Background
Mast cells are primarily known for their role in Immunoglobulin E (IgE)-dependent hypersensitivity reactions, although their role in a wide range of physiological functions including neuroimmune interactions has been increasingly appreciated [
1]. Activation of mast cells usually results in their degranulation, releasing preformed mediators, most notably histamine and chemokines while longer lasting activation results in the release of newly formed mediators such as cytokines, prostaglandins, and leukotrienes [
2]. Mast cells are often found in proximity to sensory nerve endings and vasculature and their degranulation can modulate the excitability of nociceptive nerve endings [
3]. They can themselves be activated by neurogenically-generated mediators such as substance P [
4], various antigens, and inflammatory mediators released in pathophysiological states [
5].
Peripheral mast cell activation is generally considered pro-inflammatory and pro-nociceptive. Inhibition of peripheral mast cells by chronic degranulation using intraplantar compound 48/80 blocks rodent capsaicin hyperalgesia [
6]. Direct pharmacological inhibition by the often used mast cell "stabilizer" cromolyn also inhibits nociceptive behaviours in the second phase of the formalin test [
7]. Cromolyn further inhibits neurogenic inflammation and edema induced by substance P or by low-frequency saphenous nerve electrical stimulation [
8]. Multiple substances such as cyclooxygenase (COX) inhibitors [
9] or spleen tyrosine kinase (Syk) inhibitors [
10] are also known to inhibit mast cell activation and degranulation.
While a functional involvement of peripheral mast cells in inflammatory pain is well established, the role of central nervous system (CNS) mast cells is not well understood. Mast cells are not found in the spinal cord but mostly in the thalamus [
11]. Significant sex differences and asymmetrical distribution which are estrogen-dependent are noted in thalamic mast cells in naive animals [
12,
13], and after nerve injury [
14]. Mast cells have also been particularly investigated in the cerebral dura mater in relation to migraine pathophysiology and are thought to promote neurogenic inflammation and activation of meningeal nociceptors [
15,
16]. At the spinal level, dural mast cells are found at a significant density at the cervical, thoracic, and lumbar regions [
17,
18], although their functional roles have not been studied.
Since little to no white matter separates the lumbar dorsal horn from the subarachnoid and dura mater, it is hypothesized that mediators released from dura mast cells can reach the superficial laminae, a key relay station for nociception [
19], to modulate synaptic transmission and nociception. While some of the individual mast cell mediators also produced by other immune cells are implicated in nociception and spinal synaptic transmission [
20‐
22], the combination of mediators some of which serve anti-inflammatory and homeostatic functions [
1], has not been specifically studied. To examine this, mast cells were cultured and activated
in vitro and the mast cell supernatant was injected intrathecally to assess nociception. In addition, the supernatant was applied on the lumbar spinal cord to measure C-fiber evoked field potentials. Since intraplantar capsaicin induces spinal long-term potentiation (LTP) [
23], a form of synaptic plasticity that amplifies nociception [
19,
23], we next examined spinal dural mast cell degranulation after intradermal capsaicin at various timepoints in awake conscious rats. We then examined whether intraplantar carrageenan, a model of non-neurogenic sterile peripheral inflammation, induces changes in number and degranulation of spinal dura and thalamic mast cells in male and female rats. Finally, we tested whether two intrathecally administered mast cell inhibitors, sodium cromoglycate (cromolyn) and the Syk inhibitor BAY-613606, are antinociceptive against capsaicin or carrageenan hyperalgesia.
Discussion
Dural mast cells have been mainly investigated in the cerebral dura mater, where density is high and they are close to sensory nerve endings [
24]. Substantial evidence suggests that they are involved in migraine pathophysiology [
16]. Mast cell degranulation with systemically-administered Compound 48/80 activates meningeal nociceptors, phospho-ERK expression, and spinal trigeminal nucleus cFos activation [
15]. Mast cell derived mediators such as serotonin, prostaglandins, and histamine can activate and sensitize meningeal nociceptors [
22] and COX-1 is expressed in dural mast cells [
25]. It is known that lumbar dural mast cell number varies during development [
17], although dural mast cell function remains under speculation. The current study is the first systematic examination of lower spinal dura mast cells in male and female adult rats using two different pain paradigms.
