Dual involvements of cyclooxygenase and nitric oxide synthase expressions in ketamine-induced ulcerative cystitis in rat bladder


  • Karl-Erik Andersson led the peer-review process as the Associate Editor responsible for the paper.
  • Shu-Mien Chuang and Keh-Min Liu contributed equally to this manuscript.
  • Conflict of interest: none.



The aims of the present study were to investigate voiding patterns, tissue constituents and the expressions of cyclooxygenase-2 (COX-2) and nitric oxide synthase (NOS) involved in ketamine-induced ulcerative cystitis in rat urinary bladder.


Thirty Sprague–Dawley rats were distributed into three groups which received saline or ketamine (25 mg/kg/day) for a period of 14 and 28 days. In each group, cystometry was performed weekly and the concentration of ketamine and its metabolites (norketamine) was assayed. Paraffin-embedded sections were stained with Masson's trichrome stain, and ketamine-induced morphological changes were examined. Western blot analyses were carried out to examine the expressions of COX-2 and different NOS isoforms in bladder tissues. Immunofluorescence study was done to evaluate the expressions of COX-2 and macrophage infiltration (stained with ED-1 macrophage cell surface antigen) within the bladder.


Ketamine treatment resulted in bladder hyperactivity and the non-voiding contractions were significantly increased. The urine concentrations of ketamine and norketamine were much higher in ketamine-treated group. Moreover, ulcerated urothelium and mononuclear cell infiltration were noted in ketamine-treated group. These alterations in urodynamic functions and tissue constituents were accompanied by increases in the expression of COX-2. Two NOS isoforms (iNOS and eNOS) were also overexpressed, but no significant change was observed for nNOS. COX-2 positive stained cells were significantly increased. Meanwhile, increased amounts of ED-1 positive stained macrophages were present and most of COX-2 expressed cells were co-stained with ED-1 in the early stage of ketamine treatment.


Ketamine treatment affected bladder tissues by enhancing interstitial fibrosis and accelerating macrophages infiltration. Ketamine also initiated the up-regulations of COX-2 and iNOS and eNOS expressions. These up-regulated enzymes might play an important role in contributing to ketamine-induced alterations in micturition patterns and ulcerative cystitis. Neurourol. Urodynam. 32:1137–1143, 2013. © 2013 Wiley Periodicals, Inc.


Ketamine is a non-competitive N-methyl-D-aspartic acid (NMDA) receptor antagonist used as an anesthetic drug in human and veterinary procedures.[1] Ketamine is also used for pediatric anesthesia and conscious sedation in asthmatics.[1] Recently, it has been used as a recreational and dissociative drug, especially in nightclubs and at dance parties. Nowadays, increasing numbers of ketamine abusers have been found with severe lower urinary tract syndromes and ulcerative cystitis.[2, 3] Previous studies have shown that ketamine addiction affects lower urinary tract function, resulting in an increase in urinary frequency, nocturia, urgency, suprapubic discomfort, and at times hematuria.[4, 5] In other ketamine addition cases, vesicoureteral reflux, ureteral obstruction, and hydronephrosis were also reported. These studies demonstrated that ketamine-induced inflammatory changes in the urinary bladder were similar to interstitial cystitis.[3-5]

Despite increasing reports of ketamine abuse and ulcerative cystitis cases, the question concerning how ketamine addiction produces bladder dysfunction symptoms is still poorly understood. Most of these findings primarily relied on clinical observations and radiological images. A recent study using ketamine cystitis mouse model showed that there were enhanced non-cholinergic contractions and P2X1 receptor expression in the ketamine bladder, indicating that dysregulation of purinergic neurotransmission might underlie ketamine-induced detrusor overactivity.[6] However, further research is needed to elucidate the initiation, the location, and underlying factors involving bladder dysfunctions caused by ketamine.

Cyclooxygenase (COX) is an enzyme that is responsible for the formation of prostanoids (PGs), which are important biological mediators. Two isoforms of COX have been identified. COX-1 is constitutively expressed in most tissues where it synthesizes PGs at low levels to maintain physiological functions. On the other hand, cyclooxygenase-2 (COX-2) is inducible in response to inflammatory stimuli, cytokines, and mitogens, resulting in exaggerated PGs release. In an animal study with chemical-induced hemorrhagic cystitis, it was found that the expression of COX-2 was increased in the bladder,[7] suggesting that COX-2 might play an important role in bladder inflammation.

