Inhibition of FLT3 signaling targets DCs to ameliorate autoimmune disease

Abstract

Autoimmune diseases often result from inappropriate or unregulated activation of autoreactive T cells. Traditional approaches to treatment of autoimmune diseases through immunosuppression have focused on direct inhibition of T cells. In the present study, we examined the targeted inhibition of antigen-presenting cells as a means to downregulate immune responses and treat autoimmune disease. Dendritic cells (DCs) are the central antigen-presenting cells for the initiation of T cell responses, including autoreactive ones. A large portion of DCs are derived from hematopoietic progenitors that express FLT3 receptor (CD135), and stimulation of the receptor via FLT3 ligand either in vivo or in vitro is known to drive expansion and differentiation of these progenitors toward a DC phenotype. We hypothesized that inhibition of FLT3 signaling would thus produce an inhibition of DC-induced stimulation of T cells, thereby inhibiting autoimmune responses. To this end, we used small-molecule tyrosine kinase inhibitors targeted against FLT3 and examined the effects on DCs and their role in the promulgation of autoimmune disease. Results of our studies show that inhibition of FLT3 signaling induces apoptosis in both mouse and human DCs, and thus is a potential target for immune suppression. Furthermore, targeted inhibition of FLT3 significantly improved the course of established disease in a model for multiple sclerosis, experimental autoimmune encephalomyelitis, suggesting a potential avenue for treating autoimmune disease.

The generation of a T cell immune response depends on effective processing and presentation of antigens by activated antigen-presenting cells (APCs). Although this process is necessary for beneficial immune responses, APCs can also inappropriately activate T cells, resulting in autoimmune disease. Modulating the degree of immune activity has become a significant goal of immunebased therapies, whether for increasing immunity to fight tumors or decreasing immunity to lessen the severity of autoimmune disease. Traditionally, T cells have been the primary target for therapies for autoimmune disease. Although the nature of dendritic cells (DCs) responsible for generating autoreactive responses remains extremely controversial, more data have been generated that implicate the potential for DCs to contribute to pathology. Thus, the possibilities of targeting antigen processing and presentation are beginning to be more extensively investigated.

DCs have been manipulated for use as immunostimulatory therapies for some time, and the immune system can be boosted by administering growth factors that lead to the differentiation and/or maturation of DCs. For example, FL treatment, either in vitro or in vivo, leads to a large expansion of DCs that can present antigen (1) and, furthermore, can lead to enhanced antitumor responses (2). FL binds to and signals through its receptor, FLT3, which is a receptor tyrosine kinase (TK) that plays important roles in hematopoietic stem/progenitor cells and can lead to the differentiation of progenitors into DCs. Upon binding FL, FLT3 homodimerizes and its kinase domain is activated. FLT3 kinase activation has several consequences. FLT3 directly phosphorylates a number of substrate proteins on tyrosine residues, which in turn activates these proteins. In addition, FLT3 phosphorylates itself on several tyrosine residues, and these residues are then bound by a number of adapter proteins containing SH2 domains (3-11). The end result is the transduction of signals that act on the nucleus to alter the genetic program of the cell. Some of these signals stimulate proliferation, whereas others appear to protect the cell from apoptosis or drive differentiation (12). Because of the importance of this signaling pathway, it has been investigated in a number of settings. Although FLT3 is known to be expressed on DC progenitors, and FL/FLT3 signaling results in large increases in the numbers of functional DCs, its role in mature DCs has not been extensively evaluated (13-18). FLT3 is highly expressed by mature DCs, suggesting that signaling through this receptor could be an important aspect of maintaining DC function. Interestingly, whereas relatively large amounts of exogenous FL are necessary to significantly change the numbers of DCs, most cells secrete FL such that stimulation of the receptor is likely occurring on a more frequent basis. Thus, it seemed possible that the same stimulatory pathway may also be a target for down-modulating potentially harmful immune responses. We thus sought to further investigate the role of FLT3 signaling on the function of mature DCs and to assess whether targeted inhibition would lead to a decrease in pathologic immune responses.

