* Department of Gastroenterology, Institut de Malalties Digestives, and
Liver Unit, Hospital Clínic, and
Department of Medical Bioanalysis, IIBB-CSIC, Institut d’Investigacions Biomèdiques August Pi i Sunyer (IDIBAPS), University of Barcelona, Spain; and
Department of Pathology, Hospital Mutua of Terrassa, Barcelona, Spain
Correspondence: Julián Panés, Gastroenterology Department, Hospital Clínic, Villarroel 170, 08036 Barcelona, Spain. E-mail: panes{at}medicina.ub.es
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: adhesion molecules • inflammatory bowel disease • endothelium • leukocyte
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The selectin family of CAMs, which includes endothelial P- and E-selectin and leukocyte L-selectin, is essential for the initial phase of leukocyte recruitment designed as rolling. This process greatly slows the transit of leukocytes through inflamed venules, allowing time for leukocytes to sample the local environment or the endothelial cell surface for activating or chemoattractant signals [2 ].
To determine the function of each selectin in normal physiology and in pathological conditions, several experimental approaches have been used. Immunoblockade of a specific selectin and the generation of knockout mice by gene targeting have been the most widely used. Assessment of the role of P-selectin in the pathophysiology of experimental colitis using these two approaches has produced conflicting results. Thus, a study on trinitrobenzene sulfonic acid (TNBS)-induced colitis in the rat showed that P-selectin immunoblockade ameliorated some of the parameters of disease severity [2 ], whereas another study using P-selectin-/- mice [3 ] observed an increased mortality associated with induction of colitis by TNBS. Discrepancies between these observations might be related to animal species or the different experimental approach to test the effects of ablation of P-selectin function.
In the present study, we compared the effects of two P-selectin blocking strategies, genetic deficiency and blocking antibodies, on leukocyte recruitment and disease severity in dextran sulfate sodium (DSS)-induced colitis in mice. We show that P-selectin immunoblockade affords higher protection in terms of disease severity than genetic deficiency of the selectin and that induction of colitis in P-selectin-/- mice is associated with higher mortality as compared with wild-type (WT), untreated mice. To elucidate possible mechanisms that worsen the disease course in P-selectin-/- mice, we studied the expression of other adhesion molecules that mediate leukocyte recruitment in colitis and the involvement of distant organs by the inflammatory response.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
P-selectin immunoblockade
To test the effects of continuous P-selectin immunoblockade, groups of colitic mice were treated from the time of colitis induction and up to the time of study at day 6 with an anti-P-selectin monoclonal antibody (mAb) P-SEL.KO.2.12 (1 mg·kg-1·day-1) or the same dose of the nonbonding, isotype-matched control mAb P-23. mAb P-SEL.KO.2.12 (provided by Dr. Engel, Hospital Clínic, Barcelona, Spain) is a mouse immunoglobulin (Ig)G1 against mouse P-selectin [5
], and P-23 (Pharmacia Upjohn, Kalamazoo, MI) is a nonbonding, murine IgG1 directed against human, but not mouse P-selectin [6
]. Monoclonal antibodies were administered intraperitoneally. The dose of 1 mg·kg-1·day-1 was chosen because, in preliminary experiments, P-SEL.KO.2.12 markedly reduced leukocyte rolling in colonic venules of colitic mice, and higher doses (up to 3 mg·kg-1) did not produce further blocking effects.
Evaluation of inflammatory changes
The following groups of mice were studied: WT and P-selectin-/- noncolitic (control) animals, WT and P-selectin-/- mice at day 6 after the induction of colitis, and WT colitic animals treated with anti-P-selectin mAb or the nonbinding mAb. Body weight, presence of blood in excreta, and stool consistency were determined daily. When no gross blood was seen, stools were tested for the presence of occult blood (HemoFEC; Boehringer Manheim, Barcelona, Spain). Based on these parameters, a disease activity index (DAI) was calculated according to the system published previously [7
, 8
]. At the time of the study, mice were anesthetized with a subcutaneous injection of ketamine (150 mg·kg-1 body weight; Ketolar, Parke-Davies Inc., Morris Klein, NJ) and xylacine (7.5 mg·kg-1 body weight; Sigma Chemical Co., St. Louis, MO). Plasma samples were taken after centrifugation at 1000 g during 10 min at 4° of 1 ml blood obtained from each mouse by cardiac puncture with a heparinized syringe for measurements of circulating interleukin-6 (IL-6). Colons were removed quickly in order to measure the length from cecum to rectum. Distal colon samples (
100 mg) were then excised, snap-frozen in dry ice, and stored at -80°C for later assay of myeloperoxidase (MPO) activity as a measure of neutrophil infiltration [9
].
