Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 6, 61

Protease Activated Receptors 1 and 4 Sensitize TRPV1 in Nociceptive Neurones

Affiliations

Protease Activated Receptors 1 and 4 Sensitize TRPV1 in Nociceptive Neurones

Vittorio Vellani et al. Mol Pain.

Abstract

Protease-activated receptors (PAR1-4) are activated by proteases released by cell damage or blood clotting, and are known to be involved in promoting pain and hyperalgesia. Previous studies have shown that PAR2 receptors enhance activation of TRPV1 but the role of other PARs is less clear. In this paper we investigate the expression and function of the PAR1, 3 and 4 thrombin-activated receptors in sensory neurones. Immunocytochemistry and in situ hybridization show that PAR1 and PAR4 are expressed in 10 - 15% of neurons, distributed across all size classes. Thrombin or a specific PAR1 or PAR4 activating peptide (PAR1/4-AP) caused functional effects characteristic of activation of the PLCβ/PKC pathway: intracellular calcium release, sensitisation of TRPV1, and translocation of the epsilon isoform of PKC (PKCε) to the neuronal cell membrane. Sensitisation of TRPV1 was significantly reduced by PKC inhibitors. Neurons responding to thrombin or PAR1-AP were either small nociceptive neurones of the peptidergic subclass, or larger neurones which expressed markers for myelinated fibres. Sequential application of PAR1-AP and PAR4-AP showed that PAR4 is expressed in a subset of the PAR1-expressing neurons. Calcium responses to PAR2-AP were by contrast seen in a distinct population of small IB4+ nociceptive neurones. PAR3 appears to be non-functional in sensory neurones. In a skin-nerve preparation the release of the neuropeptide CGRP by heat was potentiated by PAR1-AP. Culture with nerve growth factor (NGF) increased the proportion of thrombin-responsive neurons in the IB4- population, while glial-derived neurotropic factor (GDNF) and neurturin upregulated the proportion of thrombin-responsive neurons in the IB4+ population. We conclude that PAR1 and PAR4 are functionally expressed in large myelinated fibre neurons, and are also expressed in small nociceptors of the peptidergic subclass, where they are able to potentiate TRPV1 activity.

