,
,
,,
Hepatic Hemodynamic Laboratory, Liver Unit, Hospital Clínic, Barcelona, Spain;
* Immunology Unit, Department of Cellular Biology and Pathology, Medical School, University of Barcelona, Spain;
Consejo Superior de Investigaciones Científicas, Barcelona, Spain; and
Institut d’Investigacions Biomèdiques August Pi i Sunyer, Barcelona, Spain
Correspondence: Dr. Pilar Pizcueta, Fundació Clínic per a la Recerca Biomèdica, C/Villarroel 170, 08036 Barcelona, Spain. E-mail: pizcueta{at}medicina.ub.es
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: hepatitis • necrosis • T lymphocyte • adhesion molecule
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-granules of platelets and Weibel-Palade bodies of endothelial cells, is rapidly translocated to the cell surface upon activation by thrombogenic and inflammatory mediators [1
]. Prolonged surface expression of P-selectin on endothelial cells has been observed when these cells are stimulated with oxygen radicals, lipopolysaccharide (LPS), or cytokines [2
, 3
]. P-selectin supports the initial rolling of leukocytes and platelets on activated endothelium and mediates platelet-leukocyte adhesion [4
5
6
]. Furthermore, P-selectin binds to monocytes, neutrophils, mast cells, eosinophils, natural killer cells, and T cells, especially memory CD4+ T cells, via the P-selectin glycoprotein ligand (PSGL-1) molecule [7
, 8
]. Blocking P-selectin function prevents various pathologies in experimental animal models including myocardial infarction [9 , 10 ], lung injury [11 , 12 ], delayed-type hypersensitivity [13 ], and arthritis [14 , 15 ]. P-selectin is also involved in liver inflammatory disturbances such as endotoxemia [16 ] and hepatic ischemia/reperfusion [17 ]. Neutrophils and monocytes induce tissue damage in these experimental models. However, the role of P-selectin in lymphocyte cell-mediated hepatic inflammation is unclear. For instance, in human autoimmune liver disease, the predominant cells infiltrating the liver are CD4+ T lymphocytes [18 ]. Additionally, the disease activity in chronic active hepatitis B or C is associated with the T helper cell type 1 (Th1) cytokine response of intrahepatic CD4+ lymphocytes [19 , 20 ].
Concanavalin A (Con A)-induced hepatitis is an experimental model of human autoimmune hepatitis. Hepatocyte injury is associated with massive CD4+ lymphocyte activation and infiltration into the liver parenchyma, leading to secretion of the proinflammatory cytokines tumor necrosis factor (TNF-), interferon-
(IFN-), interleukin (IL)-2, IL-6, granulocyte macrophage-colony stimulating factor, and IL-1 [21
22
23
24
]. Con A also induces intrasinusoidal hemostasis by leukocyte, platelet, and erythrocyte aggregates, which contribute to the development of liver damage [25
].
Lymphocyte transmigration through the endothelial barrier into the inflamed tissue is a prerequisite for the action of effector cells. This is a multiple-step process that involves several adhesion receptors [26 ]. Few studies have examined the role of adhesion molecules in Con A-induced hepatic injury. Whereas one report describes the inhibitory effects of anti-E-selectin and anti-vascular cell adhesion molecule-1 (VCAM-1) monoclonal antibodies (mAb) [27 ], another study shows that these adhesion molecules have no known role and that intercellular adhesion molecule-1 (ICAM-1) is not involved in Con A-induced liver disease [28 ]. To our knowledge, the function of P-selectin has not been established in this experimental model.
Therefore, we examined the involvement of P-selectin in Con A-induced hepatitis. We used P-selectin-deficient mice and blocking anti-P-selectin mAb in wild-type mice to establish the therapeutic potential of this molecule in autoimmune hepatitis.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Reagents
Con A was from Sigma Chemical Co. (St. Louis, MO). The mouse anti-P-selectin-blocking mAb P-sel.ko.2.12 [immunoglobulin G (IgG)1] and anti-P-selectin nonblocking mAb P-sel.ko.2.3 (IgG1) were produced in our laboratory [30
]. Protein G-Sepharose (Pharmacia Biotech AB, Upsala, Sweden)-purified mAb were obtained from concentrated supernatant (Integra CL350 flasks, Integra Biosciences, Switzerland). Phycoerythrin (PE) anti-CD41, anti-E-selectin, fluorescein isothiocyanate (FITC) anti-CD4, PE anti-CD8a, and PE anti-CD11b mAb were from Pharmingen (San Diego, CA). CyTM3-conjugated Streptavidin (Jackson ImmunoResearch Labs, West Grove, PA), FITC-conjugated goat anti-rat IgG, biotin-conjugated goat anti-rat IgG (Caltag, San Francisco, CA), and R-PE-conjugated streptavidin (Southern Biotechnology Associates, Birmingham, AL) were used as secondary reagents for immunofluorescence or immunochemistry analysis.
