Nanoparticle compositions for generation of regulatory T cells and treatment of autoimmune diseases and other chronic inflammatory conditions

Abstract

The present invention relates to nanoparticles for the targeted delivery of antigen to liver cells, in particular, liver sinusoidal endothelial cells (LSEC) and/or Kupffer cells, and for the in vivo generation of regulatory T cells, notably CD4+CD25+FOXP3+ regulatory T cells (Treg). The invention provides pharmaceutical compositions and methods for the prevention and treatment of autoimmune diseases, allergies or other chronic inflammatory conditions, and for generation of regulatory T cells. The nanoparticles used in the invention comprise a) a micelle comprising an amphiphilic polymer rendering the nanoparticle water-soluble, and b) a peptide comprising at least one T cell epitope associated with the outside of the micelle. The micelle may or may not comprise a solid hydrophobic core.

Claims

1. A method comprising administering a composition comprising a nanoparticle to a subject in need thereof, wherein said nanoparticle comprises a) a micelle comprising an amphiphilic polymer rendering the nanoparticle water-soluble, and b) a peptide comprising at least one T cell epitope associated with the outside of the micelle, and wherein the composition generates regulatory T cells specific to said at least one T cell epitope in said subject.

2. The method of claim 1, wherein the micelle does not comprise a solid core.

3. The method of claim 1, wherein micelle coats a solid hydrophobic core.

4. The method of claim 3, wherein the core comprises a traceable inorganic material selected from the group comprising iron oxide, CdSe/CdS/ZnS, silver and gold.

5. The method of claim 3, wherein the diameter of the core is 5 to 30 nm.

6. The method of claim 1, wherein the polymer is a synthetic polymer selected from the group consisting of poly(maleic anhydride-alt-1-octadecene), poly(maleic anhydride-alt-1-tetradecene) and polyisoprene-block polyethyleneoxide block copolymer.

7. The method of claim 1, wherein the nanoparticle is negatively charged.

8. The method of claim 1, wherein the peptide is covalently linked to the micelle or non-covalently associated.

9. The method of claim 1, wherein the nanoparticle comprises a solid core comprising iron oxide having a diameter of about 9 to about 12 nm, which is encapsulated by a coating of poly(maleic anhydride-alt-1-octadecene) to which a peptide comprising a T cell epitope is covalently linked.

10. The method of claim 1, wherein the nanoparticles are suitable for transferring the peptide to liver sinusoidal endothelial cells of said subject in vivo.

11. The method of claim 1, wherein the pharmaceutical composition is suitable for inducing generation of regulatory T cells specific for the at least one epitope via presentation of said epitope by liver sinusoidal endothelial cells and/or Kupffer cells of said subject.

12. The method of claim 1, wherein the nanoparticles are formulated for administration to said subject having a disease wherein suppression of a specific immune response is beneficial.

13. The method of claim 12, wherein said disease is an autoimmune disease, or an allergy or a disease wherein inflammation is excessive, chronic or adverse.

14. The method of claim 12, wherein the disease is selected from the group consisting of multiple sclerosis, type I diabetes, rheumatoid arthritis, systemic lupus erythematosus, degeneration after trauma or stroke, graft versus host disease, transplant rejection, inflammatory bowel disease (IBD), asthma and allergy.

15. A method of generating regulatory T cells, comprising isolating regulatory T cells from a sample taken from a subject, wherein said subject has been administered a nanoparticle comprising: a) a micelle comprising an amphiphilic polymer rendering the nanoparticle water-soluble, and b) a peptide comprising at least one T cell epitope associated with the outside of the micelle.

16. The method of claim 15, further comprising: administering to said subject said nanoparticle; and taking said sample comprising T cells from said subject, prior to said isolating.

17. The method of claim 3, wherein the diameter of the core is 9 to 12 nm.

18. The method of claim 1, wherein the polymer is poly(maleic anhydride-alt-1-octadecene).

19. The method of claim 1, wherein the peptide consists of 9 to 60 amino acids.

Description

FIGURE LEGENDS

(1) FIG. 1: Efficient generation of Treg cells from CD4+CD25 non-regulatory T cells by various liver cell types, but not by splenic dendritic cells.

(2) FIG. 2: Amelioration and delayed onset of experimental autoimmune neuroinflammation by transfer of LSEC-stimulated neuroantigen-specific T cells.

(3) FIG. 3A: Schematic design of an exemplary micelle for targeted delivery of peptides to liver cells formed by an amphiphilic polymer. The hydrophobic parts of the polymer form the center of the micelle, while the hydrophilic parts are turned to the outside and render the micelle water-soluble. In the example shown, the hydrophilic parts are negatively charged.

(4) FIG. 3B: Schematic design of the micelle, formed as in FIG. 3A, coats a solid core. The solid core may be, e.g., an inorganic core such as an FeO nanoparticle stabilized by oleic acid molecules. The hydrophobic parts of the polymer interact with the hydrophobic core, thus forming a micelle around the core.

(5) FIG. 4A: Fluorescent nanoparticles of the invention are rapidly and specifically taken up from the circulation and accumulate predominantly in liver sinusoidal endothelial cells (LSEC), as demonstrated by intravital microscopy. In addition to the fluorescent nanoparticles taken up in the LSEC lining the hepatic sinusoids, nuclei (round) are stained with DAPI.

