J Physiol Volume 515, Number 1, 277-285, February 15, 1999
The Journal of Physiology (1999), 515.1, pp. 277-285
© Copyright 1999 The Physiological Society
The role of prostaglandins in the bradykinin-induced activation of serosal afferents of the rat jejunum in vitro
Karen A. Maubach and David Grundy
Department of Biomedical Science, University of Sheffield, Western Bank, Sheffield S10 2TN, UK
MS 8150 Received 20th April 1998; accepted after revision 4 November 1998.
 |
ABSTRACT |
- This study was performed to elucidate the role of prostaglandins in the action of bradykinin on serosal afferent neurones supplying the rat jejunum. Extracellular recordings of multi-unit activity were made from serosal afferents in isolation, using a novel in vitro preparation. The discharge of single afferents within the multi-unit recording was monitored using waveform discrimination software.
- All afferents tested were both mechano- and capsaicin sensitive. Application of bradykinin elicited increases in whole nerve discharge in a concentration-dependent manner. The agonist potency estimate (EC50) was 0·62 ± 0·12 µM and is consistent with an interaction at the B2 receptor subtype.
- The stimulatory effect of bradykinin on serosal afferents was antagonized by a specific antagonist of the B2 receptor, HOE140. In contrast, a selective B1 receptor antagonist, [des-Arg10]HOE140, had no effect. The IC50 estimate obtained for HOE140 was 1·6 nM and again consistent with an interaction at B2 receptors.
- The response to a submaximal concentration of bradykinin (1 µM) was significantly reduced to 24·4 ± 54·9 % of control following blockade of cyclo-oxygenase activity with naproxen (10 µM). The addition of 1 µM prostaglandin E2 (PGE2), in the presence of naproxen, had no direct effect on afferent activity, but fully restored the response to bradykinin in 15 single afferents.
- In summary, bradykinin stimulates serosal afferents by a direct action on kinin B2 receptors that are present on serosal afferent terminals. The response to bradykinin is dependent on the presence of prostaglandins, particularly PGE2. We suggest that bradykinin has a self-sensitizing action, whereby it stimulates the release of PGE2, which in turn sensitizes the endings of serosal afferent neurones responsive to bradykinin.
|
INTRODUCTION |
Bradykinin is a naturally occurring nonapeptide which plays a pivotal role in the production of pain and inflammation. Most of the physiological actions of bradykinin have been ascribed to activation of the B2 receptor, linked to intracellular events that involve the generation of diacylglycerol and inositol triphosphates (see Levine et al. 1993). The site of B1 receptors remains to be elucidated; they may be induced (de novo synthesis) during inflammation (see Dray & Bevan, 1993), although there is evidence that sensory neurones do not express these receptors even under conditions which induce them in other cells (Davis et al. 1996). Our understanding of mechanisms by which bradykinin produces neuronal excitation is derived largely from models of cutaneous and joint pain from which it is clear that bradykinin has both direct and indirect actions, with a complex interaction between nerves and immune cells (Haley et al. 1989; Lang et al. 1990; Grubb et al. 1991; Birrell et al. 1993). However, recently there has been an upsurge of interest in sensations of gastrointestinal origin relating to conditions such as inflammatory bowel disease and irritable bowel syndrome and relatively little is known of transduction pathways in afferents supplying the gastrointestinal tract.
Spinal afferents supplying the gastrointestinal tract have their mechanosensitive receptive fields in the serosal and mesenteric connections and respond to distortion of the viscera (see Grundy & Scratcherd, 1989). The stimulus- response characteristics of these afferents are consistent with a nociceptive function and in this respect these endings are sensitive to algesic agents such as bradykinin and capsaicin (Haupt et al. 1983; Longhurst et al. 1984; Longhurst & Dittman, 1987; Sengupta & Gebhart, 1994; Pan & Longhurst, 1996). A role for prostaglandins is implicated from studies in which these agents have been shown to sensitize visceral afferent nerve endings in abdominal visceral organs and thereby enhance their responsiveness to bradykinin (Stebbins et al. 1985; Longhurst & Dittman, 1987; Mizumura et al. 1987). However, it is not known whether and to what extent the action of bradykinin depends on the action of prostaglandins. Furthermore, the changes in afferent sensitivity observed in vivo may not be due to a direct effect of bradykinin, but may be secondary to changes in gastrointestinal function, e.g. motility, secretion and blood flow.
