| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
B/AP-1 signaling in human intrahepatic biliary epithelial cells
The Liver Research Laboratories, MRC Centre for Immune Regulation, and
The Department of Pathology, The University of Birmingham; and
The CRC Institute for Cancer Studies, Institute of Clinical Science, Queen Elizabeth Hospital, Birmingham B15 2TH, UK
1Correspondence: The Liver Research Laboratories-MRC Centre for Immune Regulation, The University of Birmingham Institute of Clinical Science, Queen Elizabeth Hospital, Edgbaston, Birmingham B15 2TH, UK. E-mail: s.c.afford{at}bham.ac.uk
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
B and AP-1 were both activated after CD40 ligation. Increased functional NF-B was seen early after CD40 ligation, but returned to baseline levels after 4 h. In contrast, the rapid up-regulation of AP-1 was sustained over 24 h. This study provides further functional evidence of the ability of CD40 to induce Fas/FasL-dependent apoptosis of liver epithelial cells supporting the importance of cross-talk between tumor necrosis factor (TNF) receptor family members as an amplification step in apoptosis induction. Sustained activation of AP-1 in the absence of NF-B signaling may be a critical factor in determining the outcome of CD40 engagement.—Afford, S. C., Ahmed-Choudhury, J., Randhawa, S., Russell, C., Youster, J., Crosby, H. A., Eliopoulos, A., Hubscher, S. G., Young, L. S., Adams, D. H. CD40 activation-induced, Fas-dependent apoptosis and NF-B/AP-1 signaling in human intrahepatic biliary epithelial cells.
Key Words: CD40-mediated apoptosis • Fas-mediated apoptosis • primary biliary cirrhosis • NF-B and AP-1 activation
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 2)
. These diseases are characterized by a loss of intrahepatic bile ducts associated with a portal infiltrate of T cells and macrophages. The mechanisms by which BEC are destroyed are not known, but recent reports suggest that apoptosis is responsible (3)
. Cytolytic T lymphocytes (CTL) can induce apoptosis in target cells by activation of the apoptosis receptor Fas (CD95, Apo-1) or via the granzyme/perforin pathway (4)
. Studies have shown increased expression of Fas on BEC in PBC and increased local expression of granzymes (3)
, suggesting that both Fas-dependent and independent pathways are involved. Recent studies using a murine model of GVHD also highlight the importance of Fas-mediated death of BEC (5)
.
The Fas pathway is thought to be a key mechanism for hepatocyte death in the liver (6
7
8
9)
, but there are no studies demonstrating a functional role for Fas in the apoptosis of human BEC. Hepatocytes undergo apoptosis when cell surface Fas is activated with either cross-linking antibody or Fas ligand (FasL) -bearing effector cells (10)
. Recent work from our laboratory suggests that the mechanism can be more complex, involving cooperation between Fas and another member of the TNF receptor superfamily, CD40 (11)
. Activation of CD40 is associated with a broad range of biological effects depending on the cell type and nature of the stimulus (12)
. For example, CD40 on antigen-presenting cells provides a vital signal for T cell activation and proliferation, whereas activation of CD40 on endothelium results in a proinflammatory signal that leads to chemokine secretion and increased adhesion molecule expression (13)
. CD40 is also expressed on malignant epithelial and melanoma cell lines, where its activation can affect cell proliferation and survival (14
, 15)
. We have demonstrated that treatment of human hepatocytes with monoclonal antibody (mAb) to CD40 or soluble trimeric CD40 ligand (gp39, CD154) can trigger Fas-dependent hepatocyte death in the absence of any cofactors (11)
. CD40 and Fas were coexpressed on hepatocytes undergoing apoptosis during allograft rejection, suggesting that this mechanism is important in liver diseases where epithelial cell loss occurs.
