Polyoma virus JC peptides and proteins in vaccination and diagnostic applications

Abstract

The present invention relates to the field of vaccination or immunization, in particular therapeutic vaccination, and diagnosis. Pharmaceutical compositions and kits capable of eliciting a protective immune response against polyoma virus JC (JCV) are disclosed, which may be used e.g., for therapy or for prevention of progressive multifocal leukoencephalopathy (PML) and/or progressive multifocal leukoencephalopathy-immune reconstitution inflammatory syndrome (PML-IRIS). Individuals in danger of such PML or PML-IRIS may, e.g., be immuno-compromised or immunosuppressed patients or patients having an autoimmune disease eligible for immunosuppressive treatment. The invention also relates to compositions comprising at least one CD4+ epitope of a JCV protein and to therapeutic, prophylactic and diagnostic uses thereof.

Claims

1. A method of treating progressive multifocal leukoencephalopathy (PML) or progressive multifocal leukoencephalopathy-immune reconstitution inflammatory syndrome (PML-IRIS) in a subject, the method comprising: providing a VP1 protein of polyoma virus JC (JCV), administering the VP1 protein of JCV to the subject, and administering an TLR-7 agonist or an TLR-8 agonist, thereby treating progressive multifocal leukoencephalopathy (PML) or progressive multifocal leukoencephalopathy-immune reconstitution inflammatory syndrome (PML-IRIS) in the subject.

2. The method as recited in claim 1, wherein only the TLR-7 agonist is administered.

3. The method as recited in claim 1, further comprising: administering imiquimod.

4. The method as recited in claim 1, further comprising: administering an adjuvant selected from the group consisting of MF59, aluminium hydroxide, calcium phosphate gel, lipopolysaccharides, oligonucleotide sequences with CpG motifs, stearyl tyrosine, DTP-GDP, DTP-DPP, threonyl-MDP, 7-allyl-8-oxoguanosine, glycolipid bay R1005, multi-antigen peptide system, polymerized haptenic peptides, bacterial extracts, and vit-A.

5. The method as recited in claim 1, wherein the subject has a congenital immunodeficiency, an acquired immunodeficiency resulting from a disease or pathological condition, or an acquired immunodeficiency resulting from a therapeutic intervention.

6. The method as recited in claim 5, further comprising: treating with an immunosuppressive antibody.

7. The method as recited in claim 6, wherein the immunosuppressive antibody is selected from the group consisting of natalizumab, efalizumab, rituximab, ocrelizumab and alemtuzumab.

8. The method as recited in claim 5, wherein the subject is afflicted with an autoimmune disease.

9. The method as recited in claim 8, wherein the autoimmune disease is multiple sclerosis.

10. The method as recited in claim 9, wherein the subject is to be treated with the antibody natalizumab.

11. The method as recited in claim 5, wherein the subject has a congenital immunodeficiency selected from the group consisting of idiopathic CD4+ lymphopenia and Hyper-IgE-Syndrome.

12. The method as recited in claim 5, wherein the subject has an acquired immunodeficiency resulting from a disease or pathological condition selected from the group consisting of AIDS, leukemia, lymphoma, multiple myeloma, infection with hepatitis virus B, and infection with hepatitis C.

13. The method as recited in claim 5, wherein the subject has an acquired immunodeficiency resulting from a therapeutic intervention, wherein the therapeutic intervention is selected from the group consisting of chemotherapy, radiation and immunosuppressive treatment.

Description

LEGENDS

(1) FIG. 1: a) PHA-expanded bulk mononuclear cell populations from the brain biopsy (left panel), CSF (second panel from left) and PBMC (third panel from left) as well as unmanipulated PBMC (right panel) were tested against JCV VP1/VLP protein and tetanus toxoid protein (TTx). Results show the mean SI±SEM. Note the different scales for the y-axis. b) Ex-vivo quantification of Th1-, Th2-, Th17- and Th1-2 cells in PHA-expanded CD4.sup.+ T cells from brain biopsy (left panels), CSF (second panels from left), PBMCs (third panels from left) and in unmanipulated PBMCs (right panels). Numbers represent the percentage of positive cells. Th1 were identified as CD4.sup.+ IFN-gamma.sup.+ IL-17A.sup.−/IL4.sup.−; Th17 cells as CD4.sup.+ IL-17A.sup.+ IFN-gamma.sup.−; Th2 as CD4.sup.+ IL-4.sup.+ IFN-gamma.sup.−, and Th1-2 as CD4.sup.+ IFN-gamma.sup.+ IL-4.sup.+.

(2) FIG. 2: a) Proliferative response of brain-derived PHA-expanded cells against 204 overlapping 15-mer peptides spanning all open reading frames of JCV (covering Agno, VP1, VP2, VP3, Large-T, and small-T proteins) and organized in 41 pools of 5 peptides each. Results show the mean SI±SEM. The different patterns of the bars correspond to the different open reading frames. Schematic representation of the 5 open reading frames in the JCV genome (upper right hand figure). b) Proliferative response of brain-derived bulk mononuclear cell populations against individual JCV peptides. Results show the mean SI±SEM. Note the different scales of the y-axes in panels a and b. c) Precursor frequency of T cells specific of the 5 JCV peptides inducing the strongest proliferative responses in PHA-expanded cells from brain biopsy. d) Percentage of CD8+ T cells that bind HLA-A*02:01-VP1.sub.36 tetramers (middle graph) and HLA-A*02:01-VP1.sub.100 tetramers (lower graph).

