Published online before print August 4, 2005
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
,1
* Department of Internal Medicine, The University of Texas Southwestern Medical School at Dallas; and
Dallas VA Medical Center, Texas
1Correspondence: Division of Digestive and Liver Diseases, Department of Internal Medicine, University of Texas Southwestern Medical Center at Dallas, 5323 Harry Hines Blvd., Dallas, TX 75235-9151. E-mail: geri.brown{at}utsouthwestern.edu
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
responses by 78% ± 18%. GVHD-induced "wasting disease" was reduced in Adv-TNFi recipients [4.4%±5.2% weight loss (n=11)] compared with Adv-ßgal recipients [16.1%±7.6% weight loss (n=11; P=0.0004)] 9 days post-transplant. Composite histopathological scores and individual venulitis scores were reduced with the addition of Adv-TNFi. Hepatic CD8+ T cells in the recipients of Adv-TNFi were reduced as compared with recipients of Adv-ßgal. In conclusion, Adv-TNFi reduces MHC class I disparate alloproliferative responses and hepatic GVHD.
Key Words: cytokines • bone marrow transplantation • cytotoxic T cells • interferon- • therapy
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
bm1 x B6F1 murine model have noted frank destruction of bile ducts with MHC class I and non-MHC-encoded disparities [9
]. Furthermore, in this MHC class I-only, disparate GVHD, previous publications have documented no clinically apparent diarrhea or signs of skin GVHD [9
]. Although the exact mechanism of hepatic injury is still unknown, because of the MHC class I disparity, CD8+ T cell responses likely play a critical role.
Previous murine studies have demonstrated an important role for tumor necrosis factor (TNF) in the pathogenesis of acute GVHD [10
11
12
13
14
]. Studies have shown that TNF blockade reduces severity and mortality of GHVD and markedly decreases hepatic lymphocyte infiltration [15
]. The present studies were designed to examine the effects of TNF blockade on CD8+ T cell activation and proliferation in the B6 bm1 x B6F1 murine model of hepatic GVHD. In contrast to other studies, which used an anti-TNF antibody injected intraperitoneally (i.p.) [16
], the present studies used a TNF inhibitor-encoding adenovirus (Adv-TNFi) injected intravenously, which transfects hepatocytes at high levels, leading to secretion of a TNF inhibitor protein into the serum [17
, 18
]. The TNF inhibitor protein inhibits greater than 99% of TNF/TNF receptor (TNFR) interactions for more than 4 weeks [17
, 18
].
The present studies examined whether TNF affects the development of MHC class I disparate hepatic GVHD. The need for TNF in CD8+ T cell alloresponses and interferon- (IFN-) production was determined by assessing IFN- production and alloproliferation after TNF blockade in a MHC class I disparate mixed lymphocyte culture (MLC). Of note, TNF blockade decreased alloproliferation and IFN- responses in the MHC class I disparate MLC. In addition, TNF blockade reduced the GVHD-induced "wasting disease", hepatic GVHD, the number of hepatic CD8+ T cells, and the percentage of T cells expressing IFN- in the sublethally irradiated B6 x bm1 recipients of B6 donor T cells. Furthermore, the recipients of the MHC class I disparate BM cells (BMC) and spleen cells (SpC) with the Adv-TNFi had significantly less venulitis than recipients of the Adv-ß-galactosidase (ßgal). Thus, the potentiating effects of TNF on hepatic GVHD appear to be mediated by CD8+ T cell alloproliferation and differentiation with a reduction in the inflammatory infiltrate in the venules, bile ducts, and hepatic lobules.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
MLC
B6 responder spleen cells (3x105) were cultured with irradiated MHC class I disparate bm1 (3x105) or syngeneic stimulator cells (3x105) in the presence or absence of TNF inhibitor protein (0.5 mg/ml) or control immunoglobulin (Ig) for 3 days before assay of [3H] thymidine incorporation, as described previously [19
]. In other experiments, TNFR1–/–/TNFR2–/– responder spleen cells (3x105) were cultured with MHC class I disparate bm1 or syngeneic stimulator cells. In all cases, the stimulator cells were T cell-depleted spleen cells.
