Research Paper Mediators of Inflammation, 10, 69–73 (2001)

BACKGROUND: Nitric oxide (NO) production is increased in inflammatory bowel disease (IBD), and measurement of NO metabolites may be useful for monitoring disease activity. AIMS AND OBJECTIVES: To characterise urinary nitrite levels, a stable metabolite of NO, in IBD and to evaluate its potential as a marker of disease activity. METHODS: Twelve-hour urinary nitrites were measured by the microplate assay method in 46 patients with IBD (active; n = 32). Urinary samples from 16 healthy individuals served as controls. RESULTS: Increased levels of urinary nitrites were found in patients with active IBD compared with those with inactive IBD. Twenty-eight out of 32 patients (87.5%) with active IBD had detectable levels of nitrite in their urine as compared with 2/14 (14.3%) patients with inactive IBD. None of the 16 healthy controls had detectable urinary nitrite. Twelve-hour urinary nitrite in active compared with inactive IBD: 5 0.7 versus 0.1+/-0.04 micromol (P < 0.05). There was good correlation between urinary nitrite and some markers of disease activity in IBD such as C-reactive protein and microalbuminuria but not with erythrocyte sedimentation rate. Conclusions: Increased levels of nitrite were detected in urine of patients with active IBD, consistent with increased NO synthesis. This simple assay may be exploited as a potential marker of disease activity in IBD.


Introduction
Nitric oxide (NO) is increasingly recognised to play a pivotal role in numerous physiological and pathophysiological processes since it was initially characterised as an endothelium-derived relaxing factor. [1][2][3] NO is formed from the conversion of L-arginine to citrulline via nitric oxide synthase (NOS), 4 which exists in several isoforms and can be found in numerous cell types including endothelial cells, neutrophils, macrophages, hepatocytes, enterocytes, neurons, and vascular smooth muscle. 5 -8 In the gastrointestinal tract, NO participates in neurotransmission, 9 smooth muscle relaxation, 10 intestinal secretion, maintenance of intestinal blood flow and as an inflammatory mediator. 10,11 Recent studies have demonstrated that NO production is increased in inflammatory bowel disease (IBD). 12 -17 Factors influencing a physiological or pathological outcome subsequent to increased NO production include the redox state of the molecule, its local concentration and its potential to form toxic intermediary metabolites. 2,7 NO is ultimately metabolised to stable end products that include nitrite and nitrate. 2,18,19 The relative amount of nitrite and nitrate produced by NO breakdown depends on the pH and redox state of the environment in which it is produced. 18,19 In experimental animals, urinary nitrite reflects systemic rather than renal production of NO. Nitrite has been used as a marker of NO production in culture fluids, plasma, synovial fluid and urine. 20,21 It has been found that plasma nitrite is increased in patients with rheumatoid arthritis compared with healthy controls and also that levels were significantly higher in the synovial fluid than in plasma. 21 Interleukin II immunotherapy causes a significant elevation in plasma nitrate levels. 22 Plasma nitrite levels have also been detected in other conditions such as sepsis, 23 cirrhosis 24 and fulminant hepatic failure, 25 and in the postoperative period. 26 Estimation of nitrite is a useful marker of NO synthesis and may potentially be used as an indicator of disease activity in chronic inflammatory conditions such as IBD.

Patients
Forty-six patients were enrolled for the study. Thirtytwo (69%) patients had active IBD at the time of study. Active IBD was defined as a Harvey-Bradshaw index (HBI) 27 > 3. Twenty-five patients had Crohn's disease (CD) and 21 patients had ulcerative colitis (UC) (pancolitis; n = 6). Sixteen healthy laboratory personnel served as controls. Thirty patients were receiving oral aminosalicyclates, four patients were receiving corticosteroids and 15 patients were receiving no medication at the time of urinary sampling.

Urinary nitrite estimation
Timed 12-h urine collections were performed on all patients. A microplate assay method was used to quantify urinary nitrite. An equal volume of the Griess reagent (0.1% sulphonide, 0.1% naphthylethylenediamine dihydrochloride, 2.5% phosphuric acid) was added to urine samples or to urine nitrite standards and incubated at room temperature for 10 min. Absorbance was performed by spectrophotometry at 540 nm. 27 The limit of detection of nitrite was 1 micromole per litre (1 m mol/l). Urinary nitrite concentration was determined and 12-h urinary nitrite production calculated.

Assessment of disease activity
Disease activity was measured using the HBI, erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) and microalbuminuria. CRP and microalbuminuria were measured by the immunoturbometric method (Behring, Germany) as previously described. 28

Statistics
The Wilcoxon rank sum test was used for nonparametric data and the non-paired Student t-test for parametric data. Correlations were determined using simple regression analysis. Statistics were performed using Statworks software.

Ethics
The study was approved by the Joint Ethics Committee of St. James's Hospital and Adelaid and Meath Hospital incorporating National Children's hospital Tallaght. Written informed consent was obtained from all study subjects prior to phlebotomy and urinary sampling.

