Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Severe cystitis has been reported in laboratory animals after cyclophosphamide (CYP) administration (Ahluwalia et al., 1994; Alfieri and Gardner, 1997) and in patients receiving the drug as part of their treatment (Frasier et al., 1991). CYP is a drug with a wide spectrum of clinical uses, and it has been proved to be effective in the treatment of cancer and nonmalignant disease states. However, unless precautions are taken, this drug may induce acute inflammation of the urinary bladder (Grinberg-Funes et al., 1990). The genesis of this inflammation is being examined.

Pretreatment with the tachykinin NK₁ receptor antagonist, GR203040, has been shown to reduce the magnitude of CYP-induced cystitis (Alfieri and Cubeddu, 1997; Alfieri and Gardner, 1997). Other investigators have also shown that primary afferent capsaicin-sensitive fibers (PACSF), through the release of substance P (sP), neurokinin A, and/or calcitonin gene-related peptide, play an important role in animal models of cystitis (Maggi et al., 1987; Chahl, 1988). However, the mechanism by which NK₁ receptor inhibition protects against CYP-induced cystitis, is unclear.

Nitric oxide (NO) synthesis is mediated by three different types of nitric-oxide synthases (NOS): neuronal, endothelial, and inducible (Moncada et al., 1991). The first two synthases are expressed constitutively and are calcium-dependent, whereas inducible NOS (iNOS) must be induced and is calcium-independent (Moncada et al., 1991). It is well accepted that NK₁ receptor activation can induce the synthesis and release of NO (Regoli et al., 1994; Maggi, 1997). In addition, different agents (chemical, biological, and physical) could also trigger the inflammatory signal via transcriptional factors like NF-B, or immune-related mediators such as interleukins, tumor necrosis factor , and platelet-activating factor (Pfeilschnifter et al., 1992; Souza-Filho et al., 1997). These factors are known to increase the levels of iNOS, producing large amounts of NO, vasodilation, and edema.

Based on these observations, we propose that CYP (and/or its metabolite acrolein) may stimulate PACSF to release sP and related substances, which through activation of NK₁ receptors my increase NO production inducing inflammation and damage. We propose that the reported amelioration of CYP-induced inflammatory cystitis with NK₁ antagonists is due to a reduction in the formation of NO. To evaluate this hypothesis, first we evaluated the role of NO on CYP-induced cystitis. Bladder iNOS was determined by measuring calcium-independent NOS activity in control and in animals treated with CYP. The urinary excretion of NO metabolites was used as an indicator of NO production. The effects of a NO synthesis inhibitor on NO production, protein plasma extravasation, and bladder toxicity were evaluated. Second, we investigated whether increases in NO formation mediate the protective effect of the NK₁-receptor antagonist (GR205171) on CYP-induced cystitis. To determine the relative contributions of NOS and of NK₁ receptors to the toxicity, the effects of separate and of combined treatments with these agents were evaluated.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Wistar male rats (body weight, 300-400 g) were used in all experiments. When administered, CYP was injected i.p. at a dose of 150 mg/kg. GR205171 [(2S,3S)-2-methoxy-(5-trifluoromethyltetrazol-1-yl-benzyl)-(2-phenylpiperidin-3-yl)amine hydrochloride] was used as the selective NK₁ antagonist. NO synthase was inhibited by the use of S(+)-N⁵-[imino(nitroamino)methyl]ornithine (N^G-nitro-L-arginine; L-NNA).

The animals were included in one of eight groups: group 1, control (saline, 0.1 ml/100 g, i.p.); group 2, GR205171 (10 mg/kg, i.p.); group 3, L-NNA (10 mg/kg, i.p.); group 4, CYP + saline (0.1 m1/100 g, i.p., 5 min before and 3 h after CYP); group 5, CYP + GR205171 (10 mg/kg, i.p., 5 min before CYP); group 6, CYP + GR205171 (10 mg/kg × two doses, 5 min before and 3 h after CYP); group 7, CYP + L-NNA (10 mg/kg, i.p., 5 min before CYP); and group 8, CYP + L-NNA (10 mg/kg, i.p., 5 min before) + GR205171 (10 mg/kg × two doses, 5 min before and 3 h after CYP).