Studies on dorsal root ganglion neurones in culture have shown that prostaglandin E2 can sensitize some nociceptors to bradykinin and, moreover, recruit other neurones that express substance P which are normally unresponsive to bradykinin (Stucky et al. 1996). Such sensitization and recruitment could contribute to visceral hyperalgesia but how the events in the cell soma reflect impulse generation in the afferent nerve terminal supplying the gut wall is unknown. We have therefore developed an in vitro model of serosal afferents supplying the rat jejunum in which impulse generation can be investigated in the absence of secondary events associated with actions on muscle and mucosa. Using this approach we have investigated interactions between prostaglandins and bradykinin in the activation of gastrointestinal serosal afferents.
|
METHODS |
Tissue preparation
All experiments were performed on male hooded Lister rats (350-400 g) and conform to Home Office guidelines. The rats were allowed to feed and drink ad libitum prior to experimentation. They were anaesthetized with a single intraperitoneal injection of urethane (1·5 g kg-1), a mid-line laparotomy performed and segments of jejunum, complete with mesenteric attachment, carefully excised before killing the animal with an overdose of anaesthetic. A 3 cm long segment was then placed in a Sylgard-lined recording chamber superfused using a Gilson Minipuls 3, at a flow rate of 10 ml min-1, with bicarbonate buffer, equilibrated with 95 % O2 and 5 % CO2 (mM: 117 NaCl, 4·7 KCl, 25 NaHCO3, 1·2 NaH2PO4, 1·2 MgCl2, 11 glucose, 2·5 CaCl2) that had been pre-heated to yield a chamber temperature of 34°C. The serosa was carefully dissected free and the remainder of the jejunum was discarded.
Nerve recording
The mesenteric arcade was drawn through a small aperture into a separate chamber, the aperture was sealed with Vaseline and the chamber filled with colourless heavy liquid paraffin (pre-warmed to 34°C) to insulate the recording electrodes. Under a viewing microscope, a mesenteric nerve bundle was teased out from the arcade and wrapped around one arm of a bipolar platinum recording electrode, with a length of connective tissue attached to the second, indifferent electrode. The electrodes were connected to a Neurolog headstage (NL100), then via a 500 × pre-amplifier, differentially amplified 50 × (NL104) and filtered with a band width of 100-1000 Hz (NL125). The signal was relayed to a spike processor (Digitimer D130), which discriminates action potentials from noise with a manually set amplitude and polarity window. The whole nerve recording was continuously monitored on a storage oscilloscope (Tektronix 5111A), digitized (PCM-2 A/D VCR adapter, Medical Systems Corp.), and recorded on VHS video tape for subsequent analysis (JVC HR-D500EK).
Single-unit discrimination
Multi-unit recordings containing only a few units often displayed action potentials of sufficiently different amplitude and waveform to allow accurate discrimination and hence identification of single units using computerized waveform analysis as described previously (Hillsley et al. 1998). This off-line analysis was performed using a Viglen Pentium PC running Spike 2 software (Cambridge Electronic Design, Cambridge, UK) via a micro1401 interface board (CED). The nerve signal was digitally sampled at 25 kHz, which was sufficient to allow accurate spike discrimination. Each spike above a given amplitude was used to set up templates for the individual action potentials. Action potential waveforms were automatically averaged, DC offset arising from noise removed, and the resulting spike shapes assigned to different waveform templates. The afferent recording was subsequently analysed such that each action potential was compared with the waveform template and either matched to one or left unclassified. The timing of each template matched spike was then used to calculate firing frequency and to plot discharge frequency against time. Tolerance was variable, but typically the allowed amplitude error was set at between 1 and 2 %, and for a spike to be matched at least 85 % of the data points had to fall within the template shape. These relatively rigid parameters were shown empirically to discriminate action potentials accurately, but at high firing frequencies a small proportion of individual spikes, typically < 5 %, could be missed because of summation. However, this underestimate of spike frequency was obviously preferable to less rigorous discrimination configurations in which 'cross-contamination' could occur.
Experimental protocol
All drugs were added to the perfusion medium supplying the first chamber. Two types of experiment were performed but in both bradykinin was applied for a period of 2 min. For the sequential concentration-response experiments, full recovery from bradykinin exposure was allowed before proceeding to the next higher concentration. Modifying drugs or vehicle were applied for at least 15 min prior to repeating the sequence. In experiments in which a submaximal concentration of bradykinin (1 µM) was administered before and after the addition of modifying drugs, a 30 min equilibration period was used as standard and compared with the appropriate time control.
Statistics
Basal impulse discharge frequency and the response discharge frequency (impulses s-1) were obtained by averaging afferent activity for 1 min prior to application of bradykinin and for the duration of the response, respectively. The response to bradykinin was also expressed as the overall increase in the number of impulses (area under the response curve); this was calculated as the total spike count for the duration of the response minus the basal impulse count which was predicted from the period before the response and adjusted for response duration. In experiments in which 1 µM bradykinin was applied before and after modifying drugs or vehicle (time controls) the responses were normalized by expressing subsequent responses to bradykinin as a percentage of the initial response before treatment. Time control experiments indicated that the bradykinin response does not run down in the current preparation. Data are expressed as means ± S.E.M. Student's t test was used for statistical analysis of the raw data. If the raw data did not fit a normal distribution then, where stated, the Mann-Whitney rank sum test was used. Values of P < 0·05 were considered to be significant.
