Neurogenic Inflammation of Guinea-Pig Bladder

Capsaicin, substance P, and ovalbumin, instilled into the bladders of naive and ovalbumin (OVA) sensitized guineapigs caused inflammation, as indicated by increased vascular permeability. Histological changes after exposure to these compounds progressed with time from intense vasodilatation to marginalization of granulocytes followed by interstitial migration of leukocytes. In vitro incubation of guinea-pig bladder tissue with substance P and ovalbumin stimulated release of prostaglandin D2 and leukotrienes. In vitro incubation of bladder tissue with capsaicin, OVA, prostaglandin D2, leukotriene C4, histamine, or calcium ionophore A-23587 all stimulated substance P release. These data suggest that bladder inflammation initiated by a variety of stimuli could lead to a cyclic pattern of release of inflammatory mediators and neuropeptides, which could result in amplification and persistence of cystitis after the inciting cause has subsided.


Introduction
Apparently spontaneous inflammation of the urinary bladder in the absence of infection has been reported in humans and cats. 1,2 In humans, the most widely recognized form of cystitis in the absence of infection is interstitial cystitis (IC), a chronic inflammatory disease of the lower urinary tract. Clinical symptoms commonly associated with IC include increased urinary frequency and urgency, nocturia, suprapubic pressure and pain. 1, Although numerous theories regarding the aetiology of IC have been proposed, the pathogenesis remains unknown. Traumatic injury, lymphatic obstruction and autoimmune disease are among the suggested causes. Regardless of the cause(s) of IC and other forms of non-infectious cystitis, a number of reports suggest that the nervous system is intimately involved in the development and persistence of non-infectious cystitis. [5][6][7] Neurogenic inflammation affecting various organs, including the skin, lungs, airways, gut, eye and joints, has been reported in several species, including humans. -1 A variety of stimuli, such as antigens, heat, cold, bacterial or viral infection, and direct stimulation of sensory nerves, can trigger the onset of neurogenic inflammation. -1 Neuropeptides such as substance P (SP), neurokinin A (NKA), calcitonin gene-related peptide (CGRP) and vasoactive intestinal peptide (VIP) are located in capsaicin sensitive primary afferent nerves in the bladders of humans and animals. 11,12 The release of SP, NKA and CGRP has been correlated with smooth muscle contraction, ( 1994 Rapid Communications of Oxford Ltd vasodilatation, increased vascular permeability and facilitated neurotransmitter release from intramural nerves. 1 SP, CGRP and VIP also mediate the sensory function of capsaicin sensitive nervous control of the micturition threshold and may facilitate spontaneous bladder contraction. 14 Experimentally, antidromic electrical stimulation of unmyelinated afferent nerve fibres causes extravasation of Evans blue dye in the urinary bladder of guinea-pigs. 15 Of the neuropeptides present in the afferent sensory nerve fibres, SP is thought to be the primary neurotransmatter responsible for initiation of neurogenic inflammation, and CGRP is believed to play an important role in facilitating and mediating the effects of SP. 16 Intravenous administration of SP also causes extravasation of Evans blue dye and accumulation of leukocytes in the bladder wall of experimental animals, and intravesical application of SP has a similar effect.7. 8 Capsaicin, an extract of hot pepper, is a neurotoxin for unmyelinated sensory nerve fibres which selectively depletes SP and other neuropeptides from the nerves at low doses and causes degeneration of these nerves at high doses. Pretreatment of rats with capsaicin significantly reduced the severity of xylene induced cystitis, strongly suggesting the participation of SP in the pathogenesis of this model of cystitis. 19 It has been reported that SP stimulates histamine release from various tissues and isolated mast cells, but this effect appears to be specific to the tissue source of the mast cells and the species from which these tissues and mast cells are obtained, a Preliminary studies in our laboratory have indicated that Mediators of Inflammation. Vo13. 1994 isolated guinea-pig bladder tissue releases histamine in response to exposure to capsaicin or SP. We have previously described the results of investigations using a guinea-pig model for IC in which cystitis develops in response to antigen sensitization and challenge. 21,22 Antigen challenge of sensitized bladder tissue in this model causes release of histamine, prostaglandins and leukotrienes, and this model of cystitis was used to test the hypothesis that antigen challenge of sensitized guinea-pig urinary bladder tissue would stimulate SP release from sensory afferent nerves and that exposure of guinea-pig bladder tissue to inflammatory mediators would also stimulate release of SP from sensory afferent nerves, suggesting a positive feedback mechanism for the amplification of cystitis.