MPO assay
MPO activity was assessed according to the technique described by Ricart et al. [10
] results are expressed as units per mg of protein. Samples were macerated with 0.5% hexadecyltrimethylammonium bromide in 50 mM phosphate buffer, pH 6.0. Homogenates were then disrupted for 30 s using a Labsonic sonicator (B. Braun, Melsungen Germany) and were subsequently snap-frozen in dry ice and thawed on three consecutive occasions before a final 30 s sonication. Samples were incubated at 60°C for 2 h and then spun down at 4000 g for 12 min. Supernatants were collected, and enzyme activity was assessed photometrically at 630 nm using 3,3',5,5'-tetramethylbenzidine as a substrate. The assay mixture consisted of 20 µl supernatant, 10 µl tetramethylbenzidine (final concentration, 1.6 mM) dissolved in dimethyl sulfoxide, and 140 µl H2O2 (final concentration, 3.0 mM) diluted in 80 mM phosphate buffer, pH 5.4. An enzyme unit is defined as the amount of enzyme that produces an increase of 1 absorbance unit per minute.
Protein measurement
Total protein concentration in homogenates was determined using a commercial kit from BioRad (Munich, Germany).
Histological damage
Colonic and pulmonary samples from control and colitic mice studied at day 6 were fixed in 4% formalin and were embedded in paraffin, and sections (5–7 mm) were stained with hematoxylin and eosin following standard procedures. A single pathologist (An Salas) assessed all samples in a blinded fashion. Colitis was graded quantitatively using previously defined criteria, which takes into account the percentage of mucosal injury in total colon length and lesion depth [7
]. The histological evaluation of lungs was qualitative.
IL-6 plasma levels measurement
Levels of murine IL-6 in plasma (pg/mL) were measured by enzyme-linked immunosorbent assays using kits supplied by Diaclone (Besançon, France).
Leukocyte-endothelial cell interactions in colonic venules
Leukocyte-endothelial cell interactions in colonic submucosal and lamina propria venules were assessed using intravital microscopy in six groups of mice: WT and P-selectin-/- control animals, WT and P-selectin-/- mice at day 6 after the induction of colitis, and WT colitic animals treated with anti-P-selectin mAb or the nonbinding mAb. For that purpose, mice were anesthetized by a subcutaneous injection of ketamine (150 mg·kg-1 body weight) and xylacine (7.5 mg·kg-1), and the left jugular vein was cannulated with polythene (PE)-10 tubing (Portex Ltd., Huthe, UK). The abdomen was opened via a midline incision, and a segment of the distal colon was chosen for microscopy examination, exteriorized, and covered with cotton gauze soaked with bicarbonate buffer. Mice were then placed on an adjustable microscope stage, and the colon was extended over a nonautofluorescent coverslip that allowed observation of a 2 cm2 segment of tissue. Throughout the experiments, body temperature was maintained with an infrared heat lamp. An inverted microscope (Diaphot 300, Nikon, Tokyo, Japan) with a CF Fluor 40x objective lens (Nikon) was used. A charged coupled device (CCD) camera (model XC-77, Hamamatsu Photonics, Hamamatsu, Japan) with a C2400 CCD camera control unit and a C2400-68 intensifier head (Hamamatsu Photonics), mounted on the microscope, projected the image onto a monitor (Trinitron KX-14CP1, Sony, Tokyo, Japan), and the images were recorded using a videocassette recorder (SR-S368E, JVC, Tokyo, Japan) for off-line analysis. A video date-time generator (Panasonic Digital AV Mixer WJ-AVE55, Matsushita Communication Industrial, Tokyo, Japan) displayed these parameters on recorded and live images. Leukocytes were labeled in vivo by IV injection of rhodamine-6G (Molecular Probes, Leiden, The Netherlands) as described previously [11
]. Rhodamine-6G-associated fluorescence was visualized by epi-illumination at 510–560 nm, using a 590 nm emission filter. Single unbranched submucosal and lamina propria venules with internal diameters (ID) ranging between 30 and 40 µm were selected for observation. Venular ID was measured on-line using a video caliper (Microcirculation Research Institute, Texas A&M University, College Station). The flux of rolling leukocytes, leukocyte rolling velocity, number of adherent leukocytes, venular blood flow (Vbf), and venular wall shear rate (
) was determined off-line after playback of videotapes. Rolling leukocytes were defined as those white blood cells that moved at a velocity less than that of free-flowing leukocytes (ffv) in the same vessel. The flux of rolling leukocytes was measured as the number of rolling leukocytes that passed a fixed point within a small (10 µm) viewing area of the vessel in a 1 min period. Leukocyte rolling velocity was calculated as the mean of 10 rolling leukocyte velocities and was expressed in µm·s-1. Leukocytes were considered adherent to venular endothelium when stationary for 30 s or longer and were expressed as number per 100 µm length of venule. Vbf was estimated from the mean velocity of ffv, using the empirical relationship Vbf = ffv/1.6 [12
]. Venular wall shear rate was calculated, assuming cylindrical geometry, using the Newtonian definition = 8 (Vbf/ID) [13
]. In each animal, three to six random venules were examined, and results were calculated as the mean of each parameter in all venules examined.
