• calcitonin gene-related peptide;
  • cytokine;
  • migraine;
  • mitogen-activated protein kinase;
  • trigeminal;
  • tumor necrosis factor-α


  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Expression of the neuropeptide calcitonin gene-related peptide (CGRP) in trigeminal ganglion is implicated in neurovascular headaches and temporomandibular joint disorders. Elevation of cytokines contributes to the pathology of these diseases. However, a connection between cytokines and CGRP gene expression in trigeminal ganglion nerves has not been established. We have focused on the effects of the cytokine tumor necrosis factor-α (TNF-α). TNFR1 receptors were found on the majority of CGRP-containing rat trigeminal ganglion neurons. Treatment of cultures with TNF-α stimulated CGRP secretion. In addition, the intracellular signaling intermediate from the TNFR1 receptor, ceramide, caused a similar increase in CGRP release. TNF-α caused a coordinate increase in CGRP promoter activity. TNF-α treatment activated the transcription factor NF-κB, as well as the Jun N-terminal kinase (JNK) and p38 mitogen-activated protein (MAP) kinase pathways. The importance of TNF-α induction of MAP kinase pathways was demonstrated by inhibiting MAP kinases with pharmacological reagents and gene transfer with an adenoviral vector encoding MAP kinase phosphatase-1 (MKP-1). We propose that selective and regulated inhibition of MAP kinases in trigeminal neurons may be therapeutically beneficial for inflammatory disorders involving elevated CGRP levels.

Abbreviations used

charge coupled device


calcitonin gene-related peptide




extracellular signal-regulated kinase


Jun N-terminal kinase


mitogen-activated protein


MAP kinase phosphatase-1


nuclear factor-κB


nerve growth factor


phosphate-buffered saline


temporomandibular joint disorder


tumor necrosis factor-α

Release of the neuropeptide calcitonin gene-related peptide (CGRP) from trigeminal ganglion nerve terminals is implicated in the underlying pathology of several disorders involving the cerebrovasculature and craniofacial structures. CGRP is the most potent vasodilatory peptide known and it is a major contributor to neurogenic inflammation and nociception (Brain et al. 1985; Williamson and Hargreaves 2001). Increased levels of CGRP have been reported in serum and saliva during migraine headache and cluster headache (Nicolodi and Del Bianco 1990; Goadsby and Edvinsson 1993, 1994; Fanciullacci et al. 1995). High levels of CGRP have been found in the synovial fluid of arthritic temporomandibular joints in association with spontaneous pain, impairment of mandibular mobility, and occlusal signs of destruction in patients with rheumatoid arthritis and inflammation (Appelgren et al. 1993, 1995; Sessle 2001). Results from these studies demonstrate that elevated levels of CGRP correlate with inflammation and pain. A common theme among these disorders is the apparent involvement of an inflammatory cascade of cytokines, including tumor necrosis factor-α (TNF-α) (Covelli et al. 1992; Kopp 2001).

Reciprocal communication between the immune and nervous systems has long been recognized (Hopkins and Rothwell 1995; Rothwell and Hopkins 1995). Important messengers between these systems are the multifunctional cytokine peptides that are released by both immune and non-immune cells. One of the best characterized is TNF-α, which has been implicated as a potential pathogenic mechanism for a number of inflammatory diseases (Taylor et al. 2004). TNF-α levels have been shown to positively correlate with acute and chronic joint inflammation, connective tissue destruction, and pain in temporomandibular joint disorder (TMD) (Appelgren et al. 1993, 1995). For migraine, while sampling difficulties and the complexity of migraine have led to conflicting reports on which cytokines are elevated, among the most consistent findings is elevation of TNF-α (Kemper et al. 2001; Perini et al. 2005).

Although the cellular mechanisms by which cytokines might increase CGRP synthesis and release during these diseases are not well understood, cytokines are known to activate multiple signaling pathways, including mitogen-activated protein (MAP) kinase signaling pathways. It is now known that at least four distinctly regulated groups of MAP kinases are present in mammalian cells: extracellular signal-regulated kinases (ERK-1, -2), Jun N-terminal kinases (JNK1, -2, -3), p38 proteins (p38α, -β, -δ), and ERK5, which are in turn activated by specific MAP kinase kinases (Schramek 2002). TNF-α has been reported to activate the ERK, JNK, and p38 pathways in various systems, including neurons (Barbin et al. 2001; Schafers et al. 2003). However, the TNF-α-induced paths in trigeminal ganglion neurons have not been reported.

We have focused on TNF-α induction of MAP kinases. CGRP gene transcription is regulated by MAP kinases at two sites: a distal 18-bp cell-specific enhancer and a proximal cAMP- and Ras-responsive region (deBustros et al. 1986; Tverberg and Russo 1993; Thiagalingam et al. 1996; Lanigan and Russo 1997). At least in trigeminal ganglion neurons, the stimulation by MAP kinases appears to be mediated predominantly through the cell-specific enhancer (Durham and Russo 2003). A complex of the basic helix-loop-helix zipper proteins USF-1 and -2, and the Foxa2 forkhead protein regulates the enhancer in a neuronal-like cell line (Viney et al. 2004). Whether the same or similar proteins play a role in trigeminal neurons is not known, but it is interesting that the USF proteins can be regulated by elevated Ca2+ (Tabuchi et al. 2002; Chen et al. 2003) and phosphorylated by p38 MAP kinase (Galibert et al. 2001).

In this study, we used primary cultures of trigeminal ganglion neurons to investigate the mechanisms by which TNF-α regulates CGRP gene expression. We show that TNF-α can stimulate CGRP secretion from trigeminal neurons and that the TNFR1 signaling intermediate ceramide can also increase CGRP release. Importantly, CGRP promoter activity was stimulated in response to TNF-α activation of MAP kinases. Finally, we describe the use of gene transfer to down-regulate TNF-α induction of the CGRP promoter.

Materials and methods

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Animals and cell culture

The animal care and procedures were conducted in accordance with institutional and National Institutes of Health guidelines. Adult female Sprague–Dawley rats (Charles River Laboratories Inc., Wilmington, MA, USA) were housed in clean plastic cages on a 12-h light/dark cycle with unrestricted access to food and water. Trigeminal ganglia primary cultures were established based on our published protocol (Durham and Russo 2003). Briefly, ganglia isolated from 3- to 4-day-old rats were incubated for 30 min at 37°C in L-15 (Leibovitz, Sigma, St Louis, MO, USA) media containing 10 mg/mL Dispase II (Invitrogen Life Technologies, Gaithersburg, MD, USA). Following centrifugation at 250 g for 1 min, the cells were dissociated in plating media by trituration using a 5-mL pipet, and large debris removed using a Pasteur pipet. The remaining supernatant was transferred to a new tube and then spun at 250 g for 3 min. The resultant cell pellet was re-suspended in L-15 medium containing 10% fetal bovine serum, 50 mm glucose, 250 μm ascorbic acid, 8 μm glutathione, 2 mm glutamine, and 10 ng/mL mouse 2.5 S nerve growth factor (Alomone Laboratories, Jerusalem, Israel) at 37°C at ambient CO2. Penicillin (100 units/mL), streptomycin (100 μg/mL), and amphotericin B (2.5 μg/mL, Sigma) were also added to the supplemented L15 media, which is referred to as L15 complete medium. Dissociated cells were plated on either 6-well (promoter studies), 24-well (secretion studies) tissue culture plates (BD Biosciences, Bedford, MA, USA), or 11-mm glass or plastic coverslips (immunostaining) coated with poly d-lysine (MW 30,000–70 000; Sigma). The cells from the equivalent of three ganglia were plated per 6-well plate, one per 24 well, and 0.5 per 11-mm coverslip. The medium was changed after 24 h and every other day thereafter. TNF-α and ceramide (C2 or N-acetylsphingosine) were purchased from Sigma, while SB 203580 hydrochloride and SP 600125 were purchased from Tocris Cookson (Ballwin, MO, USA).