Because of frequent constitutive activation of FLT3 through mutation in leukemia, a number of small-molecule FLT3 tyrosine kinase inhibitors (TKIs) have been developed for the treatment of acute myelogenous leukemia (19-21). These drugs work by competitively inhibiting the binding of ATP to the active site of the receptor. CEP-701 is a FLT3 inhibitor that is both potent (IC₅₀ of 2 nM) and selective (no activity on the other members of the type III RTK family until >500 nM) (21). We thus used both CEP-701 and other FLT3 inhibitors in our studies to examine the effects of this inhibition on DCs, both in vitro and in vivo.

To evaluate the effects of FLT3-targeted inhibition on one possible pathologic result of altered DC function, autoimmune disease, we used the myelin oligoglycoprotein (MOG_35-55)-induced experimental autoimmune encephalomyelitis (EAE) model system. EAE serves as a model for multiple sclerosis (MS), which is a disease of the central nervous system in which T cells recognize myelin, and the destruction of myelin proteins leads to a progressive, disabling disease.

The induction of autoimmune disease is a complex process, and the involvement of CNS immunity adds an additional complicating element. Interestingly, although both EAE and MS are diseases of the CNS, it is not clear where antigen presentation first occurs, and studies show that it may occur in the periphery as well as the CNS (22). Traditional means of treating this and other autoimmune diseases have focused on suppressing T cell responses directly. Limitations of these approaches have been toxic side effects and insufficient success in controlling disease. Because it is likely that even autoreactive preactivated T cells require antigen presentation (23) to maintain disease, additional investigations into the role of APCs in the onset and maintenance of EAE and/or MS have been conducted. One recent report indicated that increasing numbers of DCs via FLT3L-Ig administration led to an increased severity of EAE. Results of that study showed that CD11c⁺ cells were capable of presenting antigen to autoreactive T cells and, more importantly, led to disease development (24). Because of the importance of antigen presentation in the induction of immune responses, we evaluated the potential for treating EAE by decreasing DC numbers and function through inhibition of FLT3 signaling.

Results of our studies confirm a previous result showing that FLT3 expression is maintained on mature DCs (25) and extend this finding by showing that signaling through this pathway plays a role not only in proliferation of DC progenitors, but also in protecting mature DCs from apoptosis. We show that inhibition of FLT3 signaling leads to death in mature DCs and, furthermore, that it can inhibit both an in vitro and an in vivo immune response through its effects on DCs. Thus, by inhibiting the generation and possibly the maintenance of the immune response, FLT3 inhibition provides a means of intervention that may prove efficacious in the treatment of autoimmune diseases.

Materials and Methods

Mice. Mice were purchased from NCI and maintained in the Johns Hopkins Oncology Animal Facility. All procedures were conducted under approved protocols, and mice were killed at appropriate signs of distress for the studies involving EAE.

Generation of DCs. BM was flushed from the femurs and tibias and differentiated into DCs by using standard methods with granulocyte/macrophage colony-stimulating factor (GM-CSF) with or without FL.

Human DCs were generated by differentiation of peripheral blood mononuclear cells in the presence of GM-CSF (20 ng/ml) plus IL-4 (20 ng/ml) (both from Peprotech).

Western Analysis. BM DCs were grown as described above; on day 8, 0 or 50 nM CEP701 was added to the culture for 1 h. Cells (10⁷) were harvested, and their lysates subjected to immunoprecipitation with FLT3 antibody (Santa Cruz Biotechnology). SDS/PAGE gels were then run, followed by transfer to nylon membranes, which were probed with antibodies to both the phosphorylated protein (4G10) and total FLT3, as described in ref. 26. Resultant films were scanned by using an Agfa Arcus 1200 laser scanner. Densitometry was performed with the public-domain program nih image (http://rsb.info.nih.gov/nih-image).

Isolation of DCs from Mice. Spleens and lymph nodes were harvested from mice, and single-cell suspensions were prepared; in some experiments, they were collagenase-digested (Roche Molecular Biochemicals) and purified on a CD11c column as described in refs. 27 and 28 (Miltenyi Biotec, Auburn, CA).

