± SEM. Statistical tests: two-way ANOVA, Bonferroni test, or Fischer LSD post hoc analysis. When the assumptions of sphericity were violated (Mauchly's test; p < 0.05), the Greenhouse–Geisser correction was applied and independent two-tailed t tests were used to determine differences between groups. INH, Inhibitor (PD168393); NRG ANT, NRG antagonist (HBD-S-H4). The lines in a–d denote the period of pump infusion. The arrows in e and f denote the days of intrathecal injections.