(6) FIG. 4B: Confirmation of localization of nanoparticles of the invention in liver sinusoidal endothelial cells by transmission electron microscopy.

(7) FIG. 4C: Nanoparticles embedded in recombinant triglyceride rich lipoprotein liposomes prepared according to Bruns et al., 2009, are specifically taken up from the circulation and accumulate predominantly in Kupffer cells, as demonstrated by intravital microscopy (a) in control treated mice. (b) In clodronate-treated mice, wherein Kupffer cells were depleted, uptake of the liposomal nanoparticles is significantly reduced, confirming accumulation in Kupffer cells in normal mice.

(8) FIG. 5: Targeted delivery of neuroantigen peptide to liver cells by peptide-nanoparticle conjugates, but not peptide-free nanoparticles induces full protection from clinical neuroinflammatory disease.

(9) FIG. 6: TGF dependent Treg conversion by liver antigen-presenting cells. MBP-specific CD4+Foxp3 non-Tregs were stimulated with MBP peptide by liver DC, LSEC, KC or splenic DC in presence (A) or absence (B) of TGF and conversion rates to Foxp3+Tregs were determined. C) Defective Treg conversion by TGF-insensitive CD4 T cells. D) Colocalization of GARP or LAP and CD31 on liver sections of wild-type mice revealed by confocal microscopy. E) Expression of GARP and LAP (solid lines) by freshly isolated LSEC (upper panels) or splenic DC (lower panels) determined by flow cytometry (isotype controls in grey). F) MBP-specific CD4+CD25 non-Tregs were stimulated by LSEC with MBP peptide and transferred to B10.PL mice (10 5 cells/mouse; n=8), in which EAE had been induced. Control mice received PBS (n=9). Mean EAE scoress.e.m. are shown; P=0.0025.

(10) FIG. 7: Targeted delivery of MBP peptides to LSEC with nanoparticles. A) Plasma clearance rates of MBP peptide coupled to .sup.59Fe-NP (MBP-.sup.59Fe-NP) and .sup.14C-MBP peptide coupled to unlabeled NP(.sup.14C-MBP-NP). B) Organ distribution of MBP-.sup.59Fe-NP and .sup.14C-MBP-NP 60 min after intravenous injection. Mean valuess.e.m. are shown (n=4). C) Representative coronal and transversal MR images of mice pre and post injection of unloaded NP or MBP-NP. D) Intravital confocal microscopy of the liver showing uptake of MBP-NP or unloaded NP by LSEC. Nuclei were also stained. right panel: LSEC (FITC-dextran), left panel: quantum dot-labeled NP; scale bar: 20 m. E) Electron microscopy of liver, 60 min after intravenous injection of MBP-NP or unloaded NP.

(11) FIG. 8: Nanoparticle-based delivery of neuroantigen peptides to LSEC prevents EAE induction. EAE was induced in B10.PL wildtype mice (A, E) or tg4 mice (B) by immunization to MBP. In C57BL/6 wildtype mice (C) or hCD2-kTRII mice (D), EAE was induced by immunization to MOG. One day after EAE induction, mice were injected once intravenously with PBS (A-E) or neuroantigen peptide-loaded nanoparticles (MBP-NP (A, B, E) or MOG-NP (C, D,). As control, mice received equivalent amounts of unloaded NP (A, C) or unconjugated MBP peptide (B,). E, Twice per week, mice treated with MBP-NP were intraperitoneally injected with Treg-depleting PC61 antibody (MBP-NP/Treg-depleted) or isotype-matched control antibody (MBP-NP). Shown are mean EAE scoress.e.m. (n=5-7).

(12) FIG. 9: Therapy of established EAE by nanoparticle-based transfer of MOG peptide to LSEC. EAE was induced in C57BL/6 mice by immunization to MOG. After disease onset (day 8-12, indicated by arrow), mice were treated with a single injection of MOG-loaded nanoparticles (n=10, grey line) or PBS (n=17, black line). A) EAE course over an extended observation period of more than nine weeks after MOG-NP (grey) or PBS (black) treatment. Shown are mean EAE scoress.e.m. B) Change of mean individual disease score after treatment with MOG-NP (gray) or PBS (black); EAE score=score difference compared to the time of treatment.

(13) FIG. 10: Coupling efficiency of peptide to NP. Calculated number of peptide molecules coupled to NP after incubation with increasing amounts of radiolabeled MBP peptide. Mean valuess.e.m. are shown (n=3).

(14) FIG. 11: Efficient Treg conversion by LSEC in vitro using nanoparticle-conjugated MBP peptide or free MBP peptide. Splenic CD4+Foxp3 T cells from tg4Foxp3gfp.KI mice were co-cultured with LSEC in the presence of TGF (2 ng/ml) and free MBP peptide (5 ng/ml) or an equivalent dose of MBP peptide coupled to nanoparticles (MBP-NP) for four days. Foxp3 induction in CD4+ T cells was then analyzed by flow cytometry. For statistical analysis, Mann-Whitney test was performed (free MBP: 32.161.075; MBP-NP: 29.631.558; P=0.2857). n.s.=not significant.