Mice treatment
Con A was dissolved in sterile, pyrogen-free phosphate-buffered saline (PBS; Gibco-BRL, Gaithersburg, MD) and injected intravenously via tail vein at a dose of 20 mg/kg per mouse. In control-treated animals, only PBS was injected. C57BL/6 wild-type mice were pretreated intraperitoneally (i.p.) with anti-P-selectin-blocking mAb (P-sel.ko.2.12) or anti-P-selectin isotype antibody-matched, nonblocking mAb (P-sel.ko.2.3) at a dose of 100 µg/mouse, 5 min before Con A administration. None of the mAb treatments caused neutropenia. LPS (Escherichia coli serotype 0111:B4; Sigma Chemical Co.) was injected i.p. at a dose of 10 mg/kg per mouse.
Immunohistochemistry
Livers from wild-type and P-selectin-deficient mice were perfused through the portal vein with a peristaltic pump at a rate of 6 ml/min with 10 ml PBS followed by 2% paraformaldehyde (w/v) in PBS. Liver portions (1–1.5 cm3) were then fixed with 2% paraformaldehyde for 1 h at 4°C, cryoprotected overnight in a 30% sucrose solution (w/v) at 4°C, and frozen in liquid nitrogen. Liver sections of 8 µm were obtained and fixed in 4% paraformaldehyde. Sections were then evaluated for P-selectin and E-selectin expression by fluorescence immunohistochemistry. Samples were incubated overnight at 4°C with the primary mAb, biotinylated P-sel.ko.2.12 (5 µg/ml), rat anti-mouse E-selectin (5 µg/ml), or PBS-1% bovine serum albumin (Sigma Chemical Co.) as negative control. Thereafter, sections were treated with CyTM3-conjugated streptavidin (Jackson ImmunoResearch Labs; 1:6000) to detect P-selectin and with a biotin-conjugated anti-rat IgG mAb (Caltag; 5 µg/ml) followed by CyTM3-conjugated streptavidin (1:6000) to detect E-selectin. Incubations were performed for 1 h at room temperature. Samples were analyzed by confocal laser-scanning microscopy (Leica TCS NT, Heidelberg, Germany).
The number of platelets accumulated in the liver sinusoids at 2 h and 6 h after Con A treatment was determined in wild-type and P-selectin-deficient mice. Three mice were included in each group. Liver sections of 8 µm were obtained and incubated overnight at 4°C with a rat anti-mouse CD41 mAb (Pharmingen; 10 µg/ml) followed by incubation with an anti-rat IgG FITC-conjugated mAb (Caltag; 5 µg/ml). Five confocal images at 400x magnification were randomly obtained from each liver section. Images were processed with the Scion image analyzing system (Scion Corporation, Frederick, MD), and the average number of platelets per microscopic field was calculated for each sample. The mean value for the three mice receiving the same treatment was then established.
Reverse transcriptase-polymerase chain reaction (RT-PCR) analysis
Total RNA from mouse liver samples at 0 h and at 4 h of LPS treatment and 6 h of Con A treatment was prepared with Trizol reagent (Life Technologies, Rockville, MD). RNAs were reverse-transcribed into cDNA using a First-Strand cDNA synthesis kit for RT-PCR avian myeloblastis virus and were then amplified by PCR with the Taq polymerase Expand High FidelityTM (Roche Diagnostics, Otteweiler, Germany), according to the manufacturer’s instructions. Cycling conditions for this amplification were 35 cycles of 94°C for 30 s, 58°C for 1 min, and 72°C for 1 min, preceded by 94°C for 4 min. RT-PCR of mouse P-selectin was performed using the following primers: 5'TGTGAAGCTATTGCGTGTGGAACC3' (sense primer) and 5'TTGATGGCTTCACACGTGGGAGG3' (antisense primer). The amplified product was electrophoresed in 1% agarose gel and stained with ethidium bromide.