Materials
All salts used in making the bicarbonate buffer were obtained from BDH and were of AnalaR grade or better. Urethane, capsaicin, naproxen and prostaglandin E2 were obtained from Sigma Chemical Company. Bradykinin, D-Arginyl- L-arginlyl- L-prolyl- trans-4-hydroxy- L-prolylglycyl-3- (2-thienyl)- L-alanyl- L-seryl- D-1,2,3,4-tetrahydro- 3-isoquinolinecarbonyl- L(2
,3
,7a)- octahydro-1H-indole-2-carbonyl-L-arginine (HOE140) and D-arginyl-L-arginlyl- L-prolyl-trans- 4-hydroxy- L-prolylglycyl-3-(2-thienyl)-L-alanyl-L-seryl- D-1,2,3,4-tetrahydro- 3-isoquinolinecarbonyl- L(2,3,7a)-octahydro-1H-indole-2-carbonyl ([des-Arg10]HOE140) were obtained from Research Biochemicals International. All drugs were dissolved in distilled water (unless otherwise stated), made up at a stock concentration of 1 mM and stored as small aliquots in a freezer (-20°C) until use. Capsaicin was dissolved in distilled water at a stock concentration of 0·5 mM and then made up to volume with bicarbonate buffer. Prostaglandin E2 (1 mg) was dissolved in 0·1 ml of absolute ethanol and then 0·9 ml of a 2 % solution of sodium carbonate was added to give a stock solution of 1 mg ml-1.
|
RESULTS |
Afferents studied
Extracellular recordings were made from a total of 47 serosal afferent preparations. All multi-unit recordings showed a continuous pattern of ongoing discharge, the extent of which was highly variable between preparations because of differences in the number of viable units in individual paravascular nerve bundles. Some preparations with relatively few active fibres were subjected to single unit analysis while in the majority of cases only whole nerve activity was quantified. The mechanosensitivity of these afferents was routinely tested by discrete probing of the serosal receptive field with the tip of a soft paint brush, which consistently produced a short burst of impulses. All nerve bundles tested were also found to be sensitive to capsaicin (0·1 µM, 2 min) which elicited a large increase in whole nerve discharge from 12·1 ± 2·7 to 58·1 ± 9·3 impulses s-1. The time from onset of capsaicin perfusion to the peak response (latency) was 17·6 ± 5·5 s and the overall increase in nerve discharge represented by the area under the response profile was 3812 ± 1254 impulses (n = 7).
Effect of bradykinin
A 2 min application of bradykinin (1 µM) evoked a marked elevation in whole nerve discharge in all cases (Fig. 1), although the magnitude of the response showed enormous variability because of the variable number of active fibres in each bundle. Mean discharge increased from a baseline of 11·8 ± 3·5 impulses s-1 to 38·5 ± 11·5 impulses s-1, with an overall increase in nerve discharge of 4773 ± 1530 (range 955-11150) impulses (n = 47). The latency of the response was significantly greater than for capsaicin at 36·6 ± 7·9 s (P < 0·001). Sequential concentration-response curves (CRCs) for bradykinin were constructed in five multi-unit recordings and the responses were concentration dependent in the range 0·03-10 µM (Fig. 2). The response was maximal at 3 µM, with an EC50 value, determined graphically, estimated at 0·62 ± 0·12 µM.
|
 |
View larger version
[in this window]
[in a new window] |
|
|
Figure 1. Bradykinin stimulates whole nerve mesenteric afferent discharge
Typical mesenteric afferent response to the application of bradykinin (1 µM, 2 min). Above the raw nerve trace is a rate histogram of the whole nerve discharge which clearly shows a large increase in nerve discharge in the presence of bradykinin. The histogram shows the number of action potentials in consecutive 5 s bins. Below is a segment of the nerve recording which has been expanded to illustrate the multi-unit nature of these recordings. However, action potentials of different amplitude can be clearly identified which, in some instances, enable single units to be discriminated using waveform analysis.
|
|
 |
View larger version
[in this window]
[in a new window] |
|
|
Figure 2. Sequential concentration-response curve for bradykinin (0·03-10 µM)
A is a rate histogram of the whole nerve discharge from a typical recording showing the total number of action potentials in consecutive 20 s bins; B is the mean CRC from 5 experiments in which nerve discharge from individual recordings has been normalized by expressing data as a percentage of the maximum response to bradykinin.