CGRP secretion

After 24 h in culture, cells were incubated in HEPES-buffered saline (HBS; 22.5 mm HEPES, 135 mm NaCl, 3.5 mm KCl, 1 mm MgCl2, 2.5 mm CaCl2, 3.3 mm glucose, and 0.1% bovine serum albumin, pH 7.4) and the amount of CGRP released into the medium was determined using a CGRP-specific radioimmunoassay (Bachem/Peninsula Laboratories Inc., San Carlos, CA, USA) as previously described (Durham and Russo 1999). As a control, the basal (unstimulated) rate of CGRP secreted into the media in 1 h immediately prior to addition of the stimuli was determined, and these values were used to normalize for differences between dishes. Cells were treated with TNF-α, the cell-permeable form of ceramide, and/or KCl for 1 h. For each experiment, one set of wells in the 24-well plate was treated with 60 mm KCl to determine the responsiveness of the cultures to depolarizing stimuli as described previously (Durham and Russo 1999). Cultures that were found to be either unresponsive or exhibit a weak response to the depolarizing stimulus on CGRP release (< 2-fold) were not analyzed. As a control, cells were treated with equivalent amounts of the appropriate vehicle (buffered saline for TNF-α, dimethylsulfoxide for ceramide). Neuronal survival was assessed in cultures 24 h after the secretion studies by applying 1 mL of trypan blue diluted 1 : 10 in phosphate-buffered saline (PBS) for 3 min, after which the cultures were washed three times with PBS and viewed by brightfield microscopy. The ratio of viable cells to total cells was determined for each secretion condition. Neuronal viability was shown to be > 90% for untreated cultures (88/93 total cells) as well as for cultures treated with TNF-α (111/120) and ceramide (92/99). Each experimental condition was repeated in at least four independent experiments carried out in duplicate. The data were reported as mean ± SEM. Statistical analysis was performed using the Wilcoxon rank test. Results were considered significant when p was less than 0.05.


Dissociated trigeminal ganglion cells were plated at a density of 500–1000 cells on poly d-lysine coated 13-mm glass or plastic coverslips and incubated in complete L-15 medium. Untreated 2-day-old primary cultures or cultures treated with 50 ng/mL TNF-α, 0.6 m sorbitol, or vehicle were briefly rinsed with PBS and incubated in 4% paraformaldehyde for 30 min and then 0.2% trition X-100 in PBS for 15 min to fix and permeabilize the cells. Stimulation with sorbitol (30 min) was used as a positive control for p38, JNK, ERK, and NF-κB activation. For the signaling pathway studies, the cultures were subcultured in L-15 medium that contained 0.5% FBS for 24 h prior to addition of stimulatory agents. Cultures immunostained for the phosphorylated MAP kinase proteins were stimulated for 30 min, while those used to determine NF-κB activity were stimulated for 120 min. Fixed cells were incubated for 30 min in PBS containing 5% donkey serum, then for 1 h with primary antibodies and for 1 h with secondary antibodies. For TNFR1 and TNFR2 receptor localization, cultures were stained with rabbit anti-TNFR1 or anti-TNFR2 antibodies (diluted 1 : 50 in PBS, R & D Systems, Minneapolis, MN, USA), as well as with rabbit α-CGRP antibodies (Sigma, 1 : 1000). Cultures were also stained with antibodies directed against the phosphorylated (active) MAP kinase proteins p38 (Promega, Madison, WI, USA; 1 : 8000), JNK (Promega, 1 : 4000), and ERK (Promega, 1 : 8000), antibodies that recognize the unphosphorylated and phosphorylated forms (total protein) of p38, JNK, or ERK (1 μg/mL, Santa Cruz Biotechnology, Santa Cruz, CA, USA), and antibodies against the active form of NF-κB (Chemicon, Temecula, CA, USA; 1 : 1000). Immunoreactive proteins were detected following incubation with FITC-conjugated donkey anti-rabbit IgG, Texas red-conjugated donkey anti-rabbit IgG, or FITC-conjugated donkey anti-mouse IgG polyclonal antibodies (diluted 1 : 100 in PBS, Jackson Immuno-Research Laboratories, West Grove, PA, USA). No staining was observed in the absence of primary antibodies. Cultures were co-stained with 4′,6 diamidino-2-phenylindole (DAPI) to identify nuclei (Vector Laboratories, Burlingame, CA, USA). The number of neuronal and non-neuronal cells that responded to TNF-α stimulation was determined by analysis of immunopositive cells from 10 random fields of cells from three coverslips. The reported numbers are the average of the counts obtained by two laboratory members that were blinded to the experimental design. Neuronal cells were identified based on their unique cell morphology (round cell body of 30–50 μm).

Transfection of trigeminal cultures

The CGRP luciferase reporter plasmids and the cytomegalovirus (CMV) β-galactosidase reporter plasmid have been described previously (Tverberg and Russo 1993; Lanigan and Russo 1997). The 1250-bp rat CGRP promoter–luciferase reporter contains sequences from the KpnI site (−1250) to the Sau3A site (+21) in exon 1. Cultured cells were transiently transfected 1 h after plating with Lipofectamine 2000 (Invitrogen Life Technologies) following manufacturer's instructions. Approximately 3–5 × 104 cells (3–4 ganglia) per well were transfected 1 h with 1–2 μg of CGRP–luciferase reporter plasmid or with 0.5 μg of c-Jun reporter plasmids (PathDetect, Stratagene, La Jolla, CA, USA). The amount of DNA was kept constant by addition of the empty expression vector CMV-5 (Durham and Russo 1998). The DNA and Lipofectamine 2000 reagent (1 : 2 ratio) were incubated together for 20 min in L15 media. The DNA : Lipofectamine complex was then added to the trigeminal cultures maintained in the fully supplemented L15 complete medium [including serum, nerve growth factor (NGF), antibiotics, and amphotericin B] and incubated for ∼24 h before assaying for reporter activity. Under these transfection conditions using Lipofectamine 2000, > 90% of the non-neuronal cells are no longer viable 1 day after plating (Durham and Russo 2003). Based on cell counts, it is estimated that < 3% of the neurons were transfected by this method. TNF-α was added to transfected cells 2 h before harvesting. For the MAP kinase inhibitor studies, the JNK inhibitor SP 600125 or p38 inhibitor SB 203580 hydrochloride was added 30 min prior to addition of TNF-α. Luciferase activity was measured using the Luciferase Assay System (Promega) and was reported as relative light units/20 μg protein or fold change normalized to vehicle control. Protein concentrations were determined by Bradford assays (Bio-Rad Laboratories, Hercules, CA, USA). In all experiments, transfection efficiencies were normalized to CMV–β-galactosidase activity measured using Galacto-Light Plus reagents (Applied Biosystems, Foster City, CA, USA). No difference in cell viability was observed following transfection with the different plasmids or treatments as determined by trypan blue exclusion. The normalized luciferase activities are reported as means with standard errors per 20 μg protein. Each experimental condition was repeated in at least three independent experiments done in duplicate. Statistical analysis was performed using the Wilcoxon rank test.