Urinary nitrite in IBD
Twenty of 32 (87.5%) patients with active IBD had detectable urinary nitrite. In contrast, only two of 14 (14.2%) patients with inactive IBD had low levels of urinary nitrite. None of the 16 healthy controls had any nitrite in their urine. Twelve-hour urinary nitrite in active compared with inactive IBD was 5 ± 0.7 m mol versus 0.1 ± 0.04 m mol/12 h (P < 0.05) (Fig.  1). Urinary nitrite in controls was 0 ± 0 m mol (P < 0.05). There was no difference in urinary nitrite level between patients with CD and UC: 4.1 ± 0.8 m mol/ 12 h versus 2.2 ± 0.44 m mol/12 h (P = not significant (NS)). The urinary nitrite concentration was also significantly higher in patients with active IBD compared with inactive IBD (25.5 ± 6.8 m mol/l versus 0.15 ± 0.05 m mol/l, respectively; P < 0.05).

Relationship between urinary nitrite and disease activity markers
Patients with active IBD had significantly higher levels of microalbuminuria than those with inactive disease. There was a strong correlation between microalbuminuria and urinary nitrite in CD (r = 0.88, P <

Mediators of Inflammation · Vol 10 · 2001
FIG. 1. Timed 12-h urinary nitrite was assayed in patients with active and inactive inflammatory bowel disease (as defined by the Harvey-Bradshaw index (n = 32 and n = 14, respectively)). * P < 0.05 versus control, ** P < 0.05 versus active disease. 0.01) ( Table 1). In contrast, there was no correlation between microalbuminuria and urinary nitrite in UC (r < 0.1, P = NS) ( Table 1). There was a strong correlation between urinary nitrite in CD and serum CRP level (r = 0.66, P < 0.05). The correlation between urinary nitrite and serum CRP in UC did not reach statistical significance (r = 0.47, P = NS). The levels of CRP were higher in urinary nitrite-positive patients than in nitrite-negative patients (32.5 ± 4.9 m mol/l versus 5.7 ± 1.9 m mol/l; P < 0.05).
There was no significant correlation between levels of urinary nitrite and HBI. ESR was higher in the active IBD group compared with the inactive group (27.6 ± 20 versus 10.7 ± 3; P = 0.05). There was no difference in the ESR in the urinary nitrite-positive patients compared with the urinary nitrite-negative patients (22.8 ± 15 mm versus 23 ± 2.5 mm in the first hour; P = NS).

Discussion
This study demonstrates that urinary nitrite was detectable in IBD and higher levels were observed during active disease. These data support previous findings of increased NO generation in active IBD. 18,19 Previous investigations have demonstrated increased plasma nitrite/nitrate levels and increased mucosal citrulline production in active IBD. Elevated urinary nitrate and nitrite were not exclusive to IBD but have been observed in other inflammatory conditions of the bowel including infectious and radiation enterocolitis. 18 In our study, urinary nitrite levels appeared to correlate with laboratory indices of disease activity in Crohn's disease but not in ulcerative colitis. Evidence for increased NO production has also been demonstrated in experimental models of IBD. Peroxynitrite-induced colitis in experimental models can be ameliorated by inducible NOS (iNOS) inhibitors. 7,13,29,30 It has also been shown that NOS activity was increased in the colonic mucosa of patients with IBD, 15,17 and that increased nitrite levels were detected in rectal diasylate of patients with active IBD, especially in active UC. 14,31 But, in the case of CD, there is conflicting evidence: low colonic NOS activity in one study, 17 and higher NOS activity in the others. 13,15,32,33 The origin of urinary nitrite in IBD is unknown. Our observation of similar levels of urinary nitrite in UC and CD patients despite reported differences in colonic levels of NOS may suggest that systemic production of NO may be a major source of urinary nitrite in IBD. A potential source of NO production is the peripheral blood macrophages, following iNOS induction in response to pro-inflammatory cytokines and endotoxins. Furthermore, bacterial translocation  may be another source of urinary nitrite. Endotoxaemia is a recognised feature of IBD, and bacterial translocation has been associated with increased urinary nitrate production in animal studies. 34 Another source of NO may be the intestinal epithelial cells, which have been shown to possess NO activity in animal studies. 35 We have shown that the levels of urinary nitrite correlate with microalbuminuria and CRP in CD. It is possible that renal production of NO following proinflammatory stimuli may contribute to the urinary nitrite detected in these subjects. A renal source of urinary nitrite has also been recently suggested in patients with urinary tract infection. 36 Previous studies in experimental models have demonstrated that, although inflamed glomeruli are capable of producing nitrite, urinary nitrite in experimental glomerulonephritis reflects systemic NO formation as a result of immune activation rather than glomerular production of NO. 37,38 This suggests that, although conversion of nitrite to nitrate is rapid in vivo, at least some of the nitrite produced systemically as a result of NO production is cleared by the kidneys before it is converted to nitrate. NO may contribute to the pathogenesis of IBD through its pro-inflammatory effects (local cell injury, intestinal hyperaemia), its effects on neurotransmission, fluid secretion and its ability to induce smooth muscle relaxation.
There are limitations to the application of nitrite measurement as a marker of NO production and routine laboratory marker of disease activity in IBD. The additional measurement of urinary nitrate, the other stable end product of NO metabolism, may provide further information regarding in vivo NO status. 39 It is conceivable that dietary arginine or nitrite intake could have influenced urinary nitrite levels among this cohort. Nevertheless, urinary nitrite was not present in healthy individuals at levels within the detection limits of our assay.
In conclusion, patients with active IBD have elevated levels of urinary nitrite, a stable end product of NO. Measurement of urinary nitrite may serve as a useful marker for monitoring disease activity in IBD.