Regulation of Wnt signaling by nociceptive input in animal models
© Shi et al.; licensee BioMed Central Ltd. 2012
Received: 10 January 2012
Accepted: 4 June 2012
Published: 19 June 2012
Skip to main content
© Shi et al.; licensee BioMed Central Ltd. 2012
Received: 10 January 2012
Accepted: 4 June 2012
Published: 19 June 2012
Central sensitization-associated synaptic plasticity in the spinal cord dorsal horn (SCDH) critically contributes to the development of chronic pain, but understanding of the underlying molecular pathways is still incomplete. Emerging evidence suggests that Wnt signaling plays a crucial role in regulation of synaptic plasticity. Little is known about the potential function of the Wnt signaling cascades in chronic pain development.
Fluorescent immunostaining results indicate that β-catenin, an essential protein in the canonical Wnt signaling pathway, is expressed in the superficial layers of the mouse SCDH with enrichment at synapses in lamina II. In addition, Wnt3a, a prototypic Wnt ligand that activates the canonical pathway, is also enriched in the superficial layers. Immunoblotting analysis indicates that both Wnt3a a β-catenin are up-regulated in the SCDH of various mouse pain models created by hind-paw injection of capsaicin, intrathecal (i.t.) injection of HIV-gp120 protein or spinal nerve ligation (SNL). Furthermore, Wnt5a, a prototypic Wnt ligand for non-canonical pathways, and its receptor Ror2 are also up-regulated in the SCDH of these models.
Our results suggest that Wnt signaling pathways are regulated by nociceptive input. The activation of Wnt signaling may regulate the expression of spinal central sensitization during the development of acute and chronic pain.
During the development of chronic pain, spinal neurons in the spinal cord dorsal horn (SCDH) become sensitized and hyper-active (termed central sensitization). A spectrum of neuronal and glial processes has been implicated in the establishment of central sensitization. For instance, in the spinal nerve ligation (SNL) and spared nerve injury (SNI) models of neuropathic pain, the central terminals of primary sensory neurons were reported to sprout [1–4]. This sprouting may increase inputs of nociceptive signals. Indeed, increased release of neurotransmitters or neuromodulators such as glutamate, substance P, prostaglandin E2 (PGE2) and calcitonin-gene related peptide (CGRP) were reported in animal pain models (reviewed in ). Another neuronal alteration associated with central sensitization is the expression of long-term potentiation (LTP) at the synapses in superficial layers of the SCDH, which is considered to be a critical synaptic mechanism underlying chronic pain [6, 7] and a potential target for chronic pain therapy . Furthermore, loss of inhibitory functions of GABAergic and glycinergic interneurons may contribute to enhanced pain sensitivity in chronic pain [9, 10]. In addition to neuronal changes, more recent studies revealed an important role of glial cells, especially microglia and astrocytes, in central sensitization, and glia are emerging as a promising target for chronic pain treatment . Activated microglia and astrocytes facilitate the development of central sensitization by releasing chemokines, cytokines and neurotrophins [12–14]. These factors can markedly enhance the excitability of neurons processing nociceptive input. For example, tumor necrosis factor-alpha (TNFα), a key proinflammatory cytokine, was shown to increase the frequency of excitatory postsynaptic currents (EPSCs) and N-methyl-D-aspartate (NMDA) currents in lamina II neurons by stimulating TNF receptor subtype-1 and 2 (TNFR1 and TNFR2) in an inflammatory pain model . Despite significant progress in identifying various cellular processes that contribute to central sensitization and chronic pain, the molecular mechanisms by which the spectrum of cellular alterations is initiated and established remain poorly understood.