Adenoviral infections and bioluminescence imaging

The 1.25-kb rat CGRP promoter and firefly luciferase gene were subcloned from pGL3-rCAP (Lanigan and Russo 1997) by digestion with KpnI and XbaI and inserted into shuttle plasmid pacAd5 K-N pA to generate Ad CGRP-Luc. Mouse MKP-1 was subcloned from pSG5 MKP-1-myc (Durham and Russo 2000) by EcoRI and BamHI digestion and inserted into shuttle plasmid pacAd5 CMV-K-N pA to generate Ad CMV-MKP-1. Unaltered pacAd5 CMV-K-N pA was used to generate the control virus Ad CMV-Empty. The Ad-NF-κB responsive luciferase reporter virus has been described (Sanlioglu et al. 2001). Briefly, a KpnI and XbaI fragment of the pNF-κB-Luc plasmid (Clontech Inc., Palo Alto, CA, USA) containing the luciferase gene driven by four tandem copies of the NF-κB consensus sequence fused to a TATA-like promoter from the herpes simplex virus-thymidine kinase gene was inserted into a promoterless adenoviral shuttle plasmid (pAd5mcspA). The insertions were confirmed by sequencing. The University of Iowa Gene Transfer Vector Core provided shuttle plasmids and generated the replication-deficient type 5 adenoviruses. The recombinant viruses have the transgenes inserted into a deletion of the E1 region. Viral titers were determined by plaque assays on 293 cells. The purified viruses were aliquoted in PBS containing 3% sucrose and stored at −80°C.

For the infection of primary cultures, the dissociated cells were plated on 25-mm glass cover slips coated with EHS-laminin (Sigma) and incubated for 24 h in L15 complete medium at 37°C and ambient CO2. The next day, the cultures were infected with 2 μL Ad CGRP-Luc (1 × 107 pfu/μL) virus in 0.5 mL L15 complete medium. For the NF-κB reporter infections, after the first 24 h the medium was changed to neurobasal medium (Invitrogen) without NGF for infection with 2 μL of Ad NF-κB-luciferase reporter virus (4 × 107 pfu/μL) per well. The cultures with neurobasal medium were incubated in 5% CO2. After 4 h, 1.5 mL L15 complete medium was added and cultures were incubated overnight. The next day, the medium was removed and the cultures were infected with either 2 μL Ad CMV-MKP-1 (5 × 107 pfu/μL) or 2 μL Ad CMV-Empty (2 × 107 pfu/μL) virus in 0.5 mL L15 complete medium. After 4 h, 1.5 mL L15 complete medium was added and cultures were incubated overnight. The next day, the medium was removed and the cells were incubated in 2 mL L15 complete medium for 1–1.5 h. Luciferin (0.25 mm) (Xenogen, Alameda, CA, USA) was added to each well and pre-stimulus luciferase activity was measured using an IVIS 100 charged coupled device (CCD) camera with Igor pro-4.2 living image software (Xenogen). Cells were treated with 50 ng/mL mouse TNF-α or vehicle for 4 h at 37°C and ambient CO2. Fresh luciferin (0.25 mm) was added to each well and post-stimulus luciferase activity was measured. All light measurements were recorded for 3 min and reported as photons/s/steridan/cm2. Data are reported as the fold change in luciferase activity between post- and pre-stimulus readings from the same cultures. For comparison between cultures, the post-/pre-stimulus fold changes from each culture were first calculated, and then normalized to the pre-stimulus value of the control cultures from each experiment. Conditions were performed in duplicate unless otherwise noted. Statistical analysis was performed using the Wilcoxon rank test.


  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Presence of TNFR1 and TNFR2 receptors on cultured trigeminal ganglion neurons

Primary trigeminal ganglia cultures were used to investigate the signaling mechanisms by which TNF-α regulates CGRP secretion and promoter activity. Under our culture conditions, > 90% of the neurons (67/72 cells) express detectable levels of CGRP-IR based on cell morphology (round cell body of 30–50 μm) (Figs 1a and b). These data are in agreement with our previously published findings (Durham and Russo 1999; Durham et al. 2004a). This is a higher percentage than reported in human and rat trigeminal ganglia in situ, where 40% of the neurons are CGRP positive (Edvinsson et al. 1998). The basis for this bias is not known, but may reflect the inclusion of only one neurotrophin, NGF, in the culture medium. As shown in Fig. 1(c–e), robust TNFR1 receptor immunoreactivity was observed primarily on the cell body of small diameter neurons (30–40 μm) that also expressed CGRP. Faint TNFR1 staining was detectable in a small percentage of non-neuronal cells. Staining for TNFR2 revealed that this receptor was also abundantly expressed by most CGRP-positive neuronal cells, and was less robust in non-neuronal cells (Figs 1f–h), Our results are in agreement with the findings of Li et al. (2004) that neuronal cells express higher levels of TNFR1 mRNA than non-neuronal cells. However, while the authors reported exclusive TNFR2 mRNA expression in neuronal cells, our data indicated TNFR2 protein expression in neurons as well as non-neuronal cells.


Figure 1. Expression of TNFR1 and TNFR2. Day-2 trigeminal ganglia cultures were stained for expression of CGRP (a,b) and co-stained for CGRP and TNFR1 (c–e) or TNFR2 (f–h). Phase microscopy was used to identify neuronal (large arrows) and non-neuronal cells (small arrows) (a, c, f). CGRP was expressed exclusively in neuronal cells (b). The same cultures were stained for CGRP (d) and TNFR1 (e) or CGRP (d) and TNFR2 (h). Scale bar, 50 μm. TNFR1 and TNFR2 were easily detected on the cell body of CGRP-positive trigeminal neurons. TNFR1 and TNFR2 staining of some non-neuronal cells was observed.