(3) FIG. 3: a) Doughnut representing the frequency of each individual TCC in the brain biopsy. b) Proliferative response of TCC against 64 individual VP1 peptides. Results show the mean SI±SEM. c) Schematic representation of the immunodominant peptides identified for the brain-derived bulk population (upper graph, show the mean SI±SEM) and for the different TCC (lower graph the different patterns correspond to the different TCC and each single cell growing culture is represented by a square).

(4) FIG. 4: a) Representative flow cytometry analysis of intracellular IFN-gamma and IL-4 production by a Th1-2 (upper plot) and a Th1 (lower plot) VP1/VLP-specific CD4.sup.+ T cell clone. b) The dot-plot represents the percentage of VP1-specific, brain-derived single cell cultures with Th1-2 and Th1 phenotype by intracellular cytokine staining Each dot corresponds to one of the 21 single cell cultures analyzed. The doughnut represents the functional phenotype of each TCC. c) ELISA detection of IFN-gamma and IL-4 production in culture supernatants of Th1-2 TCC (n=5, black bars) and Th1 TCC (n=6, white bars) 72 h after stimulation with PHA. Results show the mean±SEM. d) RT-PCR analysis for transcription factors Gata3 and t-bet of Th1-2 TCC (n=5, black bars) and Th1 TCC (n=6, white bars). Values are relative expression compared to brain-derived PHA-expanded cells (calibrator=1). Results show the mean±SEM.

(5) FIG. 5 Treatment scheme for immunisation of an immunocompromised subject with VP1 M=month, D=day, W=week, MRI=Magnetic Resonance Imaging. Adjuvant (imiquimod) was administered directly after administration of VP1.

(6) FIG. 6 Development of VP1 immune response during treatment VLP1=virus like particle1 composed of VP1 protein, TT=Tetanus toxoid

(7) FIG. 7 Course of Treatment FIG. 7A shows viral load and mean SI after VP1 stimulation, and FIG. 7B shows leukocyte counts in the CSF. Time points of administration of IL-7 and VP1 are identical and correspond to the scheme shown in FIG. 5.

(8) FIG. 8 Characterization of VP1-specific T-cells Only CD4+ cells proliferate after VP1 stimulus (6 weeks after immunization)

(9) FIG. 9 Proliferating CD4 cells are activated memory cells

(10) FIG. 10 CD4+ T cells activated by VP1 express a high background of IL-4 and produce IFN-gamma after VP1 stimulation. CD4+ T cells were stimulated with antigen on day 0. On day 6, they were restimulated with antigen, 1 hour later, secretion was blocked with Brefeldin A and after 15 h, an intracellular cytokine staining was performed and analysed by FACS.

(11) FIG. 11 3H-thymidine incorporation assay At the time points indicated, peripheral blood mononuclear cells (PBMC) were obtained from the patients and freshly seeded (1×10e5 cells/well) with antigen (VP1/VLP). After 7 days of incubation, 3H-thymidine incorporation was measured. A stimulation index of >2 is considered a positive response. SIs are shown on the y-axis of the graph.

EXAMPLE 1—IDENTIFICATION OF IMMUNODOMINANT CD4+ EPITOPES

(12) Material and Methods

(13) Patients

(14) HLA-class II types: DRB1*13:01, −*16:01; DRB3*02:02; DRB5*02:02; DQA1*01:02, −*01:03; DQB1*05:02, −*06:03 (patient 1); and DRB1*11:03, −*15:01; DRB3*02:02; DRB5*01:01; DQA1*01:02, −*05:XX (X indicating not typed to the exact subtype); DQB1*03:01, −*06:02 (patient 2).

(15) Neuropathology

(16) Small tissue fragments of a total volume of approximately 0.1 ml, were obtained by open biopsy. Following fixation in buffered formalin for 2 hours, tissue was embedded in paraffin. Microtome sections of 4 μm were stained with hematoxiline-eosin (H&E), van Gieson's trichrome, PAS, Turnbull's stain for siderin and Luxol. Immunohistochemical staining was performed on an automated Ventana HX IHC system, benchmark (Ventana-Roche Medical systems, Tucson, Ariz., USA) following the manufacturer's instructions using the following antibodies: anti-CD45/LCA (DAKO, Glostrup, Denmark; M701), anti-CD3 (DAKO; M1580), anti-CD45R0 (DAKO; M 0742), anti-CD20 (DAKO; M0755), anti-CD79a (DAKO; M7050), anti-CD68 (Immunotech/Beckmann-Coulter, Krefeld, Germany, 2164), anti-HLA-DR (DAKO; M775), anti-NF (Zymed/Invitrogen, Darmstadt, Germany, 80742971), anti-GFAP (DAKO; Z334) and anti-p53 (DAKO; M7001).

(17) Brain Tissue Processing and Expansion of Brain-Derived, CSF-Derived and Peripheral Blood Mononuclear Cells

(18) A biopsy of approximately 0.033 ml was cut into small pieces and disrupted by incubation in a solution containing 1 mg/ml Collagenase A (Roche Diagnostics, Penzberg, Germany) and 0.1 mg/ml DNAse I (Roche) at 37° C. in a water bath for 45 min. The resulting cell suspension was washed three times, and brain-derived mononuclear cells were separated using a Percoll density gradient centrifugation (GE Healthcare, Munich, Germany). Cells were resuspended in a 30% Percoll solution and carefully underlayered with a 78% Percoll solution. After centrifugation brain-derived mononuclear cells were gathered from the interface of the gradient.