(15) FIG. 12: Characteristics of NP and MBP-NP. A) The Zeta potentials of NP and MBP-NP were measured at 20 C. using a Malvern Zetasizer Nano-ZS instrument. Both samples exhibit a comparable negative surface charge (averaged Zeta-potential of NP 58 mV; MBP-NP 61.6 mV) (MBP-NP: higher peak; NP: lower peak). No differences in size could be detected by B) size exclusion chromatography (MBP-NP: left line; NP: right line) or C) by transmission electron microscopy. For electron microscopy, samples were stained with 1% sodium silicotungstate, pH 7, so that the organic shell appears light and the iron oxide core dark.

(16) FIG. 13: Depletion of CD25+Foxp3+CD4+ Treg in vivo by repeated PC61 antibody-administration. B10.PL mice were immunized to MBP peptide and treated on the next day with PBS (n=7) or MBP peptide-coupled nanoparticles (MBP-NP). Mice, which received MBP-NP, were then either treated twice per week with Treg-depleting PC61 antibody (n=5) or non-depleting isotype-matched control antibody (n=6). Mice were sacrificed on day 29 post immunization and spleens were analyzed for Treg depletion efficiency by flow cytometry. Results are depicted as mean % of CD25+Foxp3+of CD4+ T cellss.e.m. (PBS 17.160.3422; MBP-NP+isotype 15.080.8594; MBP-NP+PC61 3.1600.4643; P<0.001). For comparison of multiple groups, one-way ANOVA and Tukey's post test were performed.

(17) FIG. 14: Initial EAE scores at the time of therapeutic MOG-NP or PBS administration. C57BL/6 mice were immunized to MOG peptide and developed clinical EAE symptoms. After disease onset (day 8-12), mice were treated with a single injection of MOG-loaded nanoparticles (n=10) or PBS (n=17); at the time of treatment, the mean initial disease scores between groups were indifferent (means.e.m.: 1.590.12 vs. 1.600.19; P=0.9195).

EXAMPLES

Example 1

Generation of Regulatory T Cells by Liver Sinusoidal Endothelial Cells and Kupffer Cells

(18) Mouse liver non-parenchymal cells were isolated applying a protocol modified from (28). Briefly, mouse livers were perfused with 0.05% collagenase IV (Sigma; Taufkirchen, Germany) in Gey's balanced salt solution, mechanically dissected and further digested in 0.05% collagenase IV in Gey's balanced salt solution for 25 minutes at 37 C. at constant rotation (240 rpm). Hepatocytes and debris were sedimented twice at 40 g and non-parenchymal cells were recovered by centrifugation over a 17% Optiprep (Sigma) gradient at 400 g.

(19) Liver sinusoidal endothelial cells (LSEC) were purified from the non-parenchymal cells as described (29) by magnetic sorting with the ME-9F1 antibody (Miltenyi Biotech, Bergisch-Gladbach, Germany). LSEC were seeded on collagen-coated culture plates (Serva, Heidelberg, Germany). After overnight culture in Iscove's modified Dulbecco's medium supplemented with 5% FCS, non-adherent cells were removed by medium change. Accordingly, Kupffer cells were purified from the non-parenchymal cells by magnetic sorting with biotinylated F4/80 antibody (eBioscience, Frankfurt, Germany) and anti-biotin microbeads (Miltenyi).

(20) Alternatively, hepatocytes were isolated from mouse livers as described (30); and dendritic cells were isolated from mouse spleen by magnetic cell separation for CD11c as described (30).

(21) 10.sup.5 LSEC, 10.sup.5 Kupffer cells, or 510.sup.3 hepatocytes were seeded into 96-well plates, and used on the following day for stimulation of CD4+CD25 T cells. Alternatively, 510.sup.4 dendritic cells, were seeded into 96-well plates, and used on the same day for stimulation of CD4+CD25 T cells. These CD4+CD25 non-regulatory T cells were isolated from mouse spleens by magnetic cell separation, and 510.sup.5 purified lymphocytes per well were then stimulated on the indicated antigen presenting cells in serum-free medium (Pan Biotech, Aidenbach, Germany) with solute antibody to CD3 (BD Bioscience, Heidelberg, Germany). After four days of culture, the T cells were harvested and analysed by flow cytometry for the percentage of CD4+CD25+FOXP3+ Treg cells. As indicated, transforming growth factor-beta (TGF) was added to some stimulation cultures.

(22) Results: As shown in FIG. 1, dendritic cells were poor inducers of Treg. In the absence of exogenous TGF, the generation of Treg cells induced by dendritic cells was negligible, and in the presence of exogenous TGF, only 7.5% of the dendritic cell-stimulated T cells were Treg. In contrast, the various liver cell typesLSEC, Kupffer cells and hepatocyteswere efficient inducers of Treg, at least in the presence of exogenous TGF, with at least 20% of the liver cell-stimulated T cells being Treg. Notably LSEC were efficient Treg generators, since, in the presence of TGF, 38% of the stimulated cells were Treg, and, even in the absence of exogenous TGF, 16.3% of the stimulated T cells were Treg.

Example 2

Suppression of Experimental Autoimmune Neuroinflammation by Liver Sinusoidal Endothelial Cell-Stimulated CD4 T Cells

(23) LSEC were isolated as described in example 1 and used to stimulate non-regulatory CD4+CD25 T cells purified from tg4 mice, which carry a transgenic T cell receptor specific for myelin-basic protein (25). After 4 days of T cell stimulation with the specific myelin basic protein-peptide (5 ng/ml) on LSEC, performed as described in example 1, 10.sup.5 LSEC-stimulated T cells were transferred to B10.PL mice that had been immunized one day before with a myelin basic protein peptide to induce the clinical symptoms of experimental autoimmune encephalomyelitis (EAE), as described in (25).