The cellular response to CD40 stimulation depends on the nature and profile of transcription factor activation. The downstream consequences of CD40 ligation are complex and depend in part on the cell types studied. Nuclear factor kappa B (NF-B) and activator protein 1 (AP-1)/JNK are two transcription factors that have been be implicated in control of epithelial cell proliferation or apoptosis with the cell fate ultimately determined by the level of activation of pro- or anti-apoptotic genes with NF-B- and AP-1-sensitive promoters (16)
.
It was against this background we investigated the expression and function of Fas, CD40, and their cognate ligands in BEC. We demonstrate for the first time that CD40 engagement on human BEC causes activation of NF-B and AP-1 resulting in Fas-dependent apoptosis, suggesting that CD40 ligation by CD40L (CD154) -bearing T cells or macrophages could amplify bile duct loss in immune-mediated liver diseases such as PBC.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Immunohistochemical studies of human liver tissue
Immunohistochemistry was performed as described (11)
on 5 µ cryostat sections using the following mouse mAbs: anti-CD40 (IgG1 isotype clone G28.5, obtained as a gift from Dr. J. Pound, Immunology, University of Birmingham, UK); anti-Fas (IgG1 isotype clone SC8009 obtained from Santa Cruz Biologicals, Santa Cruz, CA); anti-FasL (IgG1 isotype clone NOK-1 from Becton Dickinson UK Ltd., Cowley, Oxford, UK). Antibodies to CD40, Fas, or FasL were used at a dilution of 1:50 and detected using a commercially available biotin/streptavidin/horseradish peroxidase detection system according to the manufacturer’s protocol (Vectastain Elite ABC kit obtained from Vector Laboratories, Burlingame, CA). Antibody binding was visualized using diaminobenzidine tetrahydrochloride substrate. Sections for each specimen, where the primary antibody was substituted for isotype-matched control immunoglobulin were included as negative controls.
The presence of apoptotic biliary epithelial cells was assessed in normal and PBC tissue sections (n=6 for each group) using the in situ DNA end-labeling method described (17)
.
Sections of PBC liver tissue were also subjected to dual immunofluorescent staining to identify potential CD40 ligand-bearing cells. T cells and macrophages were identified using the standard immunohistochemical protocol (11)
with rabbit anti-CD3 (Dako Ltd., Cambridge, UK) or a mAb specific for CD14, respectively (Dako), followed by Texas red-conjugated goat anti-rabbit IgG or goat anti-mouse IgG1 (Dako), as appropriate. CD40L was simultaneously identified using an IgG2a subclass anti-CD40L (11)
, followed by an FITC-conjugated goat anti-mouse IgG2 subclass-specific secondary antibody (Dako). In each case, controls were included where primary antibody was omitted or substituted for nonimmune serum or isotype-matched control immunoglobulin. Confocal imaging was used to identify positively stained cells.
Detection of CD40L expression in liver-derived macrophages and T cells
Liver-derived macrophages and lymphocytes were generated as a by-product of BEC isolation described below. The protocol for monocyte/macrophage and T cell purification used positive immunomagnetic selection with specific mAb as the primary antibodies, thereby minimizing cross contamination of cell preparations. The collagenase digest was washed with EDTA-PBS (x3) layered onto Lymphoprep (Life Technologies, Inc., Paisley, Scotland) and centrifuged for 30 min at 400 g. The mononuclear cell interface layer was collected, washed three times with EDTA-PBS, and incubated with 0.5 µg/ml of mAb to CD14 (for monocytes/macrophages; Serotec, Kidlington, Oxford, UK). After washing, the CD14+ve cells were selected by immunomagnetic separation with anti-mouse IgG1-coated Dynabeads (Dynal, Liverpool, UK). The procedure was repeated with the residual cells using 1 µg/ml mAb to CD3 (pan T cell marker; Dako) and again selected using immunomagnetic separation. The macrophages and T cells were plated into 24-well culture plates in RPMI medium containing 5% v/v heat-inactivated fetal calf serum, 2 mM L-glutamine, and 100 µ/ml penicillin and streptomycin. Cells were incubated for 24 h and RNA was prepared as described for BEC (see below). Specific mRNAs for CD40L and ß-actin were detected by reverse transcriptase PCR (RT-PCR) according to the protocol defined below.
