- Email: [email protected]
Autoimmune disease induced by dendritic cell immunization against leukemia
Leukemia Research 23 (1999) 549 – 557
Autoimmune disease induced by dendritic cell immunization against leukemia Marie A. Roskrow a, Dagmar Dilloo b, Nobuhiro Suzuki a, Wanyung Zhong b, Cliona M. Rooney a,c, Malcolm K. Brenner b,c,* a
Department of Virology and Molecular Biology, St. Jude Children’s Research Hospital, Memphis, TN, USA b Di6ision of Bone Marrow Transplantation, St. Jude Children’s Research Hospital, Memphis, TN, USA c Center for Cell and Gene Therapy, Baylor College of Medicine, 1102 Bates St, Suite 1100, Houston, TX 77030, USA Received 21 September 1998; accepted 28 January 1999
Abstract Induction of an optimal immune response will likely be a prerequisite for successful immunotherapy of human leukemias and other malignancies. Dendritic cells are highly effective at inducing an immune response to antigens to which the host is unresponsive, while transgenic expression of the costimulator molecule CD40 ligand (gp39/CD154) and the T cell growth factor interleukin 2 (IL2) are also able to augment immune responsiveness. We therefore investigated whether a combination of these two distinctive approaches to immunostimulation could safely increase the anti-tumor immune response compared to each stimulus alone. We injected BALB/CBYJ mice with syngeneic dendritic cells (DC) exposed to A20 lymphoblastic leukemia cell-derived peptides and proteins which had been acid-eluted from the cell surface. In additional mice, the pulsed DC were mixed with genetically modified syngeneic fibroblasts that were expressing CD40 ligand or secreting interleukin 2 (IL2). Three days after their third, weekly, vaccination, they were challenged with parental A20 cells. Tumor growth was suppressed by responses to pulsed DC alone (PB0.02). This suppression was further enhanced when pulsed DC were coinjected with fibroblasts expressing CD40 ligand and IL2 (PB 0.0005 compared to DC alone) even though CD40 ligand and IL2-expressing fibroblasts alone offered no significant protection in this model. Mice receiving the full complement of immunostimulants either failed to develop visible tumors or developed small tumors which quickly necrosed and regressed, allowing the mice to become long term tumor-free survivors. Antibody mediated depletion of either CD4 + or CD8+ T-cell subset significantly reduced the level of protection afforded by the vaccination. However, it became evident that this intensive stimulation of the immune system lead not only to tumor eradication but also to destruction of cells bearing normal self antigens. Hence, 60 days after challenge with A20 cells all mice in the DC/IL2/CD40 ligand group developed a severe, systemic autoimmune disorder that resembled graft – versus–host disease and manifest itself by significant peripheral blood cytotoxicity against autologous fibroblasts, blood dyscrasias, gross hepatosplenomegaly, cachexia and fur loss. This phenomenon depended on CD8 + cytotoxic T lymphocytes. Our results therefore suggest that the most effective strategies of immunotherapy against leukemia may also exceed the threshold of anergic cells, leading to a loss of self tolerance to normal self-antigens and the induction of an CD8 + anti-self effector response. © 1999 Elsevier Science Ltd. All rights reserved. Keywords: Autoimmune disease; Dendritic cells; Leukemia; Tumor vaccine
1. Introduction
Abbre6iations: APC, Antigen presenting cell; ARCC, American Type Tissue Collection; DC, Dendritic cell; GM-CSF, Granulocytemacrophage colony stimulating factor; gp39, CD40 ligand; IL2, Interleukin 2; TFA, Trifluroacetic acid; WBC, White Blood Count. * Corresponding author. Tel.: + 1-713-770-4662; fax: + 1-713-7704668. E-mail address: [email protected] (M.K. Brenner)
Specialized antigen presenting cells (APC) including dendritic cells (DC), are highly effective at presenting antigen and recruiting naive or quiescent CD4+ and CD8+ T cells into immune responses [1]. This property has been exploited in several novel immunotherapeutic strategies, and might well provide a basis for effective anti-cancer therapy as our ability to manipu-
0145-2126/99/$ - see front matter © 1999 Elsevier Science Ltd. All rights reserved. PII: S 0 1 4 5 - 2 1 2 6 ( 9 9 ) 0 0 0 4 5 - 4
550
M.A. Roskrow et al. / Leukemia Research 23 (1999) 549–557
late these cells improves [2 – 7]. A growing concern, however, is that enhancement of APC-elicited responses might compromise unresponsiveness to self-antigens expressed by normal cells, thereby producing autoimmune disease. We had the opportunity to test this prediction in mice treated with DC stimulated to ensure maximal recruitment of effector T cells. The paucity of tumor-specific or tumor-associated proteins known to be presented by human cancer cells [8,9] has led to efforts to use peptides eluted from the tumor cells as a source of antigen. Vaccination of mice with bone marrow-derived DC pulsed with unfractionated tumor peptides did in fact reduce the growth of established, weakly immunogenic tumors [10], and similar results were obtained by transducing DC with tumor cell-derived mRNA [11] or by fusing dendritic cells with the tumor cells themselves [4]. No adverse effects of immunization were noted in any of these studies. We hypothesized that the anti-leukemia tumor responses mediated by antigen-laden DC could be further augmented by transgenic expression of additional immunostimulatory molecules, including the CD40 ligand (gp39/CD154) and interleukin 2 (IL2). This reasoning was based on the high levels of CD40 antigen found on professional antigen presenting cells, including dendritic cells and the key role played by CD40 – CD40 ligand interactions in augmenting the function of dendritic cells [12–14]. Such interactions also stimulate CD4 + and CD8 + T cells after they have engaged antigen on APC [13,15–19]. We recently showed that transgenic expression of the CD40 ligand molecule by accessory cells markedly enhanced the immune response directed against coinjected leukemic cells, which was further amplified by transgenic production of IL2 [20]. In the study described here, a combination of tumor – antigen pulsed DC and syngeneic fibroblasts genetically modified to express CD40 ligand or IL2 uniformly protected mice against the otherwise weakly immunogenic A20 line of murine leukemic cells. However, all of the animals subsequently developed severe, systemic autoimmune disease, illustrating the potential hazard of stimulating the immune system to an extreme level.
2. Material and methods
gen, Cambridge, MA) and IL-4 (50 ng/ml, Endogen). After 48 h in culture, the loosely adherent cells were replated in a final concentration of 1× 106 cells/ml. As before, the media contained GM-CSF (150 ng/ml) and IL-4 (50 ng/ml). After 72 h in culture the population was further purified by depletion of contaminating CD11b (Mac1)/B220 double positive cells. The resulting cells were morphologically large irregularly shaped and non adherent, expressing CD45, MHC Class I and II, CD 11c, CD11b (Mac-1), Mac-3 and CD80 (all antibodies from Pharmingen) by flow cytometry. These cells were harvested and pulsed with A20 peptides as described below.
2.2. Cell lines The pre-B lymphoblastoid cell line A20 was derived from a spontaneous neoplasm [21]. Both the A20 and the CL7.1 fibroblast cell line [22] were derived from BALB/CBYJ/c mice and were obtained from American Type Tissue Collection (ATCC, Rockville, MD).
2.3. Elution of proteins and peptides from the surface of A20 cells and pulsing of mouse dendritic cells A20 cells (1 × 109) were pelleted by centrifugation at 1500 rpm for 5 min. The cell pellet was resuspended in 10 ml of precooled 0.4% trifluroacetic acid (Sigma Chemical Co.) for 2 min on ice. The supernatant containing the eluted proteins and peptides was collected after centrifugation at 1500 rpm for 5 min. For filtration and concentration of the eluted proteins, supernatants were centrifuged over a 10 kDa Centriprep cut-off filter (Amicon Inc, Beverley, MA) at 3600 rpm for 50–60 min. The eluted proteins were analyzed by mass spectrophotometry of both the filtrate and the retentate. The latter contained proteins \10 kDa and the filtrate smaller ones. Both fractions were dried overnight by high-speed vacuum centrifugation. The dried pellets were resuspended in Tris EDTA (pH 8) and quantified by a commercially available BSA protein assay kit (Pierce, Rockford, IL). Part of the A20 protein fraction (34 g) was added to 1 ×106 enriched dendritic cells for 12–16 h; 3–4 h before vaccination, the peptide fraction was added to the cells at a concentration of 8:5 g/1 ×106 cells.
2.1. Culture and phenotyping of mouse dendritic cells 2.4. Retro6iral 6ectors Dendritic cell-enriched populations were obtained as described previously [7,11]. In brief bone marrow mononuclear cells were flushed from the femurs of BALB/CBYJ mice with RPMI 1640 media (GIBCO Life Technologies, Grand Island, NY). The eluted cells were washed twice and then cultured in RPMI/10% fetal calf serum at a final concentration of 1 × 106 cells/ml in the presence of GM-CSF (150 ng/ml, Endo-
Supernatants containing the retroviral vector G1Na [23,24] or G1NaCvIL2 [25] (Genetic Therapy Inc, Gaithersburg, MD) were used to transduce the neomycin phosphotransferase gene neo and the IL2 gene respectively into CL7.1 fibroblasts. Transduction was performed at 37°C, in 5% CO2 in the presence of 6 g/ml polybrene. The transduced CL7.1 cells were se-
M.A. Roskrow et al. / Leukemia Research 23 (1999) 549–557
lected in G418 sulfate (GIBCO, Gaithersburg, MD) at an active concentration of 1 g/ml. To generate a retroviral vector containing the murine CD40 ligand gene, we first cloned the CD40 ligand cDNA by semi-nested PCR from an activated BALB/2 T-Cell cDNA library made in lambda (kindly provided by Dr C. Coleclough [26]). The primers 5%GCGGTACCTTTCAGTCAGCATGATAGAAC and 5%CTTATTCCAGCTCT-ATGTGCCT were used for the first round of amplification and the same forward primer and the primer 5%GGATCGATATACGGAAGACTGCCAGC for the second round. The PCR product was subcloned into pGEM7Z (Pharmacia, Piscataway, NJ). DNA sequence analysis confirmed the identity of the cloned CD40L cDNA with the published murine cDNA sequence. Subsequently, the plasmid PG1a.mCD40L was made by subcloning theCD40L cDNA into the retroviral vector pGIa (Genetic Therapy Inc, Gaithersburg, MD) with the CD40L gene being driven by the LTR promoter. The producer cell line BOSC, kindly provided by Dr. M. Scott, was transfected with the plasmid pG1a.mCD40L by calcium precipitation, as previously described [27]. Supernatants were collected from the BOSC cells at confluence and then used to transfect the CL7.1 fibroblast cell line. After transduction, fibroblasts were stained with the CD40L antibody MR1 (Pharmingen, San Diego, CA) and then sorted by flow cytometry (FACS-Star Plus; Beckon Dickinson, Mansfield, MA).
