* Laboratory for Cell Biology and Histology, Vrije Universiteit Brussel (VUB), Brussels, Belgium;
Evanston Northwestern Healthcare Research Institute, Illinois;
Department of Immunohematology and Bloodtransfusion, Leiden University Medical Center, The Netherlands; and
Institute for Biochemistry, BIL Biomedical Research Center, University of Lausanne, Epilanges, Switzerland
Correspondence: David Vermijlen, Laboratory for Cell Biology and Histology, Vrije Universiteit Brussel (VUB), Laarbeeklaan 103, 1090 Brussels, Belgium. E-mail: dvermijl{at}cyto.vub.ac.be
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: Fas ligand • serine protease inhibitor PI-9/SPI-6 • liver sinusoidal endothelial cell • apoptosis • P815 • hepatocyte
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Cytotoxic lymphocytes [NK cells, LAK cells, cytotoxic T cells (CTL), NK-T cells] use the Fas ligand (FasL) and the perforin/granzyme pathway to kill target cells [3 , 14 15 16 ]. FasL on effector cells binds Fas present on the target cell membrane, which results in oligomerization of Fas and activation of caspase 8. Perforin and granzymes, of which granzyme B is the most potent, reside in granules of the cytotoxic lymphocytes and are released by exocytosis. Intracellular delivery of granzyme B results in the initiation of the caspase cascade by proteolytic activation of caspase 3, directly [17 ] or through a mitochondrium-dependent pathway [18 ]. Caspases play a central role in the execution of apoptosis [17 ]. In this study, we investigated the mechanism hepatic NK cells use to kill P815 cells.
Recently, protease inhibitor 9 (PI-9)/serine PI-6 (SPI-6), a granzyme B inhibitor, has been found to be expressed in some tumor cells and constitute a mechanism to resist granzyme B-induced apoptosis [19 , 20 ]. Therefore, we investigated PI-9/SPI-6 expression in splenic/blood NK-resistant P815 tumor cells. Furthermore, it has been suggested that cytotoxic lymphocytes express high levels of this inhibitor to protect themselves against their own granzyme B [21 ] and that low expression in endothelial cells could neutralize circulating granzyme B [22 , 23 ]. Therefore, the expression of PI-9/SPI-6 was analyzed in hepatic NK cells and LSECs as well.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Isolation of cells
NK cells from liver and spleen
Hepatic NK cells were isolated from male Wistar rats (Proefdierencentrum, K.U.L., Leuven, Belgium) of 12–16 weeks old, weighing ca. 300 g, according to a combination of the methods described by Bouwens et al. [25
] and Kanellopoulou et al. [26
]. The portal vein and the inferior vena cava were cannulated by an 18G and a 14G catheter, respectively, and the superior vena cava was clamped off. The portal cannula was connected via silicone tubing with an inner diameter of 4 mm to a 500-ml glass bottle. First, the liver was perfused with a few ml phosphate-buffered saline at 37°C through the portal vein at physiologic pressure (10–12 cm H2O) to remove the blood. Perfusion was stopped for a moment to add ethylenediaminetetraacetate (EDTA) to a concentration of 0.1%, the perfusion was resumed at an elevated pressure (50 cm H2O), and 300 ml perfusate was collected. To further increase the perfusion pressure, the outlet cannula in the inferior vena cava was clamped off several times for a few seconds during the collection of the perfusate. For the isolation of NK cells from the spleen, spleens were removed aseptically and crushed with the hub of a syringe in complete medium (CM), which consisted of RPMI 1640 (52400, Life Technologies), 10% FBS (Eurobiochem), penicillin (100 U/ml), streptomycin (100 U/ml), and L-glutamine (0.2 mmol/L; Gibco, Life Technologies). Following collection by centrifugation and resuspension in Gey’s balanced salt solution (GBSS), hepatic and splenic NK cells were purified by the same purification steps. After centrifugation over Ficoll-Paque (Pharmacia AB, Upsala, Sweden; 20 min, 450 g), the mononuclear cells present at the interface were collected and resuspended in GBSS. After