Figures

Figure 1
Figure 1
Expression of PAR1-4 in sections of adult mouse DRG. A. In situ hybridisation (ISH) for PAR1-4 carried out as described in Methods. Positive cells shown by arrowheads. Sections counterstained using hematoxylin-eosin. Scale bars 40 μm. B. Similar sections in which PAR1, 3 and 4 expression was determined using immunohistochemistry. Positive cells shown with arrows. The PAR2 antibodies available to us were found to be non-specific on Western blots and results for PAR2 are therefore not shown. Sections counterstained using hematoxylin-eosin. Scale bars 40 μm. C. Expression of PAR1 - 4 as a function of neuronal size in adult mouse DRG using ISH. Overall neuronal population (grey) is compared with those positive for each PAR isoform (black). Overall, PAR1 was found to be expressed in 15.0% of neurones, PAR2 in 21.5%, PAR3 in 49.5% and PAR4 in 14.5%. D. Similar results obtained using immunohistochemistry. PAR2 is not shown because the antibody was found to exhibit non-specific binding, and PAR4 is not shown because it proved impossible to distinguish neuronal from glial cell staining (see B). Overall PAR1 was expressed in 10.3% of neurones, and PAR3 in 42.0%.
Figure 2
Figure 2
Calcium signals elicited by PAR agonists. A. Adult mouse neuron in which an increase in [Ca]i was elicited by a specific PAR2 activator peptide (PAR2-AP, SLIGRL, 100 μM), but not by thrombin (100 nM) which activates PAR1, 3 and 4. The neuron also expresses receptors for TRPA1 and TRPV1, as shown by its responses to the specific TRPA1 agonist mustard oil (MO, 100 μM) and the specific TRPV1 agonist capsaicin (1 μM). B. Most PAR2-AP-responsive adult mouse neurons also responded to capsaicin and mustard oil but none responded to thrombin (n = 180 neurones). Staining of unfixed cells with fluorescently labelled IB4 (isolectin B4 from Griffonia simplicifolia coupled to Alexa 594, Molecular Probes) immediately after the calcium imaging experiment showed that most PAR2-AP responsive neurons were IB4-positive (grey bar). Similar results were obtained in neonatal rat neurons (88.6 ± 5.1% of cells responding to PAR2-AP also responded to capsaicin, and 82.5 ± 6.0% were IB4+). C. Adult mouse neuron in which an increase in [Ca]i was elicited by thrombin (100 nM). This neuron also expresses the ion channels TRPA1 and TRPV1, as shown by its responses to mustard oil (MO, 100 μM) and capsaicin (1 μM). Cell was identified as a neuron on morphological grounds, confirmed by calcium increase observed in response to 25 mM KCl. D. Around 25-33% of thrombin-responsive neurons (n = 455) also responded to capsaicin (1 μM), mustard oil (100 μM), the peptide Bv8 (100 nM) and bradykinin (1 μM) but none responded to PAR2-AP (100 μM). Final bar shows that no thrombin-responsive adult mouse neuron bound IB4. E. Glial cell which responded with increase in [Ca]i to PAR1-AP (100 μM) and to PAR2-AP (100 μM). Cell was identified as a glial cell on morphological grounds, confirmed by absence of calcium increase in response to 25 mM KCl. In separate experiments, cells of this morphology were also identified by the glial-specific anti-S100 antibody (not shown). F. Percentage of glial cells responding to thrombin (100 nM), PAR1-AP (100 μM) and PAR2-AP (100 μM). Deletion of PAR1 ablated responses to both thrombin and PAR1-AP (bars 4 and 5) while deletion of PAR2 was without effect on responses to thrombin and PAR1-AP (bars 6 and 7).
Figure 3
Figure 3
Desensitization of PAR1 and PAR4 ablates calcium signals in response to thrombin. A. Increase in [Ca]i recorded as in Fig. 2. Calcium increase elicited by application of PAR1-AP completely desensitizes response to a subsequent application of PAR1-AP but not to PAR4-AP. The calcium signal in response to thrombin was ablated in the large majority of cells by desensitization of both PAR1 and PAR4. All experimental details as in Fig. 2. B. Following desensitization of PAR1 and PAR4 only 1.6% of neurons gave a calcium signal in response to thrombin, compared with 15.2% in control neurons. Summary of results from n = 187 neurons from 4 separate coverslips.
Figure 4
Figure 4
Sensitization of TRPV1 by PAR activation. A - D. Heat-activated currents were significantly enhanced in c. 10% of neurons by application of PAR1-AP and PAR4-AP. Single traces in panels to left are taken from time courses shown in right hand panels. Both PAR1-AP (TFLLR at 100 μM) and PAR4-AP (AYPGKF, 200 μM) caused long-lasting sensitisation. Sensitization showed complete tachyphylaxis on a second application. E. Percentage sensitization in experiments similar to those in A. Thrombin (100 nM), trypsin (100 nM), PAR1-AP and PAR4-AP all caused approximately a doubling of the inward current elicited by heat. Thrombin-induced sensitisation was largely blocked by the PKC inhibitor Ro318220 (1 μM) and by the broad-spectrum kinase inhibitor staurosporine (1 μM, both applied throughout the experiment). F, G Calcium imaging experiments to monitor functional sensitization of TRPV1 by thrombin. F shows typical experiment in which increases in [Ca]i elicited by successive brief exposures to capsaicin (500 nM, 1 s) were enhanced by exposure to thrombin (100 nM, black bar). All experiments performed in adult mouse neurones. G shows percentage of cells sensitized in experiments similar to those shown in F on neurons from WT and PAR1-/- adult mice. Difference was significant (χ2 test, *, p < 0.05).
Figure 5
Figure 5