Materials and Methods
Female, albino guinea-pigs (450-600 g) were used in these experiments. Sensitized animals were actively sensitized to ovalbumin (OVA; chicken egg albumin, ovalbumin grade V; Sigma Chemical Company, St Louis, MO) by administering three intraperitoneal injections of OVA (10 mg/kg) at 48 h intervals.
Control animals received injections of the same volume of saline at the same time intervals. Twenty-one days after the last injection, animals were sacrificed for in vitro studies with sodium pentobarbital (100 mg/kg, i.p.) or anaesthetized for in vivo studies as described subsequently. For in vitro studies, urinary bladders were removed and placed in physiologic salt solution (PSS) of the following composition (mM): NaC1, 119; NaHiPO4, 1; KC1, 4.7; CaCl2, 2.5; MgCl2, 0.5; NaHCO 3, 25; and glucose, 11. The PSS was maintained at 37C and aerated continuously with a mixture of 95% O2 and 5% CO2 (pH 7.4).
Intravesical instillation of capsaicin, SP and OVA: Guinea-pigs were anaesthetized with ketamine HC1 (40 mg/kg, i.p.) and xylazine (2.5 mg/kg, i.m.). A polypropylene catheter with a closed end and side openings (31/2 French Tom Cat Catheter, Sherwood Medical, St Louis, MO) was introduced transurethrally into the bladder. The catheter was advanced until the first drop of urine appeared. Urine was drained from the bladder by applying light pressure on the abdomen, and 3 ml of 0.9% saline, 3 l.tM capsaicin, 100 btM SP, or OVA (1 btg/ml) was infused.
To ensure consistent contact of the bladder with saline or the test substances, the animals received a total of four instillations during a 2 h period. Some guinea-pigs were sacrificed by pentobarbital overdose 30 min after the first instillation, and the remainder were sacrificed 2, 4, 8  Leukotriene and PGD2 release from bladder tissue incubated with SP was compared with release of these compounds stimulated by sensitized tissues exposed to OVA. Bladder tissue was incubated with OVA (1 l.tg/ml) or SP (100 btM) for 30 min, and the bath solution was collected for determination of concentrations of PGD2 and leukotrienes.
To compare the capacity of various substances to stimulate SP release from the bladder, other tissues were challenged with capsaicin (3 btM), leukotriene C (LTC4, 1001ttM), PGD (100btM), histamine (100 l-tM), OVA (1 g/ml), or calcium ionophore A23587 (100 btM) for 30 min, after which the bath solution was collected for subsequent determination of SP release.
PGDe radioimmunoassay: PGD2 content was determined by radioimmunoassay as described previously. 23 This assay is sensitive to about 0.03 pmol/ 0.1 ml.