Nitric oxide (NO) is a free-radical gas that regulates several physiological processes, including vascular tone, polymorphonuclear leukocyte adhesion, and inflammation. Nitric oxide synthase (NOS) is an enzyme that catalyzes the production of NO from L-arginine. Three NOS isoforms (nNOS, eNOS, and iNOS) have been identified. The neuronal NOS (nNOS) and the endothelial NOS (eNOS) are expressed constitutively, while the inducible NOS (iNOS) is evident under pathological conditions. iNOS could be induced by cytokines released from macrophages, fibroblasts, and epithelial cells. Induction of iNOS was found to result in vast amounts of NO production in thymus tissue.[8] Animal model studies have shown that the over-production of NO is toxic to tissues and selective iNOS inhibition decreases inflammatory events.[9]

The critical role of COX-2 in cyclophosphamide-induced overactive bladder and the expression of NOS isoforms within the urothelium and the presence of NO in the micturition reflex pathway have been reported.[10] Other studies also revealed interactions between NO and COX-2.[11] NO has both inhibitory and stimulatory effects on COX-2 expressions.[11] During the inflammatory processes, large amounts of proinflammatory mediators, NO, and prostaglandin E2 (PGE2), are generated. In macrophage cells, NO stimulates the expression of COX-2, but the inhibition of COX-2 by NO is observed in cultured endothelial cells.[12] Studies have indicated that NO may directly interact with COX-2 at the transcriptional level.[13] Despite these advances, investigations into the dual involvements of COX-2 and NO in ketamine-induced cystitis are still lacking.

The major aim of our present study was to establish a novel ketamine cystitis animal model in rat and to investigate urodynamic function and bladder morphology after ketamine treatment. We also focused on the expressions of COX-2 and three NOS isoforms in the bladder after ketamine treatments. By elucidating potential factors underlying ketamine effects, the results obtained could provide valuable new insights leading to a better understanding of ketamine-induced cystitis in rat bladder.


Animals and Ketamine Administration

Experiments were performed on 30 adult female Sprague–Dawley (S-D) rats (animal center of BioLASCO, Taipei, Taiwan) weighing between 200 and 250 g. These rats were housed under a 12-hr light/dark cycle with free access to food and water at 21°C. Thirty S-D rats were distributed into three groups which received saline (0.5 ml, control group) or ketamine (25 mg/kg/day diluted in 0.5 ml saline) intraperitoneal (IP) injections daily for a period of 14 days (K-14D group) and 28 days (K-28D group), respectively. Rats were weighed once at the beginning of every week for adjustment of the amount of ketamine administrated. This study was approved by the Animal Care and Treatment Committee of Kaohsiung Medical University. All experiments were conducted according to the guidelines for laboratory animal care. All efforts were made to minimize animal stress/distress.

Cystometrogram (CMG) and Data Analysis

Cystometry was performed on the second day after saline or ketamine injection before euthanizing the animal. The duration between CMG experiment and ketamine/saline administration was 24 hr. The CMGs were carried out according to the method previously described.[14] In brief, in each experiment, rats were anesthetized with Zoletil-50 (1 mg/kg IP injection). Before the beginning of each CMG, the bladder was emptied, a urethral catheter (PE50 tube) was indwelled and used to fill the bladder and to measure bladder pressure. Then the bladder was infused with saline at a steady rate (0.08 ml/min), during which the pressure was measured in-line with the catheter. A voiding contraction was defined as an increase in bladder pressure that resulted in urine loss. CMG was recorded until the bladder pressure was stable and at least 5 filling/voiding cycles were measured on each rat. Pressure and force signals were amplified (ML866 PowerLab, ADI instrument, Colorado Springs, CO), recorded on a chart recorder, and digitized for computer data collection (Labchart 7, ADI Instruments: Windows 7 system). CMG parameters recorded for each animal included threshold pressure, peak micturition pressure, bladder capacity, and the frequency of non-voiding contractions (without urine leakage during bladder infusion). Threshold pressure was the intravesical pressure right before the initiation of micturition. Peak micturition pressure was the maximum pressure during micturition as observed in CMG. Bladder capacity was measured as the amount of saline infused into the bladder at the time when micturition commenced.[15] Bladder compliance was measured by infused volume (ml)/threshold pressure (ΔcmH2O).

