| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Published online before print May 22, 2003
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Division of Infectious Diseases, Department of Research, University Hospital, Basel, Switzerland;
Department of Biochemistry, Biozentrum, University of Basel, Switzerland;
Laboratory of Immunopharmacology of Microbial Products, Tokyo University of Pharmacy and Life Science, Japan; and
Division of Infectious Genetics, Department of Microbiology and Immunology, Institute of Medical Science, University of Tokyo, Japan
Correspondence: Professor Regine Landmann, Division of Infectious Diseases, Department of Research, University Hospital, Hebelstrasse 20, CH-4031, Basel, Switzerland. E-mail: Regine.Landmann{at}unibas.ch
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-induced NO production but no tumor necrosis factor 
, interleukin (IL)-6, or IL-10 release. The large size, surface-marker expression, and capacity to clear gram-negative and -positive bacteria indicate that the clone was derived from the periportal, large KC subpopulation. The clone allows molecular studies of anti-infective and immune functions of KCs.
Key Words: mouse • lipopolysaccharide • nitric oxide • phagocytosis
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
(TNF-) induction was observed after LPS stimulation of murine KCs [12
13
14
15
]. Scavenger receptors [16
], but not CD14, were shown to be essential in these responses [12
, 14
]. TLR4, instead, has been inferred to be important by comparison with LPS-induced TNF- production in wild-type (wt) versus TLR4-deficient C3H/HeJ mice [13
]. However, direct evidence for TLR expression in KCs has so far not been provided. NO release was studied only in KCs from rats, except in one study with mice, where NO production was found weakly induced in a subpopulation of large KCs after priming and bacterial stimulation [10
]. KCs survive in culture for only 5 weeks and do not proliferate [17 ]. Consequently, the study of their function implies the use of high animal numbers and a long and complex cell-isolation procedure for each experiment [18 ]. Furthermore, KCs are morphologically and functionally heterogeneous depending on their localization within liver sinusoids. Large KCs located in the periportal zone are essentially responsible for gut-derived, material phagocytosis [7 , 10 , 11 , 19 ]. Intermediate and small KCs are midzonally and centrally located and are involved in immune responses [10 , 19 ]. Hence, clonal KC lines would represent useful, new tools for in-depth molecular and biochemical analyses.
We took advantage of the H-2Kb–tsA58 transgenic mouse, also called "Immorto" mouse, which stably expresses a thermolabile mutant (tsA58) of the Simian virus 40 (SV40) large T antigen (TAg) under the control of the H-2Kb promoter [20 ]. Cells isolated from this mouse grow continuously at the permissive temperature of 33°C, at which the mutant tsA58 is active. Moreover, at 37°C or 39°C, two temperatures that inactivate the tsA58 Tag, cells differentiate and behave normally [20 , 21 ]. The H-2Kb–tsA58 transgenic mouse is healthy at ambient temperature, and it has previously served to generate several cell lines from nonproliferating tissues, such as an endothelial line from liver sinusoids [22 ]. In addition, a line from this type of mouse could be stably transfected, illustrating the use of this model for molecular analysis [23 ].