2.5. Vaccination with pulsed dendritic cells and tumor growth Experiments to determine the effects of vaccination with pulsed dendritic cells on tumor growth were performed in a syngeneic system, as the A20, the CL7.1 and the dendritic cells were all derived from BALB/ CBYJ mice. For assessment of the inhibitory effects of vaccinations with A20-pulsed dendritic cells and CD40 Ligand/IL2 costimulation on tumor development, BALB/CBYJ mice aged 12 – 18 weeks (The Jackson Laboratory, Bar Harbor, ME) were inoculated subcutaneously (s.q.) with a mixture of 2 × 105 CL7.1 fibroblasts, each expressing either the IL2, CD40 ligand or the neo gene, and 2– 3×105 protein-pulsed or unpulsed dendritic cells. Seven mice in each group received three such injections at different sites 2 week apart. Three days after the third injection, the mice were challenged with an s.q. injection of 1 × 105 A20 cells. The cell mixtures were not irradiated. Positive tumor growth was defined as any increase in tumor volume\10 cmm, as determined from measurements of transverse and perpendicular diameters. Animals were killed when tumor growth resulted in a 30% increase in body weight, in ulceration or in distress. Wilcoxon or Fisher’s exact tests were used to compare
551
mean tumor volumes or survival in mice given different combinations of immune stimulants.
2.6. T-Cell depletion Mice were depleted of T lymphocytes by intraperitoneal (ip) injection of the Gk1.5 (anti CD4 and 2.43 (anti-CD8) monoclonal antibodies (mAbs) [28,29], kindly provided by Dr Peter Doherty. The antibodies were injected ip 1 day before tumor challenge and then every other day for a week, followed by four weekly injections. After euthanasia of tumor-bearing animals, the depletion status was \ 98% of the relevant subset, as demonstrated by flow cytometry analysis of splenocytes and peripheral blood.
2.7. Characteristics of transduced fibroblasts By ELISA (R&D, Minneapolis, MN), IL2 production by CL7.1 fibroblasts ranged from 80 to 100 IU/106 cells over 24 h. More than half (50–60%) of the fibroblasts transduced with pG1a.m.CD40 ligand expressed CD40 ligand after selection by flow cytometry and subsequent cell culture. Control fibroblasts were transduced with the G1Na vector.
2.8. Cytotoxicity assays Animals were bled by cardiac puncture and erythrocyte depletion was performed using a standard ammonium chloride depletion method. The cytotoxicity of the peripheral blood lymphocytes was analyzed in a standard 4 h chromium-51 release assay using effector: target ratios of 20:1, 10:1 and 5:1. Target cells included autologous and HLA-mismatched fibroblasts, the parental A20 tumor cell line and Yac-1 which is sensitive to natural-killer cells. Specific cytotoxic activity was determined with the formula: % specific release= [(experimental release− spontaneous release)/(total release− spontaneous release)]×100. Standard errors of the means of triplicate cultures was less than 10%.
3. Results
3.1. Vaccination with tumor protein-pulsed dendritic cells can protect mice against subsequent challenge with A20 tumor cells Since no tumor-specific antigens are known for the A20 leukemia cell line, (or for many human tumors), we used unfractionated, concentrated proteins and peptides, acid-eluted from the surface of A20 cells, to pulse the DC-enriched population generated from syngeneic bone marrow [13]. For some groups of mice, the pulsed
552
M.A. Roskrow et al. / Leukemia Research 23 (1999) 549–557
Fig. 1. Vaccination with A20 peptide-pulsed DC, alone or in combination with gene-modified fibroblasts expressing CD40 ligand and IL2, protects mice against subsequent challenge with A20 leukemia cells. Each group (seven mice per group) received three s.q. injections, 1 week apart, with a mixture of 2 × 105 transduced CL7.1 fibroblasts expressing either CD40 ligand, IL2 or neo (control) gene and 2 – 3 × 105 peptide-pulsed dendritic cells or unpulsed DC. Three days after the third injection, the mice were challenged with 1 × 105 A20 leukemia cells. Tumor volumes (cmm) are presented as means. () control mice (A20 leukemia challenge only) (2) unpulsed DC with CD40 ligand/IL2 costimulation; () peptide-pulsed DC without costimulation; ( ) peptide-pulsed DC with CD40 ligand/IL2 costimulation.
DC were mixed with genetically modified syngeneic fibroblasts that were expressing CD40 ligand or secreting IL2. The animals were vaccinated s.q. at weekly intervals prior to challenge with A20 tumor cells. Fig. 1 shows the mean tumor volume for each group of mice at various times after challenge with A20 cells. Tumor growth was weakly but consistently suppressed
by responses to pulsed DC alone (PB 0.02). This suppression was further enhanced by the addition of fibroblasts expressing CD40 ligand or IL2 (PB 0.0005 compared to DC alone). In these experiments, the mice received three vaccinations prior to the tumor challenge. It should be noted that immunization with unpulsed dendritic cells in combination with CD40 ligand and IL2 was ineffective in this model. Mice receiving the full complement of immunostimulants either failed to develop visible tumors or developed small tumors which soon became necrotic and regressed. As a result, the seven mice vaccinated with CD40 ligand, IL2 and pulsed DC survived tumor free\ 60 days, compared with none of the controls (Fisher’s test PB 0.005). This effect could also be produced with two injections, although the benefit was less pronounced (Fig. 2).These same mice were used to assess the effect of antibody-mediated depletion of CD4+ and CD8 + T-cell subsets (Fig. 2). In vivo depletion of T lymphocytes significantly reduced the level of protection afforded by the combination vaccine, so that all CD4-depleted mice had developed tumor by day ten and all CD8-depleted mice by day 25 (PB0.001 and PB 0.01 respectively compared to non-depleted mice) confirming that the anti-leukemia response relied on the recruitment of both subsets of effector T cells, although predominantly on CD4+ cells.
3.2. Immunotherapy with tumor antigen-pulsed DC, costimulated with CD40 ligand and IL2, induces an autoimmune disease Although mice vaccinated with antigen-pulsed DC and fibroblasts expressing CD40 ligand or IL2 were
Fig. 2. In vivo depletion of CD4+ and CD8+ lymphocytes reduces the anti-leukemia response. Three groups of mice (seven per group) received two s.q. injections of peptide-pulsed DC in combination with gene-modified fibroblasts expressing CD40 ligand and IL2. Control mice were injected with unpulsed DC in the absence of any costimulation. Three days after the second injection, all mice were challenged with 1 × 105 A20 leukemia cells. Tumor volumes (cmm) are presented as means. () control mice (unpulsed DC without costimulation); ( ) peptide-pulsed DC with CD40 ligand/IL2 costimulation; (2) peptide-pulsed DC with CD40 ligand/IL2 costimulation and CD8-depletion; ( × ) peptide-pulsed DC with CD40 ligand/IL2 costimulation and CD4-depletion.
M.A. Roskrow et al. / Leukemia Research 23 (1999) 549–557
553
Table 2 Spleen and liver weights in non-vaccinated mice compared to mice receiving the combination vaccine
Range Median Mean P values vaccine v. no vaccine
Fig. 3. Photographs of mice that developed an autoimmune-like disease and their normal litter mates.
protected against tumor, they did not remain healthy. Sixty days after challenge with A20 cells, they showed signs of a systemic disorder resembling graft versus host disease (GvHD), and characterized by pronounced periorbital edema and fur loss as well as ruffled body fur (Fig. 3). Further investigation confirmed the presence of a systemic autoimmune disorder (Table 1). Although the total white blood count (WBC) was within the normal range, they all had a reversal of the normal neutrophil/lymphocyte ratio to produce a relative neutrophilia and lymphopenia. In addition the mice were anemic and thrombocytopenic. At necropsy there was evidence of gross hepatosplenomegaly and exudative ascites. As compared to mice receiving an A20 tumor
Spleen weight (g)
Liver weight (g)
Vaccine
No vaccine
Vaccine
No vaccine
0.17–0.51 0.29 0.31 B0.0005
0.08–0.1 0.09 0.09
1.13–1.72 1.33 1.39 B0.0002
0.9–1.01 0.99 0.98
challenge alone (Table 2), there was significant splenomegaly (PB 0.0005) and hepatomegaly (PB 0.0002). Dip-stick analysis of their urine showed heavy proteinuria in all cases. Immunohistochemical analysis of the affected organs did not show any evidence of cellular infiltration or immune complex deposition. Flow cytometric analysis of splenocytes (Fig. 4), showed a decrease in both CD4+ and CD8+ lymphocytes by comparison to normal controls (PB 0.03) and a corresponding increase of activated macrophages (Mac1+ Mac3+ ) (PB0.01). Fresh unstimulated peripheral blood mononuclear cells from the animals with autoimmune disease had high levels of specific cytotoxic activity against syngeneic fibroblasts without significant killing of H2-mismatsched fibroblasts or Yac-1 (NK sensitive) target cells (Fig. 5a). As expected, peripheral blood mononuclear cells from healthy mice failed to recognize any of the target cells (Fig. 5b). A feature of A20 cells is that they express class II MHC molecules, from which peptides can be eluted that are capable of stimulating CD4+ T helper cells. Since CD4+ T cells play a major role in the etiology of many autoimmune diseases [30–33], we investigated whether the syndrome could be prevented by in vivo depletion of CD4+ T cells. In fact, such depletion had no effect: all seven mice treated in this manner developed signs of autoimmune disease such as periorbital edema and fur ruffling. Of note, these signs appeared as early as 14 days after leukemia challenge which was significantly earlier than in non-depleted mice injected with the combination vaccine. In contrast, none of the
Table 1 Peripheral blood counts in mice receiving the combination vaccine
Normal Range Mean
Total WBC (×103/ul)
Neutrophils (%)
Lymphocytes (%)
Hb (g/dl)
Platelets (×103/ul)
5–12,000 4.9–18.6 9.9
5–40 66–86 74.6
63–75 11–34 24.2
11–14 8.3–12.1 10.4
80–1000 170–1090 600
554
M.A. Roskrow et al. / Leukemia Research 23 (1999) 549–557
Fig. 4. Flow cytometric analysis of splenocytes from normal mice and mice that developed an autoimmune-like disease. Data are reported as percent of mononuclear cells.