(19) CSF-derived mononuclear cells were obtained directly from a diagnostic spinal tap, and peripheral blood mononuclear cells were separated by Ficoll density gradient centrifugation (PAA, Pasching, Austria).

(20) Brain-, CSF- and peripheral blood-derived mononuclear cells were expanded in 96-well U-botton microtiter plates by seeding 2000 cells per well together with 2×10.sup.5 non-autologous, irradiated PBMC (3,000 rad) and 1 μg/ml of PHA-L (Sigma, St Louis, Mo.). Medium consisted of RPMI (PAA) containing 100 U/ml penicillin/streptomycin (PAA), 50 μg/ml gentamicin (BioWhittaker, Cambrex), 2 mM L-glutamine (GIBCO, Invitrogen) and 5% heat-decomplemented human serum (PAA). After 24 h, 20 U/ml of human recombinant IL-2 (hrIL-2, Tecin, Roche Diagnostics) were added and additional hrIL2 was added every 3-4 days. After two weeks cells were pooled and analyzed, cryopreserved or restimulated again with 1 μg/ml PHA, 20 U/ml hrIL-2 and allogeneic irradiated PBMC.

(21) Flow Cytometry Analysis of Brain-Derived Mononuclear Cells

(22) Brain-derived mononuclear cells directly from brain digestion were stained with the following antibodies for surface markers: CD45 (AmCyan, 2D1, BD Pharmingen, San Diego, USA), CD56 (Alexa 488, B159, BD Pharmingen), CD3 (PeCy7, UCHT1, eBioscience, San Diego, USA), CD4 (APC, RPA-T4, eBioscience), CD8 (PB, DK25, Dako, Glostrup, Denmark), CD45RO (FITC, UCHL1, eBioscience), CD19 (FITC, HIB19, BD Pharmingen), CD38 (APC, HIT2, BD Pharmingen), and CD27 (APC-Alexa 750, CLB-27/1, Invitrogen). Analysis was performed on a3 LSRII (BD Biosciences, Heidelberg, Germany) flow cytometer.

(23) Proteins and Peptides

(24) For the identification of JCV-specific T cells, 204 (13-16 mer) peptides covering the entire JC viral proteome were applied. Peptides were synthesized and provided by pe (peptides and elephants GmbH, Potsdam, Germany). These 204 peptides overlap by 5 amino acids and include 35 common single amino acid mutations. To account for amino acid variations, that occur among the different JCV genotypes and strains, amino acid sequences of each JCV encoded protein including Agno, VP1, VP2, VP3, Large T antigen and small t antigen from all 479 JCV genomic sequences available in GenBank (by March 2008) were aligned and those polymorphisms, which were prevalent in more than 1% of the all retrieved sequences, were defined as common mutations.

(25) In order to determine which individual peptides are recognized by CNS-derived T cells, a two-dimensional seeding scheme was applied. Peptides were arranged in a set of 82 pools, where each pool contains 5 different peptides. By the combination of different peptides in each well according to a rectangular matrix and each individual peptide appearing in exactly two pools, in which the residual peptides differ, immunogenic candidate peptides could be identified at the intersections of the positive pools.

(26) JCV VP1 protein forms virus-like (VLP) particles, and VP1 and VLP are therefore used as interchangeable terms. VP1 protein forming VLP (VP1/VLP) was generated by the Life Science Inkubator, Bonn, Germany, as previously described (Goldmann et al., 1999). 20 mer myelin peptides with an overlap of 10 amino acids and covering MBP (16 peptides), MOG (25 peptides) and PLP (27 peptides) were synthesized and provided by PEPScreen, Custom Peptide Libraries, SIGMA. Tetanus toxoid (TTx) (Novartis Behring, Marburg, Germany) was used as positive control.

(27) Proliferative Assays

(28) Recognition of JCV Peptides, VP1/VLP and TTx was tested by seeding duplicates in 96-well U-botton microtiter plates 2-2.5×10.sup.4 brain-derived, CSF-derived or peripheral blood-derived PHA-expanded cells per well and 1×10.sup.5 autologous irradiated PBMC with or without peptides for 72 hours. Unmanipulated PBMC were tested at 2×10.sup.5 cells/well in a 7-day primary proliferation. In addition to TTx, PHA-L stimulation was added as positive control. All JCV peptides were either tested in pools or as individual peptides at a final concentration of 2 μM per peptide for peptides in pools and at a concentration of 10 μM for individual peptides. VP1/VLP was tested at 2 μg/ml, Tetanus toxoid (TTx) at 5 μg/ml and PHA at 1 μg/ml. Proliferation was measured by .sup.3H-thymidine (Hartmann Analytic, Braunschweig, Germany) incorporation in a scintillation beta counter (Wallac 1450, PerkinElmer, Rodgau-Jürgesheim, Germany). The stimulatory index (SI) was calculated as SI=Mean cpm (counts per minute)(peptide)/Mean cpm (background). Responses were considered as positive when SI>3, cpm>1000 and at least three standard deviations (SD) above average background cpm. Myelin peptides were tested as individual peptides at 5 μM as described above.

