点击显示 收起
1 From the Division of Gastroenterology and Nutrition Research, Department of Medicine, Toronto General Hospital, University Health Network, and the Division of Gastroenterology, Mount Sinai Hospital, University of Toronto.
2 Presented in part at the DDW/American Gastroenterology Association Meeting in Orlando, FL, 1999, and at the DDW/American Gastroenterology Association Meeting in San Diego, 2000. 3 Supported by the Crohn's and Colitis Foundation of Canada. 4 Reprints not available. Address correspondence to JP Allard, Division of Gastroenterology, Department of Medicine, Toronto General Hospital, UHN Room 9 EN-217A, 200 Elizabeth Street, Toronto, Ontario, Canada, M5G-2C4. E-mail: johane.allard{at}utoronto.ca.
ABSTRACT
Background: In Crohn disease (CD), the increased production of reactive oxygen species from activated neutrophils may reduce plasma concentrations of antioxidant vitamins and result in increased oxidative stress.
Objective: We compared lipid peroxidation, a measure of reactive-oxygen-species production, and plasma antioxidant vitamin concentrations between CD patients and healthy control subjects.
Design: Thirty-seven nonsmoking CD patients (22 women and 15 men) were compared with an equal number of healthy control subjects who were matched by age, sex, and body mass index. In patients the mean CD activity index (CDAI) was 141.2 ± 18.7 (range: 9.0–514), and 11 of 37 patients (30%) had a CDAI 150. Seventy-eight percent of patients were taking 1 medication. Medication use by subjects included the following: 5-aminosalicylic acid (40% of subjects), antibiotics (22%), oral corticosteroids (30%), and immunosuppressants (19%).
Results: Lipid peroxidation as measured by breath pentane output (CD patients, 7.47 ± 0.98 pmol•kg-1•min-1; control subjects, 4.97 ± 0.48 pmol•kg-1•min-1; P 0.025), breath ethane output (CD patients, 11.24 ± 1.17 pmol•kg-1•min-1; control subjects, 5.46 ± 0.71 pmol•kg-1•min-1; P 0.0005) and F2-isoprostane (CD patients, 78.6 ± 8.0 ng/L; control subjects, 60.6 ± 3.7 ng/L; P 0.047) were significantly higher in CD patients than in control subjects. Plasma antioxidant vitamins (ascorbic acid, - and ß-carotene, lycopene, and ß-cryptoxanthin) were all significantly lower in CD patients than in control subjects. There were no significant differences in macro- and micronutrient intakes between groups.
Conclusion: Patients with CD are oxidatively stressed, which was observed even though 70% of patients had a CDAI 150 and 78% of them were taking medications to treat CD.
Key Words: Oxidative stress • Crohn disease • micronutrients • antioxidants • lipid peroxidation • inflammatory bowel disease • reactive oxygen species
INTRODUCTION
During Crohn disease (CD), a cascade of immunologic events occurs at the intestinal level. Increasing attention has focused on the role of reactive oxygen species (ROS) produced by activated neutrophils during the inflammatory response (1). These ROS molecules are highly reactive and can attack almost every cell component, causing further damage to surrounding tissue in a nonspecific manner. Several reports have suggested that patients with CD have increased production of ROS (2–4), and direct evidence of epithelial cell injury secondary to ROS has been detected (5). ROS can attack double bonds in polyunsaturated fatty acids (PUFA), thus inducing lipid peroxidation (6), which may result in more oxidative damage (7, 8).