Plasma Protein Extravasation. Plasma protein extravasation was measured by the Evans blue dye leakage technique (Saria and Lundberg, 1983). Anesthesia was induced by the i.p. administration of urethane (1.2 g/kg). An external jugular vein was cannulated for the injection of Evans blue dye (50 mg/kg) in a dose volume of 2.5 ml/kg. The dye was administered 15 min before the animal was exsanguinated by infusion of 50 ml of 0.9% w/v saline, at 37°C, into the left cardiac ventricle. The time of exsanguination was taken as the endpoint of the experiment. After this, the urinary bladder was removed and blotted dry before weighing, and the content of dye was determined by spectrophotometry (at 620 nm), after extraction in a known volume of formamide at 60°C for 24 h. Plasma protein extravasation was expressed as the content of Evans blue dye in micrograms per gram of tissue.

Histological Study. Histological examination of the bladder was performed in three groups of animals. Controls (group 1), CYP (group 4), and CYP + L-NNA (one dose of 10 mg/kg) and + GR205171 (two doses of 10 mg/kg each) (group 8). None of these animals received the Evans blue dye. The tissue samples were fixed overnight in buffered neutral formalin, processed to paraffin wax, sectioned at 3 to 4 µm, and stained with hematoxylin and eosin. Extents of white blood cell infiltrates were graded in a 10× field from 0 to 4 as follows: 0, no extravascular leukocytes; +, 10 leukocytes; ++, 11 through 19 leukocytes; +++, 20 through 29 leukocytes; and ++++, 30 leukocytes.

Urinary Excretion of Nitrates and Nitrites. Urines were collected in metabolic cages. Two samples of 2-h intervals were obtained; the first, from hours 2 to 4 after CYP and the second, from hours 4 to 6 after CYP. Urinary volumes were measured, and the urine samples were frozen at 60°C until assayed. After protein precipitation, nitrates were quantitatively converted to nitrites by the action of the nitrate reductase (obtained from Klebsiella pneumoniae) for 60 min under anaerobic conditions. The total nitrite concentration was then estimated by the Griess reaction and read at 540 nm in a spectrophotometer. Urine samples were processed in duplicates. Creatinine was quantified by a modification of the Jaffe reaction, with picric acid in alkaline solution. NO metabolites (nitrates + nitrites) were expressed as millimoles per gram of creatinine.

NOS Activity. NOS activity was measured in the rat bladder as the formation of L-[¹⁴C]citrulline from L-[¹⁴C]arginine (NEN Life Science Products, Wilmington, DE) (Salter et al., 1991). The bladders were minced and suspended in 10 volumes of cold 20 mM HEPES buffer (pH 7.4), containing 1 mM dithiothreitol and protease inhibitors (10 µg/ml leupeptin, 10 µg/ml soybean trypsin inhibitor, and 2 µg/ml aprotinin). Bladders were homogenized by a Polyton (Brinkmann Instruments, Westbury, NY) and subsequently centrifuged at 20,000g for 20 min, at 4°C. Endogenous L-arginine was removed by passing the supernatants through a 0.5-ml column of AG 50W-X8, Na⁺ form (Bio-Rad, Hercules, CA). Supernatants were incubated in the presence of 3 µM L-[¹⁴C]arginine (3 µM final concentration), 10 mM valine, 10 µM tetrahydrobiopterin, 10 µM FAD, 2 mM NADPH, in the presence of 1 mM EGTA. The reaction was terminated by the addition of 1 ml of 20 mM HEPES buffer (pH 5.5) containing 2 mM EDTA, and the sample was immediately passed through an AG 50W-X8, Na⁺ form column (1 ml) and eluted with HEPES (pH 5.5). The radioactivity was determined by liquid scintillation spectrometry. iNOS is the calcium-calmodulin-independent isoform of NOS (Moncada et al., 1991). Therefore, iNOS activity was calculated from the differences between samples containing EGTA (3 mM) and samples containing the NOS inhibitor, L-NMMA (1 mM). Protein concentrations were measured as described by Bradford (1976). NOS activity was expressed as picomoles of citrulline per milligram of protein per minute.