(29) Generation of Brain-Derived VP1/VLP-Specific T Cell Clones

(30) 2.5×10.sup.4 brain-derived PHA-expanded cells were seeded in 96-well U-botton microtiter plates with 1×10.sup.5 autologous irradiated PBMC with or without VP1/VLP protein. After 48 hours of culture, plates were split into mother and daughter plates. Proliferation was measured in daughter plates by methyl-.sup.3H-thymidine incorporation. VP1/VLP-responsive cultures were identified in mother plates, and IL-2 was added every 3-4 days until day 12. T cell clones (TCC) were established from positive cultures by seeding cells from VP1/VLP-responsive wells under limiting dilution conditions at 0.3 and 1 cell/well in 96-well U-botton microtiter plates, and addition of 2×10.sup.5 allogeneic, irradiated PBMC and 1 μg/ml of PHA-L in complete RPMI. After 24 h, 20 U/ml of human recombinant IL-2 were added. VP1/VLP specificity was then confirmed seeding 2.5×10.sup.4 cells from growing colonies with autologous irradiated PBMC with or without VP1/VLP protein for 72 h. Specific cultures were restimulated every two weeks with 1 μg/ml PHA-L, 20 U/ml hrIL-2 and allogeneic irradiated PBMC, and hrIL2 was added every 3-4 days.

(31) TCR Analysis

(32) TCR Vβ chain expression was assessed in PHA-expanded cells and T cell clones by 22 anti TCRBV monoclonal antibodies (Immunotech, Marseille, France, (Muraro et al., 2000)) in combination with CD4 (APC, eBioscience) and CD8 (PB, PB, DakoCytomation, Denmark).

(33) Determination of Precursors Frequency in CNS-Derived Mononuclear Cells

(34) Frequencies of VP1/VLP-specific cells were determined by limiting dilution. 20, 200, 2.000 or 20.000 brain-derived PHA-expanded cells were seeded in quadruplicates in 96-well U-botton microtiter plates with 1×10.sup.5 autologous irradiated PBMC with or without VP1/VLP protein. After 72 hours, proliferation was measured by methyl-.sup.3H-thymidine incorporation. Frequencies were calculated as previously described (Taswell, 1981). Observed data were: r, the number of negatively responding cultures or wells of each dose i; n, the total number of wells per dose i, and X, the number of cells in the dose i. Calculated data was: pi=ri/ni, the fraction of negatively responding cultures of each dose i. The frequency was calculated using the following formula: f=−(ln pi)/λi.

(35) Cytokine Production

(36) For intracellular cytokine staining, PHA-expanded cells and TCC were analyzed 12 days after last restimulation. Cells were stimulated with PMA (50 ng/ml, Sigma) and ionomycin (1 μg/ml, Sigma) in the presence of Brefeldin A (10 μg/ml, eBioscience) for 5 h. Next, cells were stained with LIVE/DEAD® Fixable Dead Cell Stain Kit (AmCyan, Molecular Probes, Invitrogen), fixed and permeabilized with the corresponding buffers (eBioscience), and stained for CD3 (PE, DakoCytomation, Denmark), CD8 (PB, DakoCytomation, Denmark), IFNgamma (FITC, BDPharmingen), IL-4 (PE-Cy7, eBioscience) and IL-17A (Alexa Fluor®-647, eBioscience) at room temperature. IFN-gamma-, IL-4- and IL-2 levels were also determined by ELISA following the manufacturer's protocol (Biosource, Camarillo, Calif.) in culture supernatants of PHA-expanded cells and in TCC 72 hours after stimulation with PHA or VP1/VLP.

(37) Quantification of mRNA Expression Levels by RT-PCR

(38) For mRNA gene expression assays, the primer and probe sets (Tbet, Hs00203436_ml and Gata3, Hs00231122_ml) were purchased from Applied Biosystems (Foster City, Calif.). 18S rRNA was used as endogenous control, and the relative gene expression was calculated by the ΔΔCt method using brain-derived PHA-expanded cells as calibrator.

(39) ELISA for VP1/VLP-Specific Antibodies

(40) The titer of VP1/VLP-specific immunoglobulin G antibodies in CSF and sera from both IRIS-PML patients was determined as described previously (Weber et al., 1997). Briefly, ELISA plates were coated with 100 ml VP1-VLP (1 mg/ml) and incubated with serial dilutions of CSF or sera. Human IgG was captured by a biotin conjugated anti-human Fc antibody (eBioscience) and detected by an avidin horseradish peroxidase (eBioscience). Antibody titers in CSF as well as serum were adjusted to the total amount of IgG in the particular compartment. Results were expressed as arbitrary units, which were standardized using always the same human serum as standard.

(41) HLA-A*0201/JCV VP1.sub.36 and VP1.sub.100 Tetramers and Tetramer Staining

(42) HLA-A*02:01 tetrameric complexes were synthesized as previously described. Briefly HLA-A*02:01, β2 microglubluin and epitope peptide were refolded and isolated using size eclusion chromatography. Site-specific biotinylation was achieved through addition of the BirA target sequence to the last C terminal extracellular domain of the HLA-A*0201 molecule. Tetrameric complexes were generated using Extravidin-PE (Sigma). PHA-expanded brain-infiltrating cells were stimulated with anti-CD2/CD3/CD28 MACs beads (Miltenyi Biotec, Auburn, Calif.) and at day 5 after stimulation cells were washed and resuspended to a concentration of 5×10.sup.6 cells/ml. 100 μl were stained with 3 μl of PE-coupled tetrameric HLA-A*02:01/JCV VP1.sub.36 or HLA-A*02:01/JCV VP1.sub.100. After 30 min incubation at 37° C. the cells were washed and stained with anti-CD3 (PB, eBiolegend, San Diego, Calif.) and anti-CD8 (FITC, Dako) for additional 30 min on ice. Then cells were washed and fixed with 0.5% paraformaldehyde before analysis by flow cytometry.