Ketamine Metabolites Assay in Urine and Serum

At the end of CMG and termination of each experiment, 1 ml of blood was obtained from the rat's tail for analyzing ketamine and norketamine. Blood was separated by centrifugation at 4°C. The concentration of ketamine and norketamine in urine and serum was determined by using a slightly modified version of the high-performance liquid chromatography (HPLC) method.[2] The urine was collected by the metabolic cage on the day before euthanizing the animal. After extraction and purification by liquid–liquid extraction using ethyl ether, urine samples underwent chromatography on a reversed-phase column and ketamine and norketamine were detected at 200 nm by UV spectrophotometry. This study was carried out according to ISO 9001:2000 requirements.

Histological Study and Collagen/Detrusor Smooth Muscle (DSM) Ratio

After cystometric studies, experimental rats were perfused with a saline solution through the left ventricle. The bladders were then removed, dissecting blood vessels and fat tissue from the bladder, cut open and further fixed overnight. Part of the bladder was dissected and frozen in liquid nitrogen for further molecular experiments. The tissue samples were embedded in paraffin blocks with the same area in different groups, and serial sections of 5-µm thick were obtained. Deparaffinized sections were stained with Masson's trichrome stain (DAKO, Glostrup, Denmark, Masson's trichrome Stain Kit). The standard Masson's trichrome staining procedure was followed. Each specimen was captured using a digital camera in five random non-overlapping frames at 400× magnification. The color setting and the image-associated quantification were determined using image analysis software (Image-Pro Plus, Media Cybernetics, MD). The blue-stained collagen and the red-counterstained DSM were highlighted in each image. The total areas occupied by collagen and DSM were then determined, and the ratios of collagen to DSM (collagen/DSM) were calculated.

Protein Isolation and Western Blot Analysis for COX-2 and NOS

Frozen tissues of bladder from each group were homogenized on ice in the buffer containing the Halt Protease Inhibitor Cocktail (Pierce, Rockford, IL) at 100 mg/ml. Protein concentration in the supernatant was determined using the bicinchoninic acid protein assay (Pierce) against a bovine serum albumin protein standard. An equal amount of protein (30 µg) from the bladders was loaded on sodium dodecyl sulfate polyacrylamide gel electrophoresis gels and transferred to polyvinylidene fluoride (Immobilon-P, Millipore, Billerica, MA) membranes with Towbin buffer. Immobilon-P membranes were then blocked and were then incubated with the primary antibody. Monoclonal antibodies to COX-2 (1:1,000, Cayman, Ann Arbor, MI), iNOS (1:1,000, Millipore, Billerica, MA), eNOS (1:1,000, NeoMarker, Kalamazoo, MI), nNOS (1:1,000, Millipore, Billerica, MA) were used to determine the changes in the expressions of COX-2 and three NOS isoforms. Band intensity was normalized with respect to β-actin (1:5,000, Upstate, Billerica, MA, Signaling). In each experiment, negative controls without the primary antibody and protein molecular weight markers were carried out to exclude non-specific bands. After treatment with the primary antibody, membranes were washed with buffer and incubated with the secondary antibody (goat anti-mouse IgG). The expression of the protein band was visualized by adding ECL-Plus (Amersham Pharmacia Biotech, Buckinghamhire, England) for 1–2 min. The membrane was sealed in a hybridization bag, scanned, and analyzed with a Kodak Image Station 440CF and Kodak ID image analysis software.