Histological examinations
The livers from wild-type and P-selectin-deficient mice were obtained 0, 2, 6, 12, 24, and 48 h after Con A treatment and were fixed in 10% formalin. Four mice were used for each group. Liver portions were dehydrated through alcohol series and xylene and embedded in paraffin. For the histological studies, 5 µm-thick sections were obtained and stained with hematoxylin and eosin. Sections were examined under a light microscope (Zeiss Axioplan, Jena, Germany).
Assessment of lymphocyte adhesion
The number of lymphocytes adhering to sinusoidal endothelium and hepatic and portal venules was determined in each histological section, 0 h, 2 h, and 6 h after Con A treatment. Lymphocytes were distinguished from other leukocytes by the morphology of their nucleus. Ten microscopic fields (x400) were taken at random from each sample, and all the lymphocytes included in the field were analyzed. The mean value in the four mice was then calculated.
Assessment of leukocyte infiltration
The number of infiltrating leukocytes into the liver parenchyma was counted in the periportal areas and around hepatic venules from hematoxilyn- and eosyn-stained sections at 24 h and 48 h after Con A treatment. Cells were counted as described above.
To characterize the infiltrating leukocytes, livers from wild-type and P-selectin-deficient mice were removed 24 h after Con A treatment. Livers were cut into small pieces, digested with Collagenase A (Roche Diagnostics; 0.1 mg/ml PBS) for 5 min at 37°C, and gently pressed through a cell strainer (100 µm; Becton Dickinson Labware, Franklin Lakes, NJ). The cell liver suspension was washed and then resuspended in RPMI-1640 medium supplemented with 2% fetal calf serum (FCS; Gibco-BRL). An aliquot of total cell suspension was stained with Methyl violet (Merck, Darmstadt, Germany) and was analyzed by light microscopy to assess the percentage of polymorphonuclear leukocytes (PMN) infiltrating the liver versus mononuclear leukocytes (MNC). MNC were then isolated by Ficoll density gradient centrifugation (1000 g for 20 min). MNC were washed twice and resuspended in RPMI-1640 medium supplemented with 2% FCS. The cells were analyzed by flow cytometry for the expression of CD4, CD8, and CD11b cell surface markers. Four animals were included in each group, and the mean value was calculated.
Evaluation of liver hemostasis
Liver hemostasis was defined as occlusion of sinusoids by aggregates formed by erythrocytes, platelets, and leukocytes. It was evaluated by calculating the percentage of sinusoidal area occluded with cell aggregates in the liver samples at baseline, 2 and 6 h after Con A injection. Ten microscopic fields (x200) were obtained from each sample using the Zeiss microscope provided with a TV camera. The Scion image analyzing program was used to determine the area of each sinusoid and the area of cell aggregates inside the sinusoids. An average of 50 sinusoids was evaluated in each sample, and four mice were used in each situation. The percentage of sinusoidal occlusion was established for each group of mice.
Histological determination of hepatic necrosis
Hepatic necrosis was assessed in each section as the percentage of liver parenchyma with necrotic damage. To this end, eight microscopic fields of each sample were obtained at 25x magnification. Images were processed with the Scion image analyzing system. The total hepatic parenchyma area and the necrotic area were measured for each sample, and the percentage of necrotic area was established for each mouse. Each value is the average of four animals.
Platelets isolation
Blood from wild-type and P-selectin-deficient mice was collected from inferior cava vein into a solution containing nine parts of PBS and one part of acid-citrate-dextrose (38 mmol/l citric acid, 75 mmol/l trisodium citrate, and 100 mmol/l dextrose), at 0 h, 2 h, and 6 h after Con A treatment. Blood was centrifuged at 280 g for 5 min at room temperature to obtain platelet-rich plasma, and platelets were then isolated by further centrifugation at 2000 g for 6 min. Platelets were washed twice in PIPES buffer, pH 6.1 (25 mmol/l PIPES, 137 mmol/l NaCl, 4 mmol/l KCl, and 0.1% wt/vol dextrose) and were resuspended in PIPES buffer, pH 7.0, for flow cytometry analysis with the CD41 and P-sel.ko.2.12 mAb.