Endothelial adhesion molecules expression
Endothelial vascular CAM-1 (VCAM-1) and intercellular adhesion molecule-1 (ICAM-1) expression was determined in WT and P-selectin-/- control and colitic mice and in WT colitic mice treated with anti-P-selectin mAb or nonbinding mAb. Expression of these endothelial adhesion molecules was measured by the radiolabeled antibody technique. The mAbs used were 429 (MVCAM.A), a rat IgG2 against mouse VCAM-1 (PharMingen, BD Biosciences, Heidelberg, Germany); YN1/1.7.4, a rat IgG2 against mouse ICAM-1 (Pharmacia Upjohn); and UPC-10 (Sigma Chemical Co.), a nonbonding, murine IgG2. The binding mAbs directed against VCAM-1 and ICAM-1 were labeled with I125, whereas the nonbinding mAb UPC-10 was labeled with I131 (Amersham Ibérica, Madrid, Spain). Radioiodination of mAbs was performed by the iodogen method as described previously [14
]. Animals were anesthetized by subcutaneous injection of ketamine (150 mg·kg-1) and xylazine (7.5 mg·kg-1). The left jugular vein and the right carotid artery were cannulated with PE-10 tubing (Portex Ltd.). For assessment of endothelial expression of VCAM-1, a mixture of 10 µg I125 -429, 20 µg unlabeled 429, and 1 µg I131-UPC-10 was used. For assessment of endothelial expression of ICAM-1, a mixture of 10 µg I125-YN1/1.7.4., 40 µg unlabeled YN1/1.7.4., and 1 µg I131-UPC-10 was used. Doses of anti-ICAM-1 proved to be saturating in previous assays [15
]. The saturating dose of anti-VCAM mAb 429 was determined in preliminary experiments in which higher doses up to 150 µg/mice did not result in higher antibody accumulation in tissues of colitic mice (data not shown). The mixture of binding and nonbinding mAb was administered through the jugular catheter. The injected activity in each experiment was calculated by counting a 3 µl sample of the mixture containing the radiolabeled mAb. The accumulated activity of each mAb in an organ was expressed as ng binding antibody per g tissue. The formula used to calculate ICAM-1 and VCAM-1 expression was as follows: Endothelial expression = [(cpm 125I organ*g-1*cpm 125I-injected-1)-(cpm 131I organ*g-1*cpm 131I-injected-1)x(cpm 125I in plasma)/(cpm 131I in plasma)] * ng-injected mAb. This formula was modified from the original method [16
] in order to correct the tissue accumulation of nonbinding mAb for the relative plasma levels of binding and nonbinding mAb [17
].
Expression of very late antigen (VLA)-4 integrin on circulating leukocytes
Blood samples were obtained by cardiac puncture with a heparinized syringe from WT and P-selectin-/- mice anesthetized previously. To measure the expression of VLA-4 on circulating leukocytes, 300 µl blood was centrifuged at 4°C for 7 min at 500 g in order to obtain the cell fraction. Cells were resuspended in 300 µl RPMI medium containing 2% fetal calf serum (FCS) and 1% rabbit serum (Life Technologies, Grand Island, NY). Duplicate aliquots of 100 µl were incubated for 45 min with the PS/2 mAb (10 µg/mL), a rat IgG against mouse VLA-4 (Pharmacia Upjohn). Negative control samples were incubated with RPMI medium. After incubation, cells were washed with phosphate-buffered saline (PBS) with 2% FCS, and a goat anti-rat IgG-fluorescein isothiocyanate conjugate (Caltag, San Francisco, CA) was used as a secondary antibody at a concentration of 7 µg/mL. Afterwards, cells were washed and red cells were lysed by using fluorescein-activated cell sorter (FACS) Brand Lysing solution (Becton Dickinson, San Jose, CA). Washed cells were resuspended in 500 µl PBS with 1.5% formalin and were kept at 4°C until analyzed. FACSscan flow cytometer (Becton Dickinson) and Cellquest software were used for acquisition and data analysis. Populations of leukocytes were identified from forward- and side-scatter characteristics on dot-plot profiles and were analyzed for proportion of positive cells and fluorescence intensity by using fixed defined gates. At least 10,000 cells per sample were acquired.