Download figure to PowerPoint

TNF-α stimulation of CGRP secretion from cultured trigeminal ganglia neurons

To determine whether TNF-α could stimulate CGRP release from the cultured neurons, cultures were treated with increasing concentrations of TNF-α. The soluble form of TNF-α used in these studies is reported to fully activate the TNFR1 receptor but have minimal effects on TNFR2 (Grell et al. 1995) TNF-α treatment at the higher concentrations (50 and 100 ng/mL) caused a ∼4-fold increase in CGRP secretion when compared with unstimulated basal values (Fig. 2). For comparison, treatment of 1-day-old trigeminal ganglia cultures with a depolarizing stimulus, 60 mm KCl, caused a 6-fold increase in CGRP secretion (Fig. 2). The vehicle control did not stimulate CGRP release.


Figure 2. TNF-α increases CGRP secretion from primary trigeminal neurons. CGRP secretion from untreated cultures (CON) or cultures treated for 1 h with vehicle (VEH), 60 mm KCl, or increasing amounts of TNF-α was determined. The mean basal rate of CGRP release was 132 ± 13 pg/h/well. The secretion rate for each condition was normalized to the basal rate for each well. The means and the SE from five independent experiments are shown. *p < 0.05 when compared with control levels; †p < 0.01 when compared with control levels.

Download figure to PowerPoint

Involvement of ceramide in CGRP release

The TNFR1 receptor stimulates neutral and acidic sphingomyelinase isoforms to produce the secondary messenger ceramide (Adam-Klages et al. 1998). To investigate the role of the TNFR1-induced secondary messenger ceramide on CGRP secretion, cultures were treated with increasing concentrations of the cell permeable form of ceramide (C2) for 1 h. As shown in Fig. 3, ceramide stimulated CGRP release from trigeminal neurons. All concentrations of ceramide tested in our studies significantly (p < 0.05) increased CGRP release when compared with control or vehicle levels. The maximum fold-increase observed with ceramide (3–4-fold) was similar to that observed following TNF-α treatment. Taken together, these data are suggestive that TNF-α stimulation of CGRP gene expression involves TNFR1 activation and induction of the ceramide pathway.


Figure 3. Stimulation of CGRP secretion by ceramide. The relative amount of CGRP secreted from control cultures (CON) or cultures treated with vehicle (VEH) or increasing amounts of ceramide for 1 h. The mean basal rate of CGRP release was 107 ± 9 pg/h/well. The secretion rate for each condition was normalized to the basal rate for each well. The means and the SE from four independent experiments are shown. *p < 0.05 when compared with control levels.

Download figure to PowerPoint

TNF-α stimulation of CGRP promoter activity

Having shown that TNF-α could stimulate CGRP secretion, transient transfections of primary trigeminal ganglia cultures were used to determine whether TNF-α could also stimulate CGRP synthesis via activation of the CGRP promoter. Trigeminal ganglia cultures were transfected with a 1.25-kb fragment that contains the cell-specific enhancer and cAMP- and Ras-responsive elements of the rat CGRP gene. We have previously shown that this promoter directs neuronal specific expression and is regulated by MAP kinases in primary trigeminal neurons (Durham and Russo 2003; Durham et al. 2004b). Treatment with 50 or 100 ng/mL TNF-α resulted in a marked increase in CGRP promoter activity (> 3-fold) compared with vehicle-treated control values (Fig. 4a). These data provide evidence that TNF-α can also stimulate CGRP synthesis via activation of its promoter in trigeminal neurons.


Figure 4. TNF-α stimulates CGRP promoter and c-Jun activity. The −1250-bp CGRP promoter fragment contains proximal cAMP- and ras-responsive regions (gray box) and a distal enhancer that contains both cell-specific (open box) and non-cell-specific (striped box) elements. (a) Trigeminal cultures transfected with the CGRP–luciferase reporter were either untreated (CON) or treated 2 h with 50 or 100 ng/mL TNF-α and reporter activity measured. (b) c-Jun reporter activity in cultures treated with 50 ng/mL TNF-α for 2 h. Mean luciferase activity (normalized to β-galactosidase activities) per 20 μg protein with SEM from four experiments is shown. *p < 0.05 when compared with control levels.

Download figure to PowerPoint

TNF-α activation of JNK and p38 MAP kinases

Based on our previous data demonstrating that activity of the CGRP promoter is stimulated by MAP kinases (Durham and Russo 2003), we wanted to determine whether TNF-α could activate MAP kinase pathways in cultured trigeminal neurons. Two approaches were taken. The first was to use a reporter gene controlled by a Gal4-c-Jun fusion protein that is directly phosphorylated by MAP kinases. Treatment of trigeminal cultures with 50 ng/mL TNF-α increased c-Jun reporter activity > 3-fold (Fig. 4b). These data demonstrate that TNF-α can stimulate the JNK MAP kinase signaling pathway in the trigeminal cultures.

The second approach to test whether TNF-α activated MAP kinases was to stain trigeminal cultures for the active forms of these signaling molecules. We used phospho-specific ERK, JNK, and p38 antibodies that recognize only the dual phosphorylated (activated) proteins. Trigeminal ganglia cultures were treated for 30 min with vehicle, 50 ng/mL TNF-α, or 0.6 m sorbitol (positive control). Sorbitol was used as a positive control as it has been reported to strongly stimulate ERK, JNK, and p38 kinase activity (Kayali et al. 2000). In the vehicle control cultures, cytoplasmic and nuclear levels of active MAP kinases were barely detectable (Figs 5a–c). However, treatment of cultures with TNF-α caused a large increase in the percentage of cells exhibiting nuclear staining for active JNK and p38 (Figs 5b and c, and Fig. 6). The observed levels were similar to those seen following treatment with sorbitol. In contrast, TNF-α did not cause an increase in active ERK. As a control, cultures treated with sorbitol had a marked increase in the number of neurons with nuclear localization of phosphorylated ERK. In addition, we stained untreated and treated trigeminal cultures with antibodies that recognize both the phosphorylated as well as unphosphorylated forms of ERK, JNK, and p38. However, the level of cytoplasmic or nuclear staining for ERK, JNK, and p38 total proteins in the TNF-α-treated cultures was not increased when compared with untreated cultures (data not shown). Taken together, these data demonstrate that TNF-α treatment leads to increased phosphorylation and nuclear localization of JNK and p38, but not ERK in trigeminal ganglion neurons.


Figure 5. TNF-α stimulation of JNK and p38 MAP kinases and NF-κB. Unstimulated trigeminal ganglia cultures (control, CON) or cultures treated with TNF-α (TNF) were stained with antibodies directed against the active (phosphorylated) forms of ERK (a), JNK (b), and p38 (c), as well as NF-κB (d) (left panels). Increased expression of active JNK and p38, but not ERK, was observed in the nucleus of TNF-α-treated cultures. Nuclear expression of NF-κB was also increased in response to TNF-α. The same cultures were co-stained with DAPI to identify nuclei (right panels). Cell bodies of neurons are indicated by arrows.