(43) Results

(44) Two Cases of Natalizumab-Associated PML-IRIS

(45) Two male patients of 41 and 43 years with relapsing-remitting MS (RR-MS) presented July 2009 and January 2010 respectively with clinical signs (visual field defect in patient 1; monoparesis in patient 2) and imaging findings suspicious of PML after 28- and 40 months respectively of natalizumab treatment. Natalizumab was stopped immediately, and several rounds of plasmapheresis performed. Both patients subsequently developed PML-IRIS with patchy or band-like areas of contrast enhancement on MRI (FIG. 1a) and worsened clinical findings of complete loss of vision in patient 1, and hemiplegia, hemianopia and neuropsychological deterioration in patient 2. With respect to diagnostic workup, patient 1 was immediately diagnosed as PML based on CSF JCV viral load, although it was low. Diagnosis in patient 2 was more complicated with repeatedly negative PCR results for CSF JCV viral load until the third testing was positive just above threshold levels; 12 copies; threshold 10 copies in the NIH reference laboratory). In contrast to the low or borderline JCV CSF viral loads, antibody testing for JCV major capsid protein (VP1/VLP)-specific antibodies in serum and CSF, which was established during this study, revealed strong intrathecal antibody response with 95-180-fold higher VP1/VLP-specific antibody titers in the CSF compared to serum after adjusting total IgG concentrations to the same levels. Hence, different from the PCR testing for viral DNA, the intrathecal antibody response left no doubt of CNS infection by JCV at the time of PML-IRIS. The analysis of the IgG subclasses in patient 2 demonstrated that intrathecal antibodies are mainly IgG1 and IgG3. These data indicate a strong JCV-specific humoral immune response that is confined to the CNS compartment and directed primarily against the major structural JCV protein VP1/VLP. Whether minor components of the antibody response target other JCV proteins remains to be studied.

(46) Due to the above difficulties to diagnose PML, patient 2 underwent a diagnostic brain biopsy to confirm or refute PML. Neuropathological examination failed to show the typical signs of PML, i.e. nuclear inclusions in hyperchromatic oligodendrocytes and bizarre astrocytes, but rather a paucity of CNS cells and massive perivascular and parenchymal lymphomononuclear infiltrates, reactive gliosis with stellate astrocytes and predominance of diffuse and destructive parenchymal infiltrates of foamy macrophages. The majority of cells stained positive for HLA-DR, which is usually exclusively found on activated microglia and absent in normal brain tissue. T cells and B cells were present in the infiltrate, and a high proportion of the latter stained positive for the plasma cell marker CD138. Part of the biopsy tissue was processed, and CNS-derived mononuclear cells were also characterized by flow cytometry. 96.5% of cells expressed the pan hematopoietic cell marker CD45 (not shown) and among them 42.4% expressed the pan T cell marker CD3.sup.+. Of these 24.1% were CD8.sup.+ and 70.4% CD4.sup.+ T cells. Almost all of these cells expressed the memory marker CD45RO. 29% of CD45.sup.+ CNS-infiltrating cells expressed the B cell marker CD19, and among these 86.1% were positive for CD27/CD38, i.e. they were memory B cells/plasma cells. Accordingly, a diagnosis of inflammatory demyelinating disease rather than PML was made. Subsequent immunohistochemistry for JCV was negative, but sparse nuclear signals for JCV DNA were found by the second attempt of in situ hybridization (data not shown), which together with the low JCV viral load and strong intrathecal antibody response confirmed the initial suspicion of PML and pointed at IRIS rather than the underlying demyelinating disease as responsible for the neuropathological findings.

(47) Antigen Specificity of Brain-Infiltrating T Cells

(48) Next the antigen specificity and frequency of brain-infiltrating T cells were characterized. Brain-derived mononuclear cells were first expanded as bulk populations by an unbiased stimulus (PHA). While our culture conditions favored the expansion of CD4.sup.+ over CD8.sup.+ T cells the relative composition of CD4.sup.+ T cells remained stable as demonstrated by staining with monoclonal antibodies against T cell receptor (TCR) variable chains V131-V1322. Due to the almost threefold excess of memory CD4.sup.+ over CD8.sup.+ T cells at the time of brain biopsy, we focused our attention on CD4.sup.+ cells and assessed their specificity for JCV. For this purpose, expanded brain T cells were tested against recombinant JCV capsid protein VP1/VLP and against tetanus toxin protein (TTx) in proliferative assays. We directly compared brain-derived versus CSF- or peripheral blood-derived T cells as well as versus unmanipulated peripheral blood mononuclear cells. As shown in FIG. 1a, brain-derived T cells responded with a stimulation index (SI)>600 against VP1/VLP protein with no response against TTx. SIs against VP1/VLP and TTx in the CSF were 7 and 14 respectively, and in PHA-expanded PBMC the responses to VP1/VLP and TTx were negative and moderately positive (SI of 6.5) respectively. Unmanipulated PBMC showed a significantly stronger response to TTx compared to VP1/VLP in a 7 days primary proliferation assay.

(49) Functional Phenotype of Brain-Infiltrating CD4.sup.+ T Cells

(50) The inventors then examined if intracerebral CD4.sup.+ T cells belonged to one of the major T helper (Th) subtypes, Th1, Th2 or Th17 cells, based on their cytokine secretion pattern. Expanded bulk T cell populations from the brain, CSF and PBMC as well as unmanipulated PBMC were examined by intracellular cytokine staining against IFN-gamma, IL-4, and IL-17, the signature cytokines of Th1-, Th2- and Th17 cells. IL-17-producing cells were hardly detectable (FIG. 1b), while IFN-gamma-secreting cells made up between 46.1 and 53.2% in cells from the brain and CSF (FIG. 1b). When combining intracellular staining for IFN-gamma and IL-4, the situation was remarkably different. In the brain and CSF-derived population, Th1-, Th2- and bifunctional Th1-2 cells (secreting both IL-4 and IFN-gamma) were similar in frequency (FIG. 1b), while Th2 cells predominated in the peripheral blood-derived, PHA-expanded cells (FIG. 1b). 32.7% of brain-derived CD4.sup.+ T cells had a bifunctional Th1-2 phenotype.