Secreted signaling proteins in the Wingless–Int (Wnt) family play essential roles in many aspects of neural development/plasticity [16, 17], such as neurogenesis, axonal and dendritic branching, synapse formation, synaptic transmission and plasticity, and memory formation [18–33]. The synthesis and secretion of neuronal Wnt proteins are controlled by synaptic activity [28, 33, 34]. Three Wnt signaling pathways are well characterized, including the canonical Wnt/β-catenin pathway, the planar cell polarity (Wnt/PCP; a.k.a. Wnt/JNK) pathway and the Wnt/Ca2+ pathway . In the canonical pathway, Wnts bind to the frizzled (Fz) receptors on plasma membranes. This interaction stimulates the disheveled (Dvl) scaffold protein, leading to the inhibition or disruption of the ‘destruction complex’, which contains glycogen synthase kinase-3β (GSK-3β), axin and adenomatous polyposis coli (APC). Consequently, the β-catenin protein is stabilized, accumulates in the cytoplasm, and is imported into the nucleus to activate the transcription of TCF/LEF (T-cell factor/lymphoid enhancer factor) target genes . In hippocampal neurons, activation of NMDA receptors (NMDARs) causes β-catenin nuclear translocation from post-synaptic regions and activation of gene expression . At synapses, β-catenin interacts with cadherin to regulate synaptic assembly, remodeling and plasticity [38, 39]. In the PCP pathway, Wnt-bound Fz signals through Dvl and the GTPase RhoA to activate c-Jun amino (N)-terminal kinase (JNK) which regulates cytoskeleton dynamics and transcription [40–42]. JNK signaling plays important roles in central sensitization induced by inflammation and nerve injury [43–45]. One mechanism by which JNK signaling contributes to chronic pain is to regulate the expression of cytokines (e.g., IL-10, TNFα, IL-1β and IL-6) in the spinal glia cells . In the Wnt/Ca2+ pathway, Fz activation leads to increased intracellular Ca2+, which thereby activates Ca2+-sensitive proteins such as protein kinase C (PKC) and calcium/calmodulin dependent protein kinase II (CaMK II) . Both PKC and CaMKII play pivotal roles in central sensitization during the development of neuropathic and inflammatory pain [47–50]. Despite of the accumulation of suggestive evidence, the involvement of Wnt signaling in pathological pain has not been directly tested.
In this study, we report the spatial distribution of specific Wnt signaling proteins in mouse spinal cords and the regulated expression of the proteins in multiple pain models. Our results reveal the expression of Wnt signaling proteins in the superficial layers of the SCDH and the up-regulation of their expression in acute and chronic pain models. These findings indicate that Wnt signaling pathways may play a role in the regulation of central sensitization and chronic pain development.
The β-catenin label was also clustered into small spots or dots (Figure 2). Because β-catenin is enriched in synapses , we next tested if the clustered β-catenin dots corresponded to synapses. To this end, we performed double-labeling experiments with β-catenin and synapsin I (pre-synaptic marker) or PSD95 (post-synaptic marker). We observed that β-catenin staining substantially overlapped with that of synapsin I or PSD95 (Figure 2 D-I). On the other hand, little β-catenin was observed in the CD11b-labeled microglia (Figure 2 J-L) or GFAP-labeled astrocytes (Figure 2 M-O). These results suggest an enrichment of β-catenin at synapses in the SCDH.
In addition, we also examined the effect of capsaicin-induced pain on proteins in the non-canonical pathways. We focused here on Wnt5a, a prototypic Wnt ligand that activates the non-canonical pathways. As shown in Figure 6 E, Wnt5a was also induced in the SCDH following i.d. injection of capsaicin. The temporal profile of capsaicin-induced Wnt5a alteration differed from that of Wnt3a and β-catenin. The Wnt5a up-regulation peaked at 2 h after capsaicin injection, but returned to baseline by 3 h (Figure 6 E). These data indicate that capsaicin up-regulates Wnt5a in a more rapid and transient manner. Furthermore, we also examined the temporal profile of Ror2, a Wnt5a receptor tyrosine kinase that activates JNK signaling . Similar to Wnt5a, Ror2 was also transiently up-regulated (Figure 6 F). Compared with Wnt5a, the Ror2 up-regulation was delayed by 1 h (Figure 6 F). The overlapping but distinct temporal profiles of Wnt5a and Ror2 indicate that Wnt5a does not solely depend on Ror2 to transmit signals.