Leukotriene radioimmunoassay: The amount of leukotriene released was measured by radioimmunoassay. 24 The limit of sensitivity of this assay is approximately 0.03 pmol as defined by that amount required to inhibit [3H]-LTC binding by 10%. The anti-peptidoleukotriene antibody is highly selective with little affinity (cross-reactivity < 1%) for a variety of heterologous eicosanoids. The antibody does not, however, distinguish markedly between LTC4, LTD and LTE 4. Therefore, the results are presented as leukotriene-like immunoreactivity (LLI). Substance P radioimmunoassay: SP release was determined by radioimmunoassay (RIA). Five l.tg of synthetic [Tyr-O]-substance P was dissolved in 50 1 0.1 M phosphate buffer, pH 7.5. One mCi (5 t.tl) of high specific activity 125I and 20 1 (20 t-tg) chloramine-T was added, and the mixture was agitated for 45 s. The reaction was stopped by addition of 200 btl saturated L-tyrosine. The labelled peptide was purified by gel filtration on a 0.9 30 cm column of G-35 Sephadex (fine) eluted with 0.02 M phosphate buffered saline, pH 7.5.
[125I]-O-Tyr-SP was nearly 100% bindable with excess antibody. Standards (0 to 2 000 pg/ml) or sample (200 1) were added to 500 1 of anti-substance P (produced by G. P. Kozlowski; 1:8 000 initial dilution) in 0.1 M tris, pH 7.5 containing 0.5% normal rabbit serum. Tracer SP (100 1 containing 10000 cpm) was added to each tube and mixed. Tubes containing only tracer, as well as nonspecific binding tubes containing PBS-NRS buffer without antibody, were also prepared at this time. Samples were mixed and incubated for 24 h at 4 %.
Bound tracer was precipitated by addition of 100 btl of 1:4 sheep anti-rabbit IgG (produced in our laboratory) and 1 000 btl of 3.3% polyethylene glycol in PBS, and samples were centrifuged at 3000xg for 20 min. The supernatant was discarded, and bound radioactivity was counted in the pellet with a gamma counter equipped with RIA data reduction (Gammatrac Model 1290, Tm Analytic, Inc., Elk Grove Village, IL). Characteristics of this assay include a minimum detectable dose less than 10 pg/ tube at ECs0, intra-assay variation less than 5%, and inter-assay variation of approximately 8%. Cross-reactivity studies (at EC50) showed 100% cross-reactivity for SP, SP (4-11) and physalaemin, and less than 0.01% for metenkephalin and somatostatin. Crossreactivity of other peptides is being evaluated.
Substance P degradation in samples was pre-Statistical analysis: Statistical analysis of the results was performed with Minitab TM 8.0 (Minitab Inc., Rosemont, PA) software using analysis of variance and Student's t-test for paired or unpaired data. 25 The n values refer to one strip of tissue per experimental animal. All values were expressed as mean + standard error of the mean (S.E.M.), and a value of p < 0.05 was considered significant. stimulating Evans blue extravasation in amounts most closely approximating that in response to 1 lttg/ ml OVA were chosen for study to allow comparison of the effects of these substances. Therefore, although extravasation of Evans blue dye induced by 3 M capsaicin was significantly higher than that induced by 100 btM SP or 1 btg/ml OVA, this observation is a function of the relative concentrations of capsaicin, SP and OVA chosen to be studied.

Results
Leukocyte migration: Histological changes in the bladder tissue after capsaicin or SP exposure (or intravesical instillation of OVA in sensitized guineapigs) appeared to follow a time course of increasing severity between 30 min and 20 h post-exposure, progressing from pronounced vasodilatation to marginalization of granulocytes, followed by diapdesis of leukocytes, initially restricted to the perivascular area and subsequently spreading through the submucosa and into the detrusor (Fig.  2). Eight hours after SP exposure, a profuse cellular migration within the submucosa near the urothelium was observed, and leukocytes were also present between muscle bundles of the detrusor. Leukocytes continued to migrate into these areas for the next 12 h, at which time the histological appearance of the tissues stabilized.