Enzyme-linked immunosorbent assay (ELISA)
Supernatants from the MLC in triplicate were assayed for IFN- 72 h after initiation of the culture. Briefly, plates were coated with 100 µl/well of the diluted capture antibody for the cytokine and incubated overnight at 4°C. After washing to remove excess capture antibody, between 3 and 5 µl of the supernatants or medium was added to triplicate wells and incubated for 2 h at room temperature. After washing, the enzyme reagent was added to the wells and incubated for 1 h, followed by an incubation with the substrate solution and concluded with the addition of the stop solution. The optical density was read at 450 nm, as described previously [19
]. Standard dilutions were above background and demonstrated linearity to 1.95 pg/ml.
Transplantation
Sublethally irradiated (600 cGy) bm1 x B6 F1 recipient mice were infused with 1 x 107 B6 spleen cells and 5 x 106 T cell-depleted BMC. Plaque-forming units (PFU; 1x109) of Adv-TNFi or Adv-ßgal were injected in the lateral vein of the transplant recipients. These recipient mice were maintained on acidified (pH 2), antibiotic (neomycin, 100 mg/ml; polymyxin B, 10 mg/ml) water for 2 days before transplantation and 1 week after transplantation [9
, 10
]. Weights were taken each day, 7–19 days post-transplantation.
Isolation of hepatic lymphocytes
After i.p. injection with 10 U heparin and CO2 narcosis, the mouse’s abdomen was entered under sterile technique. The inferior vena cava was tied above the right renal vein, the portal vein was cut, and the abdominal portion of the vena cava was perfused with 20 ml phosphate-buffered saline or saline heated to 37°C. The liver was cut, mashed, and filtered through a 100-µm mesh screen. The cell suspension was centrifuged with 35% percoll at 600 g for 15 min, and the cell pellet was used for flow cytometric studies [20
, 21
].
Histological scoring of hepatic GVHD
Liver architecture was assessed in 3 µ-thick hematoxylin and eosin (H&E) and trichrome-stained sections of formalin-fixed liver. One hundred terminal hepatic veins (central veins) were evaluated and called negative for venulitis if no inflammatory cells were identified in or immediately surrounding their walls. The venulitis numerical score was defined as the number of terminal hepatic veins that had venulitis per 100 counted. Subjectively, the severity of the venulitis was defined as minimal when only a few mononuclear cells were found in or surrounding the terminal hepatic vein, mild when a single layer of cells surrounded the vein, moderate when the inflammatory infiltrate was two or more cells deep, and severe when the cellular infiltrate was more prominent and usually confluent obscuring the vein wall.
The composite score assessed three components of hepatic GVHD including lobular hepatitis, venulitis, and bile duct inflammation (composite score=lobular hepatitis score+venulitis score+bile duct inflammation score). Lobular hepatitis was assigned a score of 0 for minimal lobular inflammation, arbitrarily defined as less than 20 inflammatory foci/100x field, score of 1 for moderate inflammation defined as between 20 and 30 inflammatory foci/100x field, and a score of 2 for severe inflammation defined as greater than 30 foci/100x field. Bile duct inflammation was assigned a score of 1 when mononuclear cells were present in the bile duct walls and 0 when no inflammation was noted in over 10 200x fields examined.
Flow cytometric analysis
SpC and intrahepatic lymphocytes (IHL) were washed and incubated for 30 min at 4°C with fluorescein isothiocyanate (FITC)-labeled anti-CD4, anti-CD8
, or the appropriate isotype control FITC mouse anti-rat (all from BD PharMingen, San Diego, CA). The cells were analyzed by fluorescence-activated flow cytometry on the FACScan.
Intracellular staining
Briefly, IHL or SpC were incubated in 37°C, 5% CO2 incubator with phorbol 12-myristate 13-acetate (50 ng/ml) and A23187 (500 ng/ml) for 4 h. Cells were labeled with the FITC-labeled CD8+ or the appropriate isotype control and then fixed with 4% formaldehyde at room temperature for 10 min. After fixation, the cells were incubated on ice for 1 h with saponin-containing medium to permeabilize the membranes. The phycoerythrin-labeled, anti-IFN- (XMG1.2) antibody or the control antibody was added on ice for 30 min. This was followed by two washes with saponin-containing medium for 2 min and one final wash with normal staining medium [20
]. The cells were analyzed by fluorescence-activated flow cytometry on FACScan.