|
The effect of bradykinin was analysed in 24 single units whose spike shape and amplitude was such as to enable their accurate discrimination. Twenty-two of these units (92 %) responded to bradykinin; of these 17 fired action potentials spontaneously (77 %) and five (22 %) were silent prior to application of bradykinin. In the majority (82 %) of afferent units, increasing concentrations of bradykinin generated an increasing afferent response with a peak response occurring at 3 µM. In each of these units the response to 0·1 µM bradykinin was < 25 % of the maximum response. In contrast, a small number of afferents (18 %) were sensitive to lower concentrations of bradykinin and appeared to approach maximal responses at concentrations of bradykinin below 1 µM. In these four units the response to 0·1 µM bradykinin was > 75 % of the maximum response. Rate histograms of two single units illustrating the two patterns of bradykinin response are shown in Fig. 3.
|
 |
View larger version
[in this window]
[in a new window] |
|
|
Figure 3. Single unit responses to bradykinin
Sequential rate histograms of 2 single afferent units discriminated from the same whole nerve recording using waveform analysis. The upper unit displays sensitivity to higher concentrations of bradykinin with a threshold around 0·1 µM and a maximum response at 3 µM. The lower unit already has a prominent response at 0·1 µM and peaks at around 1 µM. The histograms shows the number of action potentials in consecutive 20 s bins.
|
Effect of bradykinin receptor antagonists
The B2 receptor antagonist HOE140 greatly reduced responses to bradykinin. After 30 min perfusion with HOE140 at 1 nM the whole nerve response to bradykinin was reduced from 2689 ± 693 impulses to 79·8 ± 14·4 % of control at 2296 ± 861 impulses (n = 3). HOE140 significantly reduced the bradykinin-induced response to 20·8 ± 4·3 % of control at 3 nM (from 2338 ± 250 to 467 ± 62 impulses, n = 3, P < 0·001) and to 6·2 ± 3·7 % at 10 nM (from 2394 ± 755 to 231 ± 162 impulses, n = 4, P < 0·05). From these data it was possible to estimate an IC50 for HOE140 of 1·6 nM. These afferent fibres continued to fire in the presence of HOE140 (10 nM) following mechanical manipulation of their receptive field using a paint brush, thus confirming that the nerve endings were still viable. In contrast, the B1 receptor antagonist [des-Arg10]HOE140 at 100 nM had little effect on the response to bradykinin when studied in three whole nerve recordings. Mean whole nerve discharges in the absence and presence of the antagonist were 4096 ± 48 and 3894 ± 103 impulses (a reduction to 95·1 ± 3·4 %) compared with 108 ± 7·4 % in the time-matched controls (Fig. 4).
Effect of cyclo-oxygenase inhibitors
The response to bradykinin was significantly reduced by blockade of cyclo-oxygenase activity with naproxen (10 µM). After 30 min perfusion with naproxen the whole nerve response to bradykinin (1 µM, n = 8) was significantly reduced to 24·4 ± 4·9 % of control (from 4885 ± 516 to 1291 ± 279 impulses, P < 0·0001) (Fig. 5).
|
 |
View larger version
[in this window]
[in a new window] |
|
|
Figure 5. Summary of the effect of the cyclo-oxygenase inhibitor naproxen on the afferent response to bradykinin
The mean percentage of the control response to bradykinin (1 µM, 2 min) remaining in the presence of naproxen (10 µM) is compared with the time matched control.
|
The effect of naproxen (10 µM) on the CRC for bradykinin (0·03-10 µM) was studied in four multi-unit recordings. Naproxen did not affect the response to low-dose bradykinin observed at 0·1 µM bradykinin concentrations (20·6 ± 3·9 % of the maximum response compared with 18·4 ± 6·7 % in the absence of naproxen), but markedly attenuated the response to higher doses (Fig. 6). Thus, in the presence of naproxen, the response to bradykinin appeared to reach maximum at 0·1 µM.
|
 |
View larger version
[in this window]
[in a new window] |
|
|
Figure 6.
Mean CRCs for bradykinin after incubation with naproxen (10 µM) in the presence and absence of PGE2 (1 µM)
Data are expressed as the percentage of the maximum response to bradykinin prior to treatment with naproxen. The control curve obtained by non-linear regression analysis of the bradykinin response data obtained prior to naproxen using a one-binding site equation (R2 = 0·91) is shown for comparison.
|
Effect of prostaglandin E2
The effect of prostaglandin E2 (PGE2) on the submaximal response to bradykinin in the presence of naproxen was studied in five whole nerve recordings. After 30 min perfusion with naproxen at 10 µM the bradykinin-induced (1 µM) increase in whole nerve discharge was reduced to 29·8 ± 7·1 % of control (from 5002 ± 599 to 1484 ± 427 impulses, P < 0·001). Further addition of PGE2 (1 µM) had no direct effect on afferent activity, but restored the bradykinin response to 76·8 ± 10·3 % of control at 3825 ± 634 impulses. Furthermore, there was no significant change in the latency of the response to bradykinin in the presence of naproxen (44·9 ± 7·6 s vs. 44·5 ± 6·9 s), but addition of PGE2 reduced the latency of the response to bradykinin to 75 % of control at 33·6 ± 5·5 s (P < 0·001).