(24) Results: Adoptive transfer of LSEC-stimulated myelin basic protein-specific T cells induces a delayed and strongly ameliorated course of EAE, indicating that LSEC-induced neuroantigen-specific Treg cells are capable of improving autoimmune neuroinflammation.

Example 3

Selective and Specific Targeting of Nanoparticle to Liver Sinusoidal Endothelial Cells In Vivo

(25) a) Nanoparticles were generated by first synthesizing iron oxide cores of about 11 nm according to (26) with 0.18 g FeOOH, 2.31 g oleic acid and 6.4 mL (5.0 g) 1-octadecene; the protocol was modified in interrupting the reaction after one hour for yielding larger core sizes of about 11 nm.

(26) Then followed encapsulation of the cores according to (27) with some modifications: 2 mL of poly(maleic anhydride-alt-1-octadecene) solution (c=0,01 g/mL in CHCl.sub.3) was added to a solution of 2 mg nanoparticles dissolved in 2 mL of chloroform and stirred at room temperature overnight. The solvent was then evaporated by exposure to a constant stream of N.sub.2, and 2 mL of 20% TBE buffer were added. The solution was sonicated three times for 10 minutes, allowing the solution to cool in between. Subsequently, the solution was heated at 60 C. for 10 minutes. Forming aggregates were removed by centrifugation at 2400 g (three times for 10 minutes), and excess polymer was removed by ultracentrifugation (1 h, 50 000 g, 4 C.). Finally, the solution was filtered sequentially through filters with pore sizes of 0.45 m, 0.2 m and 0.1 m. The quality of the particles was confirmed by size-exclusion-chromatography and transmission electron microscopy. The iron content was measured by treating 200 L of a diluted sample with 50 L of 5 M hydrochloric acid at 70 C. for 30 minutes (31). Thereafter, 150 L of a 2 M acetate buffer (pH=4.8) containing 10% ascorbic acid was added to 50 L of each sample, followed by 100 L of a solution of 50 mg bathophenanthroline in 50 mL water. After 15 minutes, the absorption was measured at 540 nm. The general design of the nanoparticles is illustrated in FIG. 3.

(27) b) Comparative nanoparticles, nanosomes comprising a lipid bilayer, were generated by the procedure described in Bruns, et al., Nat. Nanotechnol. 2009, 4 193-201.

(28) Results:

(29) a) The obtained nanoparticles were injected into mice and, as assessed by MRI, found to specifically accumulate in the liver. Moreover, radioactive labelling of the cores with .sup.59Fe showed that about 80% of the injected particles were retained in the liver. To confirm hepatotropism and further identify specific target cells, fluorescent semiconductor CdSe/ZnS cores (quantum dots), as described in (32), were used as core instead of iron oxide, leaving the capsule and contact surface of the nanoparticles unchanged. Intravital microscopy was then performed after injection of these fluorescent nanoparticles into mice, showing that these nanoparticles rapidly accumulate predominantly in LSEC (FIG. 4A). A minor fraction of the injected nanoparticles was found in Kupffer cells or in some endothelial cells of intestinal veins that feed the portal vein.

(30) The obtained nanoparticles of the invention were further injected into mice and found to accumulate in liver sinusoidal endothelial cells as assessed by transmission-electron microscopy (FIG. 4B).

(31) b) For comparison, fluorescent nanosomes prepared according to Bruns et al., 2009, were injected into mice that two days before had been depleted of Kupffer cells through injection of chlodronate-containing liposomes (FIG. 4C: b) or not depleted of Kupffer cells through injection of sham liposomes (FIG. 4C: a). In sham-treated control mice, liposomes are internalized by Kupffer cells, as inferred from the characteristic staining pattern. In clodronate-treated mice, uptake of particles was substantially reduced, confirming that nanocrystals embedded into lipoproteins are indeed internalized by Kupffer cells. FIG. 4C shows that the nanosomes accumulate in Kupffer cells, but not in LSEC.

Example 4

Treatment of Experimental Neuroinflammation with Autoantigen-Loaded Nanoparticles In Vivo

(32) For covalent conjugation of neuroantigen peptide (encephalitogenic Ac1-9 peptide derived from the amino acid sequence of myelin basic protein), iron oxide-core nanoparticles were synthesized as described in example 3 and adjusted to a concentration of 6 mol/L in 50 mM SBB-buffer (pH=9). An equal amount of 1-Ethyl-3-(3-dimethylaminopropyl)carbodiimide (c=60 nmol/L), dissolved in a 50 mM SBB-buffer (pH=9), was added. After 5 minutes, a 1000-fold excess of the neuroantigen peptide was added and the solution was incubated under rotation for 2h at room temperature and then overnight at 4 C. On the next day, the solution was filtered under centrifugal force in a filter device (100 kDa, 2500 g, 4 C.) and the remnant resuspended in PBS. This procedure was repeated 20 times to remove all excess peptide. The obtained peptide-particle conjugates were dissolved in PBS and subjected to size-exclusion-chromatography to assure for the absence of free peptide. Then, the iron content was measured to determine the concentration of the peptide-particle suspension.