Statistical Analysis. All the results are expressed as mean ± S.E., and statistical significance was determined by ANOVA followed by a post hoc Duncan's test. Two-group analysis was determined by Student's t test. Differences were considered significant at P < .05.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

The contents of Evans blue, expressed as micrograms per gram of tissue, in control rats and the effects of CYP are shown on Fig. 1. CYP induced a marked increase (30- to 40-fold increase above control levels) in protein plasma extravasation in the rat urinary bladder. Protein plasma extravasation was significantly greater at 6 than at 4 h after CYP (Fig. 1). The effects of CYP on iNOS activity and on NO production are shown in Tables 1 and 2 and Fig. 1. Calcium-independent NOS activity, characteristic of iNOS, was undetectable in bladders from control rats; however, 6 h after treatment with CYP there was a marked increase in calcium-independent NOS activity (Table 2). Urinary NO metabolites (nitrates + nitrites) were quantitated in urines collected from 2 to 4 h and from 4 to 6 h after CYP. Administration of CYP increased the urinary excretion of NO metabolites with greater increases in samples collected from 4 to 6 h, than from 2 to 4 h (Fig. 1). Higher excretion of NO metabolites was associated with greater plasma protein extravasation (Fig. 1).

View larger version (32K): [in this window] [in a new window]   Fig. 1.   Effects of CYP on urinary bladder plasma protein extravasation and on the urinary excretion of NO metabolites. Rats were sacrificed either at 4 or at 6 h after i.p. administration of 150 mg/kg CYP. Rats were placed on metabolic cages for urine collection, and urines were collected from 2 to 4 h and from 4 to 6 h after CYP. Subsequently, the rats were anesthetized by the i.p. administration of urethane (1.2 g/kg). Evans blue dye (50 mg/kg) was injected via the jugular vein, and 15 min later the rat was exsanguinated. The urinary bladder was removed and blotted dry before the Evans blue dye accumulation on the bladder was determined and expressed as micrograms of Evans blue per gram of tissue. The NO metabolites (nitrates + nitrites) were quantitated on the urine samples (2-4 h and 4-6 h). Left ordinate: plasma protein extravasation as micrograms of Evans blue/g of tissue. Right ordinate: urinary excretion of NO metabolites as millimoles of nitrates + nitrites/g of creatinine. Shown are mean values ± S.E. of at least six rats per group. **, significantly different from control values at P < .01.

Neither L-NNA (10 mg/kg, i.p.), an inhibitor of NO synthesis, nor GR205171 (10-20 mg/kg, i.p.), a selective NK₁ antagonist, had any significant effect per se on the basal levels of protein plasma extravasation (not shown). However, GR205171 reduced CYP-induced protein plasma extravasation, with two doses being more effective than a single dose of the NK₁ antagonist (Figs. 2 and 3). Similarly, L-NNA (10 mg/kg, i.p.) exerted a protective effect on CYP-induced protein plasma extravasation. The bladder content of Evans blue after CYP was reduced by 40 to 50% by treatment with L-NNA (Figs. 2 and 3). The effects of combined treatment with GR205171 and L-NNA on protein plasma extravasation are shown in Figs. 2 and 3. Combined treatment with the NK₁ antagonist and the NO-synthesis inhibitor produced comparable reduction in protein plasma extravasation to those achieved with each drug given separately. No significant differences were observed between these groups. No additive effects were observed with the drug combination.

View larger version (39K): [in this window] [in a new window]   Fig. 2.   Effects of GR205171 and L-NNA on CYP-induced plasma protein extravasation. Experiments were conducted as described in the legend for Fig. 1. Results from rats sacrificed only 4 h after CYP administration are depicted in this graph. GR205171 was administered i.p. in two doses of 10 mg/kg each (one 5 min before the CYP and the second, 3 h later). L-NNA (10 mg/kg, i.p.) was given 5 min before the CYP administration. Shown are mean values ± S.E. of at least six rats per group. **, significantly different from control values at P < .01. No significant differences were observed between GR205171, L-NNA, and GR205171 + L-NNA groups.