The effect of PGE2 (1 µM) on the CRC for bradykinin (0·03-1 µM) in the presence of naproxen was studied in three multi-unit recordings. Figure 6 summarizes the effect of naproxen and PGE2 on the mean CRC for bradykinin. The addition of PGE2 restored the response to bradykinin giving an estimated mean EC50 that was not significantly different from the control (0·34 ± 0·15 µM, P > 0·05).
In 15 single units exposed to bradykinin, naproxen and PGE2, bradykinin (1 µM) elicited an increase in mean single afferent discharge from 1·6 ± 0·3 to 3·3 ± 0·5 impulses s-1, with a response latency of 37·3 ± 1·1 s and an overall increase in mean single afferent discharge of 490 ± 76 impulses. In the presence of naproxen (10 µM), the spontaneous discharge, the bradykinin-induced increase in mean discharge, and the overall response were all significantly reduced (Mann-Whitney, P < 0·05), although the response latency remained unchanged at 38·9 ± 0·8 s (105·4 ± 3·4 % of control). After further addition of PGE2 (1 µM), the spontaneous mean discharge remained unchanged, but the bradykinin-induced mean discharge and the overall response were restored to 66·3 ± 6·17 % and 89·1 ± 10·9 % of control, respectively. The response latency was significantly reduced to 28·9 ± 0·8 s (78·1 ± 2·3 % of control, P < 0·0001). These data are summarized in Fig. 7.
|
DISCUSSION |
This study has used a novel in vitro preparation to examine the possible mechanisms underlying the nociceptive and pro-inflammatory action of bradykinin on gastrointestinal serosal afferent sensitivity. This technique offers a number of advantages over the previous in vivo techniques used to investigate the action of bradykinin on intestinal afferents. Firstly, it permits the study of the serosa in isolation allowing the link between stimulus and response to be more closely established and, secondly, administration of bradykinin to the bathing medium allows a level of control over local drug concentration that cannot be achieved following systemic administration. Thus, the activation of serosal afferents by bradykinin in this study cannot be secondary to contraction of the muscle layers, changes in local tissue blood flow, secretion or systemic effects. The response to bradykinin must result from stimulation of elements within the serosa. It is possible that bradykinin acts directly on the serosal afferent nerve terminal (Dunn & Rang, 1990). Certainly the receptor machinery is present on dorsl root ganglion neurones in culture (Burgess et al. 1989; Naruse et al. 1992; Bauer et al. 1993, 1995; Kano et al. 1994). However, bradykinin may also act on mast cells or blood vessels present within the serosa to stimulate the release of other mediators, and in the context of the present study prostaglandins may be of particular importance.
Although many units fired action potentials spontaneously, some afferents were silent prior to application of bradykinin. Bradykinin elicited concentration-dependent increases in afferent activity. The calculated EC50 (0·6 µM) is consistent with data obtained in other systems, indicating an interaction at B2 receptors (Babbedge et al. 1995). However, there is a suggestion from the CRC profile that the response to bradykinin may be more complex, with an inflection at low doses giving a biphasic appearance to the curve. The identification of single units within the nerve bundle with differential sensitivity to bradykinin would support this view. Thus the majority of units (82 %) were sensitive only to the higher concentrations of bradykinin, although a small population of units (18 %) were sensitive at lower concentrations. However, the small contribution of these more sensitive afferents was insufficient to influence the profile of the whole nerve recording to an extent which would be reflected in a significantly improved curve fit for two- compared with one-binding site curves (R2 = 0·92 compared with 0·91, respectively). The B2 receptor antagonist HOE140 attenuated both components of the response of serosal afferents to bradykinin. The antagonist potency estimate for HOE140 (IC50 = 1·6 nM) is consistent with reports in other systems (Everett et al. 1992), indicating a homogeneous population of B2 receptors
Inhibition of endogenous prostaglandin production with naproxen reduced basal nerve discharge, indicating an involvement of prostaglandins in spontaneous activity in serosal afferent neurones, yet PGE2 itself did not influence baseline discharge. It would appear, therefore, that other prostaglandins may be involved in the basal activity of these afferents. Furthermore, a major part of the response to bradykinin, especially at the higher bradykinin concentrations, also appeared to be dependent on the presence of prostaglandins. Cyclo-oxygenase products could enhance the responses of serosal afferents either by directly stimulating afferent endings or by sensitizing them to the action of bradykinin. However, the application of exogenous PGE2 at a concentration which did not influence baseline discharge restored the response to bradykinin, indicating that