Flow cytometry analysis
MNC cells were analyzed by single and double immunofluorescence using FITC or PE directly labeled mAb (CD4-FITC, CD8a-PE, CD11b-PE). Platelets were studied by double immunofluorescence with the biotinylated P-sel.ko.2.12 and rat anti-mouse CD41 mAb (Pharmingen). After washing, the platelets were treated with goat anti-rat IgG FITC-conjugated (Caltag) and PE-conjugated streptavidin (Southern Biotechnology Associates). Platelet acquisition was performed at logarithmic scale, and the platelet population was identified using CD41 mAb. Fluorescence was analyzed using a FACSCalibur (Becton Dickinson Immunocytometry Systems, San Jose, CA) equipped with the CellQuestTM software. Fluorescence intensity was represented on a four decade log scale (1–10,000). At least 5000 cells were analyzed.
Transaminase measurements
Blood was collected from the inferior cava vein into Microtainer brand tubes with ethylenediaminetetraacetate (Becton Dickinson Vacutainer Systems, Franklin Lakes, NJ), 0, 2, 6, 12, 24, and 48 h after Con A injection. Samples were immediately centrifuged at 2000 g for 7 min. Aspartate aminotransferase (ASAT) and alanine aminotransferase (ALAT) activities in plasma were measured on the Dimension RxL® clinical chemistry system (Dade-Behring Inc., Deerfield, IL) following the International Federation of Clinical Chemistry guidelines [31
].
Statistical analysis
All data were analyzed using analysis of variance and Student’s unpaired t-test. All results are reported as mean ± SEM. Statistical significance was set at P < 0.05.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
 View larger version (109K): [in a new window] |
Figure 1. Time course of P-selectin expression on the hepatic endothelium after Con A treatment. Immunofluorescent staining of P-selectin on liver sections from wild-type mice, 0 (A), 2 (B), 6 (C), and 12 (D) h after Con A treatment was performed using the biotin-conjugated P-sel.ko.2.12 mAb and CyTM3-streptavidin. Arrows indicate the staining on endothelial cells. Original magnification, x400. Higher magnification (original, x1200) of P-selectin staining 2 h after Con A treatment is represented (E).
|
 View larger version (106K): [in a new window] |
Figure 2. E-selectin expression on the hepatic endothelium after Con A treatment in wild-type and P-selectin-deficient mice. E-selectin expression was evaluated in liver samples at 0, 2, 6, and 12 h after Con A injection. Sections were stained with anti-E-selectin mAb, and binding sites were detected with a biotin-conjugated anti-rat-IgG mAb and CyTM3-streptavidin. Arrows indicate the staining on endothelial cells. Original magnification, x400.
|
 View larger version (42K): [in a new window] |
Figure 3. RT-PCR analysis of P-selectin expression in the liver. Total liver RNA was isolated from wild-type mice at baseline (0 h), 4 h after LPS treatment, and 6 h after Con A treatment. These time points correspond to maximum P-selectin expression as assessed by immunohistochemistry. mRNA was reverse-transcribed and amplified by PCR with primers specific for murine P-selectin. Ethidium bromide staining of the PCR products after gel analysis.
|
 View larger version (32K): [in a new window] |
Figure 4. Immunofluorescence analysis of P-selectin expression on platelets after Con A treatment. Platelets were stained for two-color flow cytometry. CD41 mAb was used to select the specific platelet population. Representative flow cytometry histograms of P-selectin staining on platelets are shown. Fluorescence intensity is shown on a four-decade log scale.
|
 View larger version (18K): [in a new window] |
Figure 5. Number of adherent lymphocytes in hepatic veins (A), portal veins (B), and sinusoids (C) in wild-type and P-selectin-deficient mice. Measurements were performed 2 and 6 h after Con A injection. Four mice were included in each group. Results are expressed as mean ± SEM. *, P < 0.05 versus wild-type mice.
|
 View larger version (35K): [in a new window] |
Figure 6. Con A induced intrasinusoidal occlusion in wild-type and P-selectin-deficient mice. The percentage of intrasinusoidal area occluded by cell aggregates was determined in liver sections at baseline, 2 and 6 h after Con A injection. The increase in sinusoidal congestion from basal values is represented as mean ± SEM. *, P < 0.05 versus wild-type mice.
