| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Published online before print May 13, 2005
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
,
,1
* Beirne Carter Center for Immunology Research and Departments of
Microbiology and
Pathology, University of Virginia, Charlottesville
1 Correspondence: Department of Microbiology and Beirne Carter Center, University of Virginia, HSC Box 801386, Charlottesville, VA 22908. E-mail: ysh5e{at}virginia.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: intrahepatic lymphocytes • FasL
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Hepatitis C virus (HCV)-induced chronic liver disease significantly increases the risk of developing cirrhosis and hepatocellular carcinoma (HCC) [8
, 9
]. Livers from HCV-infected individuals feature massive T cell infiltration. Most of these infiltrating T cells have an activated T helper cell type 1 phenotype but are neither HCV-specific nor able to clear the virus [10
]. These liver-infiltrating lymphocytes could contribute to the liver damage associated with chronic hepatitis C through a yet-undefined mechanism [11
]. In addition, this high incidence of chronic HCV infection (i.e., over 80% of acutely infected patients progress to a chronically infected state) implies an effective viral evasion mechanism of the host immune system. Immune evasion may partly reflect direct effects of HCV gene product(s) on the function of immune cells. Indeed, HCV is capable of infecting and possibly replicating in immune cells, including T cells [12
13
14
15
16
]. In addition, the HCV core reportedly possesses an immunomodulatory function, characterized by alteration of T cell responsiveness [17
18
19
]. Recent data indicate that even after treatment with interferon-
(IFN-) and ribavirin, cells of lymphoid origin remain chronically infected [20
, 21
].
It is interesting that the HCV core protein is able to activate the interleukin (IL)-2 promoter by stimulating the nuclear factor of activated T cells (NFAT) pathway [22
, 23
] as well as activation of NF-
B [24
25
26
] and activated protein-1 [25
, 27
]. As a result of altered T cell activation, the expression of the HCV core in T cell lines or in transgenic mice led to increased apoptosis and altered T cell response [28
]. Lymphocytic infiltration in the liver was observed in the CD2/core mice, where the expression of core protein was directed to T cells using the CD2 promoter. In these transgenic animals, the histopathologic changes in the liver resemble the liver pathology observed in chronic hepatitis C infection in humans [28
]. Furthermore, several studies have provided evidence that apoptosis of peripheral T cells is enhanced in chronically infected HCV patients [29
30
31
] and that the extent of peripheral T cell activation and apoptosis is associated with the severity of liver injury in these patients. It remains unclear to which degree bystander killing and viral-mediated apoptosis play in these observations.
Currently, the mechanisms responsible for T cell-mediated liver damage are poorly understood. It has been reported previously that activation of T cells results in the up-regulation of cell adhesion molecules, such as lymphocyte function-associated antigen-1 [32 , 33 ] and various chemokine receptors [34 , 35 ]; these are important to facilitate trafficking of activated T cells to the liver. Indeed, studies with the T cell receptor (TCR) transgenic mice, OT-1 and DO11.10, which recognize major histocompatibility complex (MHC) classes I and II-restricted ovalbumin (OVA) epitopes, respectively, indicate that activated T cells traffic to the liver following antigenic stimulation. It is interesting that antigen (OVA)-specific CD8+ T cells from OT-1 TCR mice are capable of trafficking to the liver and cause significant liver damage [36 ]. In contrast, although there is some increased migration of CD4+ T cells to the livers of DO11.10 mice, these CD4+ T cells were unable to induce liver damage. These results suggest an intrinsic difference between CD4+ and CD8+ T cells, possibly in the expression of a cell-surface molecule(s) crucial for the interaction of T cells with hepatocytes. Indeed, the kinetics of FasL expression on antigen encounter has been reported to be different between CD4+ and CD8+ T cells [37 ]. Given the role of the HCV core in the dysregulation of T cell activation, it may be possible that the expression of the HCV core in CD4+ T cells alters the expression of a critical cell-surface molecule on CD4+ T cells, enhancing the capacity of these T cells to induce liver damage.