Studies of isolated BEC
Biliary epithelial cells were isolated from normal and PBC liver tissue as described (18)
. Liver (30 g) was finely diced and incubated with collagenase type 1A (Sigma, St. Louis, MO). The digest was layered onto a 33% and 77% iso-osmotic Percoll gradient and centrifuged at 500 g for 30 min. The interface layer was collected, washed three times in PBS, and incubated with the BEC-specific mAb to HEA 125 (TCS Biologicals Ltd., Botolph Claydon, Bucks, UK). BEC were positively selected by incubating with anti-mouse IgG1-coated Dynabeads (Dynal) and by magnetic separation. The cells were cultured in plating media: Hams F12, Dulbecco’s Eagle medium, heat-inactivated fetal calf serum (10% v/v), penicillin and streptomycin (100 µ/ml), glutamine (2 mM), epidermal growth factor (10 ng/ml), hydrocortisone (2 µg/ml), choleratoxin (10 ng/ml), tri-iodo-thyronine (2 nM), and insulin (0.124 µ/ml). After 1–2 days in culture, the medium was exchanged for media containing 5%v/v fetal calf serum and 10 ng/ml hepatocyte growth factor (R&D Systems Ltd., Abingdon, Oxford, UK). Cells were expanded in 25 ml tissue culture flasks with regular medium exchanges until they became a confluent monolayer. For passage and harvesting, cells were subjected to gentle trypsinization (5 min in 0.25% bovine pancreatic trypsin in basal media (Sigma), followed by neutralization with medium containing 10% fetal calf serum). In all subsequent experiments, cells were used between passage 2 and 5 depending on the initial yield of the primary isolate.
For the functional experiments cells were seeded at 1.0 x 104/well in 24-well tissue culture plates in media for up to 24 h in the absence or presence 1 µg/ml recombinant CD40L (Alexis Corp., Bingham, Nottinghamshire, UK) 40 ng/ml neutralizing mAb against FasL (NOK-1 from PharMingen, San Diego, CA) or combinations of either CD40L plus NOK-1 or CD40L plus a neutralizing mAb to Fas (clone ANC95.1/5E2-IgG1 isotype anti-human CD95 obtained from Ancell Immunology Research Products, Bayport, MN). In additional experiments, BEC were stimulated with 10 ng/ml recombinant TNF-
(PeproTech EC Ltd., London, UK). At the end of the incubation, cells were harvested, cytospun, and assessed for apoptosis using the in situ end-labeling method (ISEL) staining as previously defined (17)
.
Cell surface expression of CD40, Fas or FasL was assessed by flow cytometry using mAb (11)
. Cytospun preparations of BEC before or after CD40 stimulation were stained by dual immunofluorescence for CD40 and Fas or CD40 and FasL, using the antibodies described earlier to determine distribution of the receptors and ligand within the total population of BEC. Levels of soluble FasL were also assessed in culture supernatants using a commercially available ELISA (R&D Systems Ltd.).
Assessment of expression of CD40, CD40L, Fas, and FasL protein and mRNA in cultured BEC
Additional experiments were carried out in which cells were stimulated with a range of proinflammatory cytokines including TNF- (10 ng/ml), interferon 
(IFN-) (1000 units/ml; PeproTech), or transforming growth factor (TGF-ß; 0.1 ng/ml; PeproTech). BEC were then assessed for Fas, FasL, CD40, or CD40L cell surface expression by flow cytometry using a Coulter XL flow cytometer (11)
. Fas, FasL, or CD40 or CD40L antibodies (previously titrated for use in flow cytometry) were added to the cells for 1 h at room temperature, followed by FITC-labeled rabbit anti-mouse immunoglobulins at a dilution of 1:25 (Dako). Control experiments were included where primary antibodies were substituted for an isotype-matched immunoglobulin.