Immunofluorescence Studies for the COX-2 and ED-1 Expressions Within the Bladder Smooth Muscle

The blocking of deparaffinized sections was done with 10% normal goat serum (Invitrogen, Grand ISland, NY) for 30 min. Tissue sections were then stained overnight with the rabbit monoclonal IgG antibody to COX-2 (Epitomics, Burlingame, CA, 1:100) and mouse monoclonal IgG antibody to ED-1 (CD68) (Serotec, Oxford, UK, 1:100). The monoclonal antibody ED1 was used as a marker for rat macrophages. The antigen was expressed on the membranes of cytoplasmic granules as well as on the cell surface.[16] Studies have shown that the antigen recognized by ED1 was the rat homologue of human CD68. Detection was achieved using a fluorescence-conjugated secondary antibodies (Invitrogen) at a dilution of 1:1,000. For each sample, the numbers of cells immunostained with ED-1 and COX-2 were counted in 10 high power fields of 400× magnifications. Figures and images were assembled using Adobe Photoshop.

Statistical Analysis

Analysis of variance followed by the Bonferroni test and two-way analysis of variance for individual comparison was carried out for the above experiments. Calculations of mean, standard deviation (SD), and P values were performed on triplicate experiments. The Student's t-test was used to calculate P-values for comparison. The significant statistics was set at a P-value < 0.05.


The Effects of Ketamine Treatment on Cystometric Parameters

CMG parameters recorded micturition pressure, micturition frequency, voiding, and non-voiding contractions were shown in Figure 1. Voiding contractions (arrows) were significantly increased in the K-28D group compared with the control (Fig. 1A). Non-voiding contractions (arrows) between micturitions were also considerably increased, especially after 28 days of ketamine injection (3.4 ± 0.4 vs. 0, in comparison with the control, P < 0.01). Figure 1B demonstrated that bladder threshold pressure and maximal micturition pressure were significantly increased by ketamine, especially following 28 days of ketamine treatment. Moreover, the bladder capacity was significantly decreased in K-28D group (P < 0.01 in comparison with the control; Fig. 1C). In Figure 1D, ketamine appreciably increased intravesical threshold pressure, but decreased bladder capacity, implying that ketamine significantly decreased the bladder compliance in the K-14D and K-28D groups (P < 0.05 and <0.01 in comparison with the control, respectively). These results demonstrated that ketamine treatment caused bladder hyperactivity by altering maximal micturition pressure, increased micturition frequency, enhanced bladder non-voiding contraction, and lessened bladder capacity.

Figure 1.

Effects of ketamine treatment on cystometric parameters. A: Cystometry recordings illustrated micturition pressure and voiding frequency, including voiding contractions (arrows) and non-voiding contractions (stars) on anesthetized rats. B: Threshold pressure and peak micturition pressure, (C) bladder capacity, and (D) bladder compliance as calculated for the control, K-14D, and K-28D groups. Each bar was the mean ± standard deviation (SD) of the mean for n = 10. * and ** indicated significant difference from the control group (P < 0.05 and 0.01 respectively). Pdet max = the maximal micturition pressure.

The Concentration of Ketamine and Its Metabolites in the Urine and Serum

The concentrations of ketamine and norketamine (ketamine metabolite) in serum and urine were investigated by HPLC. Serum ketamine concentrations were undetectable in the control, K-14D and K-28D groups, revealing the short-acting character of ketamine. Whereas serum norketamine concentration was undetectable in the control, 4.6 ± 0.67 ng/ml in the K-14D group and 5.8 ± 0.63 ng/ml in the K-28D group. In contrast, the concentrations of ketamine were much higher in urine in both the K-14D and the K-28D groups (620 ± 86.3 and 868 ± 95.9 ng/ml, P < 0.01 compared with the control). The urine concentration of the norketamine in the K-14D and the K-28D groups were 6,184 ± 730.6 and 10,680 ± 1531.6 ng/ml, respectively. These results suggested that the toxic effects of ketamine might come from both ketamine and its metabolites in urine (Table I).

Table I. The Concentration of Ketamine and Norketamine at Different Groups in the Urine and Serum
Ketamine (ng/ml)Norketamine (ng/ml)Ketamine (ng/ml)Norketamine (ng/ml)
  1. ND, not detected.
  2. Values are mean ± SEM.
  3. * and ** indicate P < 0.05 and P < 0.01 versus control, respectively.
K14DND4.6 ± 0.67*620 ± 86.3**6,184 ± 730.6**
K28DND5.8 ± 0.63*868 ± 95.9**10,680 ± 1531.6**