Download figure to PowerPoint


Figure 6. Percentage of trigeminal neurons expressing the active form of ERK, JNK, p38 or NF-κB in response to TNF-α treatment. The number of neuronal cells in which nuclear staining was detected following sorbitol (positive control, gray bars) or TNF-α (black bars) treatment was compared with unstimulated cultures (control, open bars). Total cell counts (N = number) are indicated below each data set (x-axis). The data are from three independent experiments for each condition are shown.

Download figure to PowerPoint

TNF-α activation of NF-κB

The same two approaches were used to test whether TNF-α treatment leads to activation of the NF-κB pathway in trigeminal ganglion neurons. In the first approach, activated nuclear NF-κB was determined by immunocytochemical staining. Cultures were treated for 120 min with sorbitol or TNF-α prior to staining with a phospho-specific NF-κB antibody. Under unstimulated conditions, cytoplasmic and nuclear levels of active NF-κB were barely detectable (Fig. 5d). Treatment with TNF-α increased the percentage of neuronal cells exhibiting nuclear staining for active NF-κB in neurons (Fig. 5d and Fig. 6). The observed levels were similar to those seen following treatment with sorbitol (Fig. 6). In the second approach, we used an adenoviral vector containing an NF-κB responsive luciferase reporter gene (Sanlioglu et al. 2001). An advantage of using a viral delivery system is that we can detect luciferase activity using a sensitive CCD camera and imaging system before and after treatment in the same culture. This bioluminescent detection is possible as the viral vector delivers the reporter gene to over 10 times the number of neurons than chemical transfection (Durham et al. 2004b). The bioluminescence imaging system is advantageous to the traditional in vitro luciferase assay as the ability to acquire promoter activity from the same population of cells before and after treatment eliminates the inherent prestimulus variability of primary cultures. Hence, fewer cultures and animals are required with this approach. As a control, when the cultures are treated with vehicle, there is little or no change in the light signal from the cultures (Fig. 7a). When the cultures were treated with 50 ng/mL TNFα for 4 h, there was a robust 4–5-fold increase in NF-κB-responsive luciferase activity (Figs 7b and c). These data provide evidence that TNFα treatment of trigeminal cultures leads to activation of NF-κB signaling pathways and responsive genes.


Figure 7. TNF-α induction of an NF-κB-responsive reporter gene. Luciferase activity from trigeminal ganglion cells was detected by a CCD camera and IVIS imaging system. (a) Overlay images of the pre-stimulus and post-stimulus bioluminescence signals on the culture dish immediately before addition of the vehicle control and 4 h after vehicle treatment. The pseudocolor scale bar in photons/s/steridan/cm2 is shown. (b) Overlay images from a culture immediately before addition of 50 ng/mL TNF-α and 4 h after TNF-α treatment. The pseudocolor scale bar in photons/s/steridan/cm2 is shown. (c) Luciferase activities following vehicle or TNF-α treatments are shown as the ratio of post-stimulus to pre-stimulus signals from the same culture dishes. In these experiments, only a single vehicle control dish was analyzed in parallel with the TNF-α treatments that were performed on two dishes (mean ± range). *p < 0.05 when compared with vehicle levels.

Download figure to PowerPoint

Requirement of MAP kinases for TNFα stimulation of CGRP promoter activity

Because TNF-α treatment led to activation of both MAP kinases and NF-κB, we then asked which pathway was involved in activation of the CGRP promoter. Given our previous findings that MAP kinases activate the promoter in trigeminal neurons (Durham and Russo 2003), we reasoned that TNF-α activity was likely to involve this pathway. The approach we took was to test whether inhibition of MAP kinases would lead to complete or partial inhibition of TNF-α stimulation of CGRP promoter activity. MAP kinases were inhibited by complementary pharmacological and gene-transfer strategies.

The first strategy was to determine whether TNF-α stimulation of CGRP promoter activity could be repressed by pretreatment with selective inhibitors of JNK and p38 MAP kinases. Treatment of trigeminal cultures with 100 ng/mL TNF-α resulted in a marked increase in CGRP promoter activity (> 3-fold) compared with vehicle-treated control values (Fig. 8). Pretreatment with the p38 inhibitor SB 203580 (1 μm) or JNK inhibitor SP 600125 (1 μm) significantly repressed TNF-α stimulation, while the combination of both inhibitors further reduced promoter activity to baseline levels. No effect of the inhibitors alone was observed. These data provide evidence that the stimulatory effect of TNF-α on CGRP synthesis involves activation of the p38 and JNK MAP kinase pathways.


Figure 8. MAP kinase inhibitors repress TNF-α-mediated stimulation of CGRP promoter. Trigeminal cultures transfected with the CGRP–luciferase reporter were either untreated (CON), treated 2 h with 100 ng/mL TNF-α, or pretreated for 30 min prior to addition of TNF-α with the MAP kinase inhibitors, SB 203580 and/or SP 600125, and reporter activity measured. Mean luciferase activity (normalized to β-galactosidase activities) per 20 μg protein with SEM from five experiments is shown. *p < 0.05 when compared with control levels; †p < 0.05 when compared with stimulated levels.

Download figure to PowerPoint

The second strategy was to overexpress MAP kinase phosphtase-1 (MKP-1). Cultures were sequentially infected with the Ad-CGRP-Luc reporter, then overnight with the Ad-CMV-MKP-1 expression vector (Fig. 9). As a control, the cultures were infected with a virus lacking MKP-1 (Ad-CMV-Empty). Luciferase activity was measured in the same cultures before and after a 4 h TNF-α treatment. As described above for the NF-κB responsive reporter, the use of viral vectors insured sufficient gene transfer for detection of light production by a sensitive CCD camera imaging system. We observed a 4-fold decrease in CGRP promoter activity when MKP-1 was overexpressed (Fig. 9). Stimulation of the control cultures with 50 ng/mL TNF-α for 4 h yielded a 1.8-fold increase in CGRP promoter activity. The reason for the smaller increase than seen with the plasmid reporter is not known, but the increase was statistically significant. In contrast, when the cultures were infected with Ad-CMV-MKP-1 there was no statistical difference between pre-stimulus and post-stimulus luciferase activity following TNF-α treatment (Figs 9b and c). These results demonstrate that overexpression of MKP-1 is able to completely block TNF-α stimulation of the CGRP promoter.


Figure 9. MKP-1 overexpression blocks induction of the CGRP promoter by TNF-α. Luciferase activity from trigeminal ganglion cells was detected by a CCD camera and IVIS imaging system. (a) Schematics of the adenoviral CGRP promoter–luciferase reporter vector and the adenoviral CMV promoter–MKP-1 expression vector are shown. (b) Overlay images of the pre-stimulus and post-stimulus bioluminescence signals on the culture dish immediately before addition of 50 ng/mL TNF-α and 4 h after treatment. Cultures were infected with Ad-CGRP-Luc and the control vector Ad-CMV-Empty (left images) or the Ad-CMV-MKP-1 vector (right images), as indicated. The pseudocolor scale bar in photons/s/steridan/cm2 is shown. The light signal from the cultures containing Ad-CMV-MKP-1 was ∼10-fold above the background light from non-infected cultures. (c) Luciferase activities pre-stimulus and post-stimulus TNF-α treatments are shown as the ratio of post-stimulus to pre-stimulus signals from the same culture dishes normalized to control pre-stimulus luciferase activity, which was set at 1. The means from four culture dishes in two independent experiments are shown with the SD. The p-values between the indicated samples are given.