Statistical analysis
Student’s t-test was used to analyze GVHD-induced weight loss and the cellular experiments. Values of P < 0.05 were considered significant.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
 View larger version (21K): [in a new window] |
Figure 1. TNF blockade decreases B6 T cell alloproliferative responses in MHC class I disparate MLC. B6 splenic T cells (3x105) were cultured with irradiated MHC class I disparate bm1 or syngeneic stimulator cells for 3 days in the presence (open bars) or absence (solid) of the TNFR-Ig before assay of [3H] thymidine incorporation. The stimulator cells were T cell-depleted spleen cells. The concentration of the chimeric TNF inhibitor was 0.5 mg/mL. Statistical difference was noted by t-test for Trials 1–4, comparing allogeneic control Ig with allogeneic TNFR-Ig (P=0.0001, 0.004, 0.008, and 0.003, respectively). cpm = mean [3H] thymidine incorporation. Data points represent separate experiments with triplicate determinations in individual experiments. The means and SD are displayed.
|
responses are observed in MHC class I disparate MLC with the addition of TNF inhibitor protein production in MHC class I disparate MLC, the supernatants were collected on the third day from MLC (n=4), and ELISAs were performed to detect the amount of IFN- in the supernatants of MLC. In the pooled experiment, TNFR-Ig reduced the mean IFN- level by 78% ± 18% [14,136±3179 pg/ml vs. 1788±2298 pg/ml (P=0.005; Fig. 2
)].
 View larger version (12K): [in a new window] |
Figure 2. TNF inhibitor decreases IFN- production in MHC class I disparate MLC using control B6 responder cells. B6 spleen cells were stimulated with irradiated bm1 spleen cells for 72 h. The supernatants were harvested and assessed for IFN- content. Solid bars represent MLC with serum with control Ig, and open bars represent MLC with serum with TNF inhibitor. Statistical difference was noted by t-test (P=0.004, P=0.02, and P=0.001, respectively). The means and SD are displayed.
|
 View larger version (19K): [in a new window] |
Figure 3. TNF inhibition in vivo reduces GVHD-induced weight loss in B6 bm1 x B6 F1 GVHD. Irradiated bm1 x B6 F1 mice were infused with 1 x 107 B6 spleen cells and 5 x 106 B6 T cell-depleted BMC and ßgal (n=11)-encoding (Adv-ßgal) or TNF inhibitor (n=11)-encoding adenovirus (Adv-TNFi). Irradiated bm1 x B6 F1 mice were infused with 1 x 107 bm1 x B6 F1 spleen cells and 5 x 106 bm1 x B6 F1 T cell-depleted BMC to serve as a syngeneic (Syng; n=14) control with Adv-TNFi, Adv-ßgal, or control diluent.
|
 View larger version (101K): [in a new window] |
Figure 4. TNF blockade decreases histologic features of hepatic GVHD in MHC class I disparate GVHD induced by B6 SpC transfer to bm1 x B6 F1 recipients. Hepatitis with lobular inflammation, venulitis, and portal inflammation with bile duct involvement was observed in the livers of sublethally irradiated bm1 x B6 F1 recipients of B6 SpC and BMC and Adv-ßgal recipients (A), 21 days post-transplant, and normal liver histology was noted in recipients of B6 SpC and BMC and Adv-TNFi (B).
|
 View larger version (14K): [in a new window] |
Figure 5. TNF blockade decreases the composite histopathologic score of hepatic GVHD in MHC class I disparate GVHD. Three weeks post-transplantation, sublethally irradiated bm1 x B6 F1 recipients of B6 SpC and BMC and Adv-ßgal or Adv-TNFi were harvested. The liver was removed and fixed in formalin. Using H&E and trichrome-stained, 3 µ-thick sections, venulitis, hepatitis, and bile duct inflammation were scored by the pathologist, as described in Materials and Methods.
|
 View larger version (14K): [in a new window] |
Figure 6. TNF blockade decreases venulitis in MHC class I disparate hepatic GVHD induced by B6 SpC transfer into bm1 x B6 F1. Two to 3 weeks post-transplantation, sublethally irradiated bm1 x B6 F1 recipients of B6 SpC and BMC and Adv-ßgal or Adv-TNFi were harvested. The liver was removed and fixed in formalin. H&E stains and more specifically, a trichrome stain of liver, terminal hepatic, or central veins were identified. In all, 100 veins or venules were counted and called positive or negative for venulitis, and total numbers of positive veins or venules are represented in the graph.