The position of the spinal cord grey matter close to the CSF at the lumbar spinal cord level provided grounds to hypothesize a potential impact of mast cell-derived mediators on superficial dorsal horn neurons. Consistent with this hypothesis, application of supernatant from activated cultured mast cells, particularly from those which were longer activated and secreted newly formed mediators, induced significant mechanical hyperalgesia and LTP at spinal synapses of C-fibres. This suggests a potential contribution of preformed mediators, but a much stronger contribution of newly formed mediators. One hour stimulation of mast cells with TNP is sufficient to induce degranulation, release of histamine and other preformed mediators [
26]. Intrathecal histamine administration is pro-nociceptive in mice [
27] and spinal histamine receptor antagonists show anti-nociceptive efficacy in various animal models [
28]. Given the heterogeneity of mast cells between tissues responsible for their phenotype [
1,
29] and even within the dura [
17], we believe that using mouse bone-marrow cultured mast cells which are less homogeneous than their rat equivalents [
30] is still appropriate here. The concentration of cytokines in supernatant was within the same range to that detected by spinal in-vivo microdialysis studies [
31] and that of other electrophysiological and behavioral studies. Interestingly, spinally-applied TNFα in the picogram range does not induce
in vivo spinal cord LTP [
32], although intrathecal IL-6 in the nanogram range induces mechanical allodynia [
33]. Other mediators such as IL-4 or IL-10 produced by mast cells are not known to be pro-nociceptive [
34] and the release of mast cell mediators may also be a homeostatic mechanism [
35]. While it is interesting to speculate which specific mediators may be important in inducing the effects seen in our study, it is likely that the combination of preformed and newly formed mediators is physiologically relevant and more important. Persistent peripheral nociceptive stimuli for greater than 24 hours, such as in peripheral hindpaw inflammation, may lead to different combinations of mast cell mediators and potentially to a more important activation of dural mast cells than brief peripheral nociceptive stimuli.
Previous studies have shown that peripheral application of 1% capsaicin induces LTP at spinal synapses of C-fibers [
23]. In this study, we found that this dose induces a significantly increased percentage of mast cell degranulation at 3 hours post-application. Since no difference was seen between naïve, vehicle, and 10 minutes post-capsaicin, this suggests that the 3 hour effect is not due to injection stress or the initial nociceptive behaviors induced by capsaicin. This is interesting given that mast cells can degranulate rapidly within minutes [
36]. Nevertheless, the intense afferent barrage produced by capsaicin appears to switch granulated mast cells to become degranulated, but not to increase the mast cell number. While mast cell staining with toluidine blue allows for clear visualization of degranulation by staining the granules, it is possible that results may be influenced by loss of mast cell staining with complete degranulation [
37], although we did not detect significant decreases in cell numbers after treatments. There was no effect of mast cell inhibition on capsaicin-induced flinching and both cromolyn and BAY-613606 were unable to reduce capsaicin mechanical hyperalgesia, suggesting that this increase is not necessary for mechanical hyperalgesia. Although we were not able to completely inhibit mast cells, the effects of BAY-613606 were stronger than cromolyn reaching below basal levels.
Carrageenan injection produces a peripheral inflammatory stimulus activating afferents from the entire hindpaw that persists longer than capsaicin. Surprisingly, we could find no significant mast cell changes in male rats. However, when we examined female rats, there was a highly significant increase in lumbar dural mast cell density at 24 hours post-carrageenan. This density increase is different from capsaicin and was also seen with 1% carrageenan. Since we did not see a percent increase in mast cell degranulation and no changes to the thoracic region, this could suggest increased localized infiltration or attraction of mast cells. Interestingly, mast cells may also be activated to release mediators without obvious degranulation or in a selective manner [
38] and there exist mucosal, connective, and mixed types of mast cells in spinal dura with potentially different functional significance [
17]. It may be speculated that mast cells may infiltrate the dura and arrive intact from the peripheral circulation. While in the current study, we did not attempt to localize the source of mast cells or their anatomical position on the dura, we did not find mast cells inside the spinal cord in the 24 hour carrageenan treated females (data not shown). For example, it is known that mast cells may cross blood vessel walls and their mediators may modulate the blood brain barrier [
39,
40]. Studies of CNS mast cells on blood brain/spinal cord barrier regulation in animal models of pain are lacking.