Because ROS are produced during normal aerobic metabolism and because their production can be increased dramatically during inflammation, cells and tissues have developed a sophisticated antioxidant defense system that includes adequate amounts of antioxidant enzymes and vitamins that will decompose these potentially injurious oxidizing agents (8). During chronic inflammation when sustained production of ROS occurs, antioxidant defenses can weaken, resulting in a situation termed oxidative stress (9). The small and large bowel have the enzymatic milieu necessary to produce high concentrations of ROS from resident phagocytic cells, microvascular endothelial cells, and mucosal epithelial cells (9). It was reported that the onset of CD results in a reduction in antioxidant enzymes (10) and micronutrients in the intestinal mucosa (11) and plasma (12, 13). The literature suggests that patients with CD are at risk of increased oxidative stress from lipid peroxidation and altered antioxidant defenses. However, these studies on ROS production or antioxidant micronutrients did not control well for other confounding factors such as smoking, dietary intake, alcohol consumption, and exercise (12–14). The purpose of the present study was to concurrently compare lipid peroxidation indexes and plasma concentrations of various antioxidants between a population of CD patients and healthy control subjects who were matched by age and sex in a well-controlled manner.
SUBJECTS AND METHODS
Thirty-seven patients (22 women and 15 men) diagnosed with CD were recruited from gastroenterology clinics in the Toronto area. An equivalent number of healthy control subjects were recruited from hospital workers and students and were matched to the patients by age, sex, and body mass index (BMI). Subjects were included in the study if they were nonsmokers, were not pregnant, had no known liver disease, or were not taking any micronutrient and herbal supplements.
CD patients were excluded from the study if there was any documentation by standard clinical methods of abdominal abscess, bowel obstruction, fibrostenotic disease, active gastrointestinal bleeding, or known malabsorption. Control subjects were excluded from the study if there was any evidence of an acute viral or bacterial infection, the presence of inflammatory disorders, or the use of antiinflammatory medications (eg, oral corticosteroids, aminosalicylic acid, or nonsteroidal antiinflammatory agents) at any time during the study. Because breath ethane and pentane outputs can be influenced by the amount of PUFA in the diet, all subjects were educated about a diet regimen that provided a ratio of polyunsaturated to saturated fatty acids (P:S) of 0.4:1. Subjects were instructed to follow the diet regimen for 2 wk before breath measurements. Detailed instructions on the diet regimen, methods of recording food intake, and a tool to estimate food portion size were given to all subjects. Food records were obtained to determine whether subjects complied with the dietary advice given on fat intake and to compare the nutrient intake of CD patients with that of control subjects. Previous studies published by our lab have used a similar control diet (15, 16).
The Ethics Committee of the University of Toronto approved the study protocol. All subjects gave their written, informed consent before participating in the study.
Clinical and nutritional evaluation
All subjects underwent initial screening, which included obtaining medical, surgical, and smoking, and diet histories and anthropometric (weight and height) measurements. At the second visit, breath ethane and pentane were measured and blood was drawn to assess plasma F2-isoprostane, ascorbic acid, -tocopherol, carotenes and carotenoids, glutathione peroxidase (GSHPx), and selenium.
The severity of CD was determined for each subject by use of the CD Activity Index (CDAI; 14). A CDAI <150 indicated remission of disease. The acute phase protein, 1-acid glycoprotein (orosomucoid), was used as another indicator of disease activity when measured as >1.4 g/L.
Plasma samples
Blood samples for plasma F2-isoprostane and the antioxidant vitamins were collected in tubes containing EDTA; samples for GSHPx and selenium were collected in tubes free of trace elements. Blood obtained for the measurement of lipid peroxidation and antioxidants was promptly centrifuged at 528 x g for 10 min at room temperature; the plasma was separated and stored at -70°C until analyzed to reduce the risk of further lipid peroxidation before assays for F2-isoprostane. Plasma for ascorbic acid assays was stabilized immediately with 100 g HPO3/L (0.75 mL plasma plus 0.75 mL HPO3) before deep freezing. The protocol followed was according to published methods or the manufacturer's instructions.