|
 View larger version (87K): [in a new window] |
Figure 7. Histological observation of intrasinusoidal occlusion 6 h after Con A injection. Light micrographs of hematoxilyn-eosin-stained liver sections from wild-type mice pretreated with P-sel.ko.2.3 (isotype control) mAb (A) and P-sel.ko.2.12 (function blocking) mAb (B) and from P-selectin-deficient mice (C). Original magnification, x600.
|
 View larger version (55K): [in a new window] |
Figure 8. Involvement of P-selectin in Con A-induced intrasinusoidal occlusion. The increase in sinusoidal occlusion from baseline values was determined in wild-type mice pretreated with P-sel.ko.2.3 (isotype control) mAb and P-sel.ko.2.12 (function blocking) mAb and in P-selectin-deficient mice. Results are represented as mean ± SEM. Four animals were included in each group. *, P < 0.05 versus wild-type mice pretreated with P-sel.ko.2.3.
|
 View larger version (34K): [in a new window] |
Figure 9. Leukocyte infiltration around hepatic (A) and portal veins (B) in wild-type and P-selectin-deficient mice. The number of leukocytes was determined 24 and 48 h after Con A injection. Four mice were used for each group. *, P < 0.05 versus wild-type mice.
|
 View larger version (140K): [in a new window] |
Figure 10. Histopathological examination of Con A-induced liver injury. Light micrographs of hematoxilyn-eosin-stained liver sections from wild-type and P-selectin-deficient mice at 24 h after Con A injection. Large necrotic areas were visible in wild-type animals (A) and markedly reduced in P-selectin mice (B). Original magnification, x200.
|
 View larger version (19K): [in a new window] |
Figure 11. Evaluation of Con A-induced liver damage in wild-type and P-selectin-deficient mice. The percentage of liver parenchyma with necrotic injury was established in liver sections at 24 and 48 h after Con A injection. Results represent mean ± SEM of four animals. *, P < 0.05 versus wild-type.
|
 View larger version (19K): [in a new window] |
Figure 12. Time course of plasma transaminase activities in wild-type and P-selectin-deficient mice following Con A injection (20 mg/kg). (A) Plasma ASAT and (B) plasma ALAT levels were assessed 0, 2, 6, 12, 24, and 48 h after Con A injection. Four animals were included in each group. Results are expressed as mean ± SEM; *, P < 0.05 versus wild-type.
|
 View larger version (19K): [in a new window] |
Figure 13. Levels of plasma transaminase 12 h after Con A injection in P-sel.ko.2 3 (isotype control) mAb and P-sel.ko.2.12 (function blocking) mAb pretreated wild-type mice and P-selectin-deficient mice. Control-treated (PBS) animals were included. (A) Plasma ASAT and (B) plasma ALAT levels were represented as mean ± SEM. Four animals were included in each group. *, P < 0.05 versus wild-type + P-sel.ko.2.3.
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This study establishes that P-selectin is an important determinant of injury in Con A-induced hepatitis. This consisted of rapid lymphocyte adhesion, infiltration, marked congestion, and necrosis of hepatic tissue with a concomitant increase of serum transaminases in the wild-type mice. In contrast, P-selectin-deficient mice showed significantly less lymphocyte adhesion and infiltration, hemostasis, necrosis, and levels of transaminases. Moreover, blocking P-selectin function using mAb protected mice from Con A-induced hepatic damage.
The main infiltrating lymphocytes after Con A treatment are T cells [21 , 32 ]. CD4+ lymphocytes efficiently interact with P-selectin [33 ]. The expression of the functional PSGL-1 and specific chemokine receptors dictates the capacity of the T cell to migrate to different tissues. Moreover, the CD4+ Th1 phenotype is the T cell subset that preferentially adheres to P-selectin, thus facilitating migration through the inflamed endothelium [33 ]. Several studies have established the role of P-selectin and its ligand PSGL-1 in T cell migration to the inflamed skin in a contact-hypersensitivity model [34 35 36 ], sensitized skin [37 ], psoriatic lesions, and colitis [38 ].