To test this possibility and further identify molecule(s) crucial to T cell-mediated hepatocyte damage, we bred CD2/core transgenic mice with OVA-specific TCR mice (DO11.10). These mice are designated as core(+)TCR and core(–)TCR, according to the presence or absence of HCV core protein expression in DO11.10 TCR mice. Using these mice, we examined recruitment of core-expressing, OVA-specific CD4+ T cells to the liver and induction of liver damage after antigen (OVA) administration in vivo. In the absence of peptide administration, we observed no lymphocytic infiltration in the portal tract of the liver or elevation of alanine transaminase (ALT) level in core(+)TCR or core(–)TCR mice. After OVAII peptide administration, substantial lymphocytic infiltration of the liver was seen in core(+)TCR mice accompanied by a significant and prolonged elevation of serum ALT levels. No significant lymphocytic infiltration or hepatocyte damage was seen in core(–)TCR mice. In addition, phenotypic characterization of the liver-infiltrating T cells in the core(+)TCR mice revealed that intrahepatic T cells represent activated CD4+ T cells with increased expression of cell-surface CD44 and FasL. We also demonstrated a direct role of core-expressing CD4+ T cells in the induction of liver damage by adoptive transfer. The adoptive transfer of core-expressing CD4+ T cells into severe combined immunodeficiency (SCID) mice resulted in marked infiltration of OVA-specific CD4+ T cells into the liver and associated with elevated serum ALT levels. It is important that anti-FasL antibody treatment of these mice abrogated the liver damage by intrahepatic T lymphocytes. This suggests that enhanced FasL expression by liver-infiltrating lymphocytes plays an important role in the induction of liver inflammation.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Administration of OVA peptide into core TCR-transgenic mice and isolation of lymphocytes from liver and spleen
Peptides from chicken OVA323–339 (ISQAVHAAHAEINEAGR; OVAII) and I52 (ASFEAQGALANIAVDKA) were produced at the University of Virginia Biomolecular Research Facility and screened for endotoxin using a Limulus amoebocyte lysate kit (Charles River Laboratories, Wilmington, MA). Mice received a daily intravenous (i.v.) injection of 25 nmole peptide in 250 µL phosphate-buffered saline (PBS) solution for 2 consecutive days (see Fig. 1
).
|
Antibodies and flow cytometry
Monoclonal anti-mouse-conjugated CD3e, CD4, CD8, CD44, CD49d, and CD62L antibodies, purified and biotinylated hamster anti-mouse CD95L (FasL), CD16/CD32 [Fc receptor for immunoglobulin G (IgG)]III/II] Fc block, and appropriate isotype control antibodies were purchased from BD PharMingen (San Diego, CA). Antimurine allophycocyanin-conjugated TCR antibody (KJ1-26), streptavidin-fluorescein isothiocyanate, phycoerythrin (PE), and appropriate isotype control antibody were purchased from Caltag Laboratories (Burlingame, CA), while purified anti-CD4, antiperforin (clone JAW246), and PE-TNF- and were purchased from eBioscience (San Diego, CA). Additionally, biotin- and PE-labeled donkey F(ab)2 anti-rabbit, anti-rat, and anti-goat secondary antibodies were purchased from Jackson ImmunoResearch Laboratories Inc. (West Grove, PA).
Single-cell suspensions were prepared from spleen and liver as described above. Then cells were resuspended in fluorescein-activated cell sorter (FACS) media (containing 1% fetal bovine serum and 0.1% sodium azide) and incubated for 15 min on ice with anti-mouse Fc block. Subsequently, isotype controls or primary antibodies were added and incubated for 30 min. The cells were washed with FACS media and were fixed in 1% formaldehyde/PBS solution and stored at 4°C until they were analyzed with a FACSCalibur flow cytometer (Becton Dickinson, San Diego, CA). In the case of detection of CD95L (FasL), a similar procedure was followed as above, but a three-step, indirect, biotinylated staining was performed using primary antibody, followed by biotinylated secondary and streptavidin-PE. Two antibodies that were specific for active-caspase-3, from BD PharMingen and Cell Signaling (Beverly, MA), were used to monitor for apoptosis. Cells were stained for surface markers, then fixed and permeabilized with Cyto-Fix/Perm (BD PharMingen), and then stained for active caspase-3. Although the BD PharMingen antibody was labeled directly, the Cell Signaling antibody required the use of an anti-rabbit F(ab)2-PE secondary.
For intracellular TNF- staining, the cells were removed after in vivo stimulation and then restimulated with phorbol 12-myristate 13-acetate and ionomycin (Sigma Chemical Co.) in the presence of golgiplug (BD PharMingen) for 6 h. Then, surface markers were stained, and the cells were fixed and permeablized with the Cyto-Fix/Perm kit and stained subsequently with anti-TNF-. For perforin staining, in vivo-activated cells were isolated and fixed in formaldehyde and then stained with purified antiperforin followed by anti-rat F(ab)2-PE secondary. Cells were then stained for surface markers. Data from flow cytometry experiments were analyzed with FlowJo (Tree Star Inc., Ashland, OR) software. Negative controls were set using isotype-matched Ig.