centrifugation, the pellet was resuspended in CM and put on a nylon wool column (Wako Chemicals, Neuss, Germany) for 45 min, and nonadherent cells were collected. T cells and remaining B cells and macrophages/monocytes were removed by magnetic cell sorting (VarioMACS) using mouse anti-rat CD5 [anti-T, HIS47, immunoglobulin G (IgG)2a; Pharmingen, San Diego, CA], mouse anti-rat pan B (RLN-9D3, IgG2a), and mouse anti-rat monocytes/macrophages/granulocytes (Mar3, IgG2b) antibodies (Seikagaku, Tokyo, Japan) and a secondary antibody coupled to magnetic microbeads (anti-mouse IgG2a+b Microbeads, Miltenyi Biotech, Bergisch Gladbach, Germany). Finally, the pellet was resuspended in CM. The purity of hepatic and spleen NK cells was between 85% and 95% and 75% and 85%, respectively, as evaluated by microscopical observation of May-Grünwald-Giemsa-stained cytospin preparations (large granular lymphocyte morphology) and by flow cytometric analysis staining with 3.2.3 monoclonal antibody (mAb), which recognizes NKR-P1A [27
]. It has been shown that the rat liver contains a relative low number of NK-T cells (CD3+NKRP-1A+) compared with human and some mouse strains [28
]. To further exclude possible contamination with these NK-T cells and their possible contribution to cytotoxic effects against tumor cells [3
], we applied flow cytometry and determined that more than 95% of CD3+NKRP-1A+ cells were positive for CD5. Furthermore, after magnetic cell sorting, less than 0.3% of the purified cells was positive for CD5, showing an efficient removal of CD5+ cells (T and NK-T cells) by the VarioMACS system. The viability was more than 95%, as determined by trypan blue exclusion.
LSECs
Rat LSECs were isolated, as described in detail previously, by using a single liver perfusion with collagenase A buffer, a two-step Percoll gradient centrifugation, and selective adherence to different substrates [29
]. The purity of the LSEC cultures was greater than 95%, as less than 5% of the cells examined by electron microscopy were devoid of fenestrae. The viability was more than 95%, as determined by trypan blue exclusion.
Hepatocytes
Rat hepatocytes (purity 90–95%) were isolated according to De Smet et al. [30
] using a two-step collagenase perfusion of the liver. The viability was more than 85%, as determined by trypan blue exclusion.
The procedures used in this study for animal handling were approved by the local ethical committee.
Reagents and antibodies
3,4-Dichloroisocoumarin (DCI) and ethyleneglycol-bis(ß-aminoethylether)-N,N'-tetraacetic acid (EGTA) were purchased from ICN (Asse-Relegem, Belgium) and Z-Val-Ala-Asp(OMe)-fluoromethylketone (Z-VAD-FMK), from Bachem (Bubendorf, Switzerland). Recombinant human soluble (rhs) FasL and enhancer were purchased from Alexis (Läufelfingen, Switzerland). The apoptotic-inducing capacity of rhsFasL is reduced by more than 1000-fold compared with membrane-bound FasL [31
]. However, restoration of the cytotoxic activity of rhsFasL can be achieved with the addition of a cross-linking antibody, also known as enhancer [31
]. We preferred to use rhsFasL instead of anti-Fas antibody, as antibody agonists and natural ligand can stimulate different pathways [32
]. Perforin and granzyme B were isolated from YT cells, a human NK cell line, as described previously [33
]. Hoechst 33342 and propidium iodide were obtained from Molecular Probes (Leiden, The Netherlands). FasL rabbit polyclonal (Q-20) and granzyme B goat polyclonal (C-19) antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). The mouse monoclonal perforin antibody 2d4 was a generous gift from Dr. G. Griffiths (Sir William Dunn School of Pathology, Oxford) [34
]. The mAb 17, generated against PI-9, is shown to be specific for PI-9 (human) [23
] and efficiently cross-reacts with mouse (SPI-6) and rat (ref. [35
] and see Fig. 6
). Donkey anti-rabbit IgG and sheep anti-mouse IgG coupled to peroxidase were purchased from Amersham Pharmacia Biotech (Roosendaal, The Netherlands). Donkey anti-goat IgG coupled to peroxidase was obtained from Santa Cruz Biotechnology.