Statistical analyses
Data were analyzed using the nonparametric Kruskal Wallis or Mann Whitney U tests, where appropriate. Survival was analyzed by Kaplan and Meyer analysis. All values are expressed as the mean ± SEM. Statistical significance was set at P < 0.05.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
 View larger version (23K): [in a new window] |
Figure 1. Changes in body weight, expressed as % from the initial weight of four groups of mice: WT + DSS (n=14), P-selectin-/- (P-sel-/-) + DSS (n=10), WT + DSS + 1 mg/kg/day anti-P-selectin antibody (n=5), and WT + DSS + 1 mg/kg/day nonbinding antibody (Ctrl Ab; n=5). All mice were treated for 6 days with 4% DSS in drinking water. *, P < 0.05 versus WT colitic mice.
|
Development of DSS-induced colitis is associated with a reduction in colon length, which is proportional to the severity of colitis [7 ]. As expected, WT colitic mice had a significant reduction in colonic length (64.5±1.4 mm), relative to control animals (91.3±0.3 mm, P<0.001). The reduction in colonic length was significantly attenuated in P-selectin-/- colitic animals (72.2±2.9 mm, P<0.02) and WT colitic mice treated with anti-P-selectin mAb (71.5±0.6 mm, P<0.01), although in these groups, colonic length was still significantly reduced relative to control mice. Treatment with a nonbinding mAb did not influence the reduction in colonic length associated with development of colitis (65.6±1.9 mm). Neutrophil infiltration of the colon measured as tissue MPO activity was significantly increased (fourfold) in colitic WT animals compared with the control group (Fig. 2 ). The increase in MPO activity associated with the development of colitis was reduced significantly (50%) in P-selectin-/- mice, whereas it was abrogated completely by treatment of WT colitic mice with an anti-P-selectin mAb. Treatment with a control mAb did not modify colonic MPO activity relative to colitic WT untreated mice.
 View larger version (28K): [in a new window] |
Figure 2. MPO in the colon of six groups of mice: control WT (n=4), control P-selectin-/- (n=4), WT + DSS (n=6), P-selectin-/- + DSS (n=5), WT + DSS + 1 mg/kg/day anti-P-selectin antibody (n=5), and WT + DSS + 1 mg/kg/day nonbinding antibody (n=5). All mice controls were treated for 6 days with 4% DSS in drinking water. Results are expressed as units per mg protein. *, P < 0.05 versus control mice. #, P < 0.05 versus WT colitic mice.
|
 View larger version (34K): [in a new window] |
Figure 3. Colonic histological damage score represented as the crypt score (0–100), combining the percentage of mucosal injury extension and lesion depth as described in Materials and Methods. The groups of mice used were: control WT (n=6), control P-selectin-/- (n=5), WT + DSS (n=5), P-selectin-/- + DSS (n=5), WT + DSS + 1 mg/kg/day anti-P-selectin mAb (n=4), and WT + DSS + 1 mg/kg/day nonbinding antibody (n=5). All mice except controls were treated for 6 days with 4% DSS in drinking water. *, P < 0.05 versus control mice. #, P < 0.05 versus WT colitic mice.
|
 View larger version (36K): [in a new window] |
Figure 4. Leukocyte-endothelial cell interactions in colonic venules of six groups of mice: control WT, control P-selectin-/-, WT + DSS, P-selectin-/- + DSS, WT + DSS + 1 mg/kg/day anti-P-selectin mAb, and WT + DSS + 1 mg/kg/day nonbinding antibody. All mice except the control group were treated for 6 days with 4% DSS in drinking water. The number of animals in each group is 5–9. (A) Flux of rolling leukocytes (cells/min). (B) Number of rolling leukocytes (cells/100 µm venule). (C) Leukocyte rolling velocity (µm/seg). (D) Number of adherent leukocytes (cells/100 µm venule). *, P < 0.05 versus control mice. #, P < 0.05 versus WT colitic mice.