Intravesical saline caused minimal histological changes in the bladder wall (Fig. 3). However, sensory nerve stimulation by intravesical instillation of capsaicin induced intense vasodilatation, haemorrhage and leukocyte infiltration of the submucosa and detrusor muscle. Granulocytes, most of which were releasing their granules, were the predominant leukocyte observed in the wall of the guinea-pig bladder after capsaicin exposure, but lymphocytes were also present. Similar effects were observed after intravesical administration of substance P. Intravesical instillation of OVA had no effect on the histological appearance of the bladders of non-sensitized animals and caused less haemorrhage than SP or capsaicin exposure but a similar progression of other histological changes when instilled into the bladders of sensitized guinea-pigs. Inflammatory mediator release: Substance P induced PGD2 and LLI release from isolated sensitized and control bladder tissue; OVA had no effect on unsen sitized bladder tissue but stimulated PGD and LLI release from sensitized bladder tissue (Fig. 4). Substance P stimulated release of relatively more prostaglandins and less leukotrienes than OVA exposure of sensitized bladder tissue. Increasing the concentration of either SP (to 1 000 M) or OVA (to 100btg/ml), or increasing incubation time up to 90 min had no effect on PGD2 or LLI release (data not shown). Activation of sensory neurones by inflammatory mediators: In vitro exposure of guinea-pig bladder tissue to various mediators of inflammation, including histamine, LTC and PGD2, stimulated SP release (Fig. 5). In vitro exposure of bladder tissue from sensitized guinea-pigs (but not from control animals) to OVA also stimulated SP release. SP release in response to these substances was compared with SP release in response to activation of sensory nerves with capsaicin and exposure of bladder tissue to a non-physiological, non-immunological stimulus (calcium ionophore). Calcium ionophore stimulated a comparatively low amount of SP release at this concentration, while capsaicin stimulated significant SP release, as expected.

Discussion
The results of these experiments clearly demonstrate the effects of stimulation of sensory nerves on inflammation of the guinea-pig urinary bladder. Although the concentrations of SP evaluated in these experiments exceed expected tissue concentrations, the capacity of endogenous SP to produce similar results is confirmed by the response of the bladder to treatment with capsaicin. The concentrations of capsaicin, SP and OVA used in these experiments were selected to produce a similar degree of plasma extravasation as measured by tissue concentrations of Evans blue dye to allow direct comparison between the effects of these substances. The effects of SP and capsaicin appear to be concentration dependent, and lower concentrations used in preliminary studies caused primarily mild interstitial oedema and marginalization of leukocytes within the vasculature despite marked vasodilatation.
Using this guinea-pig model of antigen induced cystitis, the authors have previously described release of inflammatory mediators, including histamine, PGD and leukotrienes, from the bladder subsequent to antigen sensitization and challenge. 21,22 Other investigators have reported that antigen sensitization results in alterations of micturition frequency, voiding volume and cystometrographic data in guinea-pigs, as well as increased uptake of intravesical 14C-urea. 26,27 The present experiments demonstrate that OVA sensitization and subsequent challenge of guinea-pig bladder causes release of SP, and that SP is released from naive and sensitized bladder tissue in response to histamine, PGD2, LTC4 and A-23587. These results indicate that SP is released in response to inflammatory mediators, and it cannot be determined from these experiments whether or not antigen sensitization and exposure directly stimulates SP release.
In a cyclophosophamide induced cystitis model in rats, Maggi et al. demonstrated that bladder hyperreflexia was mediated through stimulation of  Substance P stimulated release of significantly more PGD2, while ovalbumin stimulated release of significantly more leukotriene.., ovalbumin, lg/ml; I, substance P, 100 M. unmyelinated afferent nerve fibres but that capsaicin pretreatment actually increased plasma extravasation as measured by Evans blue.'* Bilateral removal of the pelvic ganglia almost eliminated plasma extravasation within the bladder wall in response to cyclophosphamide and suppressed enhancement of plasma extravasation by capsaicin pretreatment. The differences between this model and the present resuits may be explained in part by the authors' observation that cyclophosphamide-induced cystitis is not accompanied by a toxic effect on the afferent nerves of the bladder but may be accompanied by stimulation of the ganglia.