Download figure to PowerPoint


  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

In this report we provide the first evidence that TNF-α induces pro-inflammatory signal cascades that coordinately increase the synthesis and release of CGRP in trigeminal ganglion neurons. The TNF-α receptors TNFR1 and TNFR2 were detected primarily on CGRP-containing neuronal cells. However, the TNF-α stimulatory effects on CGRP gene expression likely are mediated by TNFR1, as the soluble form of TNF-α used in our studies has been reported to preferentially activate TNFR1 (Grell et al. 1995). The mechanism of TNF-α action on CGRP synthesis and release appears to involve selective activation of the JNK and p38 MAP kinases. The role of MAP kinases was shown using pharmacological inhibitors and by treatment with ceramide, which is a downstream signal from TNFR1 and has been reported to activate MAP kinase pathways in non-neuronal cells (Falluel-Morel et al. 2004). The involvement of MAP kinases builds on our previous findings that NGF and 5-HT1 receptor agonists regulate CGRP gene expression by modulating MAP kinase activity (Durham and Russo 2003; Durham et al. 2004b). Given the importance of MAP kinase control of CGRP expression, we then demonstrated the efficacy of gene transfer with an adenoviral vector encoding MKP-1. The MKP-1 vector was able to block TNF-α induction of CGRP promoter activity to a similar degree as seen using pharmacological inhibitors of MAP kinases.

The connection between cytokines and CGRP may have implications for understanding the craniofacial diseases of migraine and TMD that involve elevated CGRP. Trigeminal nerves could be exposed to cytokines released from locations such as dural mast cells and ganglion microglia (Reuter et al. 2001; Mori et al. 2003). During migraine, a connection with the immune system was first proposed over 75 years ago (Vaughan 1927). Trigeminal activation and neurogenic inflammation is now a generally accepted model for migraine pathogenesis (Buzzi et al. 1995; Reuter et al. 2001; Bolay et al. 2002; Goadsby et al. 2002; Parsons and Strijbos 2003). Cytokines are known to play an important role in inflammatory and pain conditions. Recently, the levels of the cytokines TNF-α and IL-1β were reported to be significantly elevated during migraine attacks in comparison with their levels outside attacks (Perini et al. 2005). However, evidence that cytokines are elevated in migraine remains controversial, perhaps in part to the difficulties in sampling and/or the inherent heterogeneity of the disease. Genetic studies suggest an association of migraine with IL-1α (Rainero et al. 2002), TNF-α (Rainero et al. 2004), and decreased soluble TNFR1 receptor that could compromise the ability to dampen TNF-α effects (Empl et al. 2003). In addition, nitric oxide, which has been implicated in migraine, induces IL-1β in the dura mater (Reuter et al. 2001), and IL-1β contributes to nociceptive fiber sensitization (Herbert and Holzer 1994; Opree and Kress 2000). In addition to the potential relevance to migraine, the connection between cytokines and CGRP is also applicable to TMD, as elevated levels of TNF-α are associated with the inflammation and pain of TMD (Fu et al. 1995; Kacena et al. 2001). In this study, we have focused on TNF-α, which is generally viewed as ‘pro-inflammatory’ and is released in response to inflammation, tissue damage, and microbial infection (Cavaillon 2001; Hanada and Yoshimura 2002; Palladino et al. 2003). Elevated levels of TNF-α are linked to several painful conditions (Cunha et al. 1992). However, cytokines cannot be easily classified by these labels (Vitkovic et al. 2000; Cavaillon 2001). Rather, cytokines appear to have context-dependent roles in the nervous system. For example, TNF-α can induce neuronal death in the absence of neurotrophins or following injury (Barker et al. 2001; Shinoda et al. 2003), yet it can also have a neuroprotective function (Cheng et al. 1994; Bruce et al. 1996). Thus, it should prove interesting in the future to identify additional consequences of TNF-α action on trigeminal neurons beyond the pro-inflammatory induction of CGRP.

Studies in both neuronal and non-neuronal cells have shown that TNF-α can trigger activation of the JNK and p38 MAP kinases (Raingeaud et al. 1995; Chen and Goeddel 2002; Pollock et al. 2002; Schafers et al. 2003). We have previously shown that MAP kinases regulate CGRP gene expression in trigeminal neurons via the distal cell-specific enhancer (Durham and Russo 2003). In this report we show that TNF-α treatment leads to activation of JNK and p38 MAP kinases in trigeminal neurons. We also show that TNF-α leads to activation of its well-established target, the NF-κB transcription factor (Chen and Goeddel 2002). The importance of the MAP kinase activation was demonstrated by pretreating the cultures with pharmacological inhibitors and by infecting the cells with a viral vector encoding MKP-1. The ability of these complementary inhibitory schemes to completely blunt TNF-α activation of the CGRP promoter indicates that the MAP kinase, not the NF-κB pathways, are required for CGRP promoter activation. The repression of MAP kinase pathway activation as a means to regulate CGRP gene expression has previously been demonstrated in a neuronal cell line (Durham and Russo 1998, 2000). Activation of serotonin type 1 receptors caused a sustained increase in intracellular calcium levels as well as the levels of MKP-1, which is known to negatively regulate JNK and p38 activity. These results provide evidence that cell-type selective and regulated suppression of CGRP synthesis by inhibiting MAP kinase activity may be an effective treatment for inflammatory conditions of the head and face involving trigeminal nerves.


  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

We would like to thank Christine Niemann for technical assistance. Supported by grants from the NIH DE015385 and National Headache Foundation to PLD and NIH DE016511 and HL14388 to AFR.