Mast cells are known to express estrogen receptors [
41] and can be activated by estrogens which increase calcium influx and enhance degranulation and mediator release. These effects can be blocked by estrogen receptor antagonists or calcium chelators [
42]. The estrous cycle influences thalamic mast cell degranulation [
43] and hormones can increase mast cell density in rats [
13]. In our study, we did not find any significant changes in thalamic mast cells in rats after carrageenan, although we were unable to conclude about the influence of the estrous cycle. Nevertheless, the brains and dura mater were collected from the same animals, thereby further suggesting that the dural mast cell changes are not widespread. It should also be noted that thalamic mast cells are implicated in multiple functions including sexual reproduction [
44], blood brain barrier regulation [
40], and stress [
45]. Administration of aspirin, but not morphine, decreases rat thalamic mast cell numbers [
46]. Nevertheless, we report here an intriguing mast cell sex difference in the spinal dura mater. Various studies have suggested an interaction of hormones and sex with mast cells and potentially pain behaviors. Prevalence of migraine, mast cell diseases, and immediate-hypersensitivity type reactions are particularly prominent in female populations and may relate to hormonal status [
47,
48]. Changes in estrogen levels may be linked to migraine attacks in women, perhaps due to cerebral mast cell interaction [
49]. Future studies may address specifically the role of the estrous cycle on dural mast cells.
We successfully inhibited capsaicin and carrageenan-induced changes in lumbar dural mast cells with two different drugs, although these did not have significant anti-nociceptive effects. Higher intrathecal dosing or volume were not deemed appropriate as they either produced side effects upon injection in the case of cromolyn, or were unable to further reduce mast cell degranulation in the case of BAY-613606 (data not shown). Peripherally, cromolyn has been shown to be anti-nociceptive when administered in the formalin test [
7]. Cromolyn, known as a "mast cell stabilizer", has been used in the treatment of asthma and allergic rhinitis [
50,
51]. It reduces neurogenic inflammation and inhibits mast cell mediator release [
52]. Cromolyn has also been suggested to inhibit chloride transport in mast cells, epithelial, and nerve cells [
53], and also inhibit substance P binding and adenosine-induced plasma extravasation [
54]. The Syk inhibitor BAY-613606 has been shown to be highly selective for the Syk family of tyrosine kinases, which are important signaling molecules involved in mast cell activation and degranulation [
55]. When administered orally in rats, it reduces mast cell degranulation, histamine release, and inflammation, more potently than cromolyn [
56]. Other Syk inhibitors reduce inflammation in animal models and have been suggested for clinical treatment of rheumatoid arthritis and inflammatory bowel syndrome [
57]. Like cromolyn, Syk inhibitors are not fully specific for mast cells since Syk is expressed on a variety of cell types including basophils, leukocytes, platelets, dendritic cells and macrophages [
58]. Nevertheless, we found that both cromolyn and BAY-613606 administered intrathecally inhibited increased percentage of mast cell degranulation and degranulated cell density in a localized fashion. Other potentially more powerful methods of mast cell inhibition including the use of transgenic animals such as mast cell-deficient rats [
59], or the assessment of different behavioral endpoints will be interesting for use in future studies.
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
DX carried out tissue processing, mast cell counting, behavioral assays, data analysis, designed and coordinated all the studies, and drafted the manuscript. SG carried out tissue processing, mast cell counting, behavioral assays, and helped with data analysis and manuscript revisions. RD carried out in vivo electrophysiological studies and helped with their design. EN carried out tissue processing, mast cell counting and helped with manuscript revisions. AA carried out mast cell culture and activation, cytokine assay, and helped with design of mast cell supernatant studies. WE helped with design of mast cell supernatant studies. JS participated in the design of the project and all the studies and in manuscript revisions. All authors read and approved the manuscript.