The Histological Features of Ketamine-Associated Bladder Damage

The histological features of ketamine-associated bladder damage were analyzed by Masson's trichrome stain. In the control group (Fig. 2A and A′), there were three to five layers of urothelium (black arrowhead) and only sparse collagen (black arrow) distributed between the smooth muscle bundles (Fig. 2A′). In the K-14D group (Fig. 2B and B′), the bladder tissues were characterized by mucosal sloughing and denuded urothelial mucosa with a thinner layer of epithelial cells (black arrowhead), red blood cell (RBC) debris (star) under suburothelium, and increased connective tissue elements (black arrow; Fig. 2B′). In the K-28D group (Fig. 2C and C′), the ulcerated mucosa and the erythematous patches in lamina propria with mononuclear cells infiltration (yellow arrowhead) were shown prominently. The deep layer of lamina propria stained in dark blue (black arrow) was dense and fibrosis noted between the DSM bundles in the K-28D group, indicating an increase in interstitial fibrosis. In Figure 2D, the ratio of collagen to DSM showed that ketamine injection for a period of 14 and 28 days significantly increased the severity of interstitial fibrosis (P < 0.01 in comparison with the control).

Figure 2.

Representative microphotographs of (A) the control group, (B) K-14D group, and (C) K-28D group as shown by Masson's trichrome staining. The upper panels (A, B, and C) showed the microphotographs taken at 100× magnification, while the lower panels (A′, B′, and C′) were captured at 400× magnification. In the control group (A′), there were 3-5 layers of urothelium (black arrowhead) and only sparse collagen (black arrow) distributing between the smooth muscle bundles. The K-14D group (B′) contained denuded urothelial mucosa with a thinner layer of epithelial cells (black arrowhead), red blood cell debris under suburothelium (black star), and increased connective tissue fibrosis (black arrow). In K-28D group (C′), prominent mononuclear cells infiltration (white arrows) were shown. The deeper layer of lamina propria was dense and fibrotic (yellow arrowhead). (D) The ratio of collagen to DSM in different groups. Scale bar represented as 200 µm. Each bar was the mean ± SD of the mean for n = 10. ** indicated significantly different from the sham group, P < 0.01.

The Expressions of COX-2 Following Ketamine Injection

To provide valuable insights into important enzyme functions underlying the bladder destruction after ketamine administration, the expression of COX-2 was examined in Figure 3. Western blots indicated that there was a significant difference in COX-2 expression among the control, K-14D, and K-28D groups. The expression of COX-2 was increased by 1.9-fold in the K-14D group, and 9.0-fold in the K-28D group, as compared with the control (Fig. 3).

Figure 3.

Effects of ketamine treatment on the expression of COX-2. A: Representative western blots and the average enzyme expression as measured by antibody specific to COX-2. B: Quantification of the ratio of COX-2 to β-actin in the control, K-14D, and K-28D groups. Results were normalized as the control = 100%. * and ** indicated significant difference from the control group, (P < 0.05 and 0.01, respectively). Each bar was the mean ± SD of the mean for n= 10.

Immunohistochemical Analysis of the COX-2 and ED-1 Expressions

The COX-2 (pro-inflammatory marker) and ED-1 (macrophage biomarker) positive-stained cells were observed mainly between the DSM bundles. The results revealed that 6.0 ± 1.3 and 22 ± 2.9% of DAPI positive cells were co-labeled with COX-2 in the K-14D group and the K-28D group (Fig. 4A and B). Moreover, 7.0 ± 1.5 and 32 ± 3.7% of DAPI positive cells were co-labeled with ED-1 in the K-14D group and the K-28D group, respectively (Fig. 4A and B). The macrophages were significantly increased after ketamine treatment, especially in the K-28D group.

Figure 4.

A: Representative immunohistochemical microphotograph of COX-2 expressing cells double-labeled with macrophages marker (ED-1) and DAPI. B: Quantitative analysis of DAPI-positive cells co-labeled with COX-2 or ED-1 between smooth muscle bundles. The arrow indicated COX-2/ED-1 co-stained macrophage and the arrowhead referred to only ED-1 positive macrophage. C: Quantitative analyses of COX-2-expressing cells double-labeled with macrophages (ED-1positive cells) in different groups under 400× high power field. ** indicated significant difference from the control group for P < 0.01. #indicated significant difference between groups for P < 0.05. Each bar was the mean ± SD of the mean for n = 10. Scale bar represented the unit of 50 µm.