  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  • Adam-Klages S., Schwandner R., Adam D., Kreder D., Bernardo K. and Kronke M. (1998) Distinct adapter proteins mediate acid versus neutral sphingomyelinase activation through the p55 receptor for tumor necrosis factor. J. Leukoc. Biol. 63, 678682.
  • Appelgren A., Appelgren B., Kopp S., Lundeberg T. and Theodorsson E. (1993) Relation between intra-articular temperature of the arthritic temporomandibular joint and presence of calcitonin gene-related peptide in the joint fluid. A clinical study. Acta Odontol. Scand. 51, 285291.
  • Appelgren A., Appelgren B., Kopp S., Lundeberg T. and Theodorsson E. (1995) Neuropeptides in the arthritic TMJ and symptoms and signs from the stomatognathic system with special considerations to rheumatoid arthritis. J. Orofac. Pain 9, 215225.
  • Barbin G., Roisin M. and Zalc B. (2001) Tumor necrosis factor α activates the phosphorylation of ERK, SAPK/JNK, and P38 kinase in primary cultures of neurons. Neurochem. Res. 26, 107112.
  • Barker V., Middleton G., Davey F. and Davies A. (2001) TNFα contributes to the death of NGF-dependent neurons during development. Nat. Neurosci. 4, 11941198.
  • Bolay H., Reuter U., Dunn A., Huang Z., Boas D. and Moskowitz M. (2002) Intrinsic brain activity trigger trigeminal meningeal afferents in a migraine model. Nat. Med. 8, 136142.
  • Brain S., Williams T., Tippins J., Morris H. and MacIntyre I. (1985) Calcitonin gene-related peptide is a potent vasodilator. Nature 313, 5456.
  • Bruce A., Boling W., Kindy M., Peschon J., Kraemer P., Carpenter M., Holtsberg F. and Mattson M. (1996) Altered neuronal and microglial responses to excitotoxic and ischemic brain injury in mice lacking TNF receptors. Nat. Med. 2, 788794.
  • DeBustros A., Baylin S., Levine M. and Nelkin B. (1986) Cyclic AMP and phorbol esters separately induce growth inhibition, calcitonin secretion, and calcitonin gene transcription in cultured human medullary thyroid carcinoma. J. Biol. Chem. 261, 80368041.
  • Buzzi M., Bonamini M. and Moskowitz M. (1995) Neurogenic model of migraine. Cephalalgia 15, 277280.
  • Cavaillon J. (2001) Pro- versus anti-inflammatory cytokines: myth or reality. Cell Mol. Biol. 47, 695702.
  • Chen G. and Goeddel D. (2002) TNF-R1 signaling: a beautiful pathway. Science 296, 16341635.
  • Chen W., West A., Tao X., Corfas G., Szentirmay M., Sawadogo M., Vinson C. and Greenberg M. (2003) Upstream stimulatory factors are mediators of Ca2+-responsive transcription in neurons. J. Neurosci. 23, 25722581.
  • Cheng B., Christakos S. and Mattson M. (1994) Tumor necrosis factors protect neurons against metabolic-excitotoxic insults and promote maintenance of calcium homeostasis. Neuron 12, 139153.
  • Covelli V., Munno I., Pellegrino N., Marinaro M., Gesario A., Massari F., Savastano S. and Jirillo E. (1992) In vivo administration of propranolol decreases exaggerated amounts of serum TNF-α in patients with migraine without aura. Possible mechanism of action. Acta Neurol. 14, 313319.
  • Cunha F., Poole S., Lorenzetti B. and Ferreira S. (1992) The pivotal role of tumour necrosis factor α in the development of inflammatory hyperalgesia. Br. J. Pharmacol. 107, 660664.
  • Durham P. and Russo A. (1998) Serotonergic repression of mitogen-activated protein kinase control of the calcitonin gene-related peptide enhancer. Mol. Endocrinol. 12, 10021009.
  • Durham P. and Russo A. (1999) Regulation of calcitonin gene-related peptide secretion by a serotonergic antimigraine drug. J. Neurosci. 19, 34233429.
  • Durham P. and Russo A. (2000) Differential regulation of mitogen-activated protein kinase-responsive genes by the duration of a calcium signal. Mol. Endocrinol. 14, 15701582.
  • Durham P. and Russo A. (2003) Stimulation of the calcitonin gene-related peptide enhancer by mitogen-activated protein kinases and repression by an antimigraine drug in trigeminal ganglia neurons. J. Neurosci. 23, 807815.
  • Durham P., Cady R. and Cady R. (2004a) Regulation of calcitonin gene-related peptide secretion from trigeminal nerve cells by botulinum toxin type A: implications for migraine therapy. Headache 44, 3542.
  • Durham P., Dong P., Belasco K., Kasperski J., Gierasch W., Edvinsson L., Heistad D., Faraci F. and Russo A. (2004b) Neuronal expression and regulation of CGRP promoter activity following viral gene transfer into cultured trigeminal ganglia neurons. Brain Res. 997, 103110.
  • Edvinsson L., Mulder H., Goadsby P. and Uddman R. (1998) Calcitonin gene-related peptide and nitric oxide in the trigeminal ganglion: cerebral vasodilation from trigeminal nerve stimulation involves mainly calcitonin gene-related peptide. J. Auton. Nerv. Syst. 70, 1522.
  • Empl M., Sostak P., Riedel M., Schwarz M., Muller N., Forderreuther S. and Straube A. (2003) Decreased sTNF-RI in migraine patients? Cephalalgia 23, 5558.
  • Falluel-Morel A., Aubert N., Vaudry D., Basille M., Fontaine M., Fournier A., Vaudry H. and Gonzalez B. (2004) Opposite regulation of the mitochondrial apoptotic pathway by C2-ceramide and PACAP through a MAP-kinase-dependent mechanism in cerebellar granule cells. J. Neurochem. 91, 12311243.
  • Fanciullacci M., Alessandri M., Figini M., Geppetti P. and Michelacci S. (1995) Increase in plasma calcitonin gene-related peptide from the extracerebral circulation during nitroglycerin-induced cluster headache attack. Pain 60, 119123.
  • Fu K., Ma X., Zhang Z. and Chen W. (1995) Tumor necrosis factor in synovial fluid of patients with temporomandibular disorders. J. Oral Maxillofac. Surg. 53, 424426.
  • Galibert M., Carreira S. and Goding C. (2001) The Usf-1 transcription factor is a novel target for the stress-responsive p38 kinase and mediates UV-induced tyrosinase expression. EMBO J. 20, 50225031.
  • Goadsby P. and Edvinsson L. (1993) The trigeminovascular system and migraine: studies characterizing cerebrovascular and neuropeptide changes seen in humans and cats. Ann. Neurol. 33, 4856.
  • Goadsby P. and Edvinsson L. (1994) Human in vivo evidence for trigeminovascular activation in cluster headache. Neuropeptide changes and effects of acute attacks therapies. Brain 117, 427434.
  • Goadsby P., Lipton R. and Ferrari M. (2002) Migraine – current understanding and treatment. N. Engl. J. Med. 346, 257270.
  • Grell M., Douni E., Wajant H. et al. (1995) The transmembrane form of tumor necrosis factor is the prime activating ligand of the 80-kDa tumor necrosis factor receptor. Cell 83, 793802.
  • Hanada T. and Yoshimura A. (2002) Regulation of cytokine signaling and inflammation. Cytokine Growth Factor Rev. 13, 413421.