Biliary epithelial cell CD40/CD40L and Fas/FasL mRNA expression was measured by RT-PCR after incubation of cells with CD40L or cytokines TNF-, IFN-, or TGF-ß for up to 24 h. Total RNA was extracted with Rnasol B solution (Biogenesis Ltd., Bournemouth, UK) according to the manufacturer’s protocol. CD40/CD40L, Fas/FasL, and ß-actin mRNA were detected using RT-PCR). Positive PCR controls were provided by isolated peripheral blood mononuclear cells (PBMNC) stimulated with 10/µg/ml phytohemagglutinin (4 days), 50/µ/ml interleukin 2 (3days), 1 ng/ml phorbol myristate acetate, and 500 ng/ml concanavalin A (1 day). Primer pairs for Fas FasL were obtained from R&D Systems Ltd. Primers for CD40 CD40L were obtained from Stratagene (Cambridge, UK). Primers for ß-actin synthesized by Alta Bioscience Ltd. (Birmingham, UK). Reverse transcription was carried out using a standard protocol incubating the RT mix at room temperature for 10 min, followed by 1 h at 42°C and 5 min at 95°C. Samples were held at 4°C before PCR amplification according to one of the following cycling programs: 1) CD40 and CD40L, 5 min denaturation at 95°C, 5 min annealing at 60°C, followed by 35 cycles of 1 min 55 s at 72°C, 1 min 10 s at 94°C, and 1 min 10 s at 60°C. Final extension was 10 min at 72°C. Samples were then held at 4°C. 2) Fas and FasL, 2 min denaturation at 96°C, followed by 35 cycles of 30 s at 96°C, 1 min at 55°C, and 1 min at 72°C. Final extension was 10 min at 72°C. Samples were held at 4°C. Beta-actin RT-PCR was possible using either protocol. After PCR amplification, samples were electrophoresed in 2% agarose gel and stained with ethidium bromide; the photographic negatives were used for assessment.
Assessment of NF-B and AP-1 activation in cultured BEC
Nuclear extracts were prepared according to standard protocol (19)
from BEC that had been cultured in the presence or absence of either CD40L or TNF- for up to 24 h. The extracts were then subjected to standard electrophoretic mobility gel shift assays (19)
. The protein content of nuclear extracts was determined by the Bicinchoninic acid protein assay (Pierce, Rockford, IL) according to the manufacturer’s instructions. 20 µg of each specimen was incubated with 20,000 cpm of 32P end-labeled NF-B or AP-1 binding consensus oligoduplex probe (Promega, Madison, WI; Santa Cruz Biologicals, respectively) for 30 min on ice. Protein–DNA complexes were separated by electrophoresis in a 5% polyacrylamide gel. Gels were dried under vacuum on Whatman 3MM chromatography paper (Whatman International Ltd., Maldstone, UK) and autoradiographed for 18–24 h before development and examination. The specificity of the bands was confirmed by supershift assays, where hybridized nuclear extracts were incubated with specific rabbit polyclonal antisera to NF-B p65 or AP-1 (Santa Cruz). Each antibody (2 µg) was added 30 min before addition of the DNA probe and kept on ice. Additional controls included incubation with an excess (100-fold) of unlabeled oligonucleotide probe to compete with the radiolabeled probe. Experiments were repeated on at least four separate occasions with different preparations of cells to confirm reproducibility.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
a–c). In PBC liver tissue, increased membranous and cytoplasmic staining for CD40, Fas, and FasL was observed in most of the surviving small intermediate or large ducts in every case (Fig. 1d-i
). Periseptal hepatocytes, sinusoidal endothelium, and portal lymphoid aggregates also stained positively for CD40. Portal vessels were generally negative. No staining was observed where primary mAb was substituted with isotype-matched control immunoglobulin. Apoptotic BEC were not detected in bile ducts in normal tissue but were identified frequently in PBC tissue using ISEL staining. (Fig. 2
a, b, respectively).