In the present study, we describe the generation and functional characterization of an immortalized line derived from periportal, large KCs. These cells show several characteristics of KCs, such as antigen-presenting molecules, LPS binding, phagocytosis of gram-positive and -negative bacteria, LPS-induced NO release, expression of the scavenger receptor A, CD14, and TLR4 proteins. Importantly, they share a specific property of KCs, as they do not express the TACO protein, which in other macrophages, is recruited to phagosomes to retain living mycobacteria. The lack of TACO allows mycobacterial degradation within these KCs [24 ]. They represent a new tool for the molecular, biochemical, and immunological analysis of periportal, large KCs.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Isolation of KCs, peritoneal macrophages, and the Raw 264.7 cell line
Animals were killed with CO2 before opening the abdominal wall. Livers were perfused 2 min in situ at a flow rate of 5 ml/min through the portal vein with 0.05% collagenase A solution (Roche Diagnostics, Mannheim, Germany) prepared in calcium-deprived medium at 37°C. Liver suspensions from 10 mice, obtained after removal and mechanical disruption of the livers, were pooled, incubated for 30 min at 37°C in Gey’s balanced solution (GBS), pH 7.4, containing 0.05% collagenase A, filtered through a nylon gauze (mesh, 300 µm), and centrifuged 15 min at 300 g. The pellet was taken up in 20 ml GBS and mixed with 28 ml 30% (w/v) metrizamide solution (Acros Organics, Geel, Belgium) prepared in NaCl-deprived GBS (final density, 1.089 g/cm3). After layering 1 ml GBS on top of the metrizamide cell mixture, the preparation was centrifuged at 1400 g for 15 min without brake. The liver sinusoidal cells from the interface were washed and further separated by centrifugal elutriation with a JE-5.0 elutriation system supplied with a standard chamber (Beckman Coulter, Fullerton, CA). The cells were loaded at a rotor speed of 3500 rpm and a flow rate of 12 ml/min. Cell fractions were collected at a rotor speed of 3200 rpm and flow rates of 12, 18, 26, 32, 41, and 50 ml/min [18
]. KCs, identified by fluorescein-activated cell sorter (FACS) staining (Becton Dickinson, San Jose, CA) with the macrophage markers F4/80 and MOMA-2, were collected from flow rates of 26–50 ml/min and selected for culture. They corresponded to 11.2% ± 2.6% of the loaded cells, and the cell sizes were between 8 and 11.5 µm, according to the JE-5.0 rotor speed and flow-rate nomogram.
Peritoneal macrophages were harvested in RPMI medium containing 5% heat-inactivated fetal bovine serum (FBS; low endotoxin; Gibco-BRL, Paisley, Scotland), 3 days after intraperitoneal injection of 4% thioglyocollate (Becton Dickinson). The Raw 264.7 macrophage cell line was obtained from American Type Culture Collection (ATCC; Manassas, VA) and was cultured in RPMI containing 10% FBS at 37°C in 5% CO2. It was stimulated with Salmonella abortus equi LPS 10 ng/ml for 3 h.
KC line culture
The cells were initially cultured at 33°C in a 5% CO2 atmosphere in RPMI containing 100 U/ml recombinant murine interferon- (rIFN-; Biosource International, Nivelles, Belgium), supplemented with 20% or 5% FBS, with or without 20% conditioned medium (CM) in 0.5 mg/ml poly-L-lysine-coated flasks. The CM was composed of supernatants of the human hepatocyte line HepG2 [26
] and the human endothelial cell line EAhy926 [27
]. rIFN- was removed after two passages. At passage 13, clones generated by the limiting dilution method were grown at 33°C in RPMI with 5% FBS and 20% CM. All functional tests were performed at 33°C or 37°C in the same medium in poly-L-lysine-coated dishes.
Proliferation assays
Cells (106) were plated on 75 cm2 flasks. After 2–5 days of culture, cells were harvested and counted after trypan blue staining.
Reverse transcriptase (RT)-PCR
Total RNA from the KC13-2 clone was prepared in TRIzol® reagent (Gibco-BRL), according to the manufacturer’s instructions and was reverse transcribed with oligo(dt)15 primers (Promega, Madison, WI) and Stratascript RT (Stratagene, La Jolla, CA). PCR was performed with the following primers: for SV40 TAg mutant tsA58, sense: 5'TCAACCTGACTTTGGAGGCTTCTG3' and antisense: 5'GTCACACCACAGAAGTAAGGTTCC3'; for CD14, sense: 5'GGTACTGAGTATTGCCCAA3' and antisense: 5'GCCCAGTGAAAGACAGATT3'; for TLR4, sense: 5'GAGCCGTTGGTGTATCTTT3' and antisense: 5'GCCGTTTCTTGTTCTTCCT3'.
Immunohistochemistry and FACS analysis
The following monoclonal antibodies (mAb; 10 µg/ml) were used: MOMA-2 (Serotec, Oxford, UK), BM8 (BMA, Augst, Switzerland), 2F8 (kindly provided by Nick Platt, University of Oxford, UK), anti-CD14 (G5A10; ref. [28
]), anti-TLR4/MD2 (MTS510; ref. [29
]), F4/80 (ATCC Number HB-198), anti-CD11b (MAC-1; PharMingen, San Diego, CA), anti-CD16/CD32 [FC receptor for immunoglobulin G (IgG; FcR)II/III, PharMingen], antifactor VIII (Dako, Zug, Switzerland), and Meca32 (PharMingen). Rat IgG2b, hamster IgG1
(PharMingen), and rabbit serum were used as isotype controls.