111Indium labeling and dual-modality imaging
Purified 107 CD4+ T cells (purity 92–98%; mean 95%) from the spleen of core(+)TCR transgenic mice and core(–)TCR transgenic littermates were washed in unsupplemented RPMI, resuspended gently in 100 µL Hanks’ balanced salt solution containing 50 µg tropolone (Sigma Chemical Co.) and 200 µCi (7.4 MBq) 111Indium (ICN, Woburn, MA), and incubated for 15 min at room temperature. Cells were washed twice with RPMI containing 10% fetal calf serum to remove unbound radioactivity, resuspended in 300 µL plain RPMI, and injected via tail vein into BALB/c recipient mice following administration of ketamine (125 µg/g body weight, Ketlar, Parke Davis, NJ) and xylazine (12.5 mg/g body weight, Phoenix Scientific, St. Joseph, MO). OVAII was administered into recipient mice at 1 h and 25 h after adoptive transfer. Dual-modality whole-body imaging was done at the time-points shown following cell transfers using a charge-coupled, device-based X-ray detector and a small field of view (FOV) 
camera [39
]. The common FOV of the X-ray detector and detectors was 
7.5 cm2. The X-ray and -ray detectors have been designed and built with spatial resolution that is appropriate for small animal research (0.05 mm and 1.8 mm pixel sizes, respectively). Following image acquisition, the interactive data language imaging software was used to merge the X-ray and -ray images to obtain a single, coregistered image containing structural (X-ray) and functional (-ray) information. Quantitative analysis of radioactivity in the spleen and liver was done on the merged images by defining the region of interest corresponding to the anatomical location of these organs.
Carboxyfluorescein diacetate succinimidyl ester (CFSE) labeling
Naïve or activated lymphocytes were isolated from the spleen. Cells were labeled with 5 µM CFSE (Molecular Probes, Eugene, OR) for 10 min at 37°C. Cells were washed to remove any unincorporated CFSE and resuspended at 2 x 106 cells/mL and stimulated with various concentrations of OVAII or I52 for various times. Cells were then collected, stained for surface markers, and analyzed by flow cytometry. Proliferation analysis was performed using Modfit software from Verity Software House (Topsham, ME).
Adoptive transfer of intrahepatic T cells into SCID mice and administration of anti-FasL antibody
Core(+)TCR transgenic mice and core(–)TCR transgenic littermates were injected with two doses of 25 nmoles OVAII in 250 µl PBS, and livers were harvested 3 days after the last peptide injection. Liver lymphocytes were purified, and SCID, BALB/c, or core(+)/(–)TCR mice were injected with 1, 2, or 4 x 106 cells of liver lymphocytes or media (control). The livers were harvested from recipient mice on days 1 and 2 after adoptive transfer for histology and FACS analysis. For the anti-FasL-blocking experiment, recipient mice were injected with 2 x 106 core(+)TCR liver lymphocytes. At the time of adoptive transfer, 200 µg anti-mouse FasL antibody (MFL3) or hamster IgG control antibody or PBS (all with no azide and low endotoxin) was simultaneously injected into SCID recipient mice. Additionally, core(+)TCR mice were treated with anti-FasL antibody by injecting 100 µg anti-FasL antibody or isotype control at the time of OVAII stimulation.
Histological evaluation, deoxyuridine triphosphate nick-end labeling (TUNEL) assay, and immunohistochemistry
Tissue samples were fixed in 10% buffered formalin solution for 24 h and then embedded in paraffin. To evaluate liver damage, 5 µm tissue sections were cut, and conventional hematoxylin and eosin (H&E) staining and TUNEL assay were performed. To detect fragmented DNA in situ, TUNEL assay was assessed on livers, according to the manufacturer’s instructions (Roche Applied Science, Indianapolis, IN). The detection of proliferating cells was performed by staining the 5-µm-thick, paraffin-embedded tissue sections with anti-Ki67 antibody. After deparaffination, sections were treated with microwave irradiation in citrate buffer (pH 6) and blocked with 2% bovine serum albumin and 10% donkey serum followed by an avidin and biotin block (Vector Laboratories, Burlingame, CA). Sections were incubated with rat anti-mouse Ki67 (clone: TEC3, Dakocytometrics, Carpinteria, CA) overnight at 4°C. Tissues were washed, and the endogenous peroxidase activity was blocked with 3% H2O2 and then incubated with biotin-conjugated donkey F(ab)2 anti-rat. The sections were washed and incubated with streptavidin-peroxidase complex (Vector Laboratories), followed by development with Nova Red substrate (Vector Laboratories), according to the manufacturer’s instructions, and then counterstained with hematoxylin. For the detection of Fas in liver tissue, the sections were prepared as above with the following alterations: Antigen retrieval was preformed with microwave irradiation using a basic retrieval solution (R&D Systems, Minneaspolis, MN), and then sections were blocked with donkey serum followed by avidin and biotin block. Sections were stained with anti-Fas goat polyclonal (R&D Systems) overnight at 4°C.