  • Herbert M. and Holzer P. (1994) Interleukin-1β enhances capsaicin-induced neurogenic vasodilatation in the rat skin. Br. J. Pharmacol. 111, 681686.
  • Hopkins S. and Rothwell N. (1995) Cytokines and the nervous system. I: Expression and recognition. Trends Neurosci. 18, 8388.
  • Kacena M., Merrel G., Konda S., Wilson K., Xi Y. and Horowitz M. (2001) Inflammation and bony changes at the temporomandibular joint. Cells Tissues Organs 169, 257264.
  • Kayali A., Austin D. and Webster N. (2000) Stimulation of MAPK cascades by insulin and osmotic shock: lack of an involvement of p38 mitogen-activated protein kinase in glucose transport in 3T3-L1 adipocytes. Diabetes 49, 17831793.
  • Kemper R., Meijler W., Korf J. and Ter Horst G. (2001) Migraine and function of the immune system: a meta-analysis of clinical literature published between 1966 and 1999. Cephalalgia 21, 549557.
  • Kopp S. (2001) Neuroendocrine, immune, and local responses related to temporomandibular disorders. J. Orofac. Pain 15, 928.
  • Lanigan T. and Russo A. (1997) Binding of upstream stimulatory factor and a cell-specific activator to the calcitonin/calcitonin gene-related peptide enhancer. J. Biol. Chem. 272, 1831618324.
  • Li Y., Ji A., Weihe E. and Schafer M. (2004) Cell-specific expression and lipopolysaccharide-induced regulation of tumor necrosis factor α (TNFα) and TNF receptors in rat dorsal root ganglion. J. Neurosci. 24, 96239631.
  • Mori I., Goshima F., Koshizuka T. et al. (2003) Iba1-expressing microglia respond to herpes simplex virus infection in the mouse trigeminal ganglion. Brain Res. Mol. Brain Res. 120, 5256.
  • Nicolodi M. and Del Bianco E. (1990) Sensory neuropeptides (substance P, calcitonin gene-related peptide) and vasoactive intestinal polypeptide in human saliva: Their pattern in migraine and cluster headache. Cephalalgia 10, 3950.
  • Opree A. and Kress M. (2000) Involvement of the proinflammatory cytokines tumor necrosis factor-α, IL-1β, and IL-6 but not IL-8 in the development of heat hyperalgesia: effects on heat-evoked calcitonin gene-related peptide release from rat skin. J. Neurosci. 20, 62896293.
  • Palladino M., Bahjat F., Theodorakis E. and Moldawer L. (2003) Anti-TNF-α therapies: the next generation. Nat. Rev. Drug Discov. 2, 736746.
  • Parsons A. and Strijbos P. (2003) The neuronal versus vascular hypothesis of migraine and cortical spreading depression. Curr. Opin. Pharmacol. 3, 7377.
  • Perini F., D'Andrea G., Galloni E., Pignatelli F., Billo G., Alba S., Bussone G. and Toso V. (2005) Plasma cytokine levels in migraineurs and controls. Headache 45, 926931.
  • Pollock J., McFarlane S., Connell M., Zehavi U., Vandenabeele P., MacEwan D. and Scott R. (2002) TNF-α receptors simultaneously activate Ca2+ mobilisation and stress kinases in cultured sensory neurones. Neuropharmacology 42, 93106.
  • Rainero I., Pinessi L., Salani G., Valfre W., Rivoiro C., Savi L., Gentile S., Giudice R. and Grimaldi L. (2002) A polymorphism in the interleukin-1α gene influences the clinical features of migraine. Headache 42, 337340.
  • Rainero I., Grimaldi L., Salani G., Valfre W., Rivoiro C., Savi L. and Pinessi L. (2004) Association between the tumor necrosis factor-α-308 G/A gene polymorphism and migraine. Neurology 62, 141143.
  • Raingeaud J., Gupta S., Rogers J., Dickens M., Han J., Ulevitch R. and Davis R. (1995) Pro-inflammatory cytokines and environmental stress cause p38 mitogen-activated protein kinase activation by dual phosphorylation on tyrosine and threonine. J. Biol. Chem. 270, 7420–7426.
  • Reuter U., Bolay H., Jansen-Olesen I., Chiarugi A., Sanchez del Rio M., Letourneau R., Theoharides T., Waeber C. and Moskowitz M. (2001) Delayed inflammation in rat meninges: implications for migraine pathophysiology. Brain 124, 24902502.
  • Rothwell N. and Hopkins S. (1995) Cytokines and the nervous system II: Actions and mechanisms of action. Trends Neurosci. 18, 130136.
  • Sanlioglu S., Williams C., Samavati L., Butler N., Wang G., McCray P. J., Ritchie T., Hunninghake G., Zandi E. and Engelhardt J. (2001) Lipopolysaccharide induces Rac1-dependent reactive oxygen species formation and coordinates tumor necrosis factor-α secretion through IKK regulation of NF-κB. J. Biol. Chem. 276, 30 18830 198.
  • Schafers M., Svensson C., Sommer C. and Sorkin L. (2003) Tumor necrosis factor-α induces mechanical allodynia after spinal nerve ligation by activation of p38 MAPK in primary sensory neurons. J. Neurosci. 23, 25172521.
  • Schramek H. (2002) MAP kinases: from intracellular signals to physiology and disease. News Physiol. Sci. 17, 6267.
  • Sessle B. (2001) New insights into peripheral chemical mediators of temporomandibular pain and inflammation. J. Orofac Pain 15, 5.
  • Shinoda S., Skradski S., Araki T., Schindler C., Meller R., Lan J., Taki W., Simon R. and Henshall D. (2003) Formation of a tumour necrosis factor receptor 1 molecular scaffolding complex and activation of apoptosis signal-regulating kinase 1 during seizure-induced neuronal death. Eur. J. Neurosci. 17, 20652076.
  • Tabuchi A., Sakaya H., Kisukeda T., Fushiki H. and Tsuda M. (2002) Involvement of an upstream stimulatory factor as well as cAMP-responsive element-binding protein in the activation of brain-derived neurotrophic factor gene promoter I. J. Biol. Chem. 277, 35 92035 931.
  • Taylor P., Williams R. and Feldmann M. (2004) Tumour necrosis factor α as a therapeutic target for immune-mediated inflammatory diseases. Curr. Opin. Biotechnol. 15, 557563.
  • Thiagalingam A., DeBustros A., Borges M., Jasti R., Compton D., Diamond L., Mabry M., Ball D., Baylin S. and Nelkin B. (1996) RREB-1, a novel zinc finger protein, is involved in the differentiation response to ras in human medullary thyroid carcinomas. Mol. Cell Biol. 16, 53355345.
  • Tverberg L. and Russo A. (1993) Regulation of the calcitonin/calcitonin gene-related peptide gene by cell-specific synergy between helix-loop-helix and octamer-binding transcription factors. J. Biol. Chem. 268, 15 96515 973.
  • Vaughan W. (1927) Allergic migraine. J. Am. Med. Assoc. 88, 13831386.
  • Viney T., Schmidt T., Gierasch W., Sattar A., Yaggie R., Kuburas A., Quinn J., Coulson J. and Russo A. (2004) Regulation of the cell-specific calcitonin/calcitonin gene-related peptide enhancer by USF and the Foxa2 forkhead protein. J. Biol. Chem. 279, 4994849955.
  • Vitkovic L., Bockaert J. and Jacque C. (2000) ‘Inflammatory’ cytokines: neuromodulators in normal brain? J. Neurochem. 74, 457471.
  • Williamson D. and Hargreaves R. (2001) Neurogenic inflammation in the context of migraine. Microsc. Res. Tech. 53, 167178.