|
Under baseline conditions, adherent leukocytes in colonic venules of control WT mice were almost absent, whereas a few adherent leukocytes were observed in the venules of P-selectin-/- mice (Fig. 4D) . A marked and significant increase in leukocyte adhesion was observed in colitic WT untreated mice. P-selectin-/- colitic mice had a significant reduction in leukocyte adhesion in comparison with WT colitic mice, although this reduction was mild (29%) when compared with the reduction in leukocyte rolling observed in the same experimental group. Treatment of WT colitic mice with an anti-P-selectin mAb induced a more pronounced decrease in leukocyte adhesion (62%). The number of adherent leukocytes in the latter group was about 50% of that observed in P-selectin-/- colitic mice, but this difference did not reach statistical significance (P=0.15). Treatment with a nonbinding mAb did not influence leukocyte adhesion in colonic venules of WT colitic mice.
Expression of adhesion molecules
As cases of genetic compensation in CAM knockout mice by up-regulation of other molecules have been described [18
19
20
], we explored this possibility in P-selectin-/- colitic mice. It has been demonstrated that the other two molecules of the selectin family (E-selectin and L-selectin) are not overexpressed in P-selectin-/- mice after lipopolysaccharide or tumor necrosis factor 
stimulation [4
, 18
], and P-selectin-/- colitic mice exhibited an increase in firm adhesion. Therefore, we chose to investigate expression of ICAM-1 and VCAM-1, which are involved in leukocyte recruitment in various models of colitis [15
, 21
]. Endothelial expression of ICAM-1 and VCAM-1 in the colon was examined in WT and P-selectin-/- control and colitic mice (n=5–7 per group) using the radiolabeled antibody technique (Fig. 5
). ICAM-1 expression increased twofold in the colon of WT colitic mice relative to control mice. P-selectin-/- colitic mice had an up-regulation of ICAM-1 of similar magnitude.
 View larger version (49K): [in a new window] |
Figure 5. Endothelial ICAM-1 expression in the colon of four groups of mice: control WT, control P-selectin-/-, WT + DSS, and P-selectin-/- + DSS. All mice except the control group were treated for 6 days with 4% DSS in drinking water. Each group includes 5–7 animals. Results are expressed as bound ng antibody/g tissue. *, P < 0.05 versus control mice.
|
 View larger version (27K): [in a new window] |
Figure 6. Endothelial VCAM-1 expression in the colon of six groups of mice: control WT, control P-selectin-/-, WT + DSS, P-selectin-/- + DSS, WT + DSS + 1 mg/kg/day anti-P-selectin antibody, and WT + DSS + 1 mg/kg/day nonbinding antibody. All mice except the control group were treated for 6 days with 4% DSS in drinking water. Each group includes 6–7 animals. Results are expressed as bound ng antibody/g tissue. *, P < 0.05 versus control mice. #, P < 0.05 versus WT colitic mice.
|
Given the increased expression of VCAM-1 in P-selectin-/- mice, we measured expression of the VCAM-1 ligand VLA-4 in peripheral blood leukocytes of WT and P-selectin-/- control mice (n=4 per group). As shown in Table 1 , expression of VLA-4 was absent or very weak in neutrophils, whereas most lymphocytes and monocytes have a robust expression of this integrin. Expression of VLA-4 in the different cell types was similar in WT and P-selectin-/- mice.
|
View this table: [in a new window] |
Table 1. Expression of VLA-4 in Peripheral Blood Leukocytes from Wild-Type and P-Selectin-/- Mice as Assessed by Flow Cytometry
|
 View larger version (11K): [in a new window] |
Figure 7. Kaplan and Meyer survival analysis (Cum. Survival) of two groups of colitic mice: WT + DSS and P-selectin-/- + DSS. Mice were treated for 6 days with 4% DSS in drinking water (n=7 per group).
|
 View larger version (31K): [in a new window] |
Figure 8. MPO in the lung of six groups of mice: control WT, control P-selectin-/-, WT + DSS, P-selectin-/- + DSS, WT + DSS + 1 mg/kg/day anti-P-selectin antibody, and WT + DSS + 1 mg/kg/day nonbinding antibody. All mice except the control group were treated for 6 days with 4% DSS in drinking water. Each group includes 5–7 animals. Results are expressed as units per mg protein. *, P < 0.05 versus control mice. #, P < 0.05 versus WT colitic mice.
|
Because P-selectin-/- colitic mice had a marked increase in colonic endothelial VCAM-1 expression relative to WT colitic mice, we investigated whether altered expression of VCAM-1 or ICAM-1 was also occurring in the lung. No increase in ICAM-1 expression was detected in WT (n=5) or in P-selectin-/- (n=6) colitic mice relative to their controls (n=6). By contrast, lung VCAM-1 expression, which was not increased in WT colitic mice, was up-regulated significantly in P-selectin-/- colitic mice (Fig. 9 ). VCAM-1 up-regulation in P-selectin-/- mice was not only detected in lung tissue, but also in all other organs studied, including small intestine, stomach, mesentery, pancreas, heart, and kidney (Table 2 ). No significant changes in extracolonic VCAM-1 expression were detected in WT colitic mice compared with controls.