For CD4 immunohistochemistry, sections were placed in optical cutting temperature compound and immediately frozen on dry ice. Sections (5 µm) were cut and air-dried and then fixed with acetone for 10 min at 4°C. Sections were rehydrated and then blocked with normal serum, followed by avidin and biotin. Sections were incubated with anti-mouse CD4 overnight at 4°C. Tissues were washed and treated with biotin-conjugated donkey F(ab)2 anti-rat. The endogenous peroxidase activity was then blocked with 0.3% H2O2 and 0.1% NaN3 in Tris-buffered saline. The sections were washed and incubated with streptavidin-peroxidase complex, followed by development with Nova Red substrate, and then counterstained with hematoxylin. Two blinded observers analyzed stained sections. For each tissue section, five 400x high-power fields (HPF) were analyzed, and positive hepatocytes and nonhepatocytes were enumerated.
Statistical analysis
A paired Student’s t-test or ANOVA was used to evaluate the significance of the differences. Statistical analysis was performed with the SPSS Version 11.5 (SPSS Inc., Chicago, IL), and a value of P < 0.05 was regarded as statistically significant.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
To examine the recruitment of HCV core-expressing CD4+ T cells to the liver, we administered 25 nmole OVAII peptide or the irrelevant peptide I52, i.v. into core(+)TCR transgenic and core(–)TCR littermate mice (Fig. 1A
). At various time-points after the last injection, organs were harvested and stained with H&E. The tissue histology shown is representative of at least five independent experiments per time-point. We examined the accumulation of lymphocytes in the periportal and lobular regions of the liver after in vivo antigen stimulation and compared the degree of liver lymphocyte accumulation between core(+)TCR and core(–)TCR. As shown in Figure 1B
, the liver histology of core(+)TCR and core(–)TCR mice was normal prior to OVAII administration. Consistent with previous studies [36
], no significant infiltration was seen in core(–)TCR mice after peptide injection. However, the livers of the core(+)TCR mice showed increasing lymphocytic infiltration specifically around the periportal area (as indicated by arrows). The lymphocytic infiltrate involved multiple periportal areas as well as foci of inflammation in lobular areas of core(+)TCR mice. Specifically, we can see that hepatocytes at the interface of the lymphocytes have been damaged by their eosinophilic appearance. Additionally, although the livers from core(–)TCR mice demonstrated a limited increase in lymphocytes, these lymphocytes did not form aggregates, and only sections from the core(+)TCR mice demonstrate the foci of inflammation involving several lymphoid cells. The administration of the irrelevant peptide I52 did not alter liver histology or ALT levels (data not shown). The recruitment of CD4 lymphocytes to the liver is also demonstrated by specific staining of CD4 lymphocytes by immunohistochemistry (Fig. 1C)
. The presence of CD4+ cells was found to localize in areas of lymphocytic infiltration and hepatic damage.
To confirm the increased infiltration of lymphocytes seen in the histological studies, we isolated and determined the number of viable intrahepatic lymphocytes (IHLs) from core(+)TCR and their core(–)TCR littermates. Following T cell stimulation, a significantly greater lymphocytic infiltrate was detected in core(+)TCR mice as compared with their core(–)TCR littermates (Fig. 2A ), and the significant trend was also demonstrated in CD4+ KJ1–26+ lymphocytes infiltrating the liver (Fig. 2B) . In addition, ALT levels were elevated in core(+)TCR mice upon TCR stimulation (Fig. 2C) . It is notable that the kinetics of ALT elevation upon TCR stimulation is similar to the appearance of liver-infiltrating lymphocytes in core(+)TCR mice. Furthermore, the elevation in ALT corresponds to the appearance of damaged hepatocytes as demonstrated in Figure 1B (and see Fig. 4B and 4C ), thus suggesting that the recruitment of activated, core-expressing CD4+ T cells to the liver might be responsible for inducing hepatic damage.