Analyses and measurements
Breath analysis was performed according to the method of Lemoyne et al (17). Subjects were required to breathe hydrocarbon-free air for 6 min to wash contaminating hydrocarbons from their lungs. During the last 2 min of the breath analysis, expired air was collected and subsequently analyzed by gas chromatography (Shimadzu 6-AM GC; Shimadzu Seisgkusho Ltd, Kyoto, Japan). Fifty milliliters of expired air was passed through a stainless steel loop packed with alumina and cooled to -95°C to adsorb the injected sample. The loop was then heated to desorb the gas thermally. Pentane and ethane were analyzed with a Porasil D column (Chromatographic Specialities Inc, Brockville, Canada) by use of a calibration curve derived from known concentrations of the gases. Concentrations of breath pentane and ethane were expressed in pmol•kg-1•min-1. Plasma F2-isoprostane was measured with the use of an enzyme-linked immunoassay method (Cayman Chemical Company, Ann Arbor, MI) (18).
The presence of GSHPx in plasma was measured by using an enzyme-linked immunoassay method (Bioxytech pl-GPx Enzyme Immunoassay; Oxis International Inc, Portland, OR) (19). Plasma selenium was measured by atomic absorption spectrophotometry, using a modification of the method by Pleban et al (20).
Retinol and other carotenoids, including ß-carotene, were analyzed by HPLC (21). A reversed-phase C18 column was used with an isocratic solvent system (methanol:acetonitrile:tetrahydrofuran, 50:45:5 by vol) after hexane extraction with 200 µL plasma. With the use of a Waters 490 Programmable Multiwavelength Detector (Waters, Milford, MA), retinoids were detected at 325 nm and carotenoids were detected at 450 nm. -Tocopherol was analyzed by isocratic reversed-phase HPLC and fluorescence spectrophotometry at 295 nm (22). Samples were analyzed for total biologically active vitamin C by spectrophotometry. The method was adapted from Bessey et al (23), using 2,4-dinitro-phenylhydrazine as the chromogen. The color density was read at 521 nm against a reagent blank and compared with a calibration curve of known standards. The 7-d food intake records completed by study participants were computer analyzed by use of the Nutrient Data System (NDS) 32, version 2.9 (Nutrition Coordinating Centre, University of Minnesota, Minneapolis).
Statistical analysis
The data were normally distributed; thus the unpaired 2-tailed t test was used to compare data between the CD population and the control subjects. In CD patients, Pearson correlations were also performed between various variables. Data are expressed as the mean value ± SEM. All statistical analyses were performed with the program SPSS 7.5 (SPSS Inc, Chicago). Statistical significance was defined as P 0.05.
RESULTS
Patient characteristics are described in Table 1. The mean CDAI was in the normal range. Seventy percent of the 37 patients had inactive disease as defined by a CDAI <150. Seventy-eight percent of patients were taking 1 medication.
View this table:
TABLE 1.. Characteristics of Crohn disease patients and healthy control subjects
Lipid peroxidation measured by breath ethane and pentane output and plasma F2-isoprostane was significantly higher in CD patients than in control subjects (Table 2). Plasma micronutrients are shown in Table 3. CD patients had significantly lower plasma concentrations of ascorbic acid, - and ß-carotene, and the carotenoids lycopene and ß-cryptoxanthin than did healthy control subjects. There was no significant difference in plasma -tocopherol, GSHPx, and selenium concentrations between groups. To rule out possible subclinical malabsorption secondary to small-bowel resection, all indexes were compared between CD patients with and without resection (Table 4). No significant difference in BMI, lipid peroxidation indexes, or plasma antioxidant concentration was noted between these 2 CD groups, except for ß-carotene.