(33) Results: The thus obtained peptide-loaded nanoparticles or, as control, unconjugated peptide-free nanoparticles were then injected into B10.PL mice (n=6), one day after induction of experimental autoimmune encephalomyelitis (EAE) by immunization to Ac1-9 myelin basic protein peptide (25). As shown in FIG. 5, only the neuropeptide-loaded nanoparticles induced complete protection from clinical EAE. In contrast, peptide-free nanoparticles did not protect from EAE and the mice developed clinical EAE scores that were similar to those of PBS-treated mice.

Example 5

T-reg-Mediated Control of Autoimmunity by Nanoparticle-Based Antigen Delivery to Tolerogenic Liver Cells

(34) Liver dendritic cells (DC), liver sinusoidal endothelial cells (LSEC) and Kupffer cells (KC) are liver resident antigen-presenting cells constitutively expressing MHC II molecules and capable of stimulating CD4.sup.+ T cells. To test whether these liver cells are involved in Treg induction, primary liver DC, LSEC or KC were used to stimulate CD4.sup.+Foxp3.sup. non-Treg cells from the spleen of (tg4Foxp3gfp.KI) F1 mice. These non-Treg cells specifically recognize myelin basic protein (MBP), and allow monitoring of Treg conversion based on the Foxp3-linked expression of green fluorescent protein (GFP). The stimulation of non-Treg cells was performed in the presence of the specific MBP peptide and exogenous TGF, which is required for Treg conversion. All tested liver cell types were capable of inducing MBP-specific Foxp3.sup.+ Tregs; however, LSEC were more efficient in inducing Tregs than KC and liver DC (FIG. 6A). The ability of the liver cells to provide TGF-signals in the absence of exogenous TGF was then tested. Liver DC and KC were not efficient in inducing Tregs (FIG. 6B), whereas LSEC could induce Treg conversion at increased rates (FIG. 6B). To confirm that LSEC can provide TGF-signals that promote Treg conversion, CD4.sup.+CD25.sup. non-Tregs from (hCD2-kT RIItg4) F1 mice were stimulated on LSEC; these T cells carry a dominant-negative TGF type II receptor and thus display reduced sensitivity to TGF. The inventors found strongly reduced conversion rates of TGF-insensitive T cells to Foxp3.sup.+ Tregs, as compared to wild-type T cells (FIG. 6C), indicating that the efficiency of LSEC to induce Tregs depends on TGF-signaling. Although LSEC seem to provide TGF-signals for Treg induction (FIG. 6B), active TGF in LSEC supernatants was not detected (not shown). Therefore, the inventors tested whether LSEC have bound TGF to their membrane; in complex with latency-associated peptide (LAP) (40, 41), TGF can be tethered to the surface of cells through glycoprotein-A repetitions predominant (GARP, also known as LRRC32) (42). Such membrane-bound TGF was shown to induce a tolerogenic phenotype and Foxp3-expression in stimulated T cells (41). Indeed, the inventors found by immunohistochemistry that GARP as well as LAP co-localized with CD31.sup.+ LSEC (FIG. 6D), whereas GARP and LAP expression was undetectable on other liver cells. Quantitation of GARP and LAP expression by flow cytometry confirmed cell surface expression of both GARP and LAP by LSEC, but not by splenic DC (FIG. 6E). These findings indicate that LSEC can efficiently induce antigen-specific Tregs through membrane-bound TGF-.

(35) Next, in vitro LSEC-induced MBP-specific Tregs were generated, and their suppressive functionality tested in vivo in a mouse model of multiple sclerosis, experimental autoimmune encephalomyelitis (EAE). Whereas the control group developed clinical EAE, mice that received LSEC-induced MBP-specific Tregs (FIG. 6F; T(LSEC)) were almost completely protected from MBP-induced EAE. However, this suppressive effect was limited to about three weeks, indicating that the transferred Tregs were not stable. The inventors reasoned that a more sustained Treg stability and protection from EAE could be achieved by antigen-specific Treg induction by LSEC in vivo.

(36) To develop a method for selective delivery of autoantigen to LSEC in vivo, the inventors made use of nanoparticles (NP) that are rapidly cleared from the circulation through uptake by liver cells. Specifically, the inventors took advantage of a polymer-coated nanoparticle (43), which they observed to accumulate selectively in liver sinusoidal endothelium (data not shown). To generate an effective antigen carrier system, a simple one-step coupling strategy using a carbodiimide crosslinking approach was chosen to attach the MBP peptide to polymer-coated nanoparticles with superparamagnetic iron oxide core. Each nanoparticle was loaded maximally with 100 peptide molecules per nanoparticle as demonstrated by binding assays with .sup.14C-radiolabeled MBP peptide (FIG. 10). Peptide-loading did not influence size, shape and charge of nanoparticles as determined by measurement of the Zeta potential, size exclusion chromatography and electron microscopy (FIG. 12). To investigate whether MBP peptide-loading altered pharmacokinetics and tropism of our nanoparticles, kinetic turnover studies with .sup.14C-radiolabeled MBP peptide-nanoparticle conjugates (.sup.14C-MBP-NP) were performed in comparison with .sup.59Fe-radiolabeled nanoparticles conjugated to non-labeled MBP peptide (MBP-.sup.59Fe-NP). Both radiolabels were cleared with similar kinetics (FIG. 7A), predominantly by the liver (FIG. 7B). Low amounts of radioactivity could be detected in spleen and kidney, whereas uptake into other organs was negligible. In addition, organ uptake was visualized by dynamic magnetic resonance imaging, revealing only hepatic targeting of both MBP peptide-loaded and unloaded nanoparticles (FIG. 7C, Table 1 (below)). Thus, MBP peptides are not excreted via the kidney, but transported by the desired targeting effect of the nanoparticles of the invention to liver cells.