 View larger version (51K): [in a new window] |
Figure 6. Analysis of PI-9/SPI-6 and granzyme B expression by Western blotting. Equal protein loading was checked by Ponceau S staining. LSEC, liver sinusoidal endothelial cells; HEP, hepatocytes; H-NK, hepatic NK cells; MFF-SPI-6, MFF cell line transfected with SPI-6 (mouse homologue of human PI-9).
|
Transmission electron microscopy (TEM)
P815 cells (5x104) were incubated with or without hepatic NK cells (5x105) in 35-mm Petri dishes. After incubation, the cells were collected and pelleted in 15-ml tubes (300 g, 10 min). The pelleted cells were resuspended in 1.5% glutaraldehyde in 0.15 M cacodylate buffer and transferred to a 1.5-ml microcentrifuge tube, fixed for 30 min at 4°C, and ultracentrifuged. Cells were postfixed in 1% OsO4 for 1 h at 4°C, dehydrated in a graded ethanol series, and embedded in epon. Subsequently, 60-nm ultrathin sections were made and stained with uranyl acetate and lead citrate before examining with a Philips EM 400 TEM at 80 kV [29
].
Quantitative DNA fragmentation assay
To label the DNA, target cells were incubated for 2 h with 28 µCi/ml [methyl-3H]thymidine (Amersham, Buckinghamshire, UK), after which the cells were washed three times. Target cells (104 cells in 100 µl) were placed in 1.5-ml microcentrifuge tubes. Freshly isolated hepatic or splenic NK cells were added to yield the desired effector/target (E/T) cell ratio, and the volume was brought to 200 µl. After 3 h coincubation, the tubes were centrifuged at 300 g for 10 min, and the medium was withdrawn. The pelleted cells were lysed with 0.5 ml cold (4°C) lysis buffer (5 mM Tris, 2 mM EDTA, 1% Triton X-100, pH 7.4) for 30 min at 4°C, and the lysates were ultracentrifuged (10,000 g, 15 min, 4°C) to separate fragmented from intact DNA. The radioactivity [counts per minute (cpm)] in the incubation medium, in the 10,000 g supernatant, and in the 10,000 g pellet was determined in a beta counter (Beckman, Fullerton, CA). The percentage-fragmented DNA was calculated using the following formula:
 |
51Cr release assay
Cytolysis was measured in a 4 h 51Cr release assay as described previously [37
]. Results were expressed as percentage of specific lysis according to the following formula:
 |
exp = experimental (target cells with effector cells); spont = spontaneous (target cells and medium only); and max = maximal, obtained after the addition of sodium dodecyl sulfate (SDS; 1% final concentration). All data are triplicate determinations.
Real time quantitative reverse transcriptase-polymerase chain reaction (RT-PCR)
Total RNA was isolated using the RNeasy Mini Kit (Qiagen, Hilden, Germany) according to the instructions of the manufacturer. A DNase treatment was included using the RNase-Free DNase set (Qiagen). Equal amounts of total RNA were reverse transcribed using random hexamers with Taqman reverse transcription reagents (Applied Biosystems, Lennik, Belgium). In real time PCR, reactions are characterized by the point in time during cycling when amplification of a PCR product is first detected (threshold cycle: CT) rather than the amount of PCR product accumulated after a fixed number of cycles. The higher the starting copy number of the cDNA, the sooner a significant increase in fluorescence is observed and the lower the CT value. We used SYBR Green PCR Master Mix to perform real time PCR on the ABI Prism 7700 sequence detection system, running under SDS software version 1.7 (Applied Biosystems). Primers for FasL, perforin, and granzyme B (RNKP-1) were designed using Primer Express 1.5 software (Applied Biosystems). The sequences were as follow: 5'-GGCTGGGTGCCATGCA and 5'-GGCACTGCTGTCTACCCAGTAGA for FasL, 5'-CGGAAGCAAACGTGCATGT and 5'-GGCCTTCTCGGCTGCAA for perforin, and 5'-CGCTGTGAAGCCTCTCAATCT and 5'-CCAGCCACATAGCACACATCTC for granzyme B. We used 18S ribosomal RNA as endogenous control, using the primerset from Taqman ribosomal RNA control reagents (Applied Biosystems). Total RNA was diluted 500 times to quantify 18S rRNA to have a CT value higher than 15. Standard curves for the different molecules were generated using total RNA isolated from RNK-16 cells. Using these standard curves and endogenous control, relative quantification of FasL, perforin, and granzyme B mRNA was performed (User Bulletin #2, Applied Biosystems). Specificity of the PCR reaction was checked by agarose gel electrophoresis and/or the dissociation curve method using Dissociation Curves 1.0 software (Applied Biosystems). No template controls were included. Reactions were performed in triplicate.