We also examined the expression profiles of Wnt5a and Ror2 in the gp120 pain model. The results showed that Wnt5a rapidly increased and peaked at 15–30 min after gp120 injection (Figure 7 E). Similar to the Wnt5a expression in the capsaicin pain model (Figure 6 E), the up-regulation of Wnt5a was relatively transient and returned to baseline by 2 h (Figure 7 E). Like Wnt5a, Ror2 also rapidly increased and peaked at 15–30 min after gp120 injection (Figure 7 F). Unlike Wnt5a, expression levels of Ror2 were maintained at significantly higher levels over baseline for 3 h (Figure 7 F). In addition, the magnitude of the Ror2 increase (3.6 fold) was higher than that of Wnt5a (1.9 fold).
We describe here the expression of Wnt signaling proteins in the SCDH and their change in expression in three pain models. We show that β-catenin is enriched in neurons in lamina II, and that Wnt3a is abundant in neurons in the superficial layers (laminae I-III). We also show that these and other Wnt signaling proteins (Wnt5a and Ror2) are up-regulated in the SCDH in these pain models. Our data suggest potential involvement of Wnt signaling pathways in the regulation of central sensitization in acute and chronic pain. Future studies are warranted to directly test this hypothesis.
Wnt signaling may contribute to chronic pain via multiple routes. We found that β-catenin is enriched in SCDH lamina II, especially at synaptic regions. Lamina II neurons, which include both excitatory and inhibitory interneurons, play crucial roles in central sensitization [48, 64, 65]. Because β-catenin is known to regulate synaptic transmission, synapse/spine assembling and remodeling [38, 39], the observation of enriched β-catenin protein at the synapses in lamina II suggests that canonical Wnt/β-catenin signaling may regulate synaptic plasticity in the neurocircuitry processing nociceptive input in the SCDH. Consistent with its role in central sensitization, β-catenin is up-regulated in capsaicin, HIV gp120, and SNL pain models. While we found β-catenin is up-regulated at 7 days after SNL, a recent study showed that this protein significantly increases at 1 and 3 days and returns to baseline at 7 days in the rat SCDH after unilateral spared nerve injury (SNI) . These findings indicate that the regulated expression of β-catenin in the SCDH of different neuropathic pain models follows different temporal patterns. In further support of a role of β-catenin signaling, Wnt3a, a prototypic Wnt ligand for the canonical Wnt/β-catenin pathway, is also expressed in the superficial laminae (including lamina I) and is up-regulated in these pain models. Previous studies show that activation of NMDA receptors by synaptic stimulation elicits Wnt3a secretion from hippocampal synapses to activate β-catenin signaling and facilitate long-term potentiation . One may conceive that activation of NMDA receptors in the SCDH by nociceptive stimuli could also cause Wnt3a secretion to facilitate central sensitization via β-catenin. Signaling proteins in the non-canonical pathway, including Wnt5a and Ror2, are also up-regulated in the pain models. Recent studies have shown that Wnt5a is an NMDAR-regulated protein  and critical for the differentiation and plasticity of excitatory synapses [21, 23]. Ror2 may mediate the activity of Wnt5a in the regulation of synapse differentiation . In addition, Wnt5a also regulates GABA receptor recycling at inhibitory synapses . These previous findings suggest that the observed up-regulation of Wnt5a and Ror2 in these pain models may also contribute to synaptic remodeling during the development of chronic pain.