Experiments involving double-labeled with COX-2 with ED-1 were also performed. In the control, the amount of ED-1 and COX-2 co-staining was very low. In contrast, the percentage of ED-1 positive cells that was double-labeled with COX-2 positive was 97 ± 5.5 and 80 ± 8.0% in the K-14D group and K-28D group, respectively (Fig. 4C). In the K-14D group, most of COX-2 expression cells were co-stained with ED-1 (Fig. 4A and C), suggesting that the COX-2 was synthesized by macrophages in the early stage of inflammation. However, at the late stage, some of macrophages were not co-stained with COX-2 (Fig. 4C).

The Expressions of Different NOS Isoforms Following Ketamine Treatment

Figure 5A and B showed that the expression of iNOS protein was increased by 3.9-fold in the K-14D group, and 8.0-fold in the K-28D group as compared with the control. Similarly, the expression of eNOS protein was increased by 3.0-fold in 14 days and 9.8-fold in 28 days following ketamine administration, respectively (Fig. 5C). In contrast, there was no significant difference in the expression of nNOS between the control and the ketamine-treated rats (Fig. 5D).

Figure 5.

Effects of ketamine treatment on the expressions of different isoforms of nitric oxide synthases (NOS). A: Representative Western blots and the average expressions as measured by antibody specific to iNOS, eNOS, and nNOS in different groups. BD: Illustrated the quantification of the ratio of iNOS, eNOS, or nNOS to β-actin, respectively. Results were normalized as the control = 100%. ** indicated a significant difference from the control group, P < 0.01. Each bar was the mean ± SD of the mean for n= 10.


Ketamine treatment significantly increased micturition frequency and non-voiding contractions, while it decreased bladder capacity. Ketamine also enhanced bladder interstitial fibrosis, submucosal hemorrhage, and macrophages infiltration, but decreased urothelium thickness. At the protein level, ketamine induced over-expressions of iNOS, eNOS, and COX-2. In the early stage of ketamine cystitis, immunofluorescence studies revealed that COX-2 was mainly synthesized by macrophage. However, at the later phase of ketamine-induced inflammation, macrophages might produce other cytokines and were not completely co-stained with COX-2.

COX-2 enzyme is an inflammatory early response gene, which becomes active in response to the presence of pro-inflammatory cytokines and growth factors. It has been suggested that COX-2 is the key mediator following the induction of cyclophosphamide-induced cystitis.[7] A number of studies have demonstrated the role of COX-2 in bladder overactivity associated with inflammation and/or hypertrophy.[7, 10] An important role of COX-2 in altering micturition patterns has been reported. Although COX-2 has little or no direct pro-inflammatory effects, the products of COX-2, prostaglandins, had strong inflammatory effects. In sensory neuron culture and chemical-induced cystitis, PG-E2 could induce sensitization of sensory neurons, revealing an important role of COX-2 and PGs in altered micturition patterns. Treatment with COX-2 inhibitor was found to ameliorate ifosfamide-induced bladder damages and alter micturition overactivity.[17] In the present results, COX-2 expression increased to ninefold of the control level after 28 days ketamine treatment. This prominent overexpression of COX-2 might suggest that COX-2 or its downstream product (PG) have an important effect on altering micturition patterns and bladder interstitial fibrosis in ketamine-induced cystitis.

Macrophages play a central role in inflammation induction and regulation of the immune response. There was an increase in macrophages infiltration of the bladder tissues after ketamine treatment. The present study also showed that COX-2 expression was observed mainly within the bundles of DSM in rats treated with ketamine. In the early stage of inflammation, the percentages of ED-1 positive cells double-labeled with COX-2 were as high as 97%, showing that COX-2 was synthesized mainly by macrophages. However, at the late stage of ketamine-induced inflammation, macrophages were not completely co-stained with COX-2, demonstrating that activated macrophages might produce other cytokines. Further experiments in the future is required in order to understand the relevance of COX-2 overexpression in relation to macrophage activation.