FACS Analysis. Cells were harvested and subjected to FACS analysis on a FACSCalibur system (BD Immunocytometry Systems), by standard methods. All antibodies were purchased from BD Pharmingen.

T Cell Proliferation Responses. DCs were irradiated at 3,000 rad, then incubated with responder T cells at the ratios shown in each figure. T cells were purified from spleens of mice by nylon wool enrichment (Polysciences). The number of T cells was kept constant, whereas the number of DCs was varied, as described in each figure. Cytokine ELISA kits were obtained from Endogen (IFN-γ) or R & D Systems (TNF-α).

In Vivo Autoreactive Responses. The transgenic HA137 mice (described in ref. 29) were treated with CEP-701 (20 mg/kg; Cephalon, West Chester, PA) twice daily for 3 days before the introduction of the 6.5 T cells, which have transgenic T cell receptors for hemagglutinin (HA). T cells were harvested from 6.5 mice and injected i.v. at 2.5 × 10⁶ per mouse. The clone 6.5 T cells also express the T cell congenic marker thy1.1. Five days after T cell transfer, mice were killed, and their spleens were harvested for analysis. T cell expansion was determined by staining with CD4 and Thy1.1 (both from BD Pharmingen).

Listeria monocytogenes Challenge. L. monocytogenes was obtained from Eric Pamer (Memorial Sloan-Kettering Cancer Center, New York) and injected at a dose of 5 × 10⁴ colony-forming units per mouse. Mice were injected for 5 days before challenge with either CEP-701 or vehicle and continued to be injected during the course of the experiment. Mice were checked daily for survival, and the experiment terminated at the time of plateau.

EAE. EAE was induced by standard methods described in refs. 30 and 31. Briefly, the MOG_35-55 peptide was synthesized at the Core Peptide Synthesis Facility at the Johns Hopkins Medical Institutions, then emulsified in complete Freund's adjuvant (Sigma), supplemented with an additional 400 μg of Mycobacterium tuberculosis (Difco), and injected at 100 μg per mouse. Pertussis toxin (200 ng per injection; List Biological Laboratories, Campbell, CA) was administered i.v. at the time of injection and then again 48 h later. Mice were injected with either vehicle control or CEP-701 (20 mg/kg, twice a day, 12 h apart), a protocol based on previous studies (20). Mice were scored daily by a blinded observer. Scoring was conducted by using the following scale: 0, no abnormalities noted; 0.5, loss of tail tonicity; 1, loss of tail reflex; 2, affected gait; 3, full hind limb paralysis; 4, front and hind limb paralysis; 5, moribund.

Pathology. Mice were killed, and the brain and spinal cord were removed and fixed in formalin for analysis. Paraffin-embedded sections were stained with hematoxylin/eosin and luxol fast blue. A pathologist (C.M.) evaluated degrees of myelin loss, in a blinded fashion, in the two groups, which were coded before analysis.

Results

Mature DCs Express FLT3. To determine expression and activity of FLT3 and potential inhibition of signaling via CEP-701, mature BM DCs were generated by a standard 8-day culture in GM-CSF + IL-4. Exogenous FL was added for 15 min before assaying the cells. Parallel cultures were established in which diluent was added to one set and 50 nM CEP-701 was added to the other, for 1 h, followed by lysis and Western analysis. Fig. 1 shows these results: Mature DCs express functional FLT3, and its activation is significantly decreased by the addition of CEP-701, thus providing a target for inhibition.

Fig. 1.

Mature DCs express FLT3, and CEP-701 inhibits its phosphorylation. Shown is a Western blot analysis of murine BM DCs demonstrating that mature DCs express high levels of FLT3 and that CEP-701 effectively inhibits its activation. Relative quantities of phospho-FLT3 are shown in the densitometry graph, normalized to total FLT3.