Western blot analysis
Cells were lysed in lysis buffer (5% SDS, 80 mmol/L Tris, pH 6.8, 5 mmol/L EDTA, 10% glycerol, 5% ß-mercaptoethanol, bromophenol blue, 1 mmol/L phenylmethylsulfonyl fluoride) and were sonicated (Sonifier 250, Branson Corp., Danbury, CT). Samples were loaded on a 10% polyacrylamide gel, separated by electrophoresis at 100 V for 2 h, and electroblotted onto a nitrocellulose membrane (BA38, Schleicher and Schuell, Dassel, Germany) at 25 V overnight. Following blotting, the membrane was stained with 0.2% Ponceau S (Sigma Chemical Co., Bornem, Belgium) to check homogenous transfer of proteins. After blocking with 5% skimmed milk, the membrane was reacted with primary and peroxidase-conjugated secondary antibody, and signals were detected by enhanced chemiluminescence using Renaissance Western blotting substrate (NEN Life Science Products, Zaventem, Belgium).
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
 View larger version (127K): [in a new window] |
Figure 1. Hepatic NK cells induce apoptosis in P815 cells as shown by Hoechst 33342/propidium iodide staining and fluorescence microscopy (A) and TEM (B). P815, P815 cells in medium only; P815+H-NK, P815 cells coincubated with hepatic NK cells at an E/T ratio of 10/1. When P815 cells are coincubated with hepatic NK cells (thin arrow), nuclei of P815 cells become condensed and fragmented (thick arrow). Note the granules present in the hepatic NK cells (arrowhead, B). Scale bar (A), 20 µm; scale bar (B), 1 µm.
|
 View larger version (15K): [in a new window] |
Figure 2. Hepatic NK cells, but not splenic NK cells, induce DNA fragmentation and 51Cr release in P815 cells. E/T ratio, 10/1. Values are means of three independent experiments. Error bars, SD.
|
 View larger version (86K): [in a new window] |
Figure 3. P815 cells are sensitive to the FasL (A) and perforin/granzyme (B) pathway. P815, P815 cells in medium only; P815+FasL, P815 cells treated with 100 ng/ml rhsFasL and 1 µg/ml enhancer; P815+PFN, P815 cells treated with sublytic concentration perforin (=induce less than 10% propidium iodide-positive cells); P815+GrB, P815 cells treated with 1 µg/ml granzyme B; P815+PFN/GrB, P815 cells treated with PFN (sublytic concentration) and granzyme B (1 µg/ml). When P815 cells are treated with FasL or with PFN/GrB, their nuclei become condensed and fragmented (arrow). Scale bar, 20 µm.
|
 View larger version (20K): [in a new window] |
Figure 4. Expression of FasL, perforin, and granzyme B in splenic and hepatic NK cells on RNA (A) and protein (B) level as determined by real time RT-PCR and Western blotting, respectively. The relative RNA expression quantification was performed as described in Materials and Methods. Error bars, SD. Western blot probings for FasL (rabbit Q-20 antibody), perforin (mouse 2d4 antibody), and granzyme B (goat C-19 antibody) were performed on the same blot. Equal protein loading was checked by Ponceau S staining. S-NK, splenic NK; H-NK, hepatic NK; PFN, isolated human perforin; GrB, isolated human granzyme B. Experiments shown are representative for three independent experiments.
|
 View larger version (32K): [in a new window] |
Figure 5. The effect of inhibitory compounds on the killing of P815 cells by hepatic NK cells as determined by DNA fragmentation and 51Cr release (A) and Hoechst 33342/propidium iodide staining (B). E/T ratio, 10/1. CON, Control; EGTA, 5 mmol/L EGTA present during coincubation; DCI, preincubation of hepatic NK cells with 50 µmol/L DCI for 30 min, no DCI present during coincubation; VAD, preincubation of P815 cells with 160 µmol/L Z-VAD-FMK for 30 min, 80 µmol/L present during coincubation. Values are means of three independent experiments. Error bars (A), SD. Scale bar (B), 10 µm.
|
 View larger version (13K): [in a new window] |
Figure 7. The granzyme B inhibitor PI-9/SPI-6 prevents hepatic NK cell-mediated apoptosis. When MFF tumor cells are transfected with SPI-6 (mouse homologue of PI-9), hepatic NK cell-induced DNA fragmentation is completely inhibited. The results shown are representative for three independent experiments. Error bars, SD.