|
|
T cells and macrophages in PBC liver tissue express CD40L
Dual immunohistochemistry demonstrated the presence of CD40L-positive cells within the portal tracts in PBC liver tissue. Both CD3+ T cells and CD14+ macrophages within the portal tract were positive for CD40L (Fig. 3
a, b). Freshly isolated unstimulated T cells and macrophages isolated from PBC liver tissue were positive for CD40L mRNA by RT-PCR (Fig. 3c
) confirming that these cell types can transcribe CD40L.
|
BEC in culture express Fas, FasL, and CD40 and ligation of CD40 with CD40L-induced transcriptional up-regulation of FasL
Flow cytometry demonstrated that cultured BEC express Fas, FasL, and CD40 protein constitutively (Fig. 4
A). Dual immunofluorescent staining confirmed the flow cytometry data and demonstrated that CD40, Fas, and FasL were expressed on the same population of cells. CD40L stimulation did not appear to affect the distribution of CD40 or FasL expression within the total BEC population (Fig. 4B
). Culture supernatants were negative for soluble FasL, showing that FasL was not shed after CD40 stimulation (data not shown).
|
Activation of CD40 with CD40L led to a threefold up-regulation of FasL mRNA (Fig. 5
), which was reproduced on four occasions using different cell preparations. RT-PCR confirmed that BEC contained mRNA for CD40, Fas (data not shown), and FasL but were consistently negative for CD40L even after cytokine stimulation (Fig. 5)
.
|
Activation of cells with TNF-, anti-Fas, or anti-FasL had no effect on levels of Fas or CD40 mRNA (data not shown) and antibodies to Fas or FasL had no effect on FasL mRNA (Data not shown). They did, however, effectively prevent induction of apoptosis after stimulation with CD40L (see below).
Cytokine stimulation experiments showed that cell surface CD40 expression was increased modestly from 20.5% positive cells ± SD 7.7 to 27.4% ± SD 7.74 after treatment with TNF- (P<0.002, paired t test). The increase in percentage of positive cells was also reflected in a shift in median channel fluorescence. Incubation with TGF-ß or IFN- did not effect cell surface expression of CD40 over a period 48 h (data not shown).
CD40 activation-induced BEC apoptosis is mediated via Fas/FasL
Activation of cell surface CD40 with CD40L induced apoptosis (Fig. 6
). CD40 engagement was a potent proapoptotic signal, which led to an approximate three- to fourfold increase in the number of ISEL-positive apoptotic cells after 24 h. Addition of the FasL-blocking mAb NOK-1 or the anti-Fas-neutralizing/antagonistic mAb completely prevented the apoptosis induced by CD40L, demonstrating that 1) CD40 apoptosis was FasL dependent, 2) FasL expressed by BEC is functionally active (Fig. 6A-C
), and 3) CD40L-mediated apoptosis was dependent on functional Fas expression.
|
Activation of BEC CD40 results in up-regulation of functional NFB and AP-1
Electrophoretic mobility gel shift assays demonstrated that BEC in culture express low levels of functional NF-B and AP-1 (Fig. 7
A, B (lanes 3), respectively). Incubation with TNF- for 2 h resulted in up-regulation of both NF-B and AP-1 (Fig. 7
A , B
, lanes 2). A similar increase in NF-B and AP-1 was seen after 2 h incubation with CD40L (Fig. 7
A, lane 4; and Fig, 5
b, lane 4). Hybridization of the radiolabeled NF-B oligonucleotide resulted in visualization of multiple bands. The functional p65 moiety of NF-B was identified by hybridization with a specific mAb that resulted in a shift in the specific p65 band, allowing positive identification (Fig. 7
A, lane 5). NF-B and AP-1 probes failed to hybridize in the presence of a 100-fold molar excess of unlabeled oligonucleotide (Fig. 7
A, B
, lanes 6). After 24 h stimulation with CD40L, NF-B activity had diminished to levels similar to unstimulated cells (Fig. 7
A, lane 7; C). AP-1 was maintained over 24 h (Fig. 7
B, lane 5; C). This coincided with the time when large numbers of ISEL-positive apoptotic cells were detected (see Fig. 6
).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. The coexpression of Fas, FasL, and CD40 on bile ducts in patients with PBC suggests that cooperative interactions between these molecules may be involved in apoptosis of BEC in immune-mediated biliary diseases.