mice depleted of CD8 + T cells developed the syndrome. Necropsy of four mice from both, the CD4-depleted and the CD8-depleted groups was performed for analysis of additional signs of autoimmune disease. Examinations confirmed that CD8-depleted mice did not develop the disorder (Table 3). Thus, induction of autoimmune disease in our model is independent of CD4 + helper cells, relying instead on CD8 + cytotoxic T lymphocytes [34].
produce organ damage, the pattern of which is dictated by the class of effector cells whose anergy is reversed, and the target antigens they recognize. Our data indicate that combination therapy with pulsed DC and fibroblasts expressing a costimulator molecule or a T-cell growth factor is not only an effective way to generate antitumor immunity, but also exceeds the activation threshold of anergic cells required to induce a CD8+ effector response. Of note, unpulsed dendritic cells neither prevented tumor growth, nor induced autoimmune disease even in the presence of CD40 ligand/
4. Discussion There has been considerable interest in using antigen presenting cells including dendritic cells (DC) as therapeutic agents in patients with cancer because of their capacity to capture tumor specific antigens and recruit T cell dependent anti-tumor immunity [2 – 6,35,36]. In our own model, we combined leukemia-peptide pulsed dendritic cells with two powerful costimuli, CD40L and IL2, delivered by syngeneic fibroblasts. While this combination enhanced anti-tumor activity, it also induced a severe systemic autoimmune disorder. What mechanisms could account for autoimmune disease in mice receiving combination therapy of the type described here? Physiologically, most autoreactive T-cell clones are eliminated in the thymus [37,38]. Those that reach the periphery are functionally inactivated (anergized), probably as a result of stimulation by self-antigen in the absence of appropriate costimulatory molecules [39,40]. These anergized cells are not eliminated, so that self-antigens might be rendered immunogenic by any stimulus that exceeds the anergy threshold [30,41]. The resulting activated effector cells may then
Fig. 5. Fresh peripheral blood cells from mice with autoimmune disease exhibit high levels of specific cytotoxic activity against syngeneic fibroblasts without significant killing of H2-mismatched fibroblasts or Yac-1 (NK sensitive) target cells. Cytotoxic activity was measured in a standard chrom-release assay in mice with the autoimmune-like disorder (a) and in normal mice (b) at various effector to target ratios. Cytotoxic activity of unstimulated peripheral blood mononuclear was measured against several targets: ( ) Yac-1 cells, () H2-mismatched fibroblasts, () matched fibroblasts and ( × ) A20 leukemia cells.
M.A. Roskrow et al. / Leukemia Research 23 (1999) 549–557 Table 3 Autoimmune disease in CD4-depleted versus CD8-depleted mice Signs
CD4-depletion
CD8-depletion
Periorbital edema/ruffled fur Hepatosplenomegaly Proteinuria Ascites Lymphopenia
4/4 4/4 4/4 4/4 4/4
0/4 0/4 0/4 0/4 0/4
IL2. Since these DC already present an array of processed self-peptides [1], the induction of autoimmunity by tumor peptide-pulsed DC implies that the self peptides obtained from tumor cells are either quantitatively or qualitatively different from those normally processed by dendritic cells. In most murine models of autoimmune disease, it is CD4+ T lymphocytes that mediate disease development [42]. In our study, the autoimmmune-like disorder was mediated by CD8+ T cells suggesting that the tumor cell-derived self-antigens were presented by class I molecules expressed on the dendritic cells. Conventionally, class I associated peptides derive from endogenous antigen processing. However, there are now several reports indicating that exogenous peptides can access class I molecules. While some studies suggest that it is specialized APCs that can shunt exogenous peptide into the class I pathway [43,44] others have attributed class I presentation of exogenous peptide to the fact that overfeeding of APCs leads to particle overloading of the phagosomal compartments with subsequent release of antigen into the cytosol [45]. Most studies analyzing the possible mechanisms that enable class I presentation of exogenous peptides focus on expression of non-self peptides with analysis of crosspriming events being the primary example [46]. A more recent report demonstrates that processing of exogenous self-antigens through the class I pathway is also possible [47]. Since leukemic cells may overexpress certain self-antigens such as minor histocompatibility antigens [48], antigen excess may contribute to the observed autoimmune-like effects. The contribution of CD40 ligand to autoimmunity in this model likely comes from its known immunostimulatory effects on both APC and T lymphocytes. Engagement of the CD40 counter receptor expressed at high densities on dendritic cells [13] induces DC maturation with enhanced expression of adhesion molecules and the costimulatory molecule B7.1 [14]. Upregulation of B7 molecules is one of the critical events in CD40/ CD40 ligand-mediated immune-activation. Binding of CD40 ligand to DC also promotes IL-12 release [49,50], which increases the function of IL2-activated cytotoxic T cells and expands NK cell populations. Finally, CD40 ligand also stimulates the effector cells of the
555
immune response. Under physiological conditions, Tcell responses can be directly amplified by CD40–CD40 ligand interactions between antigen-activated T cells [15,17,18]. The presence of IL2 complements these activities, bypassing a requirement for CD4+ mediated helper function to induce CD8+ effector cells [20]. Hence a CD8+ -dependent autoimmune disorder develops in these mice even after CD4 + helper cells have been depleted. There are several other lines of evidence suggesting that costimulatory molecules play an important role in the breakdown of self-tolerance. In human autoimmune disease such as rheumatoid arthritis and autoimmune thyroiditis, upregulation of B7.1 on monocytes, T cells and B cells has been demonstrated [51,52]. The impact of aberrant expression of costimulatory molecules on potential target tissue of the autoimmune response has been examined in transgenic mice. In these models, aberrant expression of one costimulatory molecule such as B7 alone was not sufficient to induce autoimmune disease [53]. Crossbreeding with transgenic mice expressing a second signal with broad immunostimulatory capacity such as TNF-alpha, however, resulted in breakdown of self-tolerance and the development of severe diabetes [54]. Similarly, in our leukemia vaccine model, three components were required for induction of the autoimmune-like disorder, namely the leukemiaderived peptides, the dendritic cells and CD40 ligand /IL2 costimulation. In addition to CD40/CD40 ligandmediated upregulation of B7 molecules on both target as well as immune effector cells and the implication of B7 in autoimmune disease, it has been shown that treatment of autoimmune mice with anti-CD40 ligand antibody can prevent disease development. In murine models of lupus nephritis, collagen-induced arthritis and experimental allergic encephalomyelitis [55–58], injections of anti-CD40 ligand antibodies proved protective, underscoring the impact of CD40 ligand on the development of autoimmune-disease. A similar mechanism, may contribute both to spontaneous autoimmune disease and to the perpetuation of chronic graft–versus–host disease. In both settings, APC may be stimulated to take up large quantities of host antigens following cell destruction by microorganisms or by alloreactive T lymphocytes. These APC are in an environment in which there is a high level of expression of costimulatory molecules such as CD40L on activated T cells, and in which cytokines such as IL2 are readily available. If this combination exceeds the threshold of anergy, self-perpetuating autoimmune/ chronic GvH disease may result, in an analogous manner to that we describe in the tumor model. Because human tumors likely express a circumscribed and heterogeneous range of tumor antigens of limited immunogenicity, there is strong incentive to prepare tumor vaccines that can overcome non-responsiveness
556
M.A. Roskrow et al. / Leukemia Research 23 (1999) 549–557
to weak immunogens. While we have shown that strategies combining peptide-pulsed DC and costimulatory molecules may fulfill these requirements, our results also indicate that the most effective immunostimulation may interfere with tolerance to normal self antigens. Although generally undesirable, such an effect, if controlled, could have therapeutic benefit in some circumstances. If, for instance, antigens known to be specific for certain organs and tissues [36] [48,59,60] were used to pulse dendritic cells in the strategy described here, one might anticipate the development of autoimmunity against the relevant organ. Destruction of both the normal tissue and antigenically related malignant cells derived therefrom could offer an acceptable therapeutic strategy if the cancer were disseminated and the targeted organ not essential for life.
Acknowledgements We would like to thank John Gilbert for scientific editing and Dr Helen Heslop for helpful scientific discussion. This work was supported in part by NIH grants CA 61384, CA 20180, CA 75014 and the Cancer Center Support grant, and by the American Lebanese Syrian Associated Charities (ALSAC). M. Roskrow and D. Dilloo equally provided the concept, design, analysis of data, drafting and critical review of the manuscript, collection of data and statistical expertise. N. Suzuki and W. Zhong assisted with data collection and technical support. C.M. Rooney provided critical revision of the article and funding. M. Brenner provided critical revision of the article, final approval and funding.
References [1] Steinman RM. The dendritic cell system and its role in immunogenicity. Annu Rev Immunol 1991;9:271. [2] Hsu F, Benike C, Fagnoni F, Lines TM, Czerwinski D, Taidi B, et al. Vaccination of patients with B-cell lymphoma using autologous antigen-pulsed dendritic cells. Nat Med 1996;2:52. [3] Murphy G, Tjoa B, Ragde H, Kenny G, Boynton A. Phase I clinical trial: T-cell therapy for prostate cancer using autologous dendritic cells pulsed with HLA-A0201-specific peptides from prostate-specific membrane antigen. Prostate 1996;29:371. [4] Gong J, Chen D, Kashiwaba M, Kufe D. Induction of antitumor acitvity by immunization with fusions of dendritic and carcinoma cells. Nat Med 1997;3:558. [5] Henderson RA, Nimgaonkar MT, Watkins SC, Robbins PD, Ball ED, Finn OJ. Human dendritic cells genetically engineered to express high levels of the human epithelial tumor antigen mucin (MUC-1). Cancer Res 1996;56:3763. [6] van-Elsas A, van-der-Burg SH, van-der-Minne CF, Gorghi M, Mourer JS, Melief CJ, et al. Peptide-pulsed dedritic cells induce tumoricidal cytotoxic T lymphocytes from healthy donors against stably HLA-A*0201-binding peptides from Melan-A/ MART-1 self antigen. Eur J Immunol 1996;26:1683.