 View larger version (32K): [in a new window] |
Figure 9. Endothelial VCAM-1 expression in the lung of six groups of mice: control WT, control P-selectin-/-, WT + DSS, P-selectin-/- + DSS, WT + DSS + 1 mg/kg/day anti-P-selectin antibody, and WT + DSS + 1 mg/kg/day nonbinding antibody. All mice except the control group were treated for 6 days with 4% DSS in drinking water. Each group includes 6–7 animals. Results are expressed as bound ng antibody/g tissue. *, P < 0.05 versus control mice. #, P < 0.05 versus WT colitic mice.
|
|
View this table: [in a new window] |
Table 2. Endothelial VCAM-1 Expression in Different Organs of Wild-Type and P-Selectin-/- Controls and Colitic Mice
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Apparently, the observations of the current study may be at odds with previous work from our group showing that in TNBS-induced colitis in the rat, treatment with anti-P-selectin antibodies affords very slight protection [2 ]. However, McCafferty et al. [3 ] have demonstrated variable responses to abrogation of P-selectin function in different models of experimental colitis. These authors showed that P-selectin-/- mice were protected from acetic acid-induced colitis, but not from TNBS-induced colitis. Therefore, the protective effects of modulation of leukocyte recruitment on intestinal inflammation seem to depend on the pathophysiological mechanisms involved in a particular condition. The acetic acid-induced colitis [21 ] and the initial phase of DSS-induced colitis are acute models of colitis [23 ], whereas TNBS-induced colonic inflammation is a more chronic type. In keeping with previous studies, we observed focal crypt lesions and submucosal inflammation with granulocytes and macrophages in the acute phase of DSS-induced colitis [7 ]. These two cell types appear to play a key role in the pathophysiology of acute colitis. The absence of selectins specifically impairs neutrophil recruitment, but leaves mononuclear cell recruitment relatively intact [22 ]. By contrast, CD4+ (helper) T cells with increased resistance to apoptosis [24 ] play a more prominent role in the chronic phase of DSS colitis and in TNBS-induced colitis [23 ]; this may explain why blockade of selectin function in this setting has little effect on the course of the colonic inflammatory lesions. Nevertheless, even in T helper cell type 1 inflammatory conditions, blockade of other adhesion molecules such as VCAM-1 [17 ] or VLA-4 [25 ] may afford significant protection.
One of the intriguing findings of the current study is the observation that although P-selectin immunoblockade and genetic deficiency afforded protection against DSS-induced colonic lesions, chronic treatment with the anti-P-selectin antibody was more potent in reducing the DAI score, neutrophil infiltration (measured as MPO activity), and leukocyte adhesion (although the latter did not reach statistical significance: P=0.15) in the colon. Previous studies comparing the effects of blocking antibodies with those of gene knockout mice have produced diverging observations. In peritonitis, neutrophil emigration is prevented to a similar degree when the function of P-selectin is blocked by mAb or by gene mutation [4 , 26 ]. By contrast, in an acute lung injury model induced by cobra venom factor, P-selectin-/- mice were not protected against injury, whereas treatment with an anti-P-selectin antibody inhibited leukocyte recruitment and injury by 57% and 60%, respectively [27 ].
Taken together, all these observations raise the question whether the use of the mAb may overestimate the function of P-selectin or whether the use of mutant mice may lead to underestimates of P-selectin function. mAb binding to the endothelium might result in alteration of another molecule that mediates adhesion. The signal transduction in neutrophils or endothelial cells that occurs following adhesion may be altered by the presence of a mAb. For example, it has been shown that antibody blockade of P-selectin reduces platelet-induced tyrosine kinase activation and Mac-1 up-regulation in neutrophils [28 ]. Therefore, the effects of blocking antibodies against P-selectin may not only depend on abrogation of the adhesive function of P-selectin, but also on the prevention of downstream signaling that may ultimately affect the function of other adhesion molecules. An additional shortcoming in the assessment of CAM function using blocking mAb is the potential perturbation of cellular function via nonspecific ligand interactions. Here, we conducted experiments using an isotype-matched control, nonbinding antibody, which was without any effect. Therefore, it is unlikely that the effects of the anti-P-selectin antibody result from unspecific actions.