The expression of P-selectin is limited to the portal tract (arterial and venous) and central vein endothelia in normal and inflamed human livers [39 ]. In two reports, no P-selectin expression was observed after Con A injection in mice [27 , 28 ]. Although conflicting results on the expression of several adhesion molecules in Con A-induced hepatitis have been reported [27 , 28 ], we demonstrate that P-selectin is not constitutively present but transiently expressed in portal and hepatic venules and negative in the sinusoids. P-selectin protein expression was detected in a narrow time frame from 2 to 6 h after Con A injection. These immunohistochemical data were corroborated by the finding of elevated levels of P-selectin mRNA. The kinetics of P-selectin expression is identical to that observed in other mouse experimental models such as ischemia, reperfusion, and endotoxemia [16 , 17 ]. However, P-selectin staining in livers of Con A-treated mice was less intense than that observed in the liver endothelium after LPS treatment using the same P-selectin mAb [30 ]. Consistent with this observation, endotoxin also caused higher P-selectin mRNA levels than Con A. The main difference is that LPS induces the expression of P-selectin on the sinusoidal endothelium, whereas this was not observed after Con A treatment. We also observed transient expression of P-selectin on platelets.
Lymphocyte adhesion in hepatic and portal veins was lower in the P-selectin-deficient mice than in wild-type mice. Hepatic veins and to a lesser extent portal veins are the main sites of lymphocyte adhesion/transmigration during Con A-induced hepatitis in mice [27 ]. Our results confirm that lymphocytes started to adhere to the endothelium 2 h after Con A injection, sharply increased in number at 6 h, quickly decreased at 12 h, and were almost absent at 24 h [27 ]. Therefore, there was a high correlation between P-selectin expression and lymphocyte adhesion. Thus, the inhibition of P-selectin-dependent lymphocyte recruitment to the hepatic venules may be responsible for the protective effects observed in P-selectin-deficient mice and after anti-P-selectin antibody treatment in wild-type mice.
Lymphocyte adhesion to the sinusoids is lower than to the hepatic veins [27 ]. Adhesion to the sinusoids was maximum at 2 h and very low at later time points. However, if we consider the large area of sinusoidal beds involved, the total number of infiltrating lymphocytes is quantitatively very important. The number of lymphocytes adhering to the sinusoids decreased significantly 2 h after Con A injection in P-selectin-deficient mice. Whereas the reduction in lymphocyte adhesion to hepatic venules in the P-selectin-deficient mice can be explained by the lack of expression of this adhesion molecule on the endothelium, this is unlikely for the decrease in lymphocyte adhesion to the sinusoids, as sinusoids do not express P-selectin after Con A treatment. Moreover, leukocyte adhesion to hepatic sinusoids appears to be less dependent on selectins [40 ]. As E-selectin is expressed at later time points, a role of this adhesion molecule in the lymphocyte adhesion to the sinusoids cannot be ruled out.
P-selectin on activated platelets may be involved in the delivery of lymphocytes to sinusoids shortly after the injection of Con A. This idea is consistent with our observation that Con A induced the expression of P-selectin on platelets. The decrease in lymphocyte adhesion to the sinusoids, observed in the P-selectin-deficient mice, is correlated in time (2 h) with the expression of P-selectin on platelets in wild-type animals. No differences in the number of adhering lymphocyte to the sinusoids in wild-type and P-selectin-deficient mice were observed at 6 h after treatment when P-selectin was absent on platelets. Therefore, the presence or the absence of P-selectin expression on platelets may determine the lymphocyte adhesion to the sinusoids. Several reports show that the interaction of P-selectin on activated platelets with its ligand(s) on leukocytes induces integrin-mediated leukocyte adhesion to the endothelium [41 , 42 ]. Moreover, P-selectin expressed on activated platelets enhances lymphocyte binding to endothelial cells [6 ]. Thus, the absence of platelet P-selectin may affect the adhesion of lymphocytes to sinusoids, which do not express P-selectin.
We emphasize the role of P-selectin in the generation of intrasinusoidal hemostasis caused by Con A treatment. This process includes erythrocyte agglutination, platelet deposition, and degranulation and lymphocyte/neutrophil adhesion, contributing to the development of hepatic damage [25
]. TNF- and IFN- are also involved by inducing sinusoidal alterations and serotonin-mediated sinusoidal constriction [25
]. After hemostasis, confluent hepatic necrosis occurs within the congested area of the liver parenchyma. Anticoagulant (heparin) and antiserotonin (cyproheptadine) agents mitigate hemostasis and liver injury [25
]. We measured the sinusoidal area occupied by cell aggregates in Con A-induced hepatitis. The occlusion of hepatic sinusoids by leukocyte and erythrocyte aggregates in the P-selectin-deficient mice significantly decreases after 6 h of Con A treatment. Moreover, the blocking P-selectin mAb also reduces intrasinusoidal occlusion.