Neuroinflammation in the SCDH is a constant manifestation of chronic pain in animal models. Pro-inflammatory factors such as IL-6, IL-1β, TNF-α and MCP-1 play important roles in the initiation and maintenance of chronic pain [11, 15, 69]. Recent studies have suggested that Wnt5a signaling may regulate the peripheral inflammatory response in chronic disorders, including sepsis , rheumatoid arthritis , atherosclerosis , melanoma , and psoriasis . Wnt5a is known to activate CaMKII signaling to modulate the macrophage-mediated inflammatory response . Our previous studies revealed that Wnt5a evokes the expression of proinflammatory cytokines (IL-1β and TNF-α) in primary cortical cultures, indicating a role of Wnt5a in the regulation of neuroinflammation in CNS . Wnt5a is up-regulated by i.t. gp120 and peripheral nerve injury, and each of these is known to induce persistent neuroinflammation in SCDH [76–78]. We propose that one potential mechanism by which up-regulated Wnt5a may facilitate chronic pain development is by promoting neuroinflammation.
The temporal expression of proteins in canonical and non-canonical pathways appears to follow differential profiles after pain induction. β-catenin and Wnt3a, which are in the canonical pathway, displayed a gradual increase following capsaicin or HIV-gp120 administration. Their gradual up-regulation is correlated with the progressive development of capsaicin-induced and gp120-induced mechanical allodynia and stays at a peak level when mechanical sensitivity starts decreasing. On the other hand, Wnt5a and Ror2 in the non-canonical pathway showed a more rapid but transient increase; their up-regulated expression came back to baseline when capsaicin-induced allodynia was still at a maximal level. These observations suggest that the canonical and non-canonical Wnt signaling pathways may have distinct biological functions in different phases of chronic pain development.
Young adult male C57 BL/6 J mice (8–10 weeks), purchased from Jackson Laboratory (Bar Harbor, Maine, USA) were used for all studies. Animals were housed in a constant-temperature environment with soft bedding and free access to food and water under a 12/12-h light–dark cycle. All animal procedures were performed in accordance with an animal protocol approved by the Institutional Animal Care and Use Committee at the University of Texas Medical Branch (protocol #: 0904031) and adhered to the guidelines of the International Association of the Study of Pain for the ethical care and use of laboratory animals .
The mouse capsaicin pain model was generated as described . Briefly, mice were anesthetized with isoflurane (2% for induction and 1.5% for maintenance) in a flow of O2 and placed in a prone position. For each mouse, 5 μl of capsaicin (0.5% in saline containing 20% alcohol and 7% Tween 80; purchased from Sigma) was injected intradermally (i.d) into the plantar region of hind paw using a 30 gauge needle attached to a Hamilton Syringe. Mice injected with vehicle were used as controls. Five minutes later, injected mice were returned to their home cages.
The recombinant HIV-gp120 protein (HIV Bal gp120; NIH AIDS Research and Reference Reagent Program) in PBS was stored in a −80°C freezer. At the time of injection, gp120 was slowly thawed, diluted to a concentration of 20 ng/μl in ice-cold PBS and maintained on ice. For gp120 administration, mice were anesthetized under 2% isoflurane, and 5 μl gp120 (100 ng) was intrathecally (i.t) injected into the subarachnoid space between the L5 and L6 vertebrae using a 30 gauge needle attached to a Hamilton Syringe [57, 80]. Mice injected with vehicle were used as controls.
Peripheral neuropathy in mice was produced by a unilateral L5 spinal nerve ligation as previously reported [60, 81]. Briefly, mice were anesthetized with 2% isoflurane, and the left L5 spinal nerve was isolated and tightly ligated with 7-0 silk thread. Mechanical sensitivity was assessed 7 days after ligation.