Cells (2x104/well) were plated 24 h in 16-well glass LabTekTM chamber slides (Nalge Nunc, Naperville, IL) at 37°C. After fixation in periodate-lysine-paraformaldehyde (PLP) solution (periodate, 2.14 g/l; lysine, 13.7 g/l; paraformaldehyde, 2%), 15 min at room temperature, slides were washed with phosphate-buffered saline (PBS) containing 0.1% Tween 20 and were incubated 30 min with 0.3% H2O2 in methanol, followed after washing by a 20-min incubation with 1.5% normal rabbit or goat serum and a 2-h incubation with the primary antibodies. After washing, cells were incubated 45 min with biotinylated secondary antibodies (Vector Laboratories, Burlingame, CA) and 30 min with avidin/biotinylated-peroxidase complexes (Vectastain Elite ABC, Vector Laboratories). After washing, cells were incubated 5 min with peroxidase substrate 3-amino-9-ethyl carbazole (Vector Laboratories). The slides were counterstained with hematoxylin (Medite, Nunningen, Switzerland), mounted, and examined microscopically.
FACS was also used to analyze phenotypes with fluorescein-conjugated goat anti-rat IgG, goat anti-rabbit IgG, or goat anti-Armenian hamster IgG (Jackson ImmunoResearch Laboatories, West Grove, PA) as a secondary antibody. Anti-SV40 TAg was a kind gift of Loren Field (Riley Hospital, Indianapolis, IN). Anti-I-Ab, -H-2Kb, -CD40, -CD80, and -CD86 mAb were provided by PharMingen. Anti-CD1d mAb (ATCC Number HB-322) was a kind gift of Gennaro de Libero (University Hospital, Basel). In selected experiments, primary sinusoidal cells were sorted with anti-CD14 and MOMA-2 antibodies in a FACS Vantage® sorter.
Western blotting
The proteins were solubilized by lysing macrophages in 10 mM Tris, pH 8.0, containing 1% Triton, 60 mM n-octyl-ß-D-glucopyranoside (Sigma Chemical Co., St. Louis, MO), 150 mM NaCl, and 1 mM phenylmethylsulfonyl fluoride. Protein (15 µg) was separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis in a 10% acrylamide gel and transferred onto a nitrocellulose membrane. This membrane was blocked with 10% milk in PBS, incubated overnight at 4°C with rabbit antibody directed against TACO diluted 1/1000 in 1% milk in PBS, washed three times with 0.1% Tween 20 in PBS, incubated 2 h at room temperature with horseradish peroxidase-labeled donkey-anti-rabbit IgG (Jackson ImmunoResearch Laboatories) diluted 1/10,000 in 1% milk in PBS, washed five times, incubated 1 min with enhanced chemiluminescence substrate (Amersham, Uppsala, Sweden), and exposed to a double-emulsion film (Kodak, Rochester, NY).
Enzymatic activities
Nonspecific esterase activity was assessed in PLP-fixed cells incubated 1 h at 37°C with paranosaline (80 µg/ml) and -naphtyl acetate (400 µg/ml; Sigma Chemical Co.). Sodium fluoride (NaF; 1.5 mg/ml) was added in control wells. For endogenous peroxidase activity, fixed cells were incubated 1 h at 37°C with 1% diaminobenzidine tetrahydrochloride and 0.01% H2O2. Cells were counterstained with hematoxylin and mounted.
Dextran-uptake and LPS binding
Cells (2x105) were incubated for 45 min at 37°C or 4°C with 10 µg/ml fluorescein isothiocyanate (FITC)-labeled dextran (Sigma Chemical Co.) before FACS analysis. Cells (2x105) were incubated for 1 h at 37°C with or without FITC-labeled LPS from Serratia marcescens (20–100 ng/ml) with 0.5% murine EDTA plasma or 0.5% human serum albumin (HSA) before FACS analysis or confocal microscopy. The effect of anti-CD14 (4C1) [30
] and anti-TLR4/MD2 mAb (MTS510) [29
] was assessed by a 30-min preincubation with 50 µg/ml antibody before LPS–FITC treatment.
