Changes in calcitonin gene-related peptide (CGRP) receptor component and nitric oxide receptor (sGC) immunoreactivity in rat trigeminal ganglion following glyceroltrinitrate pretreatment

Abstract

Background: Nitric oxide (NO) is thought to play an important role in the pathophysiology of migraine. Infusion of the nitrovasodilator glyceroltrinitrate (nitroglycerin, GTN), which mobilizes NO in the organism, is an approved migraine model in humans. Calcitonin gene-related peptide (CGRP) is regarded as another key mediator in migraine. Increased plasma levels of CGRP have been found during spontaneous as well as nitrovasodilator-induced migraine attacks. The nociceptive processes and interactions underlying the NO and CGRP mediated headache are poorly known but can be examined in animal experiments. In the present study we examined changes in immunofluorescence of CGRP receptor components (CLR and RAMP1) and soluble guanylyl cyclase (sGC), the intracellular receptor for NO, in rat trigeminal ganglia after pretreatment with GTN.

Methods: Isoflurane anaesthetised rats were intravenously infused with GTN (1 mg/kg) or saline for four hours and two hours later the trigeminal ganglia were processed for immunohistochemistry. Different primary antibodies recognizing CLR, RAMP1, CGRP and sGC coupled to fluorescent secondary antibodies were used to examine immunoreactive cells in serial sections of trigeminal ganglia with epifluorescence and confocal laser scanning microscopy. Several staining protocols were examined to yield optimized immunolabeling.

Results: In vehicle-treated animals, 42% of the trigeminal ganglion neurons were immunopositive for RAMP1 and 41% for CLR. After GTN pretreatment CLR-immunopositivity was unchanged, while there was an increase in RAMP1-immunopositive neurons to 46%. RAMP1 and CLR immunoreactivity was also detected in satellite cells. Neurons immunoreactive for sGC were on average smaller than sGC-immunonegative neurons. The percentage of sGC-immunopositive neurons (51% after vehicle) was decreased after GTN infusion (48%).

Conclusions: Prolonged infusion of GTN caused increased fractions of RAMP1- and decreased fractions of sGC-immunopositive neurons in the trigeminal ganglion. The observed alterations are likely immunophenotypic correlates of the pathophysiological processes underlying nitrovasodilator-induced migraine attacks and indicate that signalling via CGRP receptors but not sGC-mediated mechanisms may be enhanced through endogenous NO production.

Figure 1

Confocal images of trigeminal ganglion sections treated with the rabbit anti-rat CLR antibody and the secondary goat anti-rabbit antibody conjugated to Cy3 after different fixation methods. (A) Short Zamboni perfusion ~5 min, post-fixed in Zamboni for two hours. (B) Perfusion with 4% paraformaldehyde (PFA) for 20 min and postfixation in 4% PFA (2 h). (C) Perfusion with 4% PFA for ~5 min, postfixation in Zamboni (2 h). (D) Cryofixation, postfixation with methanol-acetone (7:3) for 10 min. (E) Cryofixation without any postfixation. (F) Low-power micrograph of a trigeminal ganglion section, stained with azur methylene blue, with the ophthalmic (V1), maxillary (V2) and mandibular division (V3), in which separate cell counts were made. Scale bars A-E = 100 μm; F = 1 mm.

Figure 2

Comparison of the CLR and RAMP1 immunofluorescence in trigeminal ganglia using different antibodies. (A) Rabbit anti-human CLR antibody coupled to Alexa 488 (green) shows partly nuclear staining while neuronal cell bodies lack immunofluorescence. (B) Anti-human RAMP1 antibody coupled to Alexa 488 (green) produced homogenous or granulated staining of neuronal cell plasma but no nucleus staining. This antibody was used for the main experiments with cell counting. (C) Rabbit anti-rat CLR antibody coupled to Cy3 (red), resulting in mostly homogenous staining. It was finally used for the experiments with cell counting. (D) Rabbit anti-rat RAMP1 coupled to Cy3 (red) caused weak and less specific staining of neurons. Scale bars = 100 μm.