|
|
As shown in Figure 3A , adoptively transferred CD4+ T cells initially localize to the lungs of recipient mice by 1 min after transfer. The labeled cells begin to migrate from the lungs to the spleen by 50 min (yellow outline in Fig. 3A ) after adoptive transfer of cells and have migrated out of the lungs by 4 h after cell transfer. Specifically, as early as 4 h after the adoptive transfer of radiolabeled CD4+ T cells, a significantly larger population of cells begins to appear in the liver (white outline in Fig. 3A ) of recipient mice receiving core(+)TCR cells rather than those receiving core(–)TCR. However, accumulation of labeled lymphocytes within the liver gate became more evident at later time-points (25 h, 48 h) after adoptive transfer. In contrast, as early as 4 h post-adoptive transfer, the trafficking of radiolabeled CD4+ T cells to the liver was less pronounced in recipient mice receiving core(+)TCR T cells without peptide injection and in recipients of core(–)TCR T cells with or without peptide injection. Percentage of radioactivity (i.e., labeled CD4+ T cells) localized in the livers relative to total body radioactivity was 2.5-fold higher in response to OVAII administration to donor mice receiving core(+)TCR T cells than in those receiving core(–)TCR T cells (Fig. 3B) .
|
Increased apoptosis and cellular proliferation in core(+)TCR mice following OVAII peptide administration
To assess whether the elevated ALT level in core(+)TCR mice was associated with increased apoptosis of hepatocytes, a TUNEL assay was performed on liver tissues obtained from core(+)TCR or core(–)TCR mice after OVAII injection. Representative TUNEL sections from at least five independent experiments per time-point (Fig. 4A
4B
4C
4D
) demonstrate a statistically significant, higher incidence of apoptosis in the IHL population and the hepatocytes of core(+)TCR mice as compared with their core(–)TCR littermates. This confirms the correlation between liver damage and elevation of ALT levels (Fig. 2C)
. Most of the apoptotic hepatocytes were located in the periportal areas along with liver-infiltrating lymphocytes in core(+)TCR mice, similar to the damaged hepatocytes seen in Figure 1B
. However, only scattered apoptotic hepatocytes and lymphocytes were detectable in core(–)TCR mice. In contrast, there was no significant difference in apoptotic cells in the spleen of core(+)TCR and core(–)TCR littermates (Fig. 4A)
. In addition, ex vivo FACS staining for active caspase-3 in splenic and hepatic lymphocytes demonstrated an increase in the activation of caspase-3 in the CD4+KJ1-26+ populations from the liver (Fig. 4E)
. In agreement with the in situ TUNEL assay, the active capase-3 stain suggested an increase in the percentage and number of lymphocytes undergoing cell death in the liver of the core(+)TCR mice. In addition, this increase in cell death was only observed in the liver and not the spleen, suggesting a compartmentalization of cell death. In vitro, OVA stimulation of splenic T cells further demonstrates the increased death of dividing cells, as the SSC versus FSC plots demonstrate an increased percentage of lymphocytes transitioning from the live gate to the dead gate after 5 µM OVAII stimulation (Fig. 4E)
. Taken together, the in vitro data demonstrate increased cellular death of activated lymphocytes from core(+)TCR, and the active caspase-3 and in situ TUNEL assays demonstrate that these dying cells localize to the liver of core(+)TCR mice.
It has been reported that apoptosis can also be accompanied by vigorous cellular proliferation. Therefore, we examined cellular-proliferative responses in core(+)TCR mice upon antigenic stimulation by staining spleen and liver tissues with Ki67 antigen, which is expressed exclusively in dividing cells [40 ] and provides a marker for cell proliferation. Core(+)TCR and their core(–)TCR littermates were stimulated with OVAII peptide or irrelevant peptide, and spleen and liver tissues were harvested and stained for Ki67. Figure 5B 5C 5D , demonstrates an increased number of hepatocytes/lymphocytes positive for Ki67 in the liver sections of core(+)TCR mice. In contrast, a slight increase in Ki67+ lymphocytes but not the hepatocytes was detectable in the liver of core(–)TCR littermates. It is interesting that the lymphocytes in core(+)TCR mice were found to be located in foci of inflammation in the periportal and the lobular areas, and the Ki67+ lymphocytes in core(–)TCR are scattered throughout the tissue. However, similar levels of Ki67-positive cells are detectable in the spleen of core(+)TCR and core(–)TCR mice following OVAII stimulation (Fig. 5A) . This suggests that the increased proliferation of lymphocytes in the liver of core(+)TCR mice might be a result of the recruitment of activated, core-expressing CD4+ T cells to the liver following antigen stimulation and further activation/proliferation in the liver environment. In addition, the increase in Ki67-positive hepatocytes indicates an increase in hepatocellular death, as the sounding hepatocytes are regenerating to replace the dead or dying cells.