(37) To further investigate the cellular tropism of administered peptide-nanoparticle conjugates, conventional as well as intravital confocal fluorescence microscopy of the liver was performed, using nanoparticles with a fluorescent quantum dot core (44). Prior to administration of fluorescent MBP peptide-nanoparticle conjugates, recipient mice were injected with FITC-dextran that is rapidly absorbed by LSEC. After intravenous injection of MBP-nanoparticles or unloaded nanoparticles, the inventors observed a rapid accumulation of both unloaded and MBP peptide-loaded nanoparticles in the liver, which clearly co-localized with FITC-dextran stained LSEC (FIG. 7D). To further confirm uptake of peptide-loaded and unloaded nanoparticles by LSEC, transmission electron microscopy of liver slides was performed, and a sizeable internalization of MBP peptide-nanoparticle conjugates in endosomal compartments of LSEC detected (FIG. 7E). To confirm that delivered peptide can be presented to T cells, Treg conversion in vitro by LSEC in the presence of free MBP peptide or of MBP peptide bound to nanoparticles (MBP-NP) was studied. The inventors found that Treg generation occurred with similar efficacy (FIG. 11), indicating that MBP peptide delivered by nanoparticles was efficiently presented to T cells. Together, these findings demonstrate that these nanoparticles, with or without peptide cargo, are selectively taken up by LSEC in vivo, and thus facilitate selective delivery of antigen peptides to LSEC.

(38) TABLE-US-00001 TABLE 1 Detailed parameters of MRI-sequences used for in vivo animal measurements The measurements were done in the imaging- mode at 22 C. room temperature. slice effective total thick- voxel dynamic scan FoV ness volume FA TR TE interval time [mm] matrix [mm] (m).sup.3 slices [] [ms] [ms] NSA [s] [min] T2*- 100 35 448 1 223 22 30 602 6.9 4 N/A 3:59 weighted 223 coronal 1000 multisclice FFE T2*- 35 35 288 1 208 20 80 353 6.9 4 N/A 3:02 weighted 208 transverse 1000 multislice FFE T2*- 34 34 160 1.3 213 1 2.0 29 6.9 8 1.42 17:16 weighted 213 transverse 1100 FFE Abbrevations: Fast Field Echo sequence (FFE), Field of View (FoV), excitation flip angle (FA), Repetition Time (TR), Echo Time (TE), Number of Signal Averages (NSA).

(39) The proposed immunosuppressive effect of MBP-nanoparticles in vivo was first tested in MBP-induced EAE in B10.PL mice (FIG. 8A). A single intravenous injection of MBP peptide-loaded nanoparticles one day after EAE-induction provoked a lasting and almost complete protection from clinical EAE; in contrast, control mice treated with PBS or with unloaded nanoparticles developed clinical EAE symptoms (FIG. 8A). Remarkably, even tg4 mice (45), which feature MBP-specific CD4.sup.+ T cells and develop more severe EAE, were protected from disease by a single MBP-NP injection one day after EAE-induction (FIG. 8B); control mice receiving an equivalent dose of free MBP peptide, in contrast, developed severe EAE (FIG. 8B). To confirm the efficacy of the inventive treatment approach in another independent disease model, the effect of nanoparticles loaded with myelin oligodendrocyte glycoprotein (MOG) peptides on the development of MOG-induced EAE was tested in C57BL/6 mice. Consistent with the inventors' observations in MBP-induced EAE of B10.PL mice, administration of MOG peptide-loaded nanoparticles to C57BL/6 mice one day after EAE induction lastingly rescued recipient mice from developing any clinical disease symptom; in contrast, control mice receiving unloaded nanoparticles developed EAE (FIG. 8C).

(40) As the inventors have shown above that tolerance induction by LSEC required TGF signaling to T cells, they reasoned that nanoparticle-based targeting of autoantigen peptides to LSEC should be ineffective in mice with TGF-insensitive T cells. Indeed, MOG peptide-loaded nanoparticles induced only a minor degree of protection from EAE in hCD2-kTRII mice (46), which feature TGF-insensitive T cells (FIG. 8D). Thus, it can be concluded that also in vivo, the efficacy of nanoparticle-mediated tolerance induction is dependent on TGF signaling to T cells. To confirm that nanoparticle-mediated protection from EAE was dependent on Treg function in vivo, Tregs were depleted in mice that had received MBP-nanoparticles by repeated administration of the PC61 antibody that recognizes CD25 (47) (FIG. 8E); the efficacy of depletion was confirmed by flow cytometry (FIG. 13). Although administration of MBP-nanoparticles induced protection from EAE in control mice, MBP-nanoparticles could not prevent the development of disease in Treg-depleted mice (FIG. 8E). In fact, the disease severity precipitated by Treg depletion was similar to the disease severity of the control group receiving PBS instead of MBP-nanoparticles (FIG. 8E), demonstrating that Tregs are critically involved in nanoparticle-mediated tolerance induction in vivo.