The calcium ionophore A-23587 is a non-immunologic secretagogue for mast cell products. 29 Although SP release in response to this compound was relatively small at the concentration used, this finding supports the capacity of mast cell degranulation to stimulate SP release. Degranulating mast cells release histamine, prostaglandins and leukotrienes, among other inflammatory mediators. It is apparent that, at least in the guinea-pig bladder, these compounds stimulate release of SP, most likely from unmyelinated afferent nerve fibres, which can then promote and amplify inflammation. These effects may be concentration dependent, and it is probable that SP release could be altered by varying the concentration of agonists used.
The capacity of SP to stimulate histamine release from mast cells has been demonstrated repeatedly, but sensitivity of mast cells to the effects of SP seems to be specific to the species and tissue of origin of the mast cells. 2 32 While leukotrienes and other metabolites of arachidonic acid released in response to SP may come from a variety of sources, including mast cells, vascular endothelium and smooth muscle, it appears that mast cells are the sole source of histamine. ," Neuropeptide-induced increased vascular permeability clearly appears to be in part due to interaction of SP with mast cells.
The role of mast cells in the pathogenesis of IC has been the subject of considerable debate. IC is characterized by chronic inflammation, and several reports suggest a role for increased mast cell numbers (and possibly activity) in the initiation and propagation of IC, 33'34 although others question whether or not mast cells are involved. 35 Participation of neuropeptides in the pathogenesis of IC may not rely solely on the presence and degranulation of mast cells. SP and CGRP also have the capacity to modify the function of other cells of the immune system. SP acts as a chemoattractant and migratory stimulant for monocytes and neutrophils, and SP receptors have been identified on the surface of these cells. 36,7 SP receptors have also been identified on the surface of lymphocytes, 8 and functional receptors for CGRP have been characterized on lymphocytes and macrophages in mice and rats. 7,9,4 SP stimulates lymphocyte proliferation, migration and immunoglobulin production. 7,8 CGRP stimulates lymphocyte migration but may inhibit lymphocyte proliferation. 37,9 SP stimulates neutrophil antibody dependent, cell mediated cytotoxicity and may facilitate neutrophil production of toxic oxygen species. 41 Several reports have described increased nerve density in the bladder wall in the presence of IC, including Hand in 1949 and more recent investigators. <7 Lasanen et a1.12 described a marked, extensive depletion of VIP, neuropeptide Y, and SP-immunoreactive nerve fibres after distension of the rat bladder wall. This depletion of SP-containing sensory nerves was reversible as early as 21 days later. A decrease in active sensory nerves or decreased concentrations of neurotransmitters may contribute to transient relief of pain provided IC patients by bladder distension. Increased SP activity was observed during the recovery phase in previously distended rat bladders; this may be related to neurogenic inflammation and resumption of clinical symptoms. Intravesical instillation of capsaicin in five patients with hypersensitivity disorders of the lower urinary tract provided temporary relief of symptoms in all patients, strongly suggesting participation of unmyelinated afferent nerve fibres in sensation of pain in the lower urinary tract. 42 Systemic administration of capsaicin to newborn rats decreased detrusor hyperreflexi and vascular permeability associated with xylene-induced cystitis. 19 In humans, local application of capsaicin has been shown to block the axonflare response and the flare response to intradermal allergen and vasosactive inflammatory mediators. 4 The results of the experiments described suggest that bladder inflammation in the absence of infection may at least in part be due to initiation of a cyclic sequence of events which entails release of inflammatory mediators by a variety of stimuli which subsequently themselves stimulate, directly or indirectly, release of neuropeptides. Neuropeptides, particularly SP, contribute to continued release of inflammatory mediators, perpetuating and amplifying inflammation. Chronic bladder inflammation may thus be the result of a variety of initiating factors. This also suggests that, to be effective, therapeutic interventions should be directed at more than one phase of this cyclic process.