Further evidence that these observations are functionally important comes from our studies with cultured human BEC. We have shown for the first time that human BEC express functionally active Fas, FasL, and CD40 in culture and that CD40 expression can be further increased by treatment with TNF-. Induction of CD40 expression in vitro has been demonstrated in keratinocytes (20)
, vascular endothelium (21)
, and aortic smooth muscle cells where TNF- increased CD40 expression via activation of NF-B (22)
. The apparently modest increase in CD40 expression we observed after TNF- stimulation probably reflects the fact that cultured BEC showed comparatively high basal levels of CD40 expression, implying some degree of activation of cells in culture. Nevertheless, other cytokines tested (including TGF-ß and IFN-) had no effect, suggesting that the effect of TNF- was real. In contrast to the reported studies of vascular endothelium, we did not see an increase in CD40 protein or mRNA in BEC in response to either IFN- or TGF-ß, suggesting that CD40 expression may be regulated by different cytokines in different cell types.
We found similar levels of Fas, FasL, and CD40 on cultured BEC isolated from normal or PBC liver but no detectable CD40 or Fas on bile ducts in normal liver in vivo. This probably reflects partial activation of the cells induced by isolation and culture procedures. The expression of FasL in unstimulated BEC (Fig. 4A
) presumably also accounts for the basal levels of apoptosis in the control cells (Fig. 6C
). The cultured cells were in cycle due to stimulation with growth factors in vitro, whereas cells in tissues normally are not proliferating and cells in the proliferative cycle are vulnerable to apoptosis (23)
.
Examination of the dual stained cytospun BEC preparations suggested that CD40, Fas, and FasL proteins were expressed on the same population of BEC in both resting cells and those that had been activated via CD40. The increase in FasL was at the transcriptional level with RT-PCR data, showing a threefold increase in BEC FasL mRNA expression (Fig. 5)
. The increased levels of FasL mRNA were observed at 24 h, suggesting a sustained up-regulation. This may explain why the proapoptotic effects of CD40L are so potent. Thus, CD40 activation might provide an important amplification signal to maintain increased levels of functional FasL and thereby Fas activation. Although our study cannot determine whether CD40-mediated, Fas-dependent apoptosis is an autocrine or paracrine mechanism, the fact that CD40, Fas, and FasL are expressed on the same subset of cells within the population suggests that autocrine or fratricrine mechanisms could be operating. We did not detect soluble FasL, which implies that CD40 stimulation does not result in shedding of FasL protein. That the induction of apoptosis by CD40L could be completely blocked with antagonistic mAb to Fas or FasL shows that CD40-mediated apoptosis in BEC is entirely Fas dependent.
Induction of BEC apoptosis via activation of cell surface CD40 was accompanied by rapid activation of NF-B and AP-1. At 24 h, when significant numbers of apoptotic cells were detected, there was a reproducible loss of the NF-B signal but maintenance of AP-1. The early activation of NF-B may be critical for the induction of Fas-L expression (24)
and may also induce anti-apoptotic mechanisms that protect against apoptosis (25)
. Thus, loss of a sustained NF-B signal could allow AP-1 to act unopposed in order to promote BEC apoptosis via Fas activation by autocrine Fas-L.
Our data demonstrate that activation of CD40 results in direct induction of apoptosis of BEC via a Fas-dependent mechanism. TNFR family members possessing the cytoplasmic death domain (Fas, TNFR1 DR3, DR6, TRAIL-R1 and 2) can trigger cellular apoptosis when activated directly with cognate ligand or agonistic antibodies. Our study provides further evidence that TNF receptors that lack the cytoplasmic death domain can trigger autocrine death by induction of cell-bound ligands, including TNF- and FasL (11
, 14
, 15
, 18
, 26)
.