|
Liver-infiltrating lymphocytes show activated phenotype with enhanced cell-surface expression of FasL but no increased expression of perforin
To characterize the phenotype of liver-infiltrating lymphocytes in core(+)TCR mice, we performed FACS analysis on liver-infiltrating lymphocytes isolated from core(+)TCR transgenic or core(–)TCR littermate mice on day 3 after the last OVAII injection. To establish OVA-peptide specificity of lymphocytes, they were stained with TCR-specific antibody and antibodies for T cell activation markers. As expected, the liver-infiltrating lymphocytes in the core(+)TCR and their core(–)TCR transgenic littermates demonstrated activated phenotype, as determined by decreased expression of CD62L and CD49d and increased expression of CD44. Of specific interest is the increased percentage of CD44hiCD3+KJ1-26+ T cells in the liver of core(+)TCR mice as compared with that in core(–)TCR littermates, 40.6% versus 14.4%, respectively (data not shown). This indicates a considerable influx of OVAII peptide-specific T lymphocytes into the liver. In contrast to the lymphocyte populations in the liver, differences in splenic lymphocyte populations between core(+)TCR and core(–)TCR mice were not significant. In addition, without OVAII administration or with irrelevant peptide, the levels of these activation markers were comparable in core(+)TCR and core(–)TCR mice with a low percentage of CD44+KJ1-26+ cells (data not shown). These findings suggest that core expression in CD4+ T cells leads to a higher frequency of activated CD4+ T cell migration to the liver and rapid accumulation of these T cells in the liver compartment.
As T cell activation is accompanied by up-regulation of FasL, and Fas/FasL engagement plays a pivotal role in inducing apoptosis of hepatocytes [41
42
43
], we determined the expression of Fas and FasL in the lymphocytes recruited to the liver. The expression level of Fas was similar among lymphocytes from the spleen and liver of core(+)TCR and core(–)TCR mice (Fig. 6A
). Additionally, the level of Fas detected by immunohistochemistry demonstrates the increased expression on hepatocytes after antigen stimulation, but no significant difference is detectable between core(+)TCR mice and core(–)TCR littermates (Fig. 6B)
. However, more intense Fas staining was found at sites of inflammation as compared with the noninflamed region during liver inflammation. The lack of baseline differences in Fas expression between core(+)TCR and core(–)TCR mice suggests that the increased apoptosis in the liver of core(+)TCR mice cannot be restricted to the altered expression of Fas. It is important that FasL expression in the liver lymphocytes was elevated in core(+)TCR mice (Fig. 6C) but not in core(–)TCR mice. Therefore, core-expressing T cells may not only exhibit an increased frequency of migration to the liver, but core expression in CD4+ T cells may further alter and enhance the activation state of the CD4+ T cells infiltrating into the liver. Next, we examined two other hepatotoxic molecules, perforin and TNF-. There were no significant differences in the percentage of infiltrating T cells positive for the expression of perforin or TNF- (Fig. 6D)
between core(+)TCR mice and their core(–)TCR littermates. These findings suggest that Fas/FasL-dependent interactions between liver-infiltrating T cells and hepatocytes may be a prominent mechanism of cell death, resulting in the induction of liver damage.
|
|
Blocking effect of anti-FasL antibody on CD4+ T cell-dependent liver inflammation
Hepatocytes express high levels of Fas that are increased under conditions of inflammation and are thus susceptible to Fas-mediated apoptosis. As core-expressing CD4+ T cells express an increased level of FasL, it is likely that the FasL expressed on the cell surface of liver lymphocytes from core(+)TCR mice might be responsible for the observed liver damage. To test this possibility, we attempted to block the Fas/FasL interaction by using a FasL-specific antibody and examined whether anti-FasL antibody inhibited liver damage by liver-infiltrating lymphocytes. Core(+)TCR mice were injected with OVAII twice, and livers were harvested after 3 days to purify IHLs. SCID mice were injected with 2 x 106-purified T cells from the core(+)TCR mice along with media (control), an isotype-matched control antibody, or the anti-FasL antibody, respectively. ALT determination and histology were performed on day 2 after the adoptive transfer of cells from core(+)TCR mice into SCID.