[7] Celluzzi CM, Mayordomo JI, Storkus WJ, Lotze MT, Falo LDJ. Peptide-pulsed dendritic cells induce antigen-specific CTLmediated protective tumor immunity. J Exp Med 1996;183:283. [8] Henderson RA, Finn OJ. Human tumor antigens are ready to fly. Adv Immunol 1996;62:217. [9] Boon T, van Der Bruggen P. Human tumor antigens recognized by T lymphocytes. J Exp Med 1996;183:725. [10] Paglia P, Chiondoni C, Rodolfo M, Colombo MP. Murine dendritic cells loaded in vitro with soluble protein prime cytotoxic T lymphocytes against tumor antigen in vivo. J Exp Med 1996;183:317. [11] Boczkowski D, Nair SK, Synder D, Gilboa E. Dendritic cells pulsed with RNA are potent antigen-presenting cells in vitro and in vivo. J Exp Med 1996;184:645. [12] Alderson MR, Armitage RJ, Tough TW, Strockbine L, Fanslow WC, Spriggs MK. CD40 expression by human monocytes: regulation by cytokines and activation of monocytes by the ligand for CD40. J Exp Med 1993;178:669. [13] Freudenthal PS, Steinman RM. The distinct surface of human blood dendritic cells as observed after an improved isolation method. Proc Natl Acad Sci USA 1990;87:7698. [14] Caux C, Massacrier C, Vanbervliet B, Dubois B, van Kooten C, Durand I, Banchereau J. Activation of human dendritic cells through CD40 cross-linking. J Exp Med 1994;180:1263. [15] Cayabyab M, Phillips JH, Lanier L L. CD40 preferentially costimulates activation of CD4 + T lymphocytes. J Immunol 1994;152:1523. [16] Wagner DH, Stout RD, Suttles J. Role of the CD40–CD40 ligand interaction in CD4 + T cell contact-dependent activation of monocyte interleukin-1 synthesis. Eur J Immunol 1994;24:3148. [17] Fanslow WC, Clifford KN, Seaman M, Alderson MR, Spriggs MK, Armitage RJ, et al. Recombinant CD40 ligand exerts potent biologic effects on T cells. J Immunol 1994;152:4262. [18] Armitage RJ, Tough TW, Macduff BM, Fanslow WC, Spriggs MK, Ramsdell F, et al. CD40 ligand is a T cell growth factor. Eur J Immunol 1993;23:2326. [19] Paul WE, Seder RA. Lymphocyte responses and cytokines. Cell 1994;76:241. [20] Dilloo D, Brown M, Roskrow M, Zhong W, Holden W, Holladay M, et al. CD40 ligand acts as a co-stimulator molecule for the generation of an anti-tumor immune response. Blood 1997;90:1927. [21] Kim KJ, Kanellopoulos-Langevin C, Verwin RM, Sachs DH, Asofsky R. Establishment and characterization of BALB/c lymphoma lines with B cell properties. J Immunol 1979;122:549. [22] Patek PQ, Collins JL, Cohn M. Transformed cell lines susceptible or resist to in vivo surveillance against tumorigenesis. Nature 1978;276:510. [23] Bender MA, Palmer TD, Gelinas RE, Miller AD. Evidence that the packaging signal of Moloney, Murine Leukemia Virus extends into the gag region. J Virol 1987;61:1639. [24] Brenner MK, Rill DR, Holladay MS, Heslop HE, Moen RC, Buschle M, et al. Gene marking to determine whether autologous marrow infusion restores long-term haemopoiesis in cancer patients. Lancet 1993;342:1134. [25] Brenner MK, Furman WL, Santana VM, Bowman L, Meyer W. Phase I study of cytokine-gene modified autologous neuroblastoma cells for treatment of relapsed/refractory neuroblastoma. Hum Gene Ther 1992;3:665. [26] Coleclough C. Cell-free expression of large collections of cDNA clones using positive-selection I phage vectors. Meth Enzymol 1993;217:152. [27] Pear WS, Nolan GP, Scott ML, Baltimore D. Production of high-titer helper-free retroviruses by transient transfection. Proc Natl Acad Sci USA 1993;90:8392.
M.A. Roskrow et al. / Leukemia Research 23 (1999) 549–557 [28] Dialynas DP, Quan ZS, Wall KA, Pierres A, Quintans J, Loken MR, et al. Characterization of the murine T cell surface molecule, designated L3T4, identified by monoclonal antibody GK1.5; similarity of L3T4 molecule. J Immunol 1983;131:2445. [29] Sarmiento M, Glasebrook AL, Fitch FW. IgG of IgM monoclonal antibodies reactive with different determinants on the molecular complex bearing LYT 2 antigen block T cell-mediated cytolysis in the absence of complement. J Immunol 1980;125:2665. [30] Fulcher DA, Lyons AB, Korn SL, Cook MC, Koleda C, Parish C, et al. The fate of self-reactive B cells depends primarily on the degree of antigen receptor engagement and availability of T cell help. J Exp Med 1996;183:2313. [31] Ferrara JLM, Deeg HG. Graft–versus–host disease. N Engl J Med 1991;324:667. [32] Korngold R. Biology of graft–versus–host disease. Am J Ped Hem/Onc 1993;15:18. [33] Truitt RL, Atasoylu AA. Contribution of CD4 +and CD8 +T cells to graft – versus-host-disease and graft–versus–leukemia reactivity after transplantation of MHC-compatible bone marrow. Bone Marrow Transpl 1991;8:51. [34] Maldonado MA, Eisenberg RA, Roper E, Cohen PL, Kotzin BL. Greatly reduced lymphoproliferation in lrp mice lacking major histocompatibility complex class I. J Exp Med 1995;181:41. [35] Grabbe SS, Beissert S, Schwartz T, Granstein RD. Dendritic cells as initiators of tumor immune responses: a possible strategy for tumor immunotherapy. Immunol Today 1995;16:117. [36] Tjoa B, Boynton A, Kenny G, Ragde H, Misrock SL, Murphy G. Presentation of prostate tumor antigens by dendritic cells stimulates T-cell proliferation and cytotoxicity. Prostate 1996;28:65. [37] Marrack P, Hugo P, McCormack J, Kappler J. Death and T cells. Immunol Rev 1993;133:119. [38] Pardoll D, Darreara A. Thymic Selection. Curr Opin Immunol 1992;4:162. [39] Scollay RG, Butcher EC, Weissman IL. Thymus cell migration: Quantitative aspects of cellular traffic from the thymus to the periphery in mice. Eur J Immunol 1980;10:210. [40] Tough DF, Spren J. Turnover of naive-and memory-phenotype T cells. J Exp Med 1994;179:1127. [41] Cyster JG, Hartley SB, Goodnow CC. Competition for follicular niches excludes self-reactive cells from the recirculating B-cell repertoire. Nature 1994;371:389. [42] Kroemer G, Hirsch F, Gonzalez-Garcia A, Martinez C. Differential involvement of Th1 and Th2 cytokines in autoimmune disease. Autoimmunity 1996;24:25. [43] Harding CV, Song R. Phagocytic processing of exogenous particulate antigen by macrophages for presentation by class I MHC molecules. J Immunol 1994;153:4925. [44] Grant EP, Rock KL. MHC class I restricted presentation of exogenous antigen by thymic antigen-presenting cells in vitro and in vivo. J Immunol 1992;148:13. [45] Reis E, Sousa C, Germain RN. Major histocompatibility complex class I presentation peptides derived from soluble exogenous antigen by a subset of cells engaged in phagocytosis. J Exp Med 1994;182:841.
.
557
[46] Bevan MJ. Cross-priming for a secondary cytotoxic response to minor H antigens with H-2 congenic cells which do not cross-react in the cytotoxic assay. J Exp Med 1976;143:1283. [47] Kurts C, Heath WR, Carbone FR, Allison J, Miller JFAP, Kosaka H. Constitutive class I-restricted exogenous presentation of self-antigens in vivo. J Exp Med 1996;184:923. [48] Den Haan JMM, Sherman NE, Blokland E, Huczko E, Koning F, Dijfhout JW, et al. Identification of a graft versus host disease-associated human minor histocompatibility antigen. Science 1995;268:1476. [49] Cella M, Scheidegger D, Palmer-Lehmann K, Lane P, Lanzaveccia A, Albert G. Ligation of CD40 on dendritic cells triggers production of high levels of interleukin-12 and enhances T cell stimulatory capacity:T-T help via APC activation. J Exp Med 1996;184:747. [50] Koch F, Stanzl U, Jennewein P, Janke K, Heufler C, Kampgen E, et al. High level IL-12 production by murine dendritic cells: upregulation via MHC class Il and CD40 molecules and downregulation by IL-4 and IL-10. J Exp Med 1996;184:741. [51] Ranheim EA, Kipps TJ. Elevated expression of CD80 (B7.1) and other accessory molecules on synovial fluid mononuclear cell subsets in rheumatoid arthritis. Arthritis Rheum 1994;37:1637. [52] Tandon N, Metcalfe R, Barnett D, Weetman A. Expression of the costimulatory molecule B7/BB1 antigen in autoimmune thyroid disease. Quart J Med 1994;87:231. [53] Sharpe AH. Analysis of lymphocyte costimulation in vivo using transgenic and ‘knockout’ mice. Curr Opin Immunol 1995;7:389. [54] Guerder S, Picarella DE, Linsley P, Flavell RA. Costimulator B7-1 confers antigen-presenting-cell function to parenchymal tissue and in conjunction with tumor necrosis factor (leads to autoimmunity in transgenic mice. Proc Natl Acad Sci USA 1994;91:5138. [55] Early GS, Zhao W, Burns CM. Anti-CD40 ligand antibody treatment prevents the development of lupus-like nephritis in a subset of New Zealand black ×New Zealand white mice. Response correlates with the absence of an anti-antibody response. J Immunol 1996;157:3159. [56] Kalled SL, Cutler AH, Datta SK, Thomas DW. Anti-CD40 ligand antibody treatment of SNF1 mice with established nephritis: preservation of kidney function. J Immunol 1998;160:2158. [57] Durie FH, Fava RA, Foy TM, Aruffo A, Ledbetter JA, Noelle RJ. Prevention of collagen-induced arthritis with an antibody to gp39, the ligand for CD40. Science 1993;261:1328. [58] Gerrits K, Laman JD, Noelle RJ, Aruffo A, Ledbetter JA, Boersma WJ, et al. CD40-CD40 ligand interactions in experimental allergic encephalomyelitis and multiple sclerosis. Proc Natl Acad Sci USA 1996;19:2499. [59] Molldrem J, Dermime S, Parker K, Jiang Y, Mavroudis Z,D, Hensel N, et al. Targeted T-cell therapy for human leukemia: Cytotoxic T lymphocytes specific for a peptide derived from Proteinase 3 preferentially lyse human myeloid leukemia cells. Blood 1996;88:2450. [60] Celis E, Tsai V, Crimi C, Demars R, Wentworth PA, Chesnut RW, et al. Induction of anti-tumor cytotoxic T lymphocytes in normal-humans using primary cultures and synthetic peptide epitopes. Proc Natl Acad Sci USA 1994;91:2105.