The use of the P-selectin-/- mice may have underestimated the function of P-selectin during the inflammatory process. Studies have indicated that chronic deficiency of some CAMs in knockout mice results in perturbation of basal and cytokine-induced endothelial cell surface expression of other CAMs. In particular, it has been shown that ICAM-1 knockout mice have an enhanced VCAM-1 [20 ] and E-selectin [18 ] up-regulation in response to endotoxin challenge compared with WT mice. In a previous study, it has been shown that basal and stimulated E-selectin expression in the intestine of P-selectin-/- mice is similar to that of WT mice [18 ]. Expression of VCAM-1 in P-selectin-/- mice had not yet been assessed. We believed that alterations in the regulation of expression of this CAM might be especially relevant in DSS-induced colitis, given the key role of VCAM-1 in leukocyte recruitment in this experimental model [15 ]. We observed that under baseline conditions, expression of VCAM-1 was slightly but significantly increased in P-selectin-/- mice, and this was accompanied by an increased number of adherent leukocytes in colonic venules. When animals were rendered colitic by administration of DSS, the magnitude of VCAM-1 up-regulation was considerably higher in the colon of P-selectin-/- mice than in WT mice. By contrast, expression of ICAM-1 in colonic endothelium was similar in P-selectin-/- and WT mice, suggesting that increased VCAM-1 expression in the colitis may contribute to sustaining higher adhesion and to dampening, to a certain extent, the protection afforded by abrogation of P-selectin function.
VCAM-1 overexpression was not observed in anti-P-selectin, mAb-treated mice. This is an important observation if anti-P-selectin therapy is considered for treatment of human inflammatory bowel disease. In the current study, animals received a daily injection of antibody for 6 days. We do not know whether treatment for more prolonged periods of time might also lead to development of compensatory mechanisms for leukocyte adhesion, but so far, this has only been shown for lack of adhesion molecule function already present in embryonic life.
P-selectin-/- mice not only had a higher expression of VCAM-1 in the inflamed colon, but also had an up-regulation of endothelial VCAM-1 expression in the extracolonic organs studied, including lung, heart, small intestine, stomach, mesentery, pancreas, and kidneys. Up-regulation of VCAM-1 in extracolonic organs was not present in WT colitic mice. Although elucidation of the mechanisms leading to this up-regulation are beyond the scope of this study, it is conceivable that the genetic alteration in regulation of expression of VCAM-1 and a higher systemic inflammatory response, as evidenced by increased levels of circulating IL-6 levels in colitic P-selectin-/- mice as compared with colitic WT mice, may contribute to VCAM-1 up-regulation. This up-regulation seems to have a functional relevance, as histological studies showed infiltration of lung tissue by lymphocytes and neutrophils, and lung MPO activity was elevated only in P-selectin colitic mice, but not in WT colitic animals. It is interesting that a previous study in an allergic model of inflammation has shown that P-selectin deficient, but not WT mice, exhibit 4-integrin-dependent rolling [29
]. Although ligands for VCAM-1 (VLA-4 and 4ß7 integrins) are characteristically expressed in lymphocytes, recent evidence shows that activated neutrophils can also express VLA-4 and adhere through a VCAM-1-dependent mechanism [30
]. However, we could not directly prove the role of VCAM-1 in the lung inflammatory changes because that would require administration of an anti-VCAM-1 antibody, which by itself would markedly reduce intestinal inflammatory lesions, and in the absence of intestinal inflammation, P-selectin-/- mice do not develop lung inflammatory changes.
A previous study conducted on P-selectin-/- mice with TNBS-induced colitis [3 ] showed that these mice have an increased mortality despite that histological damage score and colonic MPO activity were not increased relative to WT colitic mice. In our experiments performed at day 6, no mortality was observed in any of the experimental groups, but P-selectin-/- colitic mice had the clinical appearance of severe disease (ruffled hair, inactivity). Therefore, we made additional experiments just prolonging the observation period, without administering any further DSS, and mortality was significant by day 10 (60%) in P-selectin-/- colitic mice. None of the animals in other experimental groups died. Because colitis was even milder in P-selectin-/- mice than in WT mice, mortality cannot be ascribed to colonic inflammatory changes, but is probably related to inflammatory changes in vital distant organs such as the lungs.