As shown in Figure 8A , although liver-infiltrating lymphocytes were detectable in mice receiving lymphocytes from core(+)TCR mice, treatment with anti-FasL antibody reduced the lymphocytic inflammation. In addition, treatment with anti-FasL reduced hepatic damage, as determined by TUNEL and Ki67 as well as ALT levels, which in mice receiving IHL alone or IHL and control antibody, showed elevated mean ALT levels of 430 and 250 U/L, respectively, indicating significant liver damage. It is interesting that the ALT level in mice receiving the FasL antibody demonstrated significantly lower levels (96 U/L) of hepatocyte damage compared with media alone and control antibody. This could be indicative of a Fas/FasL-related mechanism in the observed liver damage from core(+)TCR intrahepatic CD4+ T cells.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
One molecular mechanism for the increased T cell activation and recruitment of activated T cells to the liver is the HCV core’s ability to activate IL-2/NFAT [22
, 23
] and the NF-B factors [24
, 25
, 44
45
46
]. Although TCR engagement and activation of NF-B result in the production of antiapoptotic genes, TCR engagement and activation of NFAT lead to the production of proapoptotic genes and activation-induced cell death. The activation of these pathways along with TCR engagement and the influence of the liver environment would allow for an increase in proliferation as well as the observed increase in lymphocyte cell death. As a result of these events, the expression of the HCV core in T cells becomes immunosuppressive, and core-expressing T cells also act as an inflammatory stimuli in the liver environment. In addition, the FasL promoter contains NFAT [47
] and NF-B [48
] elements; thus, it is likely that the HCV core is able to up-regulate the expression of FasL. Indeed, the HCV core protein has been reported to activate and up-regulate the expression of FasL in infection and transfected cells [49
].
At present, it remains to be determined how the expression of core protein in CD4+ T cells facilitates trafficking of activated CD4+ T cells into the liver. Several possibilities might explain this finding. First, the expression of the HCV core protein in CD4+ T cells may alter the activation state of T cells and enhance the trafficking of activated T cells to the liver following OVAII administration in vivo. Indeed, as discussed above, the expression of the HCV core has been shown to enhance transcription of the IL-2 promoter of T cells, leading to increased T cell activation/apoptosis [22 ]. Second, the association of the HCV core protein with the cytoplasmic domain of the TNF receptor (TNFR) family in CD4+ T cells may directly increase the susceptibility of these cells to TNFR-mediated apoptosis and facilitate the recruitment of these cells to the liver [28 , 50 ]. Lastly, the expression of core protein in CD4+ T cells may up-regulate the expression of chemokine receptors such as CXC chemokine receptor 3, which has been reported to be responsible for trafficking of peripheral lymphocytes to the liver [51 , 52 ]. We are currently examining the effect of the HCV core protein on recruitment of activated CD4+ T cells to the liver and which molecules may be altered through inflammation.
Several reports indicate that the liver is a destination for activated lymphocytes, many of which are destined to undergo apoptosis [53 ]. This would support the possibility that altered T cell activation by HCV core protein might facilitate the trafficking of core-expressing CD4+ T cells into the liver. It is interesting that using CD8 and CD4 transgenic mice has demonstrated that the infiltration of the liver by CD8+ T cells was more pronounced than CD4+ T cell infiltration, and significant liver inflammation was only observed to result from the CD8+ T cell population, suggesting a primarily CD8+ T cell-mediated hepatocyte damage [36 ]. These studies suggest that there might be an intrinsic difference in the ability of activated CD4+ and CD8+ T cells to induce liver inflammation.
The molecular mechanism by which IHLs induce hepatic damage is not well understood. Hepatocytes have been demonstrated to be remarkably sensitive to Fas-induced apoptosis following in vivo administration of anti-Fas antibody [54
]. However, studies of lymphocyte-mediated liver inflammation have demonstrated different results, depending on which agent was used for the induction. Liver damage by concanavalin A-stimulated T cells has been reported to be mediated through the release of perforin [55
] as well as IFN- [56
, 57
] but was not Fas-dependent. In addition, in a murine model of hepatitis B infection, there was little involvement of perforin in hepatocyte damage by CD8+ T cells [41
], and the FasL expression by CD8+ T cells was found to be responsible for the resulting liver damage. The acute activation of CD8+ T lymphocytes in OT-1 transgenic mice also demonstrates the requirement of FasL for the induction of acute hepatic damage [58
]. Thus, the actual levels of cytolytic activity or expression of cytolytic mediators from the infiltrating lymphocytes may be a determinant for induction of liver damage.