Phagocytosis
Cells (5x104/well) were cultured 24 h in 48-well plates or eight-well slides and were incubated 1 h at 37°C with opsonized Bodipy® FL-labeled Escherichia coli (K-12 strain) BioParticles® (Molecular Probes, Eugene, OR) at particle-to-cell ratios of 20/1. Controls included incubation at 4°C or 2 h preincubation with 8 µg/ml cytochalasin D (Sigma Chemical Co.). Anti-CD14 and -TLR4/MD2 effects were studied by a 30-min preincubation with 50 µg/ml antibodies. Cells were analyzed in the confocal microscope or detached with trypsin/EDTA, and propidium-negative cells (live) were FACS analyzed with or without trypan blue.
Alternatively, phagocytosis and killing of live Streptococcus pneumoniae were assessed with a method adapted from Netea et al. [31 ]. Briefly, 105 KC13-2 cells or primary KCs were incubated in Dulbecco’s modified Eagle’s medium (DMEM; Gibco, Basel) containing 0–40% fresh, human pooled serum in polylysine-coated, 16-well glass LabTekTM chamber slides with 106 colony forming units (cfu) of a clinical isolate of S. pneumoniae serotype 3. After 1 h, supernatants were aspirated, and monolayers were gently washed with DMEM to remove noningested microorganisms. The supernatants and washings containing uningested bacteria were combined and plated in serial dilutions on blood agar plates. The percentage of phagocytosed microorganisms was defined as [1–(number of uningested cfu/cfu at the start of incubation)] x 100. Killing of S. pneumoniae by KC13-2 was assessed in the same culture. After removal of the noningested bacteria, 200 µl DMEM was added. After a further hour of incubation, the cells were washed twice with 100 µl distilled H2O to achieve lysis. To quantify remaining viable, intracellular S. pneumoniae, tenfold dilutions of each sample were spread on blood agar plates and incubated at 37°C for 24 h. The percentage of killed bacteria was calculated as follows: [1–(cfu after incubation/number of phagocytized cfu)] x 100. In selected experiments, bacteria were preopsonized for 30 min in DMEM medium containing 10% human serum.
Double-immunofluorescence staining was also performed with E. coli particles and anti-CD14 mAb labeled with phycoerythrin-conjugated donkey anti-rat IgG (Jackson ImmunoResearch Laboatories).
NO and cytokine production
Cells (2x105/well) were stimulated 24 h at 37°C with 10 ng/ml–10 µg/ml S. abortus equi LPS (purified and kindly provided by Chris Galanos, Max Planck Institute Freiburg im Breisgau, Germany) and/or 100 U/ml rIFN- in 96-well plates. Anti-CD14 and -TLR4/MD2 effects were studied by a 30-min preincubation with 50 µg/ml antibodies. NO was assessed using the Griess reagent [32
]. TNF- levels were determined by bioassay [33
]. Interleukin-6 (IL-6) and IL-10 were measured by enzyme-linked immunosorbent assay, according to the manufacturer’s instructions (OptEIATM, PharMingen).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
was added to activate the H-2Kb promoter controlling mutant tsA58 TAg expression. Thereafter, cells of the three lines continued to grow in the absence of rIFN-. All further cultures of KC lines and clones were performed in the presence of 20% CM from the human hepatocyte line HepG2 [26
] and from the endothelial cell line EAhy926 [27
] to provide paracrine factors. In the absence of the CM, cells lost their phenotype and the capacity to release NO. Similar to earlier observations [6 , 7 , 10 , 11 , 18 ] and to primary KCs from C57BL/10 mice (Fig. 1A ), which is the mouse strain used to generate the transgenic mouse [20 ], the three lines remained to be composed during 3 months of culture of three subpopulations, with small, round cells; long, spindle-shaped cells; and large, stellate cells (Fig. 1B) . As the three lines behaved the same, we chose to continue the study only with one of them, KC13. Four clones composed exclusively of large, stellate cells with a large nucleus were generated, and one of them, clone KC13-2 (Fig. 1C) , was further evaluated. It proliferated regularly and rapidly at 33°C and 37°C. At 33°C, cell numbers increased fivefold within 3 days and ninefold within 5 days. Cells maintained for more than 1 week at 37°C showed a three- and sevenfold multiplication after 3 and 5 days, respectively (Fig. 2 ). These results were confirmed by the tritiated thymidine incorporation assay (data not shown).