View this table:
TABLE 2.. Lipid peroxidation indexes in Crohn disease patients and healthy control subjects1
View this table:
TABLE 3.. Plasma antioxidants in Crohn disease patients and healthy control subjects1
View this table:
TABLE 4.. Comparison of Crohn disease patients with small-bowel (SB) resection with Crohn disease patients with no bowel resection1
Lipid peroxidation indexes and plasma antioxidant micronutrients were also compared in CD patients taking (n = 29) or not taking (n = 8) medication. There were no significant differences between these subgroups (data not shown). CD patients with a CDAI <150 (n = 24) and 150 (n = 13) and with an orosomucoid concentration 1.4 g/L (n = 30) and >1.4 g/L (n = 5) were also compared (Table 5). In general, there was no significant difference between these subgroups. However, some antioxidant vitamin concentrations were low during active disease: -carotene in those with a CDAI 150 and ascorbic acid, ß-carotene, lutein and zeaxanthin, and retinol in those with an orosomucoid concentration >1.4 g/L. Several plasma antioxidant vitamins were also found to inversely correlate with the CDAI and orosomucoid concentrations (Table 6). This suggests that as CD activity increases, the plasma concentration of antioxidant vitamins decreases. However, when lipid peroxidation indexes (breath ethane and pentane and F2-isoprostane) were compared with the CDAI, no significant associations were found. However, there was a weak negative correlation between F2-isoprostane and lutein and zeaxanthin (r = -0.325, P 0.05) and between ethane and retinol (r = -0.368, P 0.05). Seven-day food intake records were completed by study participants to compare actual intake between the 2 groups (Table 7). There was no significant difference in the consumption of macronutrients and antioxidant vitamins. Overall, in CD patients, no correlation was found between macro- or micronutrient intake and disease activity indexes.
View this table:
TABLE 5.. Subgroup analysis of Crohn disease patients according to Crohn Disease Activity Index (CDAI) and orosomucoid concentration1
View this table:
TABLE 6.. Pearson correlation (r): disease activity variables compared with plasma antioxidants in Crohn disease patients
View this table:
TABLE 7.. Daily intake of nutrients in Crohn disease patients and healthy control subjects1
DISCUSSION
The results of this study indicate that lipid peroxidation, determined by breath ethane and pentane outputs and F2-isoprostane, was significantly higher in CD patients than in healthy control subjects. In addition, the plasma concentrations of several antioxidant vitamins were significantly lower in CD patients than in healthy control subjects.
This study is unique because it is the first time that these indexes of lipid peroxidation and plasma antioxidant micronutrients were measured in a controlled fashion in CD patients and compared with matched healthy subjects. Oxidative stress was increased despite the presence of a low CDAI (
Subgroup analysis of the CD population comparing those taking medication with those who were not indicated no significant difference when lipid peroxidation and antioxidant micronutrient indexes were compared. Interestingly, among the 14 patients taking oral 5-aminosalicylic acid, which is known to have antioxidant properties, 50% were still oxidatively stressed with a breath pentane output of >6.0 pmol•kg-1•min-1. These data suggest that either ongoing disease activity or perhaps a weaker antioxidant defense system was present in this population of CD patients.
The increase in lipid peroxidation in this CD population was also associated with significantly lower plasma concentrations of antioxidant vitamins such as ascorbic acid, - and ß-carotene, and the carotenoids ß-cryptoxanthin and lycopene when compared with the HC population. The difference in plasma antioxidant vitamins may be due to several factors. The majority of CD patients studied had minimal disease activity (CDAI <150) and were not a malnourished cohort (BMI: 23.2 ± 0.69) and there was no evidence of recent weight loss. Inadequate food intake or malabsorption is unlikely because of the entry criteria for this study. In addition, there was no overall significant difference between resected and nonresected CD patients for all indexes, except for ß-carotene. Although ß-carotene can be used as an index of malabsorption, other carotenoids in this study were similar in both CD groups. Other investigators have also reported a low concentration of antioxidant vitamins (13, 27). However, in contrast with our study, these results may have been confounded by the presence of smokers. Smoking is well known to increase oxidative stress and reduce plasma antioxidant micronutrients (16, 28, 29).