Treatment of Mature DCs with FLT3 TKI Induces Apoptosis in Mature Murine and Human DCs. We tested the hypothesis that FLT3 signaling helps maintain the survival of DCs and therefore that FLT3 inhibition might induce apoptosis in at least a fraction of DCs. After 8 days of culture in GM-CSF/IL-4 or GM-CSF/IL-4 plus FL (however, it should be noted that because numerous cell types secrete FL, even those cells grown in GM-CSF are subject to some stimulation through FLT3), nonadherent mature murine DCs were harvested and replated in the absence or presence of CEP-701 at 5 or 50 nM. After 24 h, CD11c⁺ cells were assessed for MHC class II expression and Annexin V binding by FACS analysis. CEP-701 induced a dose-dependent induction of apoptosis in a significant fraction of DCs generated either in GM-CSF alone or in the combination of GM-CSF and FL (Fig. 2 A and B). Similar results were observed when other FLT3 inhibitors were used (data not shown).

Fig. 2.

Treatment with FLT3 TKI induces apoptosis in mature mouse and human DCs. (A) DC survival was measured after exposure to CEP-701. Mature DCs were harvested and replated in the absence or presence of CEP-701 (0, 5, or 50 nM). CD11c⁺ cells were then assessed for MHC class II expression and Annexin V binding. CEP-701 induced apoptosis in DCs generated either in GM-CSF/IL-4 alone or in the combination of GM-CSF/IL-4 and FL. Shown are representative FACS plots (from four experiments). (B) Graph of the averages and SEM. (C) Treatment with FLT3 TKI induces apoptosis in mature human DCs. Shown are the effects of CEP-701 on human DCs. Either DMSO vehicle or CEP-701 was added to the cultures for 48 h, and then the cells were stained for CD11c and Annexin V binding. As was the case with the murine DCs, a significant increase in Annexin V⁺ cells was observed in the cultures exposed to CEP-701. Shown are representative plots from two separate studies. (D Left) FLT3 inhibition decreases proliferation of T cells through its effect on DCs. For these experiments, BM DCs from BALB/c mice were treated with CEP-701 or vehicle control, then plated with C57BL/6 splenocytes at the ratios of DC:spleen shown. After 3 days, proliferation of T cells was determined. As shown, CEP-701 significantly decreased proliferation. (D Right) Cytokine analysis for IFN-γ and TNF-α secretion was also conducted, and these results show a corresponding decrease in cytokine secretion for these inflammatory cytokines, shown as a percentage of control cytokine secretion in vehicle-treated cultures from the 1:10 DC:T cell ratio group. (E) FLT3 TKI does not directly inhibit T cells. Shown is the absence of a direct effect of FLT3 TKI on T cells. T cells were plated on anti-CD3-coated plates, followed by the addition of anti-CD28 antibody. As shown, the FLT3 TKI had no significant effect on direct stimulation of T cells. These studies were repeated twice.

The effects of FLT3 inhibition on human DCs were also evaluated. DCs were generated by differentiation of peripheral blood mononuclear cells in the presence of GM-CSF + IL-4. After incubation with 5 nM CEP-701 for 48 h, cells were harvested and stained for CD11c and Annexin V and then analyzed by FACS. As Fig. 2C shows, CEP-701 also induced apoptosis in a significant fraction of human DCs.

Treatment of Mature DCs with FLT3 TKI Decreases T Cell Stimulation by Murine DCs. We next evaluated the effects of FLT3 inhibition on stimulation of T cells by DCs. Increasing ratios of DC:T cells were tested in an allogeneic stimulation assay. CEP-701 (5 nM) or DMSO diluent was added to the cultures of BALB/c DCs, along with C57BL/6 T cells. [³H]Thymidine incorporation was used to determine T cell stimulatory activity. As shown in Fig. 2D, a decrease in the stimulation of allogeneic T cell responses as well as cytokine secretion by T cells and DCs was observed. Mature T cells do not express FLT3 and thus should not respond to FLT3 inhibitors directly. Therefore, we tested the possibility that the observed inhibition of T cell proliferation was due to nonspecific, toxic effects on T cells. For these studies, we incubated C57BL/6T cells in the presence or absence of CEP-701 on plates coated with anti-CD3 antibody and then added anti-CD28 stimulation to the cultures. There was no difference in the response of the T cells to direct T cell receptor stimulation, indicating that CEP-701 did not inhibit T cells directly (Fig. 2E).