(41) Having determined that autoantigen peptide-loaded NP efficiently prevented autoimmune disease development, the inventors addressed the even more important issue of whether peptide-loaded nanoparticles are an effective treatment of established autoimmune disease. Therefore, they first induced clinical EAE in C57BL/6 mice by MOG-immunization. At the time of treatment, both the recipients of MOG peptide-loaded nanoparticles and control mice showed manifest clinical symptoms with similar disease scores (FIG. 14). The control group progressed in EAE severity; in contrast, the mean clinical scores of MOG-NP treated mice improved rapidly and substantially (FIG. 9A, FIG. 9B). Of note, tolerance induced by a single therapeutic injection of MOG peptide-loaded nanoparticles continued throughout an extended follow-up period of 9 weeks (FIG. 9A, FIG. 9B). In conclusion, nanoparticle-mediated autoantigen delivery to LSEC is therapeutically effective in autoimmune disease.

(42) Taken together, the inventors' results indicate that nanoparticle-based targeted delivery of autoantigen peptides to Treg-inducing LSEC in vivo can restore immune tolerance to self and serve the effective treatment of autoimmune disease. Therefore, this approach offers a solution to the problem of effectuating a Treg-based therapy for autoimmune disease by providing an efficient methodology for inducing antigen-specific Tregs in vivo. LSEC are particularly well-suited as target cells for tolerogenic antigen-delivery, since they induce T cell tolerance and do not support inflammatory T cell effector responses. Nanoparticles targeting splenic DC are also being explored as tools for antigen-specific tolerance induction. However, DC display a high degree of plasticity and readily can differentiate to inflammatory cells, whereas LSEC display a robust tolerance-inducing phenotype even in the presence of pro-inflammatory signals. Therefore, autoantigen delivery to robust tolerogenic LSEC provides increased safety, notably when attempting to treat individuals with on-going inflammation.

(43) Nanoparticles are most commonly modified by the attachment of specific ligands to their surface in order to avoid their rapid clearance by the liver and to redirect them to specific targets, e.g. to cancer cells. Nevertheless, the targeting efficacy of such redirection is still limited by undesired uptake in the liver. By taking advantage of the highly selective nanoparticle uptake by LSEC, the inventive approach avoids mistargeting of autoantigen to potentially inflammatory cells. Indeed, this efficient and selective delivery to LSEC may explain the stable and potent treatment efficacy induced by only a single administration of peptide-loaded nanoparticles.

(44) The inventors' data provide proof-of-principle that antigen-specific tolerance induction in vivo can be attained through nanoparticle-based antigen-delivery to LSEC. These findings enable the development of an effective Treg-based treatment for human autoimmune and inflammatory diseases in which the driving antigens have been identified.

REFERENCES

(45) 1. Hogquist, K. A., Baldwin, T. A., and Jameson, S. C. 2005. Central tolerance: learning self-control in the thymus. Nat. Rev. Immunol. 5: 772-782 2. Goodnow, C. C., Sprent, J., Fazekas de St Groth, B., and Vinuesa, C. G. 2005. Cellular and genetic mechanisms of self tolerance and autoimmunity. Nature. 435: 590-597. 3. Jonuleit, H., and Schmitt, E. 2003. The regulatory T cell family: distinct subsets and their interrelations. J. Immunol. 171: 6323-6327. 4. Cohen, I. R. 2001. Tending Adam's garden: evolving the cognitive immune self. Academic Press. New York, USA. 266 pp. 5. Viglietta, V., Baecher-Allan, C., Weiner, H. L., and Hafler, D. A. 2004. Loss of functional suppression by CD4+CD25+ regulatory T cells in patients with multiple sclerosis. J. Exp. Med. 199: 971-979. 6. von Herrath, M. G., and Harrison, L. C. 2003. Antigen-induced regulatory T cells in autoimmunity. Nat. Rev. Immunol. 3: 223-32. 7. Fontenot, J. D., Gavin, M. A., and Rudensky, A. Y. 2003. Foxp3 programs the development and function of CD4+CD25+ regulatory T cells. Nat. Immunol. 4: 330-336. 8. Sakaguchi, S. 2005. Naturally arising Foxp3-expressing CD25+CD4+ regulatory T cells in immunological tolerance to self and non-self. Nat. Immunol. 6: 345-352. 9. Wan, Y. Y., and Flavell, R. A. 2006. The roles for cytokines in the generation and maintenance of regulatory T cells. Immunol. Rev. 212: 114-130. 10. Huber, S., and Schramm, C. 2006. TGF-beta and CD4+CD25+ regulatory T cells. Front. Biosci. 11: 1014-1023. 11. Shevach, E. M. 2009. Mechanisms of Foxp3+ T regulatory cell-mediated suppression. Immunity 30: 636-645. 12. Von Boehmer, H. 2005. Mechanisms of suppression by suppressor T cells. Nat. Immunol. 6: 338-344. 13. Gavin, M. A., Clarke, S. R., Negrou, E., Gallegos, A., and Rudensky, A. 2002. Homeostasis and anergy of CD4(+)CD25(+) suppressor T cells in vivo. Nat. Immunol. 3: 33-41. 14. Walker, L. S., Chodos, A., Eggena, M., Dooms, H., and Abbas, A. K. 2003. Antigen-dependent proliferation of CD4+ CD25+ regulatory T cells in vivo. J. Exp. Med. 198: 249-258. 15. Liang, S., et al. 2005. Conversion of CD4+ CD25 cells into CD4+ CD25+ regulatory T cells in vivo requires B7 costimulation, but not the thymus. J. Exp. Med. 201: 127-137. 16. Knoechel, B., Lohr, J., Kahn, E., Bluestone, J. A., and Abbas, A. K. 2005. Sequential development of interleukin 2-dependent effector and regulatory T cells in response to endogenous systemic antigen. J. Exp. Med. 202: 1375-1386. 17. Kretschmer, K., et al. 2005. Inducing and expanding regulatory T cell populations by foreign antigen. Nat. Immunol. 6: 1219-1227. 18. Quintana, F. J., et al. 2008. Control of T(reg) and T(H)17 cell differentiation by the aryl hydrocarbon receptor. Nature. 453: 65-71. 19. Crispe, I. N. 2003. Hepatic T cells and liver tolerance. Nat. Rev. Immunol. 3: 51-62. 20. Limmer, A., et al. 1998. Failure to induce organ-specific autoimmunity by breaking of tolerance: importance of the microenvironment. Eur. J. Immunol. 28: 2395-2406. 21. Knolle, P. A., et al. 1999. Induction of cytokine production in nave CD4 (+) T cells by antigen-presenting murine liver sinusoidal endothelial cells but failure to induce differentiation toward Th1 cells. Gastroenterology 116: 1428-1440. 22. Limmer, A., et al. 2000. Efficient presentation of exogenous antigen by liver endothelial cells to CD8+ T cells results in antigen-specific T cell tolerance. Nat. Med. 6: 1348-1354. 23. Caine, R. Y., et al. 1969. Induction of immunological tolerance by porcine liver allografts.