|
 View larger version (17K): [in a new window] |
Figure 7. TNF inhibition reduced the number of infiltrating CD8+ T cells in the liver. Lymphocytes were isolated from the liver of sublethally irradiated bm1 x B6 F1 recipients of B6 SpC and BMC and the control Adv-ßgal (n=10) or Adv-TNFi (n=10) 14–21 days post-transplantation. Similarly, lymphocytes were isolated from bm1 x B6 F1 recipients of syngeneic BMC and control Adv-ßgal (n=4) or Adv-TNFi (n=4). Flow cytometry assessed the percentage of CD8+ T cells. Absolute numbers were derived from the percentage of CD8+ T cells x the total number of cells.
|
in the recipients of the Adv-TNFi [2.3x106±1.0x106 (n=9) vs. 0.63x106±0.26x106 (n=6; P=0.001; Fig. 8
)]. Hence, 
50% of the CD8+ T cells were positive for IFN-. Optimal IFN- production in the liver, necessary for the development of MHC class I disparate hepatic GVHD, is dependent on TNF.
 View larger version (9K): [in a new window] |
Figure 8. TNF inhibition reduced the number of IFN--expressing CD8+ T cells in the liver. Lymphocytes were isolated from the liver of sublethally irradiated bm1 x B6 F1 recipients of B6 SpC and BMC and the control Adv-ßgal (n=9) or Adv-TNFi (n=6) 14–21 days post-transplantation. Intracellular stains of IFN- were performed as described in Materials and Methods. Flow cytometry assessed the percentage of CD8+ T cells, which were expressing IFN-. Absolute numbers of double-positive cells were derived from the percentage of CD8+ T cells, which were expressing IFN- x the total number of cells.
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The present studies report a reduced number of CD8+ T cells in the liver and a reduction in the venulitis, suggesting a similar mechanism resulting in a reduction in a MHC class I disparate hepatic GVHD. It is important that this reduction was induced by the TNF blockade, implicating the critical role of TNF in CD8+ T cell alloproliferation and allospecific cytokine responses and the development of hepatic GVHD.
Of note, TNF blockade has been shown to be critical for the induction of acute GVHD in MHC class I and II disparate models and for the development of intestinal manifestations in MHC class II disparate GVHD, but little data are known regarding the role of TNF in MHC class I disparate hepatic GVHD [10 11 12 ]. In a MHC class I and II disparate murine GVHD, induced by the transfer of B6 donor SpC and BMC into the B6 x CBF1 GVHD model, TNF blockade improved the intestinal GVHD but not hepatic GVHD [16 , 23 ]. Conversely, TNF blockade reduced hepatic GVHD of the lethally irradiated B6D2F1 recipient of B6 donor SpC and BMC [14 , 24 ]. The present studies report the role of TNF in the induction of a MHC class I disparate hepatic GVHD. Specifically, multiple variables of GVHD were examined, including the perfusion of the liver of BMT recipients, which provided numbers of infiltrating hepatic lymphocytes during the development of the disease, the cytokine responses of the infiltrating lymphocytes, multiple scores of histological features, as well as weight changes. The combination of the scoring of specific histological features and detailed counts of hepatic lymphocytes revealed important mechanisms of the effect of TNF on hepatic GVHD.
This laboratory has previously shown that TNF/TNFRI interactions on responder CD8+ T cells are critical for optimal alloproliferation and for interleukin-2 responses by T cells in response to alloantigenic stimuli in vitro [19
]. The present experiments demonstrated that TNF inhibition decreases alloproliferation responses and IFN- production in MHC class I disparate MLC. These findings indicate that TNF is critical for the proliferation and activation of CD8+ T cells in vitro. Furthermore, the present studies have shown decreases in the number of hepatic lymphocytes expressing IFN- at Day 15 after BMT occurs with the addition of Adv-TNFi. Therefore, optimal IFN- production in the liver during the development of MHC class I disparate GVHD is dependent on TNF. Other effects of TNF may also play a role in the development of a MHC class I disparate GVHD. For example, indirect effect of TNF on CD8+ T cell activation and proliferation through enhanced cytokine production by Th1 CD4+ T cells may be critical in the initiation and development of hepatic GVHD. Furthermore, TNF may be important in effector functions of CD8+ T cells, which may induce specific pathological features characteristic of hepatic GVHD.