Translocation of PKCε to neuronal surface membrane caused by thrombin. A. Translocation of PKCε to neuronal surface membrane in control conditions (left) and following exposure to thrombin (100 nM, 30 and 60 sec). PKCε translocated rapidly to the surface membrane following application of thrombin (arrow in middle panel) and at longer times became progressively internalised (arrowhead in middle panel and right panel). Adult mouse neurons cultured in 10% FBS in absence of NGF and neurturin. Scale bars 5 μm. B. Percentage of neurons showing translocation to the plasma membrane as a function of time of exposure to thrombin (number of neurons > 2000 for each point). C. Peak percentage of neurons in which PKCε was translocated, as a function of thrombin concentration (number of neurons > 2000 for each point). Continuous curve shows a Hill equation with n = 0.7 and K1/2 = 2 nM. D. Size distribution of thrombin-responsive neurons. Grey bars show size distribution of overall neuronal population, and black bars show neurons in which PKCε translocation was observed following exposure to thrombin (100 nM, 30 s). E. Activation of PKCε translocation by proteases and specific PAR activator peptides. Thrombin, trypsin and cathepsin G (all 100 nM, 30 s) caused translocation of PKCε in a similar percentage of adult mouse neurons but tryptase and collagenase IV were ineffective. *, p < 0.05, ***, p < 0.001, t test compared to thrombin. PAR1-AP (TFLLR, 100 mM) caused translocation similar that of thrombin. PAR4-AP (AYPGKF, 200 mM, bar 2) caused translocation in a significantly smaller proportion of neurons when compared to PAR1-AP. Increased concentrations of activating peptides did not cause increased translocation (not shown). The effects of PAR1-AP and PAR4-AP (both at 100 mM) were partially but not completely additive. PAR2-AP (SLIGRL-NH2) had no effect. *, p < 0.05, ***, p < 0.001, t test compared to PAR1 alone.
Figure 6
Figure 6
Co-localisation of thrombin-induced translocation of PKCε with other neuronal markers. A. PKCε translocation (green) following exposure to thrombin (100 nM, 30 s) colocalises with other neuronal markers as shown. PKCε translocation was co-localised in c. half of cells with expression of the neuropeptide CGRP and with the neurofilament marker N52, and in a smaller proportion of cells with the neuropeptide substance P (SP) (panels on right). PKCε translocation was not in general co-localised with IB4 binding nor with parvalbumin (Prv) or COX-1 (panels on left). Neurones from adult mice cultured in the absence of NGF, with the exception of the COX-1 experiment which was carried out in neonatal rat sensory neurons cultured in NGF (100 ng/ml) as the antibody available to us did not bind mouse COX-1. Scale bars all 5 μm. B. Summary of results from experiments similar to those shown in A. First bar shows percentage of cells showing translocation of PKCε in response to thrombin (100 nM, 30 s). Remaining bars show percentages of these thrombin-responsive cells which co-expressed the neuronal markers noted beneath each bar. White bar in N52 column shows the proportion of the N52 positive neurons in which TRPV1 expression had been demonstrated by recording a calcium increase in response to capsaicin prior to fixation (c.f. Fig. 2). Final white bar shows overall fraction of thrombin-responsive neurons in which TRPV1 expression had been demonstrated by calcium imaging.
Figure 7
Figure 7
Heat-induced CGRP release from isolated rat skin is facilitated by PAR1 activation. A. CGRP release elicited by heat (closed circles, n = 16) was increased approximately two-fold by the rat PAR1-AP SFLLRN-OH (100 μM, open circles, applied during minutes 5 to 20, n = 12,). Points show mean ± SEM; **, p < 0.01 (ANOVA + Scheffé).
Figure 8
Figure 8
Upregulation of proportion of thrombin-responsive cells by neurotrophic factors. A. Percentage of neurons in which PKCε translocation was observed following exposure to thrombin (100 nM, 30 s) (white bars) was increased significantly by NGF (100 ng/ml, 3 days, p < 0.05) and by neurturin (NTN, 50 ng/ml, 3 days, P < 0.01). Effects were partially additive (final bar). Grey bars show percentage of neurons that were IB4+; neurturin caused significant upregulation of expression of thrombin responsiveness in the IB4+ population but NGF had no significant effect. *; p < 0.05; **, p < 0.01 compared to control. B. Deletion of PAR1 reduces but does not eliminate responsiveness to thrombin. C. Deletion of PAR2 does not affect proportion of neurons responsive to thrombin.

Similar articles

See all similar articles

Cited by 29 articles

See all "Cited by" articles

References

    1. Bunnett NW. Protease-activated receptors: How proteases signal to cells to cause inflammation and pain. Seminars in Thrombosis and Hemostasis. 2006;32:39–48. - PubMed
    1. Ossovskaya VS, Bunnett NW. Protease-activated receptors: Contribution to physiology and disease. Physiological Reviews. 2004;84:579–621. - PubMed
    1. Dale C, Vergnolle N. Protease signaling to G protein-coupled receptors: Implications for inflammation and pain. Journal of Receptors and Signal Transduction. 2008;28:29–37. - PubMed
    1. Ramachandran R, Hollenberg MD. Proteinases and signalling: pathophysiological and therapeutic implications via PARs and more. British Journal of Pharmacology. 2008;153:S263–S282. - PMC - PubMed
    1. Coughlin SR. How the protease thrombin talks to cells. Proceedings of the National Academy of Sciences of the United States of America. 1999;96:11023–11027. - PMC - PubMed

Publication types

MeSH terms

Feedback