(46) Nature 223: 472-476. 24. Cantor, H. M., and Dumont, A. E. 1967. Hepatic suppression of sensitization to antigen absorbed into the portal system. Nature 215: 744-745. 25. Lath, S., et al. 2008. Ectopic expression of neural autoantigen in mouse liver suppresses experimental autoimmune neuroinflammation by inducing antigen-specific Tregs. J Clin. Invest. 118: 3403-3410. 26. Yu, W. W., Falkner, J. C., Yavuz, C. T., and Colvin, V. L. 2004. Synthesis of monodisperse iron oxide nanocrystals by thermal decomposition of iron carboxylate salts. Chem. Commun. 20: 2306-2307. 27. Shtykova, E. V., et al. 2008. Hydrophilic monodisperse magnetic nanoparticles protected by an amphiphilic alternating copolymer, J. Phys. Chem. C 112: 16809-16817. 28. Wiegard, C., Frenzel, C., Herkel, J., Kallen, K. J., Schmitt, E., and Lohse, A. W. 2005. Murine liver antigen presenting cells control suppressor activity of CD4+CD25+ regulatory T cells. Hepatology 42:193-199. 29. Schrage, A., et al. 2008. Murine CD146 is widely expressed on endothelial cells and is recognized by the monoclonal antibody ME-9F1. Histochem. Cell. Biol. 129:441-451. 30. Wiegard, C., et al. 2007. Defective T helper 1 response by hepatocyte-stimulated CD4 T cells impairs anti-viral CD8 response and viral clearance. Gastroenterology 133: 2010-2018. 31. Huberman, A., and Perez C. 2002. Nonheme iron determination. Anal. Biochem. 307: 375-378. 32. Wu, X. et al. 2002. Immunofluorescent labeling of cancer marker Her2 and other cellular targets with semiconductor quantum dots. Nat. Biotechnol. 21: 41-46. 33. Nguyen et al., 2004, Gene Ther. 11 (Suppl 1); S76-S84 34. Carambia et al. 2010, J. Hepatol 52, S59-S182. 35. Minchin 2008. Nature Nanotechnol. 3: 12-13. 36. Giri et al. 2011, Acta Biophys Sin, 1-7. 37. Shen et al. 2011, Int. J. Nanomedicine 6: 1229-1235. 38. Wu et al. Mol. Pharm. 6:1506-1517. 39. Bruns et al., Nat. Nanotechnol. 2009, 4:193-201. 40. H. L. Weiner, A. P. da Cunha, F. Quintana, H. Wu. Immunol Rev 241, 241-259 (2011). 41. J. Andersson, et al. J Exp Med 205, 1975-1981 (2008). 42. D. Q. Tran, et al. Proc Natl Acad Sci USA 106, 13445-13450 (2009). 43. T. Pellegrino, et al. Nano Letters 4, 703-707 (2004). 44. A. Bartelt, et al. Nat Med 17, 200-205 (2011). 45. G. Y. Liu, et al. Immunity 3, 407-415 (1995). 46. S. Huber, et al. J Immunol 173, 6526-6531 (2004). 47. J. Shimizu, S. Yamazaki, S. Sakaguchi. J Immunol 163, 5211-5218 (1999).