View larger version (34K): [in this window] [in a new window]   Fig. 3.   Effects of GR205171 and L-NNA on CYP-induced plasma protein extravasation. Experiments were conducted as described in legend for Fig. 1. Results from rats sacrificed only 6 h after CYP administration are depicted in this graph. GR205171 was administered i.p. in two doses of 10 mg/kg each (one 5 min before the CYP and the second, 3 h later). L-NNA (10 mg/kg, i.p.) was given 5 min before the CYP. Shown are mean values ± S.E. of at least six rats per group. **, significantly different from control values at P < .01. No significant differences were observed between GR205171 and GR205171 + L-NNA groups. Values obtained with GR205171 + L-NNA were significantly lower than those obtained with L-NNA alone (P < .05) but not from values obtained with GR205171 alone.

L-NNA markedly decreased the basal urinary excretion of NO metabolites, as well as CYP-induced increases in urinary NO metabolites (Table 1). The increases in NO metabolites induced by CYP were also reduced by GR205171, with two doses being more effective than a single dose of GR205171 (Fig. 4). GR205171 had no effect on basal NO-metabolite excretion.

View larger version (38K): [in this window] [in a new window]   Fig. 4.   Effects of CYP on NO metabolite excretion: interaction with GR205171. Experiments were conducted as described in the legend for Fig. 1. Urinary excretion of NO metabolites (nitrates + nitrites), expressed as millimoles of NO metabolites per gram of creatinine, was quantified in urine samples collected from 2 to 4 h and from 4 to 6 h after CYP. Urinary NO metabolites were also measured in control rats, 2 to 4 h and 4 to 6 h after being treated with saline. Dotted lines across the figure were drawn to facilitate observing changes above control values. Shown are mean values ± S.E. of at least six rats per group. **, significantly different from CYP-treated rats at P < .01.

The histological appearance of bladders obtained from control animals (group 1), animals treated with CYP (group 4), and of rats treated with CYP + L-NNA + GR205171 (group 8), are shown on Fig. 5. CYP induced marked urothelial damage, edema, vascular congestion, and white blood cell infiltrate (Fig. 5B). Combined treatment with L-NNA and GR205171 reduced the histological damage and the inflammatory changes induced by CYP in the rat bladder (Fig. 5C). Lesser edema, congestion, and white blood cell infiltrates were observed in the group treated with the NO synthesis inhibitor and NK₁ antagonist.

View larger version (110K): [in this window] [in a new window]   Fig. 5.   Histological changes in the rat bladder induced by CYP: effects of combined treatment with GR205171 and L-NNA. All sections were stained with H&E and photographed at 25×. A, control rats (group 1, saline injection): intact urothelium, blood vessels with thin walls, no edema or infiltrate present (0/+++), two smooth muscle layer without alterations. B, CYP rats (group 4): extensive urothelial damage, urothelium was absent from many regions of the bladder, marked inflammatory infiltrate with abundant lymphocytes, and polymorphonuclear white blood cells (+++/++++), marked edema separating the smooth muscle layers, and marked congestion. C, CYP + L-NNA + GR205171 (group 8): urothelium present, mild to moderate vascular congestion, mild edema, and few white blood cell infiltrates (+/++++). C shows considerable improvement in comparison with histological findings of B.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

CYP is known to produce hemorrhagic cystitis (Grimberg-Funes et al., 1990). In the present study, treatment with CYP produced marked plasma protein extravasation, vascular congestion and edema of the bladder, extensive leukocyte infiltration, and damage of the urothelium. Interestingly, these changes were associated with the appearance of calcium-independent NOS activity, which characterizes iNOS expression, and with increases in the urinary excretion of NO metabolites. These findings together with the observation that treatment with NOS inhibitors markedly ameliorates CYP-induced cystitis, indicates a fundamental role for NO in the pathogenesis of this form of drug-induced toxicity.