Adult mice were deeply anesthetized with 4% isoflurane and perfused transcardially with 50 ml of D-PBS, followed by 50 ml of paraformaldehyde (PFA; 4% in 0.1 M phosphate buffer). The L4 and L5 DRG, and lumbar spinal cord tissues were dissected out, post-fixed in the same PFA solution for 3 hr at 4°C, and then cryoprotected in sucrose solution (30% in 0.1 M phosphate buffer) overnight at 4°C. Transverse sections (15 μm) were prepared on a cryostat (Leica CM 1900) and thaw-mounted onto Superfrost Plus microscope slides. For immunostaining, sections were incubated in blocking buffer (5% BSA and 0.3% Triton X-100 in 0.1 M phosphate buffer) for 1 h at room temperature, followed by overnight incubation with anti-β-catenin (1:500, BD: 610153), anti-substance P (SP, 1:1000, Abcam: ab10353), anti-IB4 (1:400, Sigma: L2895), anti-PKCγ (1:1000, Santa Cruz: SC211), anti-NeuN (1:200, Millipore: MAB377), anti-MAP2 (1:400, Millipore: MAB378), anti-Wnt5a (1:200, Abcam: ab72583), anti-Synapsin I (1:400, Millipore: AB1543), anti-PSD95 (1:400, Cell Signaling: 2507), anti-GFAP (1:500, Millipore: 04-1062 and MAB360), anti-CD11b (1:100, AbD: MCA74GA) or Wnt3a (1:200, Millipore: 09162) antibody. After five washes with PBS (0.1 M phosphate buffer), the sections were incubated with FITC or Cy3-conjugated secondary antibody (1:200, Jackson ImmunoResearch Laboratories), followed by incubation with DAPI (Sigma). IgG from the same animal sources was used as negative controls for immunostaining. Images were captured using a laser confocal microscope (Zeiss).
Mice were anesthetized and sacrificed and the L4-L6 lumbar spinal cord segments were collected. The dorsal halves were dissected on an ice-chilled plate, and the dorsal roots were cut off under dissecting microscopes. The collected dorsal spinal tissues were homogenized in RIPA lysis buffer (1% Nonidet P-40, 50 mM Tris–HCl, pH 7.4, 1% sodium deoxycholate, 150 mM NaCl, 1 mM EDTA, pH 8.0) with a protease inhibitor cocktail (Sigma). Equal amounts of protein were loaded and separated by SDS-PAGE. Protein was transferred to pure nitrocellulose membranes, which were then blocked and incubated with anti-total-β-catenin (TBC, 1:5000, BD: 610153), anti-Active-β-catenin (ABC, 1:1000, Millipore: 05665), anti-Wnt3a (1:1000), anti-Wnt5a (1:1000), or anti-Ror2 (1:1000, a gift form Dr. Roel Nusse, Stanford University School of Medicine  ) primary antibodies. Protein bands were visualized by an Enhanced Chemiluminescence kit (Pierce).
For the capsaicin and gp120 pain models, a series of calibrated von Frey filaments (0.1 to 2.0 g) were applied to the plantar surface of the mouse hind paw using the “up and down paradigm” described previously [82, 83]. Mechanical allodynia was assessed by changes in paw withdrawal threshold in response to von Frey stimuli. For the SNL neuropathic pain model, mechanical sensitivity was assessed before and seven days after ligation by paw withdrawal frequencies in response to von Frey stimuli as previous reported [60, 81].
Densitometry of Western blotting was conducted and quantified using the ImageJ software (NIH) with β-actin as the loading control. Values were represented as mean ± SEM of 3 separate experiments. Statistical analysis was performed using Prism 5 (GraphPad) software. One-way ANOVA or student’s t-test was used to analyze data from different groups. Two-way repeated measures ANOVA with one repeated factor (time) was used for mechanical threshold data analysis (p < 0.05 was considered significant).
This work was supported by the start-up funds from the University of Texas Medical Branch and a research grant from the Whitehall Foundation to S.J.T. and NIH grants to S.M.C. (NS027910), J.M.C (NS031680) and K.C. (NS011255). We are grateful to Drs. Renee van Amerongen and Roel Nusse for their generosity to provide the anti-Ror2 antibody.
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.