FLT3 Inhibition in Vivo Decreases the DC and Natural Killer (NK) Cell Populations. The next set of studies addressed the effects of CEP-701 on DCs in vivo. Because CEP-701 decreased the survival of DCs in vitro, we hypothesized that it would induce apoptosis in vivo as well. Toward this end, we treated mice with either vehicle control or CEP-701 and then evaluated the effects on the composition of different cell types of the immune system in the spleen and lymph nodes. As Fig. 3A shows, treatment of mice with FLT3 TKI decreases both CD11c^hi and CD11c^lo/B220⁺ as well as NK populations but not T or B cells, indicating that the effects of the inhibitor are somewhat restricted to cells expressing FLT3. Further subsetting of DCs shows a loss in all subsets in spleen and LN, with no discernible bias toward any particular subset and, as expected, monocytes were not significantly affected (Fig. 3B).

Fig. 3.

Treatment of mice with FLT3 TKI decreases CD11c but not CD3 or B220 populations. (A) Spleens from BALB/c mice treated with CEP-701 or vehicle control, twice a day for 5 days, were harvested and analyzed by FACS for immune cell constitution. Shown is the FACS analysis for the presence of T cells (CD3⁺/DX5^-), NK cells (CD3^-, DX5⁺), NK T cells (CD3⁺, DX5⁺), B cells (B220⁺, CD11c^-), or DCs (myeloid, CD11c^hi/B220^-, or plasmacytoid CD11c^lo/B220⁺). As shown, whereas this treatment led to a decrease in NK and DC populations (FLT3 expressing populations), no significant change was observed in the B or T cell (not FLT3 expressing) populations. (B) In vivo treatment of murine DCs with CEP-701 leads to a loss of all major DC subsets in the lymph nodes and spleen. Shown are the relative percentages of some of the subsets, compared with controls (set at 100%). No significant change was observed in total cellularity or in CD11b⁺ CD11c^- macrophages.

Treatment of Mice with FLT3 TKI Decreases an in Vivo Autoreactive Immune Response. Because systemic administration of CEP-701 led to a decrease in DCs, we hypothesized that inhibition would also result in the down-regulation of an immune response. To test this hypothesis, we selected a model of autoimmunity in which we could track the expansion of antigen-specific T cells. The host HA-expressing mice (termed HA137 mice) were pretreated with either vehicle control or CEP-701, followed by adoptive transfer of 6.5 CD4⁺ T cells. In the absence of manipulations, adoptive transfer results in expansion of the T cells in response to the endogenous antigen (29). As shown in Fig. 4, in mice treated with vehicle control, there was a substantial activation of the 6.5 T cells; however, there was a significant inhibition of T cell expansion in the presence of CEP-701, indicating that CEP-701 suppressed an autoreactive response.

Fig. 4.

Treatment of mice with FLT3 TKI decreases an in vivo autoreactive immune response. Transgenic HA-expressing mice (termed HA137) were injected with HA-specific CD4⁺ T cells (6.5), which express the congenic T cell marker thy1.1, to determine the effect of CEP-701 on expansion of autoreactive T cells. In untreated mice, the transferred transgenic T cells expand and induce an autoimmune process. For these experiments, HA137 mice were treated with CEP-701, then analyzed for expansion of the 6.5 T cells. As shown, the CEP-701 treatment led to a decrease in the autoreactive T cell expansion, as measured by thy1.1. Shown are the data from two separate experiments, with six mice per group.

Treatment of Mice with FLT3 TKI Decreases the Severity of Established EAE. Although these results provide evidence for a mechanism of immune suppression, we wanted to test whether it would impact on an established disease process. To determine whether the downregulation of the autoimmune response observed in the transgenic model would result in an improvement in a recognized model of autoimmune disease, we determined the effects of FLT3 inhibition on the progression of established EAE. In these experiments, EAE was induced by immunization of mice with MOG_35-55, which were allowed to become symptomatic before the initiation of therapy. After the onset of symptoms, mice were randomized, then treated twice daily with CEP-701 or vehicle control (at doses previously established in the leukemia models) and scored daily, in a blinded fashion, for disease symptoms. A significant improvement in disease progression was observed in the mice receiving FLT3 inhibition (Fig. 5A).