(51) Fine Specificity and Frequency of Brain-Infiltrating T Cells

(52) To determine which specific JCV peptides are recognized by brain-infiltrating T cells, 204 15-mer peptides spanning all JCV proteins (Agno, VP1, VP2, VP3, Large-T, and small-T) were synthesized and arranged in a set of 82 pools, where each peptide appears twice, but in two different pools (see methods). Brain-derived T cells responded to multiple pools (FIG. 2a; pools 1-41). The inventors identified 15 immunogenic candidate peptides that were then tested individually and lead to the identification of 11 stimulatory peptides (peptides with SI>10) (FIG. 4b). The response was directed against peptides 4 (Agno.sub.25, the number denotes the first amino acid (aa) of the 15-mer peptide), 20 (VP1.sub.34), 23 (VP1.sub.54), 27-29 (VP1.sub.74)(all VP1.sub.74 peptides; 28 and 29 are variants of peptide 27 with single aa mutations), 72 (VP1.sub.310), 73 (VP1.sub.319), 76 (VP1.sub.335), 191 (LTAg.sub.668), and 195 (sTAg.sub.82) (peptides with SI>25 in bold). Thus, brain-derived T cells responded to several JCV proteins (Agno, VP1, LTAg, sTAg), however, by far the strongest against VP1 (6 peptides). That VP1 is the prime target is supported by an even stronger response against entire VP1/VLP protein (FIG. 1a) and by the higher precursor frequencies of VP1-specific T cells (between 1/294 and 1/714 T cells responding to peptides VP1.sub.34, VP1.sub.319 and VP1.sub.74) when compared to cells responding to Agno.sub.25 (1/14492) and LTAg.sub.668 (1/1449) (FIG. 2c). When the inventors examined PHA-expanded CSF- and peripheral blood-derived T cells, CSF only mounted weak responses against pool 39 and peptide LTAg.sub.668 contained in this pool, and PBMC were negative. Remarkably, peptide VP1.sub.34, the peptide that elicits the strongest response with respect to SI (FIG. 2b) and precursor frequency (FIG. 2c), contains the JCV epitope VP1.sub.36, one of the two epitopes together with VP1.sub.100 that are recognized by HLA-class I-restricted CD8.sup.+ T cells in the context of HLA-A*02:01 (Du Pasquier et al., 2003b). PML-IRIS patient 2 is HLA-A*02:01.sup.+ (HLA-class I and -class II types under methods), for this reason the inventors determined the frequency of CD8+ T cells specific of these two HLA-A*02:01 JCV epitopes in the PHA-expanded brain-infiltrating CD8+ T cells by tetramer staining 0.8% of PHA-expanded brain-infiltrating CD8.sup.+ T were specific for VP1.sub.36 and 0.6% for VP1.sub.100 (FIG. 4d). With respect to HLA-class II, patient 2 expresses the MS-associated HLA-DR haplotype DRB1*15:01 and DRB5*01:01. The JCV-specific CD4.sup.+ T cell response was largely restricted by DRB1*15:01/B5*01:01, when VP1/VLP protein was presented by APCs from a DRB1*15:01/B5*01:01 homozygous donor (not shown). These data demonstrate that, similar to intrathecal antibodies, the CD4.sup.+ T cell response is mainly directed against the major structural JCV protein VP1 and that peptide VP1.sub.34 contains an epitope for both virus-specific CD4.sup.+ and CD8.sup.+ T cells. Such a focus of CD4+ and CD8+ T cells on the same immunodominant epitope has previously been shown for an influenza nucleoprotein peptide (Carreno et al., 1992).

(53) Table 1 shows the immunodominant CD4+ T cell epitopes identified by the inventors.

(54) Since PML is characterized by oligodendrocyte lysis and release of myelin and since the patient suffers from MS, it was of interest to examine if brain-derived T cells responded to myelin proteins. PHA-expanded brain-derived T cells were tested against overlapping peptides spanning the major myelin proteins, myelin basic protein (MBP), proteolipid protein (PLP), and myelin oligodendrocyte glycoprotein (MOG), but none of the myelin peptides was recognized despite a strong response against JCV VP1/VLP protein.