Correlation studies showed some weak but significant negative correlations between antioxidant vitamins and either the CDAI or orosomucoid concentrations. No correlations were found between the dietary intake of macro- and micronutrients and the CDAI. The results suggest that as disease activity increases, plasma antioxidant vitamin stores, such as ascorbic acid, -and ß-carotene, and lutein and zeaxanthin, become depleted, although there is no change in intake. This suggests an increased consumption of antioxidants during free radical scavenging in response to ROS production. In addition, the weak but significant negative correlations between retinol and breath ethane, and between lutein and zeaxanthin and F2-isoprostane, suggest a similar relation in which the concentrations of the 2 carotenoids decrease as lipid peroxidation increases.
Breath alkane output is an indirect measure of lipid peroxidation, wherein the endproducts of the process are measured. These volatile hydrocarbons are produced by the ß-scission of PUFA and are passed from the lungs into the expired air (17). Ethane evolves from the peroxidation of n-3 PUFA and pentane is the end product of n-6 PUFA. Breath alkane measures have been validated in human studies as a noninvasive measure of lipid peroxidation (17, 30, 31). F2-isoprostane is a relatively new index used to evaluate lipid peroxidation (32, 33) and, to our knowledge, has never been reported in the CD population.
Lipid peroxidation can be affected by many factors. The proportion of n-3 and n-6 PUFA intake can influence the composition of the breath alkanes produced (ethane and pentane; 34). A sustained increase in the production of free radicals per ROS in the chronically inflamed bowel might influence the consumption and requirement of antioxidant micronutrients (9, 35). However, in our studies, dietary intake data indicate that there were no significant differences in dietary fat, including percentage saturated fat or PUFA, or in the P:S between the 2 groups. Calculated food record data indicate a similar intake of macronutrients and antioxidant vitamins in the 2 groups. Health Canada surveys evaluating the actual diet consumption of fat in Eastern provinces (Quebec and Nova Scotia) indicate that the average P:S is 0.4, which is the ratio of fat all subjects in the study were advised to consume (Health Canada Nutrition Research, unpublished observations, 1997–1999).
In general, several factors that influence oxidative stress needed to be controlled for during the study. Age was similar between the 2 groups (36). Exercise was avoided on the day of the test (37). Malnourished subjects were excluded because of possible inadequate micronutrient intake or malabsorption, which would increase oxidative stress (8). Smoking (38), liver disease (39), alcohol consumption (40), and pregnancy (41) can all affect oxidative stress and breath alkane output; therefore, all subjects were carefully screened and excluded from the study if these confounders were present.
We had expected to find a significant decrease in plasma -tocopherol concentrations in the CD group, but concentrations were similar to those of control subjects. Although the plasma concentration of -tocopherol in CD patients was similar to those reported elsewhere (13), control subjects had lower values when compared with values from previous studies (13, 15, 16). Vitamin E intake of control subjects was similar to that of CD patients.
Plasma GSHPx and selenium concentrations were also similar in both groups. These results are in contrast with those of other studies (12, 13, 42, 43) in which plasma GSHPx and selenium concentrations were reduced in CD patients with active disease. In addition to differences in disease activity and nutritional status, the absence of smokers in our study may explain some of the discrepancy with other studies. There are reports in the literature regarding the reduction of mucosal glutathione with inflammatory bowel disease (10, 44–46), which is thought to increase the susceptibility of the intestine to ROS production, in addition to a reduction in the activity and availability of key precursors in the synthesis of glutathione (45). Mucosal glutathione was found to be more severely depleted in malnourished CD patients (10).
In conclusion, on the basis of increased lipid peroxidation and decreased plasma antioxidant vitamins, CD patients are oxidatively stressed. These findings were observed even though 70% of patients had a CDAI <150 and 78% of patients were taking medication to reduce intestinal inflammation. These results are consistent with the underlying hypothesis that there is an imbalance between ROS production and the antioxidant defense system in inflammatory bowel disease. The effect of antioxidant vitamin supplementation on oxidative stress and disease activity in this population needs to be studied.
REFERENCES