In this report, we demonstrated clearly that the HCV core up-regulates the expression of FasL on CD4+ T cells and induces CD4+ T cell-dependent hepatocyte damage. This finding is consistent with a recent report that HCV core increases the expression of FasL in hepatocytes [49 ]. In addition, we showed a crucial role of FasL displayed on the cell surface of intrahepatic CD4+ T cells in hepatocyte damage by abrogation of liver inflammation with the blockade of Fas/FasL interaction. These data provide compelling evidence that Fas/FasL interaction plays a pivotal role in the induction of hepatocyte damage by activated, intrahepatic CD4+ T cells. It is likely that hepatocyte damage by intrahepatic CD4+ T cells represents a bystander killing mechanism for liver injury. Specifically, antigen presentation by hepatocyte, which does not express MHC class II molecules, may not be necessary for the trafficking of lymphocytes to the liver or the subsequent hepatocyte damage induced by liver-infiltrating lymphocytes. This model might also serve to explain the liver dysfunction that is often observed after situations in which the immune system has been strongly activated outside the liver, such as a superantigen in toxic shock syndrome [59 , 60 ].
Fas/FasL engagement induces hepatocyte damage through the association of Fas-associated death domain with caspase 8, leading to activation of downstream caspase molecules, which mediate apoptosis [61 62 63 ]. It is interesting that FasL has been reported to induce inflammation by promoting chemokine secretion and facilitating recruitment of inflammatory cells (macrophages, dendritic cells) [64 , 65 ]. In addition, FasL contains the costimulatory function to induce the proliferation of CD8+ T cells to a greater extent than that of CD4+ T cells [37 ]. It is striking that blockade of FasL abolished the development of HCC, suggesting that FasL plays a pivotal role in inducing liver cancer by providing an environment of chronic inflammation and promoting the secretion of cytokines associated with the development of fibrosis [66 ]. In HCV-related liver disease, liver histology is characterized by the presence of inflammatory cells in the portal area and enhanced Fas-mediated hepatocyte damage [67 ]. Although IHLs that recognize the viral antigen on hepatocytes become activated and express cytolytic FasL molecules, hepatocytes exhibit enhanced Fas expression and become susceptible to FasL-mediated cell death by a bystander killing mechanism [68 , 69 ]. In addition, under conditions of inflammation, Kupffer cells exhibit increased expression of FasL and other proinflammatory molecules, which can further induce hepatocellular death [70 ]. The further involvement of T lymphocytes in hepatic inflammation and damage has been demonstrated through correlation of peripheral apoptosis to liver injury and inflammation [29 ]. Specifically, the expression of caspase-3 (active) correlates with the degree of liver inflammation in HCV-infected patients [71 ]. Thus, further studies about liver inflammation using core(+)TCR mice upon OVAII peptide administration in vivo will help to understand a mechanism(s) for the role of FasL displayed on IHLs in bystander killing of hepatocytes. Our finding suggests that a low level of HCV infection in T cells with concomitant HCV core expression in T cells may alter the activation state of T cells and result over time in chronic, progressive liver injury observed in patients persistently infected with HCV.
In summary, we have demonstrated that HCV core-expressing CD4+ T cells traffic to the liver upon activation by OVA peptide administration in vivo and that these intrahepatic CD4+ T cells contain a direct effector function, which damages the liver though Fas/FasL engagement. These results provide a rationale for designing a therapy for hepatitis using anti-FasL antibody or inhibitors of the Fas-mediated apoptotic pathway.
|
ACKNOWLEDGEMENTS |
|---|
Received January 7, 2005; revised March 22, 2005; accepted April 13, 2005.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
24+ T cells and rapid elimination of effector cells by apoptosis Eur. J. Immunol. 28,3448-3455[CrossRef][Medline]
B-dependent signaling through tumor necrosis factor receptor-associated factor J. Biol. Chem. 276,16399-16405B in hepatitis C virus infection: implications for pathogenesis and hepatocarcinogenesis Hepatology 31,656-664[CrossRef][Medline]