Autoimmune disease induced by dendritic cell immunization against leukemia Marie A. Roskrow a, Dagmar Dilloo b, Nobuhiro Suzuki a, Wanyung Zhong b, Cliona M. Rooney a,c, Malcolm K. Brenner b,c,* a
Department of Virology and Molecular Biology, St. Jude Children’s Research Hospital, Memphis, TN, USA b Di6ision of Bone Marrow Transplantation, St. Jude Children’s Research Hospital, Memphis, TN, USA c Center for Cell and Gene Therapy, Baylor College of Medicine, 1102 Bates St, Suite 1100, Houston, TX 77030, USA Received 21 September 1998; accepted 28 January 1999
Abstract Induction of an optimal immune response will likely be a prerequisite for successful immunotherapy of human leukemias and other malignancies. Dendritic cells are highly effective at inducing an immune response to antigens to which the host is unresponsive, while transgenic expression of the costimulator molecule CD40 ligand (gp39/CD154) and the T cell growth factor interleukin 2 (IL2) are also able to augment immune responsiveness. We therefore investigated whether a combination of these two distinctive approaches to immunostimulation could safely increase the anti-tumor immune response compared to each stimulus alone. We injected BALB/CBYJ mice with syngeneic dendritic cells (DC) exposed to A20 lymphoblastic leukemia cell-derived peptides and proteins which had been acid-eluted from the cell surface. In additional mice, the pulsed DC were mixed with genetically modified syngeneic fibroblasts that were expressing CD40 ligand or secreting interleukin 2 (IL2). Three days after their third, weekly, vaccination, they were challenged with parental A20 cells. Tumor growth was suppressed by responses to pulsed DC alone (PB0.02). This suppression was further enhanced when pulsed DC were coinjected with fibroblasts expressing CD40 ligand and IL2 (PB 0.0005 compared to DC alone) even though CD40 ligand and IL2-expressing fibroblasts alone offered no significant protection in this model. Mice receiving the full complement of immunostimulants either failed to develop visible tumors or developed small tumors which quickly necrosed and regressed, allowing the mice to become long term tumor-free survivors. Antibody mediated depletion of either CD4 + or CD8+ T-cell subset significantly reduced the level of protection afforded by the vaccination. However, it became evident that this intensive stimulation of the immune system lead not only to tumor eradication but also to destruction of cells bearing normal self antigens. Hence, 60 days after challenge with A20 cells all mice in the DC/IL2/CD40 ligand group developed a severe, systemic autoimmune disorder that resembled graft – versus–host disease and manifest itself by significant peripheral blood cytotoxicity against autologous fibroblasts, blood dyscrasias, gross hepatosplenomegaly, cachexia and fur loss. This phenomenon depended on CD8 + cytotoxic T lymphocytes. Our results therefore suggest that the most effective strategies of immunotherapy against leukemia may also exceed the threshold of anergic cells, leading to a loss of self tolerance to normal self-antigens and the induction of an CD8 + anti-self effector response. © 1999 Elsevier Science Ltd. All rights reserved. Keywords: Autoimmune disease; Dendritic cells; Leukemia; Tumor vaccine
1. Introduction
Abbre6iations: APC, Antigen presenting cell; ARCC, American Type Tissue Collection; DC, Dendritic cell; GM-CSF, Granulocytemacrophage colony stimulating factor; gp39, CD40 ligand; IL2, Interleukin 2; TFA, Trifluroacetic acid; WBC, White Blood Count. * Corresponding author. Tel.: + 1-713-770-4662; fax: + 1-713-7704668. E-mail address: [email protected] (M.K. Brenner)
Specialized antigen presenting cells (APC) including dendritic cells (DC), are highly effective at presenting antigen and recruiting naive or quiescent CD4+ and CD8+ T cells into immune responses [1]. This property has been exploited in several novel immunotherapeutic strategies, and might well provide a basis for effective anti-cancer therapy as our ability to manipu-
0145-2126/99/$ - see front matter © 1999 Elsevier Science Ltd. All rights reserved. PII: S 0 1 4 5 - 2 1 2 6 ( 9 9 ) 0 0 0 4 5 - 4
550
M.A. Roskrow et al. / Leukemia Research 23 (1999) 549–557
late these cells improves [2 – 7]. A growing concern, however, is that enhancement of APC-elicited responses might compromise unresponsiveness to self-antigens expressed by normal cells, thereby producing autoimmune disease. We had the opportunity to test this prediction in mice treated with DC stimulated to ensure maximal recruitment of effector T cells. The paucity of tumor-specific or tumor-associated proteins known to be presented by human cancer cells [8,9] has led to efforts to use peptides eluted from the tumor cells as a source of antigen. Vaccination of mice with bone marrow-derived DC pulsed with unfractionated tumor peptides did in fact reduce the growth of established, weakly immunogenic tumors [10], and similar results were obtained by transducing DC with tumor cell-derived mRNA [11] or by fusing dendritic cells with the tumor cells themselves [4]. No adverse effects of immunization were noted in any of these studies. We hypothesized that the anti-leukemia tumor responses mediated by antigen-laden DC could be further augmented by transgenic expression of additional immunostimulatory molecules, including the CD40 ligand (gp39/CD154) and interleukin 2 (IL2). This reasoning was based on the high levels of CD40 antigen found on professional antigen presenting cells, including dendritic cells and the key role played by CD40 – CD40 ligand interactions in augmenting the function of dendritic cells [12–14]. Such interactions also stimulate CD4 + and CD8 + T cells after they have engaged antigen on APC [13,15–19]. We recently showed that transgenic expression of the CD40 ligand molecule by accessory cells markedly enhanced the immune response directed against coinjected leukemic cells, which was further amplified by transgenic production of IL2 [20]. In the study described here, a combination of tumor – antigen pulsed DC and syngeneic fibroblasts genetically modified to express CD40 ligand or IL2 uniformly protected mice against the otherwise weakly immunogenic A20 line of murine leukemic cells. However, all of the animals subsequently developed severe, systemic autoimmune disease, illustrating the potential hazard of stimulating the immune system to an extreme level.
2. Material and methods
gen, Cambridge, MA) and IL-4 (50 ng/ml, Endogen). After 48 h in culture, the loosely adherent cells were replated in a final concentration of 1× 106 cells/ml. As before, the media contained GM-CSF (150 ng/ml) and IL-4 (50 ng/ml). After 72 h in culture the population was further purified by depletion of contaminating CD11b (Mac1)/B220 double positive cells. The resulting cells were morphologically large irregularly shaped and non adherent, expressing CD45, MHC Class I and II, CD 11c, CD11b (Mac-1), Mac-3 and CD80 (all antibodies from Pharmingen) by flow cytometry. These cells were harvested and pulsed with A20 peptides as described below.
2.2. Cell lines The pre-B lymphoblastoid cell line A20 was derived from a spontaneous neoplasm [21]. Both the A20 and the CL7.1 fibroblast cell line [22] were derived from BALB/CBYJ/c mice and were obtained from American Type Tissue Collection (ATCC, Rockville, MD).
2.3. Elution of proteins and peptides from the surface of A20 cells and pulsing of mouse dendritic cells A20 cells (1 × 109) were pelleted by centrifugation at 1500 rpm for 5 min. The cell pellet was resuspended in 10 ml of precooled 0.4% trifluroacetic acid (Sigma Chemical Co.) for 2 min on ice. The supernatant containing the eluted proteins and peptides was collected after centrifugation at 1500 rpm for 5 min. For filtration and concentration of the eluted proteins, supernatants were centrifuged over a 10 kDa Centriprep cut-off filter (Amicon Inc, Beverley, MA) at 3600 rpm for 50–60 min. The eluted proteins were analyzed by mass spectrophotometry of both the filtrate and the retentate. The latter contained proteins \10 kDa and the filtrate smaller ones. Both fractions were dried overnight by high-speed vacuum centrifugation. The dried pellets were resuspended in Tris EDTA (pH 8) and quantified by a commercially available BSA protein assay kit (Pierce, Rockford, IL). Part of the A20 protein fraction (34 g) was added to 1 ×106 enriched dendritic cells for 12–16 h; 3–4 h before vaccination, the peptide fraction was added to the cells at a concentration of 8:5 g/1 ×106 cells.
2.1. Culture and phenotyping of mouse dendritic cells 2.4. Retro6iral 6ectors Dendritic cell-enriched populations were obtained as described previously [7,11]. In brief bone marrow mononuclear cells were flushed from the femurs of BALB/CBYJ mice with RPMI 1640 media (GIBCO Life Technologies, Grand Island, NY). The eluted cells were washed twice and then cultured in RPMI/10% fetal calf serum at a final concentration of 1 × 106 cells/ml in the presence of GM-CSF (150 ng/ml, Endo-
Supernatants containing the retroviral vector G1Na [23,24] or G1NaCvIL2 [25] (Genetic Therapy Inc, Gaithersburg, MD) were used to transduce the neomycin phosphotransferase gene neo and the IL2 gene respectively into CL7.1 fibroblasts. Transduction was performed at 37°C, in 5% CO2 in the presence of 6 g/ml polybrene. The transduced CL7.1 cells were se-
M.A. Roskrow et al. / Leukemia Research 23 (1999) 549–557
lected in G418 sulfate (GIBCO, Gaithersburg, MD) at an active concentration of 1 g/ml. To generate a retroviral vector containing the murine CD40 ligand gene, we first cloned the CD40 ligand cDNA by semi-nested PCR from an activated BALB/2 T-Cell cDNA library made in lambda (kindly provided by Dr C. Coleclough [26]). The primers 5%GCGGTACCTTTCAGTCAGCATGATAGAAC and 5%CTTATTCCAGCTCT-ATGTGCCT were used for the first round of amplification and the same forward primer and the primer 5%GGATCGATATACGGAAGACTGCCAGC for the second round. The PCR product was subcloned into pGEM7Z (Pharmacia, Piscataway, NJ). DNA sequence analysis confirmed the identity of the cloned CD40L cDNA with the published murine cDNA sequence. Subsequently, the plasmid PG1a.mCD40L was made by subcloning theCD40L cDNA into the retroviral vector pGIa (Genetic Therapy Inc, Gaithersburg, MD) with the CD40L gene being driven by the LTR promoter. The producer cell line BOSC, kindly provided by Dr. M. Scott, was transfected with the plasmid pG1a.mCD40L by calcium precipitation, as previously described [27]. Supernatants were collected from the BOSC cells at confluence and then used to transfect the CL7.1 fibroblast cell line. After transduction, fibroblasts were stained with the CD40L antibody MR1 (Pharmingen, San Diego, CA) and then sorted by flow cytometry (FACS-Star Plus; Beckon Dickinson, Mansfield, MA).