|
|
Phenotypes and endogenous enzymatic activities of primary KCs and of the KC13-2 clone
To confirm the lineage specificity of the isolated clone, the KC13-2 cells, kept at 37°C, were stained with mAb, which recognize C57BL/10 primary KCs, such as MOMA-2 (Fig. 3A
, left), BM8 (Fig. 3B
, left), and the antiscavenger receptor A antibody 2F8 (Fig. 3C
, left). KC13-2 cells expressed all of these markers, although to a lower degree (Fig. 3A
3B
3C
, right). Comparison of staining and shape with primary KCs revealed that the clone was derived from large, stellate cells, as the staining at a lower intensity resembled the distribution of primary, large cells [18
].
|
|
To further characterize the isolated clone, the endogenous peroxidase and nonspecific esterase were investigated. KC13-2 cells maintained the enzyme pattern of primary KCs, and all expressed endogenous peroxidase (data not shown) or nonspecific esterase (Fig. 3G) . The latter activity was partially inhibited by NaF (data not shown), which is characteristic of esterase isoforms associated with macrophages [34 ].
Western blot results show that the phagosome-coat protein TACO is not expressed in KC13-2 cells (Fig. 5 , lanes 4 and 5), consistent with this molecule being absent from KCs [24 ] (Fig. 5 , lane 6). In contrast, mouse peritoneal macrophages from C57BL/10 mice harvested 3 days after thioglycollate injection (Fig. 5 , lane 1), or the RAW 264.7 macrophage cell line without or with LPS stimulation (Fig. 5 , lanes 2 and 3) showed a TACO signal.
|
|
Phagocytosis of E. coli and S. pneumoniae by KC13-2 clone
An important function of KCs is clearance by phagocytosis of bacteria and their products. In the KC13-2 clone, E. coli phagocytosis was assessed by laser confocal microscopy and cytofluorometry (Fig. 7
). Using both methods, we observed two populations of phagocytosing cells. One-third of the cells took up three or more particles (Fig. 7A)
and had a broad fluorescence distribution when analyzed by cytofluorimetry (Fig. 7C)
. Two-thirds of the cells phagocytosed one or two particles (Fig. 7A)
and showed a sharp and weak fluorescence spectrum (Fig. 7C)
. The specificity of phagocytosis was documented by blockage at 4°C and with cytochalasin D (Fig. 7B
and 7C)
. In agreement with the absence of Fc and complement receptors, the KC13-2 clone also phagocytosed nonopsonized E. coli (Fig. 7G)
.
|
|
and LPS activated KC13-2 cells synergistically, as NO release rose from 17.4 ± 6.1 µM after stimulation with rIFN- to 25.9 ± 6 µM after coincubation with LPS (Fig. 9A)
. We also studied NO production in bulk, primary KCs, which reached the amount of 48 ± 6.7 µM under the same conditions (data not shown). Sorted, large CD14++/MOMA+ primary KCs released spontaneously 1.8 + 1.4 µM NO; upon rIFN- stimulation, they produced 23 µM NO, which was similar to the NO response of the KC13-2 clone. They did not respond to LPS alone, most likely because they were preactivated by the isolation. In KC13-2 cells stimulated with rIFN- and LPS, anti-TLR4/MD2 and -CD14 mAb caused only a slight, nonsignificant reduction of NO (Fig. 9C)
. Finally, inhibition of NO production was observed with the selective, inducible NO synthase inhibitor NG-monomethyl-L-arginine, indicating that NO was derived from the L-arginine NO pathway (data not shown).
|
, IL-6, or L-10, even after stimulation with 10 µg/ml LPS + 100 U/ml rIFN-. Bulk, primary KCs produced 120 ± 30 pg/ml TNF- after stimulation with a low dose of LPS (10 ng/ml) together with rIFN- (data not shown). These findings clearly show that the isolated clone retains several functional characteristics of bulk, primary KCs, thus representing the first homogenous and expandable population of stable KCs suitable for further studies.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In contrast to most other cell lines isolated from other tissues of H-2Kb –tsA58 mice [21
, 22
, 35
, 36
], the KC13-2 clone showed an IFN--independent proliferation persisting at 37°C. This unusual property is very important, as it permits functional studies of the established KCs under nonstimulated conditions and at physiological temperature. Growth at 37°C indicates an IFN--independent regulation of the SV40 TAg, as confirmed by detection of its mRNA and protein at 37°C.