For CD40 activation to be important in diseases such as PBC requires a cellular source of CD40L to activate CD40 on the BEC cell surface. We found that liver-derived macrophages and T cells in PBC both express CD40L protein; RT-PCR data confirm they synthesize CD40L mRNA. Although the relative contributions of these cell types to CD40L expression in vivo must be interpreted with caution in light of the transient nature of CD40L expression in T cells, our data suggest that macrophages rather than T cells appear to be the dominant source of CD40L because they stained more strongly for the protein in tissue sections (Fig. 3)
. That these cells are found in close proximity to BEC in PBC (27)
suggests that they could present the CD40L required to activate cell surface CD40 in vivo. The exact mechanism of presentation of CD40 ligand in vivo is not known. Our in vitro studies were performed with soluble recombinant CD40L, implying that direct cell–cell contact would not be needed to induce CD40-mediated, Fas-dependent death. However, inflammatory bile duct lesions in vivo are surrounded by leukocytes that disrupt and infiltrate the biliary epithelial cell junctions, allowing them direct access to the lateral and luminal surfaces of the epithelial cells. Thus, either mechanism for CD40 presentation (cell-bound or soluble) may be applicable in vivo.
It is unclear why liver epithelial cells should require such a complex mechanism to control cell survival. The CD40/Fas system may act as a check to control cell proliferation and turnover in the face of liver injury and hepatic regeneration or provide a rapid and potent method to eliminate infected cells. Studies of Fas/FasL (lpr/gld) knockout mice suggest that bronchial epithelium infected with Pseudomonas aeruginosa dies via apoptosis and that the process is dependent on autocrine expression of both Fas and FasL on the target cell (28)
. Furthermore, patients with X-linked hyper-IgM syndrome who lack functional CD40L are susceptible to BEC tumors and biliary infections with cryptosporidiosis (29)
, implying that CD40 on BEC is important for T cell activation and clearance of infection. In support of this, cryptosporidial infections in CD40 deficient mice result in persistent infection associated with bile duct proliferation and an absence of BEC apoptosis (30)
. These studies support the model we are proposing in which CD40 on biliary epithelium regulates the survival of infected or damaged cells, leading to clearance of the infection and prevent malignant change and cholangiocarcinoma.
Ligation of CD40 on endothelium has been shown to increase expression of adhesion molecules, including E-selectin (CD62e), ICAM-1 (CD54), and VCAM-1 (CD106) (31)
, and CD40 activation on kidney-derived epithelium enhances cytokine and chemokine secretion (32)
. The expression of adhesion molecules and chemokines by BEC increases upon treatment with proinflammatory cytokines in culture (33
, 34)
and it is possible that signaling through CD40 might enhance this. If so, CD40 activation might have a dualistic effect, acting initially to trigger the secretion of factors that will attract effector leukocytes and then providing a mechanism for amplifying the cell death induced by such cells. Periseptal hepatocytes also stained for Fas, FasL, and CD40; it is possible that CD40/Fas-mediated apoptosis contributes to the death of periseptal hepatocytes in interface hepatitis, providing a mechanism for progressive liver damage and the development of fibrosis and cirrhosis.
We thus believe that cooperative interactions between the CD40 and the Fas pathways are important regulators of BEC survival in the liver and may contribute significantly to the destruction of bile ducts in PBC. Understanding the balance of signaling pathways activated in response to these receptors will elucidate how cooperative interactions between different members of the TNF receptor superfamily regulate cell growth and apoptosis.
|
ACKNOWLEDGMENTS |
|---|
Received for publication April 10, 2001.
Revision received July 19, 2001.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
B functions as both a proapoptotic and antiapoptotic regulatory factor within a single cell type. Cell Death Differ 6,570-582[Medline]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