2.5. Vaccination with pulsed dendritic cells and tumor growth Experiments to determine the effects of vaccination with pulsed dendritic cells on tumor growth were performed in a syngeneic system, as the A20, the CL7.1 and the dendritic cells were all derived from BALB/ CBYJ mice. For assessment of the inhibitory effects of vaccinations with A20-pulsed dendritic cells and CD40 Ligand/IL2 costimulation on tumor development, BALB/CBYJ mice aged 12 – 18 weeks (The Jackson Laboratory, Bar Harbor, ME) were inoculated subcutaneously (s.q.) with a mixture of 2 × 105 CL7.1 fibroblasts, each expressing either the IL2, CD40 ligand or the neo gene, and 2– 3×105 protein-pulsed or unpulsed dendritic cells. Seven mice in each group received three such injections at different sites 2 week apart. Three days after the third injection, the mice were challenged with an s.q. injection of 1 × 105 A20 cells. The cell mixtures were not irradiated. Positive tumor growth was defined as any increase in tumor volume\10 cmm, as determined from measurements of transverse and perpendicular diameters. Animals were killed when tumor growth resulted in a 30% increase in body weight, in ulceration or in distress. Wilcoxon or Fisher’s exact tests were used to compare
551
mean tumor volumes or survival in mice given different combinations of immune stimulants.
2.6. T-Cell depletion Mice were depleted of T lymphocytes by intraperitoneal (ip) injection of the Gk1.5 (anti CD4 and 2.43 (anti-CD8) monoclonal antibodies (mAbs) [28,29], kindly provided by Dr Peter Doherty. The antibodies were injected ip 1 day before tumor challenge and then every other day for a week, followed by four weekly injections. After euthanasia of tumor-bearing animals, the depletion status was \ 98% of the relevant subset, as demonstrated by flow cytometry analysis of splenocytes and peripheral blood.
2.7. Characteristics of transduced fibroblasts By ELISA (R&D, Minneapolis, MN), IL2 production by CL7.1 fibroblasts ranged from 80 to 100 IU/106 cells over 24 h. More than half (50–60%) of the fibroblasts transduced with pG1a.m.CD40 ligand expressed CD40 ligand after selection by flow cytometry and subsequent cell culture. Control fibroblasts were transduced with the G1Na vector.
2.8. Cytotoxicity assays Animals were bled by cardiac puncture and erythrocyte depletion was performed using a standard ammonium chloride depletion method. The cytotoxicity of the peripheral blood lymphocytes was analyzed in a standard 4 h chromium-51 release assay using effector: target ratios of 20:1, 10:1 and 5:1. Target cells included autologous and HLA-mismatched fibroblasts, the parental A20 tumor cell line and Yac-1 which is sensitive to natural-killer cells. Specific cytotoxic activity was determined with the formula: % specific release= [(experimental release− spontaneous release)/(total release− spontaneous release)]×100. Standard errors of the means of triplicate cultures was less than 10%.
3. Results
3.1. Vaccination with tumor protein-pulsed dendritic cells can protect mice against subsequent challenge with A20 tumor cells Since no tumor-specific antigens are known for the A20 leukemia cell line, (or for many human tumors), we used unfractionated, concentrated proteins and peptides, acid-eluted from the surface of A20 cells, to pulse the DC-enriched population generated from syngeneic bone marrow [13]. For some groups of mice, the pulsed
552
M.A. Roskrow et al. / Leukemia Research 23 (1999) 549–557
Fig. 1. Vaccination with A20 peptide-pulsed DC, alone or in combination with gene-modified fibroblasts expressing CD40 ligand and IL2, protects mice against subsequent challenge with A20 leukemia cells. Each group (seven mice per group) received three s.q. injections, 1 week apart, with a mixture of 2 × 105 transduced CL7.1 fibroblasts expressing either CD40 ligand, IL2 or neo (control) gene and 2 – 3 × 105 peptide-pulsed dendritic cells or unpulsed DC. Three days after the third injection, the mice were challenged with 1 × 105 A20 leukemia cells. Tumor volumes (cmm) are presented as means. () control mice (A20 leukemia challenge only) (2) unpulsed DC with CD40 ligand/IL2 costimulation; () peptide-pulsed DC without costimulation; ( ) peptide-pulsed DC with CD40 ligand/IL2 costimulation.
DC were mixed with genetically modified syngeneic fibroblasts that were expressing CD40 ligand or secreting IL2. The animals were vaccinated s.q. at weekly intervals prior to challenge with A20 tumor cells. Fig. 1 shows the mean tumor volume for each group of mice at various times after challenge with A20 cells. Tumor growth was weakly but consistently suppressed
by responses to pulsed DC alone (PB 0.02). This suppression was further enhanced by the addition of fibroblasts expressing CD40 ligand or IL2 (PB 0.0005 compared to DC alone). In these experiments, the mice received three vaccinations prior to the tumor challenge. It should be noted that immunization with unpulsed dendritic cells in combination with CD40 ligand and IL2 was ineffective in this model. Mice receiving the full complement of immunostimulants either failed to develop visible tumors or developed small tumors which soon became necrotic and regressed. As a result, the seven mice vaccinated with CD40 ligand, IL2 and pulsed DC survived tumor free\ 60 days, compared with none of the controls (Fisher’s test PB 0.005). This effect could also be produced with two injections, although the benefit was less pronounced (Fig. 2).These same mice were used to assess the effect of antibody-mediated depletion of CD4+ and CD8 + T-cell subsets (Fig. 2). In vivo depletion of T lymphocytes significantly reduced the level of protection afforded by the combination vaccine, so that all CD4-depleted mice had developed tumor by day ten and all CD8-depleted mice by day 25 (PB0.001 and PB 0.01 respectively compared to non-depleted mice) confirming that the anti-leukemia response relied on the recruitment of both subsets of effector T cells, although predominantly on CD4+ cells.
3.2. Immunotherapy with tumor antigen-pulsed DC, costimulated with CD40 ligand and IL2, induces an autoimmune disease Although mice vaccinated with antigen-pulsed DC and fibroblasts expressing CD40 ligand or IL2 were
Fig. 2. In vivo depletion of CD4+ and CD8+ lymphocytes reduces the anti-leukemia response. Three groups of mice (seven per group) received two s.q. injections of peptide-pulsed DC in combination with gene-modified fibroblasts expressing CD40 ligand and IL2. Control mice were injected with unpulsed DC in the absence of any costimulation. Three days after the second injection, all mice were challenged with 1 × 105 A20 leukemia cells. Tumor volumes (cmm) are presented as means. () control mice (unpulsed DC without costimulation); ( ) peptide-pulsed DC with CD40 ligand/IL2 costimulation; (2) peptide-pulsed DC with CD40 ligand/IL2 costimulation and CD8-depletion; ( × ) peptide-pulsed DC with CD40 ligand/IL2 costimulation and CD4-depletion.
M.A. Roskrow et al. / Leukemia Research 23 (1999) 549–557
553
Table 2 Spleen and liver weights in non-vaccinated mice compared to mice receiving the combination vaccine
Range Median Mean P values vaccine v. no vaccine
Fig. 3. Photographs of mice that developed an autoimmune-like disease and their normal litter mates.
protected against tumor, they did not remain healthy. Sixty days after challenge with A20 cells, they showed signs of a systemic disorder resembling graft versus host disease (GvHD), and characterized by pronounced periorbital edema and fur loss as well as ruffled body fur (Fig. 3). Further investigation confirmed the presence of a systemic autoimmune disorder (Table 1). Although the total white blood count (WBC) was within the normal range, they all had a reversal of the normal neutrophil/lymphocyte ratio to produce a relative neutrophilia and lymphopenia. In addition the mice were anemic and thrombocytopenic. At necropsy there was evidence of gross hepatosplenomegaly and exudative ascites. As compared to mice receiving an A20 tumor
Spleen weight (g)
Liver weight (g)
Vaccine
No vaccine
Vaccine
No vaccine
0.17–0.51 0.29 0.31 B0.0005
0.08–0.1 0.09 0.09
1.13–1.72 1.33 1.39 B0.0002
0.9–1.01 0.99 0.98
challenge alone (Table 2), there was significant splenomegaly (PB 0.0005) and hepatomegaly (PB 0.0002). Dip-stick analysis of their urine showed heavy proteinuria in all cases. Immunohistochemical analysis of the affected organs did not show any evidence of cellular infiltration or immune complex deposition. Flow cytometric analysis of splenocytes (Fig. 4), showed a decrease in both CD4+ and CD8+ lymphocytes by comparison to normal controls (PB 0.03) and a corresponding increase of activated macrophages (Mac1+ Mac3+ ) (PB0.01). Fresh unstimulated peripheral blood mononuclear cells from the animals with autoimmune disease had high levels of specific cytotoxic activity against syngeneic fibroblasts without significant killing of H2-mismatsched fibroblasts or Yac-1 (NK sensitive) target cells (Fig. 5a). As expected, peripheral blood mononuclear cells from healthy mice failed to recognize any of the target cells (Fig. 5b). A feature of A20 cells is that they express class II MHC molecules, from which peptides can be eluted that are capable of stimulating CD4+ T helper cells. Since CD4+ T cells play a major role in the etiology of many autoimmune diseases [30–33], we investigated whether the syndrome could be prevented by in vivo depletion of CD4+ T cells. In fact, such depletion had no effect: all seven mice treated in this manner developed signs of autoimmune disease such as periorbital edema and fur ruffling. Of note, these signs appeared as early as 14 days after leukemia challenge which was significantly earlier than in non-depleted mice injected with the combination vaccine. In contrast, none of the
Table 1 Peripheral blood counts in mice receiving the combination vaccine
Normal Range Mean
Total WBC (×103/ul)
Neutrophils (%)
Lymphocytes (%)
Hb (g/dl)
Platelets (×103/ul)
5–12,000 4.9–18.6 9.9
5–40 66–86 74.6
63–75 11–34 24.2
11–14 8.3–12.1 10.4
80–1000 170–1090 600
554
M.A. Roskrow et al. / Leukemia Research 23 (1999) 549–557
Fig. 4. Flow cytometric analysis of splenocytes from normal mice and mice that developed an autoimmune-like disease. Data are reported as percent of mononuclear cells.