Fig. 5.

Inhibition of FLT3 ameliorates established EAE. (A) EAE was induced in C57BL/6 mice, which were then injected twice daily with either vehicle control or CEP-701. Ten mice per group were treated in the study shown. Shown is the time course of one (of three similar) experiments. (B) Treatment with CEP-701 decreases demyelination. Shown is the histological staining on paraffin-embedded spinal cord sections of CEP-701-treated and vehicle control-treated mice after induction of EAE. As shown, significant demyelination (loss of blue stain) is found in MOG-vaccinated animals (Upper) but not in MOG-vaccinated and CEP-701-treated mice (Lower).

Pathology of EAE. Six mice from each group, MOG-vaccinated followed by CEP-701 or vehicle control treatment, were killed, then the spinal cord, cerebellum, and brain were removed from each mouse and evaluated for myelin loss by using luxol fast blue cytochemical stains. These slides were read by a pathologist in a blinded fashion. Histopathologic analysis revealed that mice from the group that received CEP-701 (Fig. 5B Lower) had significant myelin preservation as demonstrated by increased blue staining when compared with the control-treated MOG-vaccinated mice (Fig. 5B Upper) that only received vehicle.

Treatment with CEP-701 Does Not Induce Gross Immunosuppression. We next tested whether the observed decrease in an autoreactive response would be indicative of severe widespread immunosuppression. Although the mice showed no outward appearance of illness, we tested whether they would have an altered ability to ward off an infectious challenge. So that we could evaluate any alterations in immunity resulting from effects of inhibition, we pretreated mice with either vehicle control or CEP-701 for 5 days before challenge with L. monocytogenes, then continued to treat throughout the experiment. We selected an LD₂₀ so that the assay would provide us with a high level of sensitivity to assess any increases in death due to treatment. Mice were followed for signs of illness and survival. Although most mice progressed through an acute phase illness, there was no observed difference in the severity between the two groups. As the graph shows, there was no significant change in survival in the mice treated with CEP-701 (Fig. 6), indicating that any induced immunosuppression was not so severely toxic that it inhibited a response to this infectious challenge.

Fig. 6.

Treatment with CEP-701 does not inhibit the ability to respond to an infectious challenge. C57BL/6 mice were pretreated for 5 days with either vehicle control or CEP-701, then injected i.p. with an LD₂₀ of attenuated L. monocytogenes bacteria. Mice were followed for survival. This study was repeated with similar findings.

Discussion

The results of our studies suggest that inhibiting FLT3 signaling is a way to target DCs and thus could potentially serve as the basis for therapy in autoimmune diseases. Mature DCs, which are critical for the induction of immune responses, express FLT3. Maintenance of FLT3 signaling appears to be necessary for survival and activation of DCs. We show here that inhibition of FLT3 signaling induces cell death in mature DCs with a subsequent down-modulation of immune activation; we further show that treatment of mice with CEP-701 led to a significant improvement in the course of established EAE, a model for the autoimmune disease MS.

Although it has long been recognized that FLT3 signaling is important for the expansion and differentiation of hematopoietic progenitors into mature DCs (16, 17, 32, 33), and FLT3 has been shown to be present on mature DCs (25), the function of FLT3 signaling in mature DCs has not been elucidated. Our results indicate that inhibition of FLT3 signaling down-regulates immune responses mediated through DCs, both in vitro and in vivo. The effect on limiting the immune response is likely due to both the induction of apoptosis and a decrease in secretion of stimulatory cytokines from DCs, which results in corresponding decreases in T cell stimulation. Our in vivo analysis of the effects of CEP-701 on total cellularity showed a specific targeting of cells that express FLT3 in that there was a decrease in the number of DCs and NK cells but no change in the B and T cell populations after 5 days of treatment.