Hepatocyte- and endothelial cell-derived factors were required for maintenance of a differentiated phenotype in the KC clone, suggesting that a continuous stimulation is required to keep a differentiation state in this cell type. This property may offer the possibility of identifying which factors modulate gene and protein expression in KCs and to study interactions with other liver or blood cells [37 38 39 40 41 ].
Esterase and peroxidase are markers common to all KCs. They were instrumental in determining the purity of KC populations in different animal species [12
, 18
, 42
43
44
]. The strong esterase staining in 100% of our primary KCs as well as in the KC13-2 clone was therefore considered as additional proof of the KC nature of the established cell line. Furthermore, the KC13-2 cells, as the primary KCs at day 1 of culture, did not express TACO, a phagosome-coat protein, which is found in all other tissue macrophages except KCs [24
]. This is an additional proof of the KC nature of the KC13-2 clone. With regard to other phenotypes and functions, rat and mouse KCs are highly heterogeneous [6
, 7
, 10
, 11
, 18
, 19
]. Large cells, which represent 43% of the total KC population [6
, 7
], are located in the periportal zone. They show strong phagocytic activity [6
, 7
, 10
, 11
, 19
] and high NO release after S. pneumoniae stimulation with and without cytokine priming [10
]. Intermediate and small KCs are instead located in the intermediate and central zone of the lobule, respectively. They are less phagocytic and release IL-1 and TNF- after stimulation [6
, 7
, 10
, 11
, 19
] but much less NO than large cells. Our immortalized cells are most likely derived from periportal cells, as they are large and phagocytic and release NO. This conclusion is further supported by other phenotypic characteristics. The surface markers MOMA-2 and BM8, which are expressed in large, primary cells [18
, 45
], are retained in the clone. On the contrary, the F4/80 marker, which predominantly stains the small cell fraction [18
], totally disappeared from the clone. Similarly, expression of FcR and CR3 was more intense in small, primary KCs than in large cells and became negative in the clone. Finally, the particular CD14-positive phenotype of the KC13-2 clone was allowed to document that it presented the same characteristics as a subpopulation of freshly isolated, primary KCs. Indeed, large, sorted CD14++/MOMA-2+ KCs exhibited the same markers, i.e., scavenger receptors, BM8, and TLR4/MD-2, and they lacked F4/80 as the clone. Further studies will address whether the lack of detection of some markers in the immortalized cells is due to lack of gene transcription or instead, to the presence of extremely low levels of respective proteins, not detectable with specific antibodies. In conclusion, the surface antigens of immortalized cells largely resemble those of large, primary KCs and further support the periportal origin of the clone.
Importantly, KC13-2 as well as primary KCs expressed scavenger receptor A [46 ]. It is likely that this receptor participates in phagocytosis of E. coli by the clone. Indeed, scavenger receptor A mediates not only low-density lipoprotein internalization [47 ] but also phagocytosis, as recently reported [48 ]. This is indirectly supported by the observed, efficient internalization of bacteria in the absence of Fc and complement receptors, as well as in the presence of antibodies blocking CD14 and TLR4/MD2 molecules. This additional property makes the KC13-2 cells suitable to investigate the function of scavenger receptors in KCs using a variety of ligands [9 ].
Our study also shows that the KC13-2 clone possesses all the characteristics of antigen-presenting cells (APC). It constitutively expressed MHC class I and class II and CD40 and CD80 and thus, resembles primary, murine KCs, which also express the costimulatory molecules CD40 and CD80 [49 ]. MHC class II as well as CD80 expresssed by primary KCs were shown to participate in antigen presentation [50 ]. We have further observed that this KC13-2 clone makes active endocytosis, thus resembling professional APC such as dendritic cells. We also found CD1d, the MHC class I-like protein, which presents glycosphingolipids to NKT cells [51 ]. The set of antigen-presenting molecules in the clone allows study of the consequences of antigen presentation to MHC- and CD1-restricted T cells. Furthermore, it will allow investigation of in vitro consequences of antigen presentation such as T cell apoptosis [52 ]. Moreover, the study of virus [53 ] or parasite [54 ] entry into KCs and the immunological response can be followed in the clone.