P-selectin-deficient and wild-type mice showed equivalent numbers of platelet aggregates in the liver sinusoids. Therefore, the differences in sinusoidal congestion are not a result of the impaired platelet aggregation. This was unexpected, as P-selectin is involved in platelet aggregation [43 ]. Moreover, alterations in hemostasis have been reported in P-selectin-deficient mice [44 ]. In a recent study, platelet sequestration in the liver was significantly reduced in P-selectin-deficient mice following ischemia reperfusion [17 ]. We cannot role out that other P-selectin-dependent functions on platelets affect the process of hemostasis and participate in the occlusion of sinusoids.
Between 24 and 48 h, leukocyte infiltration into perivascular areas significantly decreased in the absence of P-selectin. Leukocyte infiltration around hepatic venules was less affected 48 h after Con A, indicating that other adhesion molecules were able to compensate the absence of P-selectin. This infiltration is mostly formed by mononuclear cells, as the percentage of infiltrating neutrophils was always lower than 7%. The percentage of the different mononuclear leukocytes population including the ratio CD4/CD8 T lymphocytes in wild-type and P-selectin-deficient mice was similar. A significant percentage of monocytic (CD11b+) cells was detected. However, no differences in the percentage of CD11b+ cells were observed between the two groups of animals.
No overexpression of E-selectin was observed in the P-selectin-deficient mice. The expression of these selectins on hepatic venules indicates that E-selectin would not be able to compensate the deficiency of P-selectin, as it appears later. It is conceivable that other adhesion molecules, including E-selectin, VCAM-1, and ICAM-1, participate in the rolling and adhesion leukocytes to the liver in this experimental model [27 ], as the number of adhering cells is not completely reduced in the P-selectin-deficient mice. The up-regulation of these adhesion molecules may be a secondary effect caused by the cytokines released from activated T cells [28 ]. Thus, the impaired P-selectin-dependent lymphocyte adhesion at early times may dramatically affect the lymphocyte infiltration observed at a later time.
Lymphocyte infiltration through the vascular endothelial barrier and the ischemia produced by the obstruction of sinusoids are thought to be the main causes of the liver necrosis [21 , 22 , 25 ]. Our study confirmed that the reduction in the extent of lymphocyte migration and intrasinusoidal hemostasis avoids hepatic necrosis. Few if any necrotic foci developed in animals lacking P-selectin. Therefore, transaminase levels after Con A injection are lower in P-selectin mice than in the wild-type, showing that P-selectin-deficient mice were less susceptible to Con A-induced liver damage. Anti-P-selectin mAb also attenuated liver injury, as measured by transaminase levels. This finding further supports the role of P-selectin in Con A-induced hepatic injury.
This study is the first to show the critical role of P-selectin in T cell-dependent hepatic damage using an in vivo experimental model. Although the function of P-selectin expressed on activated endothelial cells and platelets is complex, our results show that P-selectin is a potential, therapeutic target in inflammatory pathologies in which T cell infiltration causes tissue damage.
Received October 18, 2001; revised March 1, 2002; accepted March 18, 2002.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  A. Kawasuji, M. Hasegawa, M. Horikawa, T. Fujita, Y. Matsushita, T. Matsushita, M. Fujimoto, D. A. Steeber, T. F. Tedder, K. Takehara, et al. L-selectin and intercellular adhesion molecule-1 regulate the development of Concanavalin A-induced liver injury J. Leukoc. Biol., April 1, 2006; 79(4): 696 - 705. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Thorlacius and D. Klintman Response to the letter by Dr. Jaeschke J. Leukoc. Biol., December 1, 2004; 76(6): 1091 - 1092. [Full Text] [PDF]
|
|
|
|
 |
|
 C. S. Bonder, M. N. Ajuebor, L. D. Zbytnuik, P. Kubes, and M. G. Swain Essential Role for Neutrophil Recruitment to the Liver in Concanavalin A-Induced Hepatitis J. Immunol., January 1, 2004; 172(1): 45 - 53. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