(55) TABLE-US-00001 TABLE 1 List of immunodominant JCV epitopes.  Peptide Name  SEQ # (position) Sequence Length ID NO               24 VP1 (64-78) V1 ISDTFESDSPNRDML 15  4  25 VP1 (64-78) V2 ISDTFESDSPNFDML 15  5  26 VP1 (64-78) V3 ISDTFESDSPNKDML 15  6              30 VP1 (81-95) YSVARIPLPNLNEDL 15 10    32 VP1 (101-115) LMWEAVTLKTEVIGV 15 12  33 VP1 (108-122) V1 LKTEVIGVTSLMNVH 15 13  34 VP1 (108-122) V2 LKTEVIGVTTLMNVH 15 14  35 VP1 (108-122) V3 LKTEVIGVTALMNVH 15 15  36 VP1 (118-129) V1 LMNVHSNGQATH 12 16  37 VP1 (118-129) V2 LMNVHSNGQAAH 12 17  38 VP1 (118-129) V3 LMNVHSNGQASH 12 18  39 VP1 (123-137) V1 SNGQATHDNGAGKPV 15 19  40 VP1 (123-137) V2 SNGQAAHDNGAGKPV 15 20  41 VP1 (123-137) V3 SNGQASHDNGAGKPV 15 21  42 VP1 (133-147) AGKPVQGTSFHFFSV 15 22    44 VP1 (151-165) V1 ALELQGVLFNYRTKY 15 24  45 VP1 (151-165) V2 ALELQGVLFNYRTTY 15 25  46 VP1 (161-175) V1 YRTKYPDGTIFPKNA 15 26  47 VP1 (161-175) V2 YRTTYPDGTIFPKNA 15 27  48 VP1 (161-175) V3 YRTTYPHGTIFPKNA 15 28  49 VP1 (171-186) FPKNATVQSQVMNTEH 16 29  50 VP1 (182-196) MNTEHKAYLDKNKAY 15 30  51 VP1 (193-208) KNKAYPVECWVPDPTR 16 31  52 VP1 (203-217) PDPTRNENTRYFGTL 15 32  53 VP1 (210-224) NTRYFGTLTGGENVP 15 33  54 VP1 (220-234) V1 GENVPPVLHITNTAT 15 34  55 VP1 (220-234) V2 GENVPSVLHITNTAT 15 35  56 VP1 (220-234) V3 GENVPPVLHITKTAT 15 36      59 VP1 (239-253) V1 DEFGVGPLCKGDNLY 15 39  60 VP1 (239-253) V2 DEFGVRPLCKGDNLY 15 40  61 VP1 (249-263) GDNLYLSAVDVCGMF 15 41  62 VP1 (259-273) V1 VCGMFTNRSGSQQWR 15 42        78 VP2 (1-15) MGAALALLGDLVATV 15 46  79 VP2 (11-24) LVATVSEAAAATGF 14 47  80 VP2 (20-34) AATGFSVAEIAAGEA 15 48  81 VP2 (30-43) AAGEAAATIVEIA 14 49  82 VP2 (39-51) EVEIASLATVEGI 13 50  83 VP2 (47-61) TVEGITSTSEAIAAI 15 51  84 VP2 (57-71) AIAAIGLTPETYAVI 15 52  85 VP2 (67-81) TYAVITGAPGAVAGF 15 53  86 VP2 (77-88) AVAGFAALVQTV 12 54  87 VP2 (84-97) LVQTVTGGSAIAQL 14 55 153 LTAg (328-342) FADSKNQKSICQQAV 15 56 154 LTAg (338-351) CQQAVDTVAAKQRV 14 57 155 LTAg (347-361) AKQRVDSIHMTREEM 15 58 156 LTAg (357-370) TREEMLVERFNFLL 14 59 158 LTAg (376-390) IFGAHGNAVLEQYMA 15 60 159 LTAg (386-399) EQYMAGVAWIHCLL 14 61 161 LTAg (405-419) VIYDFLKCIVLNIPK 15 62 162 LTAg (415-429) LNIPKKRYWLFKGPI 15 63 168 LTAg (472-486) VFEDVKGTGAESRDL 15 64 169 LTAg (482-495) ESRDLPSGHGISNL 14 65 170 LTAg (491-506) GISNLDCLRDYLDGSV 16 66 171 LTAg (500-514) DYLDGSVKVNLERKH 15 67 172 LTAg (508-522) VNLERKHQNKRTQVF 15 68 200 StAg (123-137) RKFLRSSPLVWIDCY 15 72 201 StAg (133-146) WIDCYCFDCFRQWF 14 73 202 StAg (142-156) FRQWFGCDLTQEALH 15 74 203 StAg (149-162) DLTQEALHCWEKVL 14 75 204 StAg (158-172) WEKVLGDTPYRDLKL 15 76 271 VP1 L55F (54-68) HFRGFSKSISISDTF 15 77 272 VP1 S269Y (259-273) VCGMFTNRSGYQQWR 15 78 273 VP1 K60N (54-68) HLRGFSNSISISDTF 15 79 274 VP1 D66H (54-68) HLRGFSKSISISHTF 15 80 275 VP1 D66H (64-78) ISHTFESDSPNRDML 15 81 276 VP1 V223I (220-234) GENIPPVLHITNTAT 15 82 277 VP1 N265D (259-273) VCGMFTDRSGSQQWR 15 83 278 VP1 S267F (259-273) VCGMFTNRFGSQQWR 15 84 279 VP1 Q271H (259-273) VCGMFTNRSGSQHWR 15 85 280 VP1 S61L (54-68) HLRGFSKLISISDTF 15 86 281 VP1 K60E (54-68) HLRGFSESISISDTF 15 87 282 VP1 N265H (259-273) VCGMFTHRSGSQQWR 15 88 283 VP1 N265T (259-273) VCGMFTTRSGSQQWR 15 89 284 VP1 S267Y (259-273) VCGMFTNRYGSQQWR 15 90 285 VP1 S267L (259-273) VCGMFTNRLGSQQWR 15 91 286 VP1 S269C (259-273) VCGMFTNRSGCQQWR 15 92 The amino acid sequence and length of each JCV peptide is shown. The position of the respective peptide is related to the respective protein of the reference JCV genome NC_001699. Peptides with amino acid mutations are designated as variants (V1, V2 and so forth). Peptides recognized by T cell clones isolated from brain only are marked by bold print. Other peptides were recognized by peripheral T cell clones.
Fine Specificity and Functional Phenotype of JCV-Specific CD4.sup.+ T Cell Clones