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Although the FasL and perforin/granzyme pathway are used by cytotoxic lymphocytes to kill P815 cells [43
44
45
], we showed here that hepatic NK cells exclusively use the perforin/granzyme pathway to kill these tumor cells. Although the capacity of hepatic NK cells to kill splenic NK-resistant tumor cells may be attributed to the high expression of perforin and granzyme B, the granule exocytosis machinery is ultimately triggered by the effector-cell recognition and activation system [1
, 46
]. We recently showed that hepatic NK cells express the LAK surface markers gp42 (NK activation receptor), CD25 (IL-2 receptor 
chain), and ANK44 (unknown function) [27
] and express high levels of lymphocyte function-associated antigen-1 (LFA-1; adhesion molecule) [47
], which is involved in hepatic NK cell-mediated cytotoxicity [37
]. Therefore, the high cytotoxic activity of hepatic NK cells likely reflects the capacity to be triggered by a broader number of targets as well as possessing a higher content of granule proteins.
Previously, FasL resistance of target cells was proposed as a possible explanation for the FasL-independent killing mediated by hepatic NK cells [36 , 48 ]. However, in the present report, we demonstrated that FasL-sensitive target cells are also not killed by hepatic NK cell-mediated death receptor-induced apoptosis. Hepatic NK cells are attached to LSECs [1 , 2 ], and it has been shown that LSECs and hepatocytes are sensitive to FasL-mediated apoptosis [48 49 50 ]. As a consequence, it is conceivable that the presence of an active FasL pathway mediated by hepatic NK cells would be detrimental to LSECs and after the sinusoidal endothelial lining is damaged, to the FasL-sensitive hepatocytes as well. A possible explanation for the inhibition of an active FasL pathway mediated by hepatic NK cells is the production of soluble Fas (sFas) by hepatocytes [51 , 52 ]. Transfection of P815 cells with sFas-encoding cDNA has been shown to protect these cells from FasL-induced cell death [53 ]. sFas released in the space of Disse could pass the fenestrated endothelial lining, where it could bind the FasL present on hepatic NK cells (Fig. 8 ).
 View larger version (63K): [in a new window] |
Figure 8. Model of hepatic NK cell-mediated tumor cell killing. When a tumor cell enters a liver sinusoid, it is mechanically trapped and/or adheres to LSECs. Hepatic NK cells adhere to tumor cells by adhesion molecules such as LFA-1. sFas produced by hepatocytes blocks FasL on the hepatic NK cells, preventing possible harmful effects on the FasL-sensitive LSECs and hepatocytes. Conversely, highly expressed perforin and granzyme B, as a complex with serglycin as a scaffold, are released by granule exocytosis in the space formed between the NK-tumor conjugate. Damaging of other cells (e.g., hepatocytes) caused by leakage of granzyme B/perforin is prevented by the very efficient endocytic uptake of the granzyme B/serglycin/perforin complex by the hyaluronan receptor (HA-R) expressed on LSECs. LSECs are protected from the action of granzyme B by strong expression of the granzyme B inhibitor PI-9/SPI-6. Granzyme B, presumably taken up by the M6P-R [54
], induces apoptosis in the tumor cell by activating the caspase cascade. On the other hand, cytolysis (51Cr release) is induced by a caspase-independent mechanism. The blocked FasL pathway is shaded. Dashed lines indicate hypothetical relations. FADD, Fas-associated death domain factor; ICAM-1, intercellular adhesion molecule-1; M6P-R, mannose 6 phosphate receptor (cation-independent).