PGE2 sensitizes serosal afferent terminals to the action of bradykinin. This observation is consistent with studies in humans in which PGE2 when injected into the skin alone never elicited pain but increased the intensity of bradykinin-induced pain in a concentration-dependent manner (Kindgen-Milles, 1995). Prostaglandins may simply be required in the background to augment the action of bradykinin or may be mobilized in response to the bradykinin challenge (Gammon et al. 1989). In this respect, the latency of the response to bradykinin was significantly reduced in the presence of exogenous PGE2, indicating that part of the delay in the control response to bradykinin may result from the time required to synthesize an intermediary substance, e.g. PGE2. Certainly, PGE2 output from the gut is increased during ischaemia which sensitizes visceral nociceptors (Rendig et al. 1994). If prostaglandin synthesis were contributing to the delayed action, it may be surprising that the addition of naproxen alone did not prolong the latency. However, if it is assumed that the early part of the response is produced by bradykinin alone, then this would occur before prostaglandin release has occurred. The reduced latency in the presence of PGE2 would then be due to the expected sensitization of the nerve terminal to bradykinin. Nevertheless, the latency after PGE2 is still considerably longer than that following direct capsaicin stimulation, suggesting that other factors are involved in determining the response latency. However, this effect on latency is consistent with higher concentrations of bradykinin stimulating the release of PGE2, which in turn sensitizes the endings of the majority of serosal afferents to the action of bradykinin, i.e. has a self-sensitizing action.
The sensitivity of the bradykinin response to HOE140 in the present preparation confirms that the intestinal afferent response to bradykinin is mediated via an interaction at B2 receptors. This is consistent with data from models of ischaemia in which afferent sensitization is attenuated with B2 receptor antagonists (Pan et al. 1994). Nevertheless, there may be at least two components to this sensitivity to bradykinin. One component, evident at lower concentrations of bradykinin, appears to be unaffected by cyclo-oxygenase blockade and a second component that is dependent upon prostaglandin production. These correspond to a number of important second messenger pathways that are recognized as being responsible for the actions of bradykinin. One pathway leads to biologically active lipids and involves release of arachidonic acid and its various metabolites, including prostaglandins and leukotrienes, while another leads to activation of calcium-sensitive systems (Miller, 1987). However, an increase in permeability to extracellular calcium ions is unlikely to be involved in the bradykinin-evoked responses, since the afferents were unaffected by a calcium-free medium including the calcium chelator, EGTA (K. A. Maubach & D. Grundy, unpublished observations). Similar findings have been reported with cultured sensory neurones where bradykinin-induced membrane currents were unaffected by calcium-free solutions or by calcium channel-blocking drugs (Burgess et al. 1989).
In conclusion, as with cutaneous pain, prostaglandins play an important role in determining the sensitivity of gastrointestinal afferents to bradykinin. Our data suggest that bradykinin acts at B2 receptors to stimulate prostaglandin E2 release, which in turn sensitizes serosal afferent nerve endings to a more direct action of bradykinin at B2 receptors. This relationship between bradykinin and prostaglandin E2 could be interpreted as a positive feedback mechanism to optimize the extent of afferent activation that is evoked during certain inflammatory situations.
|
REFERENCES |
| Babbedge, R., Dray, A. & Urban, L. (1995). Bradykinin depolarises the rat isolated superior cervical ganglion via B2 receptor activation. Neuroscience Letters 193, 161-164 |
[Medline] |
| Bauer, M. B., Murphy, S. & Gebhart, G. F. (1995). Stimulation of cyclic GMP production via a nitrosyl factor in sensory neuronal cultures by algesic or inflammatory agents. Journal of Neurochemistry 65, 363-372 |
[Abstract] |
| Bauer, M. B., Simmons, M. L. & Gebhart, G. F. (1993). Bradykinin and capsaicin stimulate cyclic GMP production in cultured rat dorsal root ganglion neurons via a nitrosyl intermediate. Journal of Neuroscience Research 36, 280-289 |
[Medline] |
| Birrell, C. J., McQueen, D. S., Iggo, A. & Grubb, B. D. (1993). Prostanoid-induced potentiation of the excitatory and sensitizing effects of bradykinin on articular mechanociceptors in the rat ankle joint. Neuroscience 54, 537-544 |
[Medline] |
| Burgess, G. M., Mullaney, I., McNeill, M., Coote, P. R., Minhas, A. & Wood, J. N. (1989). Activation of glutamate cyclase by bradykinin in rat sensory neurons is mediated by calcium influx: possible role of the increase in cyclic GMP. Journal of Neurochemistry 53, 1212-1218 |
[Abstract] |