Current evidence indicates that pretreatment with selective NK₁ receptor antagonists ameliorates CYP-induced cystitis. Pretreatment with RP67580 (Ahluwalia et al., 1994), GR203050 (Alfieri and Gardner, 1997), and GR205171 (present study) ameliorated plasma protein extravasation and the histological damage of the urinary bladder of rats and ferrets treated with CYP. In addition, pretreatment with capsaicin has been shown to reduce CYP-induced cystitis (Ahluwalia et al., 1994). These observations indicate that part of the inflammatory changes induced by CYP in the urinary bladder is mediated via the activation of NK₁ receptors. Consequently, drugs with antagonistic activity on NK₁ receptors are expected to be effective against this toxicity. Although the intrinsic mechanisms accounting for NK₁ activation are not fully established, neuropeptides (sP and neurokinin A) released from PACSF may be involved in this process. This view is supported by the observation that acute administration of capsaicin as well as a variety of chemical stimuli (i.e., xylene) increases plasma protein extravasation in the bladder, only when the PACSF are intact (Maggi and Meli, 1988). In conclusion, our findings support the view that activation of NK₁ receptors, possibly by neuropeptides released from PACSF, plays a role in the pathogenesis of the inflammatory cystitis induced by CYP.

In addition to attenuating plasma protein extravasation and the histological changes indicative of inflammatory cystitis (Alfieri and Cubeddu, 1997; Alfieri and Gardner, 1997), the NK₁-tachykinin receptor antagonist markedly reduced the increase in urinary NO metabolites induced by CYP (present study). Our results suggest that the increase in NO metabolites occurring within the first 6 h of CYP administration derives mainly from activation of NK₁ receptors, possibly by sP and/or related neuropeptides. It thus appears that NK₁-mediated NO formation plays a pathogenic role in this form of toxicity. Frode-Saleh and colleagues (1999) have also shown that NO mediates sP-induced inflammatory changes in animal models of pleurisy. Interestingly, in our model GR205171 did not affect the basal levels of NO metabolites, and it was only effective in reducing the increases in NO metabolites induced by CYP. These results indicate that GR205171 does not exert a nonspecific effect on NOS activity.

Although we can not determine what proportion of the urinary NO metabolites derives from the inflamed bladder, the fact that bladders from CYP-treated rats showed a marked increase in calcium-independent conversion of L-arginine to L-citrulline, characteristic of an increased iNOS activity, suggests that at least part of the increase in urinary excretion of NO metabolites represents increased bladder production of NO. Increased urinary levels of NO metabolites have also been observed in other animal models of chemical cystitis, despite having negative cultures (Lundberg, 1996). In conclusion, iNOS activity is increased in bladders of CYP-treated rats, where it plays an important role in increasing the local production of NO, which is an important mediator of the inflammatory bladder damage.

In conclusion, the following mechanism is proposed to explain the inflammatory cystitis produced by CYP. Acrolein, its major metabolite, seems responsible for a large part of the bladder toxicity observed during CYP treatment (Phillips et al., 1961; Cox, 1979; Fraiser et al., 1991). The parent drug and/or its metabolites (acrolein) are concentrated in the urine, reaching the bladder where they would stimulate the PACSF, leading via antidromic stimulation to the release of neuropeptides, such as sP and neurokinin A, which would in turn activate NK₁ receptors (NK₂?), enhancing NO production and release. Increased vasodilation and vascular permeability, together with a possible direct irritant stimulus (acrolein), should lead to white blood cell and mast cell infiltration, leading to a local increase in the production of cytokines. The cytokines would further stimulate NO production through further induction of iNOS. High levels of NO may also sensitize the PACSF to further enhance neuropeptide release, leading to a positive feedback loop of inflammation. Blockade of NK₁ receptors or inhibition of NO synthesis would reduce NO production, and thus decrease inflammatory damage to the bladder. However, it is clear that additional undefined mechanisms are also involved in CYP-induced cystitis, because neither the NK₁ antagonist nor the NO synthesis inhibitor, either alone or in combination, were able to completely prevent the toxicity.