TNF- is an important, multifunctional cytokine, which binds to two receptors (TNF-RI p55, TNF-RII p75). The functions of TNF- include the up-regulation of adhesion molecules, such as intercellular adhesion molecule 1, which recruit lymphocytes to the site of inflammation. Recruitment of lymphocytes to the liver may be a factor in the lower number of CD8+ T cells observed in the livers of bm1 x B6F1 BMT recipients that received Adv-TNFi. Previous studies have shown that the inhibition of migration to the liver of CD8+ T cells reduces hepatic GVHD [14
, 24
, 25
]. Specifically, previous studies have shown that the inhibition of certain inflammatory chemokine production limited the migration of CD8+ T cells in the liver and reduced hepatic GVHD [25
, 26
]. Although migration and adhesion molecules may be affected by TNF blockade, in vitro data, which demonstrated decreased alloproliferation and T cell cytokines, suggested that TNF has a direct effect on T cells. Furthermore, addition of TNFR-Ig at 48 h after initiation of a MHC class I disparate MLC reduced alloproliferation, suggesting that TNF/TNFR interactions may be important, not only in the initiation but also the progression of GVHD.
From a histopathological perspective, hepatic GVHD is the least well-studied. The most recent studies characterize hepatic GVHD by the development of bile duct injuries, venulitis, and portal tract inflammation. Venulitis, which is also called endothelialitis, is a specific but less sensitive marker of acute hepatic GVHD [6 , 7 ]. Endothelialitis is described as the attachment of lymphocytes to the luminal surface of the endothelium of portal or central veins with subsequent damage to the endothelium [7 ]. The venulitis of GVHD differs from viral or alcoholic hepatitis, which is characterized by a subendothelial collection of lymphoid cells associated with hepatocellular necrosis [7 ]. Previous investigators have noted a high degree of parenchymal necrosis during acute GVHD [7 ]. In the current studies, the recipients of the Adv-TNFi had significantly lower venulitis and composite histopathological scores. Furthermore, to improve the objectivity of the effects of TNF blockade on the liver histopathology, this study’s pathologist devised a system that graded three components of hepatic GVHD, including lobular hepatitis, venulitis, and bile duct injury. The composite histopathological scores were lower in the recipients of Adv-TNFi, implicating a role for TNF in the pathogenesis of these pathological features. Of note, specific attention to the effect of the adenoviral vectors on liver histology was assessed by using syngeneic BMT recipients and Adv-ßgal or Adv-TNFi. The effect of Adv-ßgal was variable in terms of hepatitis and bile duct inflammation, resulting in some variability to the composite score of liver histology in the syngeneic BMT recipients. It is important that there was no statistical difference in venulitis amongst syngeneic BMT recipients of Adv-ßgal, Adv-TNFi, and diluent, but a statistical difference was noted in the allogeneic BMT recipients of Adv-ßgal and Adv-TNFi. Although the mechanisms of the effect of TNF on venulitis have not been elucidated, they may be associated with the effect of TNF on costimulation or migration of the lymphocytes.
In summary, TNF-TNFR interactions are important for the activation and proliferation of donor CD8+ T cells in the liver during MHC class I hepatic GVHD. In addition, the number of CD8+ T cells in the liver producing IFN- is also reduced with TNF inhibition. Furthermore, TNF is important in the pathogenesis of the histopathological lesions of hepatic GVHD and the wasting disease associated with acute hepatic GVHD.
Received December 16, 2004; revised April 8, 2005; accepted May 12, 2005.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
B6D2F1 intestinal graft-versus-host disease J. Clin. Immunol. 20,379-388[CrossRef][Medline] receptor plays a critical role in T cell alloreactivity J. Immunol. 164,656-663 antibodies on acute graft-versus-host disease pathologies Blood 91,4051-4055 responses, and intestinal graft-versus-host disease by IL-12-independent mechanisms J. Immunol. 170,5082-5088This article has been cited by other articles:
 |
|
 |
|
  M. G. M. Castor, B. Rezende, C. B. Resende, A. L. Alessandri, C. T. Fagundes, L. P. Sousa, R. M. E. Arantes, D. G. Souza, T. A. Silva, A. E. I. Proudfoot, et al. The CCL3/Macrophage Inflammatory Protein-1{alpha}-Binding Protein Evasin-1 Protects from Graft-versus-Host Disease but Does Not Modify Graft-versus-Leukemia in Mice J. Immunol., March 1, 2010; 184(5): 2646 - 2654. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Borsotti, A. R. K. Franklin, S. X. Lu, T. D. Kim, O. M. Smith, D. Suh, C. G. King, A. Chow, C. Liu, O. Alpdogan, et al. Absence of donor T-cell-derived soluble TNF decreases graft-versus-host disease without impairing graft-versus-tumor activity Blood, July 15, 2007; 110(2): 783 - 786. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