| Davis, C. L., Naeem, S., Phagoo, S. B., Cambell, E. A., Urban, L. & Burgess, G. M. (1996). B1 bradykinin receptors and sensory neurones. British Journal of Pharmacology 118, 1469-1476 |
[Medline] |
| Dray, A. & Bevan, S. (1993). Inflammation and hyperalgesia: highlighting the team effort. Trends in Pharmacological Science 14, 287-290. |
|
| Dunn, P. M. & Rang, H. P. (1990). Bradykinin-induced depolarisation of primary afferent nerve terminals in the neonatal rat spinal cord in vitro. British Journal of Pharmacology 100, 656-660 |
[Medline] |
| Everett, C. M., Hall, J. M., Mitchell, D. & Morton, I. K. (1992). Contrasting properties of bradykinin receptor subtypes mediating contractions of the rabbit and pig isolated iris sphincter pupillae preparation. Agents Actions Supplement 38, 378-381. |
|
| Gammon, C. M., Allen, A. C. & Morell, P. (1989). Bradykinin stimulates phosphoinositide hydrolysis and mobilisation of arachidonic acid in dorsal root ganglion neurones. Journal of Neurochemistry 53, 95-101 |
[Abstract] |
| Grubb, B. D., Birrell, G. J., McQueen, D. S. & Iggo, A. (1991). The role of PGE2 in the sensitizatioan of mechanoreceptors in normal and inflamed ankle joints of the rat. Exploratory Brain Research 84, 383-392. |
|
| Grundy, D. & Scratcherd, T. (1989). Sensory afferents from the gastrointestinal tract. In American Handbook of Physiology, vol. 1, Motility and Circulation, ed. Woods, J., pp. 593-620. Oxford University Press, New York. |
|
| Haley, J. E., Dickenson, A. H. & Schachter, M. (1989). Electrophysiological evidence for a role of bradykinin in chemical nociception in the rat. Neuroscience Letters 97, 198-202 |
[Medline] |
| Haupt, P., Janig, W. & Kohler, W. (1983). Response pattern of visceral afferent fibres, supplying the colon, upon chemical and mechanical stimuli. Pflügers Archiv 398, 41-47 |
[Medline] |
| Hillsley, K., Kirkup, A. J. & Grundy, D. (1998). Direct and indirect actions of 5-HT on the discharge of mesenteric afferent fibres innervating the rat jejunum. The Journal of Physiology 506, 551-561 |
[Abstract/Full Text] |
| Kano, J., Kawakami, T., Hikawa, N., Hori, H., Takenaka, T. & Gotoh, H. (1994). Bradykinin-responsive cells of dorsal root ganglia in culture: cell size, firing, cystosolic calcium, and substance P. Cell Molecular Neurobiology 14, 49-57. |
|
| Kindgen-Milles, D. (1995). Effects of prostaglandin E2 on the intensity of bradykinin-evoked pain from skin and veins of humans. European Journal of Pharmacology 294, 491-496 |
[Medline] |
| Lang, E., Novak, A., Reeh, P. W. & Handwerker, H. O. (1990). Chemosensitivity of fine afferents from rat skin in vitro. Journal of Neurophysiology 63, 887-901 |
[Medline] |
| Levine, J. D., Fields, H. L. & Basbaum, A. J. (1993). Peptides and the primary afferent nociceptor. Journal of Neuroscience 13, 2273-2286 |
[Abstract] |
| Longhurst, J. C. & Dittman, L. E. (1987). Hypoxia, bradykinin and prostaglandins stimulate ischemically sensitive visceral afferents. American Journal of Physiology 240, H539-545. |
|
| Longhurst, J. C., Stebbins, C. L. & Ordway, G. A. (1984). Chemically induced cardiovascular reflexes arising from the stomach of the cat. American Journal Physiology 247, H459-466. |
[Medline] |
| Miller, R. J. (1987). Bradykinin highlights the role of phospholipid metabolism in the control of nerve excitability. Trends in Neuroscience 210, 226-228. |
|
| Mizumura, K., Sato, J. & Kumazawa, T. (1987). Effects of prostaglandins and other putative chemical intermediaries on the activity of canine testicular polymodal receptors studied in vitro. Pflügers Archiv 408, 565-572 |
[Medline] |
| Naruse, K., McGehee, D. S. & Oxford, G. S. (1992). Differential responses of Ca-activated K channels to bradykinin in sensory neurons and F-11 cells. American Journal of Physiology 262, C453-460 |
[Medline] |
| Pan, H. L. & Longhurst, J. C. (1996). Ischaemia-sensitive sympathetic afferents innervating the gastrointestinal tract function as nociceptors in cats. The Journal of Physiology 492, 841-850 |
[Abstract] |
| Pan, H. L., Stahl, G. L., Rendig, S. V., Carretero, O. A. & Longhurst, J. C. (1994). Endogenous BK stimulates ischemically sensitive abdominal visceral C fiber afferents through kinin B2 receptors. American Journal of Physiology 267, H2398-2406 |
[Medline] |
| Rendig, S. V., Pan, H. L. & Longhurst, J. C. (1994). Brief mesenteric ischemia increases PGE2, but not PGI2, in intestinal lymph of cats. American Journal of Physiology 266, R1662-1696. |
|
| Sengupta, J. N. & Gebhart, G. F. (1994). Characterization of mechanosensitive pelvic nerve afferent fibers innervating the colon of the rat. Journal of Neurophysiology 71, 2046-2060 |
[Medline] |
| Stebbins, C. L., Smith, R. C. & Longhurst, J. C. (1985). Effect of prostaglandin on bradykinin-induced visceral-cardiac reflexes. American Journal of Physiology 249, H155-163 |
[Medline] |
| Stucky, C. L., Thayer, S. A. & Seybold, V. S. (1996). Prostaglandin E2 increases the proportion of neonatal rat dorsal root ganglion neurons that respond to bradykinin. Neuroscience 74, 1111-1123 |
[Medline] |
Acknowledgements
We thank the BBSRC for supporting this research.