Taken together, our results show that P-selectin is a key molecule in the pathophysiology of DSS colitis, as both the administration of anti-P-selectin mAb and genetic deficiency of P-selectin diminished leukocyte recruitment and inflammatory lesions in the colon and that this protection may be overshadowed by alterations in expression of other CAM in P-selectin-/- mice in the colon and in distant organs that lead to distant organ damage and increased mortality. Therefore, generalizations regarding the role of adhesion molecules in different inflammatory conditions and organs may require considerable caution. Studies of adhesion molecule function using different approaches may be crucial in order to overcome the limitations inherent to each particular approach.
Received December 19, 2001; revised February 8, 2002; accepted February 22, 2002.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
4ß 1-integrin under flow conditions Blood 89,3837-3846This article has been cited by other articles:
 |
|
 |
|
  O. Schreiber, J. Petersson, M. Phillipson, M. Perry, S. Roos, and L. Holm Lactobacillus reuteri prevents colitis by reducing P-selectin-associated leukocyte- and platelet-endothelial cell interactions Am J Physiol Gastrointest Liver Physiol, March 1, 2009; 296(3): G534 - G542. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Vowinkel, K. C. Wood, K. Y. Stokes, J. Russell, A. Tailor, C. Anthoni, N. Senninger, C. F. Krieglstein, and D. N. Granger Mechanisms of platelet and leukocyte recruitment in experimental colitis Am J Physiol Gastrointest Liver Physiol, November 1, 2007; 293(5): G1054 - G1060. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Panes, M. Aceituno, F. Gil, R. Miquel, J. M Pique, A. Salas, and P. McLean Efficacy of an inhibitor of adhesion molecule expression (GI270384X) in the treatment of experimental colitis Am J Physiol Gastrointest Liver Physiol, October 1, 2007; 293(4): G739 - G748. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Angulo, M. Llopis, M. Antolin, M. Gironella, M. Sans, J. R. Malagelada, J. M. Pique, F. Guarner, and J. Panes Lactobacillus casei prevents the upregulation of ICAM-1 expression and leukocyte recruitment in experimental colitis Am J Physiol Gastrointest Liver Physiol, December 1, 2006; 291(6): G1155 - G1162. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Anthoni, M. G. Laukoetter, E. Rijcken, T. Vowinkel, R. Mennigen, S. Muller, N. Senninger, J. Russell, J. Jauch, J. Bergmann, et al. Mechanisms underlying the anti-inflammatory actions of boswellic acid derivatives in experimental colitis Am J Physiol Gastrointest Liver Physiol, June 1, 2006; 290(6): G1131 - G1137. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. M. Lord, R. M. Rao, H. Choe, B. M. Sullivan, A. H. Lichtman, F. W. Luscinskas, and L. H. Glimcher T-bet is required for optimal proinflammatory CD4+ T-cell trafficking Blood, November 15, 2005; 106(10): 3432 - 3439. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Mori, J. W. Salter, T. Vowinkel, C. F. Krieglstein, K. Y. Stokes, and D. N. Granger Molecular determinants of the prothrombogenic phenotype assumed by inflamed colonic venules Am J Physiol Gastrointest Liver Physiol, May 1, 2005; 288(5): G920 - G926. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. M. Rijcken, M. G. Laukoetter, C. Anthoni, S. Meier, R. Mennigen, H.-U. Spiegel, M. Bruewer, N. Senninger, D. Vestweber, and C. F. Krieglstein Immunoblockade of PSGL-1 attenuates established experimental murine colitis by reduction of leukocyte rolling Am J Physiol Gastrointest Liver Physiol, July 1, 2004; 287(1): G115 - G124. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Szanto, I. Gal, A. Gonda, T. T. Glant, and K. Mikecz Expression of L-Selectin, but Not CD44, Is Required for Early Neutrophil Extravasation in Antigen-Induced Arthritis J. Immunol., June 1, 2004; 172(11): 6723 - 6734. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. Haddad, C. J. Cooper, Z. Zhang, J. B. Brown, Y. Zhu, A. Issekutz, I. Fuss, H.-o. Lee, G. S. Kansas, and T. A. Barrett P-Selectin and P-Selectin Glycoprotein Ligand 1 Are Major Determinants for Th1 Cell Recruitment to Nonlymphoid Effector Sites in the Intestinal Lamina Propria J. Exp. Med., August 4, 2003; 198(3): 369 - 377. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