mice depleted of CD8 + T cells developed the syndrome. Necropsy of four mice from both, the CD4-depleted and the CD8-depleted groups was performed for analysis of additional signs of autoimmune disease. Examinations confirmed that CD8-depleted mice did not develop the disorder (Table 3). Thus, induction of autoimmune disease in our model is independent of CD4 + helper cells, relying instead on CD8 + cytotoxic T lymphocytes [34].
produce organ damage, the pattern of which is dictated by the class of effector cells whose anergy is reversed, and the target antigens they recognize. Our data indicate that combination therapy with pulsed DC and fibroblasts expressing a costimulator molecule or a T-cell growth factor is not only an effective way to generate antitumor immunity, but also exceeds the activation threshold of anergic cells required to induce a CD8+ effector response. Of note, unpulsed dendritic cells neither prevented tumor growth, nor induced autoimmune disease even in the presence of CD40 ligand/
4. Discussion There has been considerable interest in using antigen presenting cells including dendritic cells (DC) as therapeutic agents in patients with cancer because of their capacity to capture tumor specific antigens and recruit T cell dependent anti-tumor immunity [2 – 6,35,36]. In our own model, we combined leukemia-peptide pulsed dendritic cells with two powerful costimuli, CD40L and IL2, delivered by syngeneic fibroblasts. While this combination enhanced anti-tumor activity, it also induced a severe systemic autoimmune disorder. What mechanisms could account for autoimmune disease in mice receiving combination therapy of the type described here? Physiologically, most autoreactive T-cell clones are eliminated in the thymus [37,38]. Those that reach the periphery are functionally inactivated (anergized), probably as a result of stimulation by self-antigen in the absence of appropriate costimulatory molecules [39,40]. These anergized cells are not eliminated, so that self-antigens might be rendered immunogenic by any stimulus that exceeds the anergy threshold [30,41]. The resulting activated effector cells may then
Fig. 5. Fresh peripheral blood cells from mice with autoimmune disease exhibit high levels of specific cytotoxic activity against syngeneic fibroblasts without significant killing of H2-mismatched fibroblasts or Yac-1 (NK sensitive) target cells. Cytotoxic activity was measured in a standard chrom-release assay in mice with the autoimmune-like disorder (a) and in normal mice (b) at various effector to target ratios. Cytotoxic activity of unstimulated peripheral blood mononuclear was measured against several targets: ( ) Yac-1 cells, () H2-mismatched fibroblasts, () matched fibroblasts and ( × ) A20 leukemia cells.
M.A. Roskrow et al. / Leukemia Research 23 (1999) 549–557 Table 3 Autoimmune disease in CD4-depleted versus CD8-depleted mice Signs
CD4-depletion
CD8-depletion
Periorbital edema/ruffled fur Hepatosplenomegaly Proteinuria Ascites Lymphopenia
4/4 4/4 4/4 4/4 4/4
0/4 0/4 0/4 0/4 0/4
IL2. Since these DC already present an array of processed self-peptides [1], the induction of autoimmunity by tumor peptide-pulsed DC implies that the self peptides obtained from tumor cells are either quantitatively or qualitatively different from those normally processed by dendritic cells. In most murine models of autoimmune disease, it is CD4+ T lymphocytes that mediate disease development [42]. In our study, the autoimmmune-like disorder was mediated by CD8+ T cells suggesting that the tumor cell-derived self-antigens were presented by class I molecules expressed on the dendritic cells. Conventionally, class I associated peptides derive from endogenous antigen processing. However, there are now several reports indicating that exogenous peptides can access class I molecules. While some studies suggest that it is specialized APCs that can shunt exogenous peptide into the class I pathway [43,44] others have attributed class I presentation of exogenous peptide to the fact that overfeeding of APCs leads to particle overloading of the phagosomal compartments with subsequent release of antigen into the cytosol [45]. Most studies analyzing the possible mechanisms that enable class I presentation of exogenous peptides focus on expression of non-self peptides with analysis of crosspriming events being the primary example [46]. A more recent report demonstrates that processing of exogenous self-antigens through the class I pathway is also possible [47]. Since leukemic cells may overexpress certain self-antigens such as minor histocompatibility antigens [48], antigen excess may contribute to the observed autoimmune-like effects. The contribution of CD40 ligand to autoimmunity in this model likely comes from its known immunostimulatory effects on both APC and T lymphocytes. Engagement of the CD40 counter receptor expressed at high densities on dendritic cells [13] induces DC maturation with enhanced expression of adhesion molecules and the costimulatory molecule B7.1 [14]. Upregulation of B7 molecules is one of the critical events in CD40/ CD40 ligand-mediated immune-activation. Binding of CD40 ligand to DC also promotes IL-12 release [49,50], which increases the function of IL2-activated cytotoxic T cells and expands NK cell populations. Finally, CD40 ligand also stimulates the effector cells of the
555
immune response. Under physiological conditions, Tcell responses can be directly amplified by CD40–CD40 ligand interactions between antigen-activated T cells [15,17,18]. The presence of IL2 complements these activities, bypassing a requirement for CD4+ mediated helper function to induce CD8+ effector cells [20]. Hence a CD8+ -dependent autoimmune disorder develops in these mice even after CD4 + helper cells have been depleted. There are several other lines of evidence suggesting that costimulatory molecules play an important role in the breakdown of self-tolerance. In human autoimmune disease such as rheumatoid arthritis and autoimmune thyroiditis, upregulation of B7.1 on monocytes, T cells and B cells has been demonstrated [51,52]. The impact of aberrant expression of costimulatory molecules on potential target tissue of the autoimmune response has been examined in transgenic mice. In these models, aberrant expression of one costimulatory molecule such as B7 alone was not sufficient to induce autoimmune disease [53]. Crossbreeding with transgenic mice expressing a second signal with broad immunostimulatory capacity such as TNF-alpha, however, resulted in breakdown of self-tolerance and the development of severe diabetes [54]. Similarly, in our leukemia vaccine model, three components were required for induction of the autoimmune-like disorder, namely the leukemiaderived peptides, the dendritic cells and CD40 ligand /IL2 costimulation. In addition to CD40/CD40 ligandmediated upregulation of B7 molecules on both target as well as immune effector cells and the implication of B7 in autoimmune disease, it has been shown that treatment of autoimmune mice with anti-CD40 ligand antibody can prevent disease development. In murine models of lupus nephritis, collagen-induced arthritis and experimental allergic encephalomyelitis [55–58], injections of anti-CD40 ligand antibodies proved protective, underscoring the impact of CD40 ligand on the development of autoimmune-disease. A similar mechanism, may contribute both to spontaneous autoimmune disease and to the perpetuation of chronic graft–versus–host disease. In both settings, APC may be stimulated to take up large quantities of host antigens following cell destruction by microorganisms or by alloreactive T lymphocytes. These APC are in an environment in which there is a high level of expression of costimulatory molecules such as CD40L on activated T cells, and in which cytokines such as IL2 are readily available. If this combination exceeds the threshold of anergy, self-perpetuating autoimmune/ chronic GvH disease may result, in an analogous manner to that we describe in the tumor model. Because human tumors likely express a circumscribed and heterogeneous range of tumor antigens of limited immunogenicity, there is strong incentive to prepare tumor vaccines that can overcome non-responsiveness
556
M.A. Roskrow et al. / Leukemia Research 23 (1999) 549–557
to weak immunogens. While we have shown that strategies combining peptide-pulsed DC and costimulatory molecules may fulfill these requirements, our results also indicate that the most effective immunostimulation may interfere with tolerance to normal self antigens. Although generally undesirable, such an effect, if controlled, could have therapeutic benefit in some circumstances. If, for instance, antigens known to be specific for certain organs and tissues [36] [48,59,60] were used to pulse dendritic cells in the strategy described here, one might anticipate the development of autoimmunity against the relevant organ. Destruction of both the normal tissue and antigenically related malignant cells derived therefrom could offer an acceptable therapeutic strategy if the cancer were disseminated and the targeted organ not essential for life.
Acknowledgements We would like to thank John Gilbert for scientific editing and Dr Helen Heslop for helpful scientific discussion. This work was supported in part by NIH grants CA 61384, CA 20180, CA 75014 and the Cancer Center Support grant, and by the American Lebanese Syrian Associated Charities (ALSAC). M. Roskrow and D. Dilloo equally provided the concept, design, analysis of data, drafting and critical review of the manuscript, collection of data and statistical expertise. N. Suzuki and W. Zhong assisted with data collection and technical support. C.M. Rooney provided critical revision of the article and funding. M. Brenner provided critical revision of the article, final approval and funding.
References [1] Steinman RM. The dendritic cell system and its role in immunogenicity. Annu Rev Immunol 1991;9:271. [2] Hsu F, Benike C, Fagnoni F, Lines TM, Czerwinski D, Taidi B, et al. Vaccination of patients with B-cell lymphoma using autologous antigen-pulsed dendritic cells. Nat Med 1996;2:52. [3] Murphy G, Tjoa B, Ragde H, Kenny G, Boynton A. Phase I clinical trial: T-cell therapy for prostate cancer using autologous dendritic cells pulsed with HLA-A0201-specific peptides from prostate-specific membrane antigen. Prostate 1996;29:371. [4] Gong J, Chen D, Kashiwaba M, Kufe D. Induction of antitumor acitvity by immunization with fusions of dendritic and carcinoma cells. Nat Med 1997;3:558. [5] Henderson RA, Nimgaonkar MT, Watkins SC, Robbins PD, Ball ED, Finn OJ. Human dendritic cells genetically engineered to express high levels of the human epithelial tumor antigen mucin (MUC-1). Cancer Res 1996;56:3763. [6] van-Elsas A, van-der-Burg SH, van-der-Minne CF, Gorghi M, Mourer JS, Melief CJ, et al. Peptide-pulsed dedritic cells induce tumoricidal cytotoxic T lymphocytes from healthy donors against stably HLA-A*0201-binding peptides from Melan-A/ MART-1 self antigen. Eur J Immunol 1996;26:1683.