Analysis of the immortalized cells also has provided conclusive data on the still-debated expression and function of the pattern recognition receptors CD14 and TLR4 in KCs. As previously found in KCs isolated from rat liver [16
] and in our primary KC lines, the KC13-2 clone constitutively expresses CD14. In addition, our study for the first time directly shows the expression of TLR4/MD2 on the subpopulation of large KCs. Previous studies only inferred the functional involvement of TLR4 in LPS-induced TNF- release by KCs, without investigating the presence of TLR4 mRNA or protein [13
].
The functional role of CD14 and TLR4 on KCs appears different from that on other types of macrophages. Indeed, our cells, in agreement with the properties of primary, large KCs reported earlier [10
, 19
], do not release cytokines in response to LPS and rIFN-. This lack of cytokine secretion may reflect a physiological, important property of KCs, although the possibility that it could also reflect a transcriptional repression of cytokine genes by the SV40 TAg should be considered [55
].
A significant finding is that neither receptor is involved in phagocytosis of bacteria, which is a very important function of periportal KCs. Furthermore, neither CD14 nor TLR4 is responsible for LPS binding, as shown by persistent binding in the presence of inhibitory antibodies. Our experiments also showed that neither receptor triggers the NO release, in agreement with the results obtained with rat KCs [15 ]. Taken together, these results support the possibility that large KCs interact with bacteria and LPS using receptors other than CD14, as previously suggested [12 , 14 , 16 ], and add indirect evidence that this function is also independent from TLR4 expression.
In conclusion, we have established a stable and proliferating KC line, which resembles large, stellate, periportal KCs in their phenotypic and functional characteristics. This clone will be useful to perform studies of the anti-infective and immune functions of KCs.
|
ACKNOWLEDGEMENTS |
|---|
|
FOOTNOTES |
|---|
Received January 23, 2003; accepted February 19, 2003.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  K. K. Eriksson, L. Cervantes-Barragan, B. Ludewig, and V. Thiel Mouse Hepatitis Virus Liver Pathology Is Dependent on ADP-Ribose-1''-Phosphatase, a Viral Function Conserved in the Alpha-Like Supergroup J. Virol., December 15, 2008; 82(24): 12325 - 12334. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Ashare, A. B. Nymon, K. C. Doerschug, J. M. Morrison, M. M. Monick, and G. W. Hunninghake Insulin-like Growth Factor-1 Improves Survival in Sepsis via Enhanced Hepatic Bacterial Clearance Am. J. Respir. Crit. Care Med., July 15, 2008; 178(2): 149 - 157. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Ashare, M. M. Monick, A. B. Nymon, J. M. Morrison, M. Noble, L. S. Powers, T. O. Yarovinsky, T. L. Yahr, and G. W. Hunninghake Pseudomonas aeruginosa Delays Kupffer Cell Death via Stabilization of the X-Chromosome-Linked Inhibitor of Apoptosis Protein J. Immunol., July 1, 2007; 179(1): 505 - 513. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Svensson, S. Zubairi, A. Maroof, F. Kazi, M. Taniguchi, and P. M. Kaye Invariant NKT Cells Are Essential for the Regulation of Hepatic CXCL10 Gene Expression during Leishmania donovani Infection Infect. Immun., November 1, 2005; 73(11): 7541 - 7547. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Zhai, X.-d. Shen, R. O'Connell, F. Gao, C. Lassman, R. W. Busuttil, G. Cheng, and J. W. Kupiec-Weglinski Cutting Edge: TLR4 Activation Mediates Liver Ischemia/Reperfusion Inflammatory Response via IFN Regulatory Factor 3-Dependent MyD88-Independent Pathway J. Immunol., December 15, 2004; 173(12): 7115 - 7119. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Vazquez-Torres, B. A. Vallance, M. A. Bergman, B. B. Finlay, B. T. Cookson, J. Jones-Carson, and F. C. Fang Toll-Like Receptor 4 Dependence of Innate and Adaptive Immunity to Salmonella: Importance of the Kupffer Cell Network J. Immunol., May 15, 2004; 172(10): 6202 - 6208. [Abstract] [Full Text] [PDF]
|
|
| |||||||