The present results of up-regulating iNOS and eNOS levels after treatment with ketamine also suggested that there might be an important role of these NOS isoforms in bladder damage. iNOS is inducible and involves the synthesis of NO in a great amount for a longer period.[18] It is generally recognized that iNOS plays a detrimental role in sepsis-induced organ failure and inflammation. iNOS was up-regulated in cyclophosphamide or ifosfamide induced cystitis.[19] eNOS is a constitutive form of NOS. eNOS-produced NO plays a role in inhibiting platelet aggregation and leukocyte adhesion, and is released for a short period in response to stimulation.[20] Therefore, eNOS is critical in the survival against I/R injury. In LPS-induced bladder inflammation, eNOS was up-regulated in the early phase of inflammation.[21] Regarding the location of eNOS, in cyclophosphamide-induced cystitis, immunohistochemistry studies showed that eNOS was up-regulated in the urothelium and suburothelial layer.[22] The present results also demonstrated that ketamine induced significant submucosal hemorrhage, enhanced macrophages infiltration, but decreased urothelium thickness. These findings suggested that the urothelium might play a role in the pathogenesis of ketamine-induced cystitis. Up-regulated eNOS expression after ketamine injection might have a protective effect on regulating bladder microcirculation.

nNOS is another constitutive form of NOS. This constitutively expressed NOS is thought to mediate synaptic plasticity, neuronal signaling, and regulate vascular tone.[20] Generally, nNOS is up-regulated following neuronal injury, inflammation and sepsis. However, in contrast to iNOS and eNOS, the present results revealed that the expression of nNOS showed little change after ketamine injection for a period of 14 and 28 days. It was speculated that different NOS isoforms might be expressed at different time points during the inflammatory process. Inhibition of nNOS was found to exhibit anti-inflammatory effects when the inhibitor is administered during the early inflammatory stage. On the contrary, selective inhibition of iNOS by aminoguanidine was reported to exhibit anti-inflammatory effects only at a late stage of the inflammatory process.[23]

The specific study aim of this experiment is to determine the effects of ketamine addition on voiding patterns and at which changes can occur on COX-2 and different NOS in rat bladder after ketamine treatment. As previous study showed, inflammatory mediators such as TNF-α and IL-1β could induce both the expressions of both COX-2 and iNOS.[24] In pancreatic β cells, NOS inhibitor was found to completely inhibit IL-1β-induced NO formation and attenuate PGE2 production.[25] NOS inhibitor could also inhibit IL-1β-induced promoter activity, gene transcription and protein expression of COX-2.[25] Our results showed there were significant changes in COX-2, iNOS, and eNOS expression in the bladder tissue. These results suggested that possibly there is a crosstalk between the NO and COX enzymes. However, the specific localization of these NOS expressions after ketamine injection was still unclear. Further studies such as using immunofluorescence technique to localize the expressions of different NO synthases after ketamine treatment are needed. In addition, the elucidation of the changes in the level of NO and the utilization of NOS and/or COX inhibitor may have potential therapeutic value for ameliorating bladder dysfunction in the ketamine-abused population. Future research is needed to investigate the effects of specific NOS inhibitors and to elucidate their alterations of NO following ketamine addiction.


Ketamine treatment caused bladder hyperactivity by altering micturition pressure, rising micturition frequency, increasing non-voiding contraction, and decreasing bladder capacity. Moreover, ketamine enhanced bladder interstitial fibrosis, decreased urothelium thickness, augmented submucosal hemorrhage, and accelerated macrophages infiltration. Long-term ketamine administration-induced over-expressions of iNOS, eNOS and COX-2 at protein levels. These enzyme up-regulations might play an important role in contributing to ketamine-induced alterations in micturition patterns and ulcerative cystitis. Future investigations involving COX-2 or NOS inhibitors may provide valuable information leading to an understanding of clinical significance in treating ketamine-induced cystitis.


We are grateful to Professor Chang-Hwei Chen of the University at Albany for his valuable comments on this manuscript. This research is supported in part by the Department of Medical Research, Kaohsiung Medical University Hospital grant (KMUH-100-0R44), in part by the Kaohsiung Medical University grant (KMU-Q098015) and Kaohsiung Municipal Hsiao-Kang Hospital grant (KNHK-100-025).