Corresponding author
D. Grundy: Department of Biomedical Science, University of Sheffield, Western Bank, Sheffield S10 2TN, UK.
Email: d.grundy{at}sheffield.ac.uk
Author's present address
K. A. Maubach: Merck, Sharp and Dohme Research Laboratories, Terlings Park Research Centre, Harlow, Essex CM21 0QR, UK.
This article has been cited by other articles:
|
|
|
|
 
E. Aihara, M. Hayashi, Y. Sasaki, A. Kobata, and K. Takeuchi
Mechanisms Underlying Capsaicin-Stimulated Secretion in the Stomach: Comparison with Mucosal Acidification
J. Pharmacol. Exp. Ther.,
October 1, 2005;
315(1):
423 - 432.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
B. Wang, J. Glatzle, M. H. Mueller, M. Kreis, P. Enck, and D. Grundy
Lipopolysaccharide-induced changes in mesenteric afferent sensitivity of rat jejunum in vitro: role of prostaglandins
Am J Physiol Gastrointest Liver Physiol,
August 1, 2005;
289(2):
G254 - G260.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
W. Rong, K. Hillsley, J. B Davis, G. Hicks, W. J Winchester, and D. Grundy
Jejunal afferent nerve sensitivity in wild-type and TRPV1 knockout mice
J. Physiol.,
November 1, 2004;
560(3):
867 - 881.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
H.-Z. Hu, N. Gao, S. Liu, J. Ren, Y. Xia, and J. D. Wood
Metabotropic Signal Transduction for Bradykinin in Submucosal Neurons of Guinea Pig Small Intestine
J. Pharmacol. Exp. Ther.,
April 1, 2004;
309(1):
310 - 319.
[Abstract]
[Full Text]
|
|
|
|
|
|
|
B. T. Green, A. Calvin, S. M. O'Grady, and D. R. Brown
Kinin-Induced Anion-Dependent Secretion in Porcine Ileum: Characterization and Involvement of Opioid- and Cannabinoid-Sensitive Enteric Neural Circuits
J. Pharmacol. Exp. Ther.,
May 1, 2003;
305(2):
733 - 739.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
C Kreiss, L A Birder, S Kiss, M M VanBibber, and A J Bauer
COX-2 dependent inflammation increases spinal Fos expression during rodent postoperative ileus
Gut,
April 1, 2003;
52(4):
527 - 534.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
M. J. Carr, M. Kollarik, S. N. Meeker, and B. J. Undem
A Role for TRPV1 in Bradykinin-Induced Excitation of Vagal Airway Afferent Nerve Terminals
J. Pharmacol. Exp. Ther.,
March 1, 2003;
304(3):
1275 - 1279.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
A. J. Kirkup, A. M. Brunsden, and D. Grundy
Receptors and Transmission in the Brain-Gut Axis: Potential for Novel Therapies: I. Receptors on visceral afferents
Am J Physiol Gastrointest Liver Physiol,
May 1, 2001;
280(5):
G787 - G794.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
U. C. Kopp, D. M. Farley, M. Z. Cicha, and L. A. Smith
Activation of renal mechanosensitive neurons involves bradykinin, protein kinase C, PGE2, and substance P
Am J Physiol Regulatory Integrative Comp Physiol,
April 1, 2000;
278(4):
R937 - R946.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Introduction
) in time control experiments, and following 30 min incubation with the B2 receptor antagonist HOE140 (1, 3 and 10 nM, 
) and the B1 receptor antagonist [des-Arg10]HOE140 (100 nM, 
).