Although there is a great deal of conflicting evidence regarding the mechanism of disease pathogenesis surrounding both MS and EAE, recent studies have highlighted the importance of APCs in the diseases. To induce immune responses, T cells must encounter antigen in the context of APCs, and the immunization of mice with myelin antigens depends on uptake of the antigen by DCs, with subsequent stimulation of T cells. Furthermore, one recent study showed that number of DCs is likely to be an important aspect of disease pathogenesis in that the treatment of mice with FLT3L-Ig increased disease severity (24).

We investigated both the effects of inhibiting FLT3 signaling on both APCs directly and also the subsequent effects on T cell activation. Together, these effects provide a mechanism by which CEP-701 may decrease the severity of autoimmune disease. To investigate the in vivo effects on immune responses of treatment with CEP-701, we used a model (HA137 mice) in which we could track the activation of specific transgenic T cells against a known antigen. Although this model has limitations in terms of its use of a transgenic T cell receptor, it allows us to specifically quantify changes in T cell stimulation, which provides information regarding the functional consequence of inhibiting APCs. Also, this model allows investigation of an immune response to a self-antigen, somewhat mimicking an autoimmune disease. Treatment of these mice with CEP-701 led to a significant decrease in expansion and activation of adoptively transferred antigen-specific T cells.

Many prior therapeutic interventions have been shown to ameliorate the course of EAE when given at day 0 before the onset of disease. A major advantage of our DC-based approach is that we show a therapeutic effect when given after the onset of disease.

One possible cause for concern is that of inducing a globally immunosuppressive state. Although additional comprehensive studies will absolutely need to be conducted before clinical use for autoimmune disease, our results began to address this possibility in two different ways. First, we evaluated whether treatment with CEP-701 would decrease the total numbers of immune cells in the mice. Our results showed that although DCs and NK cells were decreased, there was no effect on the total B and T cell number or total numbers of cells. These findings suggest first that although the DC number is decreased, they are not entirely eliminated, allowing for the initiation of some immune responses. This finding is consistent with our results measuring T cell activation in our transgenic model after treatment with CEP-701. These results showed that although T cell expansion was decreased, it also was not eliminated. Second, we tested the ability of the treated mice to respond to an infectious agent. We challenged mice that were pretreated with CEP-701 to an LD₂₀ dose of L. monocytogenes to determine whether the treatment would cause the mice to be significantly more susceptible to a toxic infection. There was no significant difference in the overall survival of the mice that were treated with CEP-701. One limitation of this study is that the elements leading to a successful clearance of this infection are multifactorial and multicellular (including the contribution of macrophages to clearance). Thus, additional studies will necessary to more comprehensively evaluate effects on immunity and different types of immune challenge responses.

An alternative possibility may be that the degree of immunosuppression necessary to decrease autoimmune pathology could be lower than that required to induce significantly toxic immunosuppression, and some other types of immunosuppressive treatments for autoimmune diseases can impact disease progression without inducing a severely immunocompromised state in patients.

An additional advantage for the development of DC-based therapies for the treatment of many autoimmune diseases is that attempts at T cell- or antigen-specific therapies often fail because of the degenerate nature of peptide recognition and the diversity of T cell responses, attributed to epitope spreading. A DC-targeted therapeutic approach may be beneficial because it avoids the absolute requirement for knowing the autoantigen(s). The use of DC inhibitors could block epitope spreading in the chronic setting, or (if this proved to be overly immunosuppressive) DC-targeted therapy could be given at the time of an exacerbation to limit the extent of disease, as was demonstrated here in EAE.

A number of inhibitors of FLT3 have been developed for the treatment of leukemia because of the frequent occurrence of FLT3-activating mutations in acute myelogenous leukemia (19, 21). These drugs have been well tolerated, are available orally, and are administered in the outpatient setting. Thus, they are ideal for the treatment of chronic diseases that may require prolonged periods of treatment and thus may be considered for investigation for the treatment of autoimmune diseases.