Figure 3

CLR and RAMP1 immunostaining of trigeminal ganglia. (A) CLR- and (B) RAMP1-negative controls (NC); primary antibodies were omitted and no specific immunostaining is visible after incubation with the secondary antibodies coupled with Cy3 (A) or Alexa 488 (B). (C) RAMP1 immunostaining (Alexa 488) without any pretreatment of the animals. (D) CLR coupled to Cy3 (red) (saline-treated animal) and (E) RAMP1 coupled to Alexa 488 (green) (GTN-treated animal) immunofluorescence in the trigeminal ganglion. The magnified inset in (D) shows a typical CLR-immunopositive- and a CLR-immunonegative neuron. The magnified part of (E) in (F) illustrates the different intensity of the RAMP1 immunostaining of neurons. The neuron marked with * is a very intense RAMP1 immunopositive neuron compared to the neuron marked with ° but both were counted as immunopositive. (G-I) Double immunostaining using CLR and RAMP1 antibodies (GTN-treated animal). (G) CLR-positive neurons in the red channel, (H) RAMP1-positive neurons in the green channel and (I) both receptor components in the merged image; CLR and RAMP1 were colocalized in some neurons (yellow). Neurons exclusively immunopositive for CLR are extremly rare, in contrast to only RAMP1-positive neurons (*). The magnification shows a neuron immunopositive for both receptor components. All images from cryofixed trigeminal ganglia without any postfixation. Scale bars = 100 μm.

Figure 4

Quantitative analysis of CLR- and RAMP1-immunopositive neurons in the trigeminal ganglion. (A) Percentage of CLR-immunopositive neurons in untreated animals (Control, n = 2), saline- (n = 6) and GTN-pretreated animals (n = 6). (B) CLR-immunopositive neurons specified for the ophthalmic (V1), maxillary (V2) and mandibular (V3) division of trigeminal ganglia in saline and GTN-treated animals. (C) Percentage of RAMP1-immunopositive neurons in the trigeminal ganglia of untreated animals (Control, n = 2), saline- (n = 6) and GTN-pretreated animals (n = 6). In the saline- and GTN-treated groups, intensely stained RAMP1-positive neurons are also counted separately. (D) RAMP1-positive neurons specified for V1, V2 and V3 in saline- and GTN-treated animals. Error bars represent SD (n=6). * significant difference to saline (Chi-square test of immunopositive and –negative neurons).

Figure 5

Immunostaining of non-neuronal cells in the trigeminal ganglion. (A-C) CLR- (red) and (D-F) RAMP1-immunostaining (green) combined with DAPI (blue) nucleus staining. The magnified inset in (A) and the DAPI staining (B) merged with the CLR staining (C) shows an immunonegative neuron with a ring of CLR-immunopositive cells, which most likely represent satellite glial cells. The magnified inset in (D) and the DAPI staining (E) merged with the RAMP1 staining (F) shows a RAMP1-immunopositive neuron with a RAMP1-immunopositive outer hemline, most likely representing satellite glial cells. Scale bars = 100 μm.

Figure 6

sGC and CGRP immunostaining of trigeminal ganglia. (A) Negative control (NC) omitting the primary sGC antibody to control the specificity of the immunostaining with Alexa 488. Neurons show no specific staining. (B) Preabsorption control (sGC-peptide: sGC-primary antibody 10:1) with no specific staining. (C) sGC immunostaining using Alexa 488 as secondary antibody (green). In the magnified inset the brightness is enhanced to show the negative neurons clearer. The image also shows that sGC-immunopositive neurons are smaller than the sGC-immunonegative neurons. (D – F) sGC and CGRP double-immunostaining. (D) sGC-positive neurons in the green channel, (E) CGRP-positive neurons in the red channel and (F) both channels merged with co-localised CGRP and sGC in some neurons. The magnified inset in (D) – (F) shows a double-positive neuron. Scale bars = 100 μm.

Figure 7

Quantitative analysis of sGC-immunopositive neurons in the trigeminal ganglion. (A) Percentage of neurons immunopositive for sGC in the whole trigeminal ganglion and (B) in V1, V2 and V3 divisions of the ganglion in control animals (n = 2), saline-treated (n = 6) and GTN-treated (n = 6) animals. (C) Double immunopositive neurons for CGRP and sGC in saline- und GTN-treated animals and neurons only immunopositive for CGRP. (D) Cell size distribution of sGC-positive and -negative neurons in the trigeminal ganglion in saline- and GTN-treated animals. The sGC-immunopositive neurons are mostly smaller than the sGC-immunonegative neurons in the saline- as well as the GTN-treated group. Error bars represent SD (n=6). * significant difference (Chi-square test of immunopositive and –negative neurons).