[7] Celluzzi CM, Mayordomo JI, Storkus WJ, Lotze MT, Falo LDJ. Peptide-pulsed dendritic cells induce antigen-specific CTLmediated protective tumor immunity. J Exp Med 1996;183:283. [8] Henderson RA, Finn OJ. Human tumor antigens are ready to fly. Adv Immunol 1996;62:217. [9] Boon T, van Der Bruggen P. Human tumor antigens recognized by T lymphocytes. J Exp Med 1996;183:725. [10] Paglia P, Chiondoni C, Rodolfo M, Colombo MP. Murine dendritic cells loaded in vitro with soluble protein prime cytotoxic T lymphocytes against tumor antigen in vivo. J Exp Med 1996;183:317. [11] Boczkowski D, Nair SK, Synder D, Gilboa E. Dendritic cells pulsed with RNA are potent antigen-presenting cells in vitro and in vivo. J Exp Med 1996;184:645. [12] Alderson MR, Armitage RJ, Tough TW, Strockbine L, Fanslow WC, Spriggs MK. CD40 expression by human monocytes: regulation by cytokines and activation of monocytes by the ligand for CD40. J Exp Med 1993;178:669. [13] Freudenthal PS, Steinman RM. The distinct surface of human blood dendritic cells as observed after an improved isolation method. Proc Natl Acad Sci USA 1990;87:7698. [14] Caux C, Massacrier C, Vanbervliet B, Dubois B, van Kooten C, Durand I, Banchereau J. Activation of human dendritic cells through CD40 cross-linking. J Exp Med 1994;180:1263. [15] Cayabyab M, Phillips JH, Lanier L L. CD40 preferentially costimulates activation of CD4 + T lymphocytes. J Immunol 1994;152:1523. [16] Wagner DH, Stout RD, Suttles J. Role of the CD40–CD40 ligand interaction in CD4 + T cell contact-dependent activation of monocyte interleukin-1 synthesis. Eur J Immunol 1994;24:3148. [17] Fanslow WC, Clifford KN, Seaman M, Alderson MR, Spriggs MK, Armitage RJ, et al. Recombinant CD40 ligand exerts potent biologic effects on T cells. J Immunol 1994;152:4262. [18] Armitage RJ, Tough TW, Macduff BM, Fanslow WC, Spriggs MK, Ramsdell F, et al. CD40 ligand is a T cell growth factor. Eur J Immunol 1993;23:2326. [19] Paul WE, Seder RA. Lymphocyte responses and cytokines. Cell 1994;76:241. [20] Dilloo D, Brown M, Roskrow M, Zhong W, Holden W, Holladay M, et al. CD40 ligand acts as a co-stimulator molecule for the generation of an anti-tumor immune response. Blood 1997;90:1927. [21] Kim KJ, Kanellopoulos-Langevin C, Verwin RM, Sachs DH, Asofsky R. Establishment and characterization of BALB/c lymphoma lines with B cell properties. J Immunol 1979;122:549. [22] Patek PQ, Collins JL, Cohn M. Transformed cell lines susceptible or resist to in vivo surveillance against tumorigenesis. Nature 1978;276:510. [23] Bender MA, Palmer TD, Gelinas RE, Miller AD. Evidence that the packaging signal of Moloney, Murine Leukemia Virus extends into the gag region. J Virol 1987;61:1639. [24] Brenner MK, Rill DR, Holladay MS, Heslop HE, Moen RC, Buschle M, et al. Gene marking to determine whether autologous marrow infusion restores long-term haemopoiesis in cancer patients. Lancet 1993;342:1134. [25] Brenner MK, Furman WL, Santana VM, Bowman L, Meyer W. Phase I study of cytokine-gene modified autologous neuroblastoma cells for treatment of relapsed/refractory neuroblastoma. Hum Gene Ther 1992;3:665. [26] Coleclough C. Cell-free expression of large collections of cDNA clones using positive-selection I phage vectors. Meth Enzymol 1993;217:152. [27] Pear WS, Nolan GP, Scott ML, Baltimore D. Production of high-titer helper-free retroviruses by transient transfection. Proc Natl Acad Sci USA 1993;90:8392.
M.A. Roskrow et al. / Leukemia Research 23 (1999) 549–557 [28] Dialynas DP, Quan ZS, Wall KA, Pierres A, Quintans J, Loken MR, et al. Characterization of the murine T cell surface molecule, designated L3T4, identified by monoclonal antibody GK1.5; similarity of L3T4 molecule. J Immunol 1983;131:2445. [29] Sarmiento M, Glasebrook AL, Fitch FW. IgG of IgM monoclonal antibodies reactive with different determinants on the molecular complex bearing LYT 2 antigen block T cell-mediated cytolysis in the absence of complement. J Immunol 1980;125:2665. [30] Fulcher DA, Lyons AB, Korn SL, Cook MC, Koleda C, Parish C, et al. The fate of self-reactive B cells depends primarily on the degree of antigen receptor engagement and availability of T cell help. J Exp Med 1996;183:2313. [31] Ferrara JLM, Deeg HG. Graft–versus–host disease. N Engl J Med 1991;324:667. [32] Korngold R. Biology of graft–versus–host disease. Am J Ped Hem/Onc 1993;15:18. [33] Truitt RL, Atasoylu AA. Contribution of CD4 +and CD8 +T cells to graft – versus-host-disease and graft–versus–leukemia reactivity after transplantation of MHC-compatible bone marrow. Bone Marrow Transpl 1991;8:51. [34] Maldonado MA, Eisenberg RA, Roper E, Cohen PL, Kotzin BL. Greatly reduced lymphoproliferation in lrp mice lacking major histocompatibility complex class I. J Exp Med 1995;181:41. [35] Grabbe SS, Beissert S, Schwartz T, Granstein RD. Dendritic cells as initiators of tumor immune responses: a possible strategy for tumor immunotherapy. Immunol Today 1995;16:117. [36] Tjoa B, Boynton A, Kenny G, Ragde H, Misrock SL, Murphy G. Presentation of prostate tumor antigens by dendritic cells stimulates T-cell proliferation and cytotoxicity. Prostate 1996;28:65. [37] Marrack P, Hugo P, McCormack J, Kappler J. Death and T cells. Immunol Rev 1993;133:119. [38] Pardoll D, Darreara A. Thymic Selection. Curr Opin Immunol 1992;4:162. [39] Scollay RG, Butcher EC, Weissman IL. Thymus cell migration: Quantitative aspects of cellular traffic from the thymus to the periphery in mice. Eur J Immunol 1980;10:210. [40] Tough DF, Spren J. Turnover of naive-and memory-phenotype T cells. J Exp Med 1994;179:1127. [41] Cyster JG, Hartley SB, Goodnow CC. Competition for follicular niches excludes self-reactive cells from the recirculating B-cell repertoire. Nature 1994;371:389. [42] Kroemer G, Hirsch F, Gonzalez-Garcia A, Martinez C. Differential involvement of Th1 and Th2 cytokines in autoimmune disease. Autoimmunity 1996;24:25. [43] Harding CV, Song R. Phagocytic processing of exogenous particulate antigen by macrophages for presentation by class I MHC molecules. J Immunol 1994;153:4925. [44] Grant EP, Rock KL. MHC class I restricted presentation of exogenous antigen by thymic antigen-presenting cells in vitro and in vivo. J Immunol 1992;148:13. [45] Reis E, Sousa C, Germain RN. Major histocompatibility complex class I presentation peptides derived from soluble exogenous antigen by a subset of cells engaged in phagocytosis. J Exp Med 1994;182:841.
.
557
[46] Bevan MJ. Cross-priming for a secondary cytotoxic response to minor H antigens with H-2 congenic cells which do not cross-react in the cytotoxic assay. J Exp Med 1976;143:1283. [47] Kurts C, Heath WR, Carbone FR, Allison J, Miller JFAP, Kosaka H. Constitutive class I-restricted exogenous presentation of self-antigens in vivo. J Exp Med 1996;184:923. [48] Den Haan JMM, Sherman NE, Blokland E, Huczko E, Koning F, Dijfhout JW, et al. Identification of a graft versus host disease-associated human minor histocompatibility antigen. Science 1995;268:1476. [49] Cella M, Scheidegger D, Palmer-Lehmann K, Lane P, Lanzaveccia A, Albert G. Ligation of CD40 on dendritic cells triggers production of high levels of interleukin-12 and enhances T cell stimulatory capacity:T-T help via APC activation. J Exp Med 1996;184:747. [50] Koch F, Stanzl U, Jennewein P, Janke K, Heufler C, Kampgen E, et al. High level IL-12 production by murine dendritic cells: upregulation via MHC class Il and CD40 molecules and downregulation by IL-4 and IL-10. J Exp Med 1996;184:741. [51] Ranheim EA, Kipps TJ. Elevated expression of CD80 (B7.1) and other accessory molecules on synovial fluid mononuclear cell subsets in rheumatoid arthritis. Arthritis Rheum 1994;37:1637. [52] Tandon N, Metcalfe R, Barnett D, Weetman A. Expression of the costimulatory molecule B7/BB1 antigen in autoimmune thyroid disease. Quart J Med 1994;87:231. [53] Sharpe AH. Analysis of lymphocyte costimulation in vivo using transgenic and ‘knockout’ mice. Curr Opin Immunol 1995;7:389. [54] Guerder S, Picarella DE, Linsley P, Flavell RA. Costimulator B7-1 confers antigen-presenting-cell function to parenchymal tissue and in conjunction with tumor necrosis factor (leads to autoimmunity in transgenic mice. Proc Natl Acad Sci USA 1994;91:5138. [55] Early GS, Zhao W, Burns CM. Anti-CD40 ligand antibody treatment prevents the development of lupus-like nephritis in a subset of New Zealand black ×New Zealand white mice. Response correlates with the absence of an anti-antibody response. J Immunol 1996;157:3159. [56] Kalled SL, Cutler AH, Datta SK, Thomas DW. Anti-CD40 ligand antibody treatment of SNF1 mice with established nephritis: preservation of kidney function. J Immunol 1998;160:2158. [57] Durie FH, Fava RA, Foy TM, Aruffo A, Ledbetter JA, Noelle RJ. Prevention of collagen-induced arthritis with an antibody to gp39, the ligand for CD40. Science 1993;261:1328. [58] Gerrits K, Laman JD, Noelle RJ, Aruffo A, Ledbetter JA, Boersma WJ, et al. CD40-CD40 ligand interactions in experimental allergic encephalomyelitis and multiple sclerosis. Proc Natl Acad Sci USA 1996;19:2499. [59] Molldrem J, Dermime S, Parker K, Jiang Y, Mavroudis Z,D, Hensel N, et al. Targeted T-cell therapy for human leukemia: Cytotoxic T lymphocytes specific for a peptide derived from Proteinase 3 preferentially lyse human myeloid leukemia cells. Blood 1996;88:2450. [60] Celis E, Tsai V, Crimi C, Demars R, Wentworth PA, Chesnut RW, et al. Induction of anti-tumor cytotoxic T lymphocytes in normal-humans using primary cultures and synthetic peptide epitopes. Proc Natl Acad Sci USA 1994;91:2105.