|
Although the presence of PI-9/SPI-6 in hepatic NK cells probably offers protection from unwanted, self-induced apoptosis by misdirected granzyme B, which may result during degranulation or via granule leakage (Fig. 8) [21 ], the strong expression of PI-9/SPI-6 in LSECs was unanticipated. It has been shown in vitro that bystander cells can be killed by the perforin/granzyme pathway [57 ], and granzyme B has been detected in blood of patients with ongoing CTL responses [58 ]. It is conceivable therefore that after interaction of hepatic NK cells with tumor cells, granzyme B is released in the sinusoid (leakage; Fig. 8 ). We recently observed that granzyme B is secreted bound to the proteoglycan serglycin [59 ] (Fig. 8) . On the other hand, LSECs are recognized for their scavenger function [60 ], and, interestingly, it has been shown that these cells efficiently take up serglycin, primarily through the hyaluronan receptor [61 ]. It is conceivable therefore that these cells remove granzyme B/serglycin/perforin complexes, minimizing the underlying PI-9/SPI-6-negative hepatocytes to the lethal effects of the granule constituents (Fig. 8) . Therefore, the strong expression of PI-9/SPI-6 by LSECs may offer protection against the anomalously directed granzyme B. Accordingly, we demonstrated that expression of the granzyme B inhibitor in target cells at comparable levels as in LSECs provided complete protection against hepatic NK cell-mediated apoptosis, as has been shown previously using CTLs as effector cells [19 ].
LSECs have been shown to be professional antigen presenting cells (APCs) and thus express the costimulatory molecules CD80 and CD86 [62 ]. Recognition of CD80 or CD86 by NK cells is sufficient to elicit lysis of dendritic cells (DCs) or other activated APCs in vitro, despite their expression of high levels of major histocompatibility complex I molecules [46 , 63 , 64 ]. PI-9/SPI-6 expression in DCs can protect these cells from the CTL they activate [35 ], and it has been suggested that this could protect APCs from NK cells as well [63 ]. LSECs tolerize CTL instead of activating them to kill [65 ]. Therefore, the high PI-9/SPI-6 expression in LSECs could serve to protect them from attached hepatic NK cells rather than from the CTL they tolerize. It has been shown recently that endotoxin can induce PI-9/SPI-6 expression in DCs [35 ]. Liver sinusoids are constantly exposed to gut-derived endotoxin, and LSECs have been shown to bind endotoxin [66 ]. Therefore, this could be a possible mediator of the high PI-9/SPI-6 expression.
In conclusion, we showed that (1) hepatic NK cells induce apoptosis in splenic, NK-resistant P815 tumor cells; (2) hepatic NK cells express FasL, perforin, and granzyme B, and P815 cells are sensitive to the FasL and perforin/granzyme pathway; but (3) hepatic NK cells use only the perforin/granzyme pathway to kill P815 cells, and (4) LSECs strongly express PI-9/SPI-6, providing a protective barrier against granzyme B. In this way, metastasizing cancer cells can be efficiently killed by the highly active perforin/granzyme pathway without adverse effects on liver function in general.
Received March 18, 2002; revised June 3, 2002; accepted June 11, 2002.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  B. M. Burt, G. Plitas, J. A. Stableford, H. M. Nguyen, Z. M. Bamboat, V. G. Pillarisetty, and R. P. DeMatteo CD11c identifies a subset of murine liver natural killer cells that responds to adenoviral hepatitis J. Leukoc. Biol., October 1, 2008; 84(4): 1039 - 1046. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Shanker, G. Verdeil, M. Buferne, E.-M. Inderberg-Suso, D. Puthier, F. Joly, C. Nguyen, L. Leserman, N. Auphan-Anezin, and A.-M. Schmitt-Verhulst CD8 T Cell Help for Innate Antitumor Immunity J. Immunol., November 15, 2007; 179(10): 6651 - 6662. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. E. Wesche-Soldato, C.-S. Chung, S. H. Gregory, T. P. Salazar-Mather, C. A. Ayala, and A. Ayala CD8+ T Cells Promote Inflammation and Apoptosis in the Liver after Sepsis: Role of Fas-FasL Am. J. Pathol., July 1, 2007; 171(1): 87 - 96. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Jiang, B. A. Orr, D. M. Kranz, and D. J. Shapiro Estrogen Induction of the Granzyme B Inhibitor, Proteinase Inhibitor 9, Protects Cells against Apoptosis Mediated by Cytotoxic T Lymphocytes and Natural Killer Cells Endocrinology, March 1, 2006; 147(3): 1419 - 1426. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. A. Parker and C. A. Picut Liver Immunobiology Toxicol Pathol, January 1, 2005; 33(1): 52 - 62. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. B. Barrie, H. W. Stout, M. S. Abougergi, B. C. Miller, and D. L. Thiele Antiviral Cytokines Induce Hepatic Expression of the Granzyme B Inhibitors, Proteinase Inhibitor 9 and Serine Proteinase Inhibitor 6 J. Immunol., May 15, 2004; 172(10): 6453 - 6459. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