(56) Prior data had shown that CD4+ differentiate into certain T helper phenotypes such as Th1 cells (IFN-gamma producers), Th17 cells secreting IL-17, Th2 cells expressing the signature cytokine IL-4, or T regulatory cells based on the expression of certain transcription factors (Zhu et al., 2010). The differentiation into Th1 or Th2 cells is considered mutually exclusive and controlled by the transcription factors T-bet (Th1) and Gata-3 (Th2) (Zhu et al., 2010). Based on these data, our finding of committed memory cells with a bifunctional (Th1-2) phenotype was highly unexpected, and we therefore established VP1/VLP-specific T cell clones to examine this point at the clonal level. VP1/VLP-specific TCC were generated as described in material and methods. Initially 21 VP1/VLP-specific single cell-derived cultures were generated by limiting dilution and characterized for TCRV beta expression, functional phenotype and fine specificity (Table 2). This characterization allowed the identification of 11 presumed different TCC. The number of single cell growing cultures corresponding to each TCC gives an idea about the frequency of each TCC in the brain infiltrate (FIG. 3a). TCC-4 was most abundant and represented by 5 colonies emerging from single growing wells, followed by TCC-2 with 3 single cell growing cultures, TCC-1, -8, -9 and -10 with 2 single cell growing cultures and finally TCC-3, -5, -6, -7 and -11 with only 1 single cell growing culture. The fine specificity of the 11 TCC is summarized in FIG. 5b. Each TCC was tested against 64 15-mer peptides spanning VP1 protein and lead to the identification of the following stimulatory peptides: VP1.sub.34 (recognized by TCC-1 and -2), VP1.sub.54 (recognized by TCC-3), VP1.sub.74 (all VP1.sub.74 peptides, recognized by TCC-4), VP1.sub.91 (recognized by TCC-5), VP1.sub.143 (recognized by TCC-6), VP1.sub.229 (recognized by TCC-7), VP1.sub.319 (recognized by TCC-8 and -9) and VP1.sub.335 (recognized by TCC-10 and -11). Taking into account both the number of different TCC recognizing a specific peptide and the frequency of each TCC in the brain infiltrate the immunodominant peptides recognized by VP1 specific brain infiltrating TCC were VP1.sub.34, VP1.sub.74, VP1.sub.319 and VP1.sub.335 confirming the fine specificity obtained using the brain-derived bulk cell population (FIG. 3c). Intracellular cytokine staining of these TCC revealed Th1-2 and Th1 phenotypes (FIG. 4a). 5 TCC representing the 57% of the brain-derived VP1 specific single cell growing cultures showed a Th1-2- and 6 TCC representing the 43% of the brain-derived VP1-specific single cell growing cultures a Th1 phenotype (FIG. 4b). Specificity and functional phenotype did not correlate. For three of the immunodominant peptides (VP1.sub.34, VP1.sub.319 and VP1.sub.335) we found TCC with both phenotypes. The T helper phenotype of Th1-2 and Th1 TCC was confirmed my measuring IL-4 and IFN-gamma protein secretion by ELISA and by determining the expression of mRNAs of the transcription factors Gata3 and T-bet. Th1-2 TCC (n=5) secreted IL-4 in addition to IFN-gamma, while IL-4 secretion was barely detectable in Th1 TCC (n=6) (FIG. 6c). Cytokine secretion profiles were not due to stimulation with PHA, since the same stable patterns were observed after specific stimulation with VP1/VLP protein by intracellular cytokine staining or ELISA measurements from culture supernatants. Transcription factor expression confirmed the phenotype of the TCC. Th1-2 TCC (n=5) expressed mRNA for Gata3 and T-bet, while Th1 TCC (n=6) only expressed t-bet (FIG. 6d).

(57) Discussion

(58) The viral etiology of PML has been shown almost 40 years ago, but still relatively little is known about the immune mechanisms that control JCV infection. CD8.sup.+ JCV-specific cytotoxic T cells have been related to recovery from PML (Du Pasquier et al., 2004a; Koralnik et al., 2002), and two viral epitopes have been identified in HLA-A*02:01-positive individuals (Du Pasquier et al., 2004a; Du Pasquier et al., 2004b). In contrast, limited information is available on the fine specificity and characteristics of JCV-specific CD4 T cells in PML and even less in PML-IRIS (Jilek et al.). The virus-specific T cell response at the site of infection, i.e. the CNS parenchyma, has not been examined at all.

(59) The inventors' data provide novel insights into this subject and lead them to propose the following pathogenetic events during PML-IRIS under natalizumab treatment. The anti-VLA-4 antibody inhibits immune surveillance of JCV infection at immunoprivileged sites such as the brain by blocking cell migration (Stuve et al., 2006) and local antigen presentation in the CNS (del Pilar Martin et al., 2008). As a result, pathologic neurotropic JCV variants may lead to PML in a small number (1/500-1/1000) of treated MS patients for reasons that are not yet understood (Major, 2010; Ransohoff, 2005). As soon as PML is suspected and natalizumab is stopped or actively removed by plasmapheresis, fully functional and activated T cells regain access to the CNS compartment, initiate the strong inflammation that is typical for PML-IRIS and effectively eliminate virus-infected cells by a number of mechanisms including CD4.sup.+ and CD8.sup.+ T cells and antibody-forming plasma cells.

(60) TABLE-US-00002 TABLE 2 Characterization of VP1 specific brain infiltrating T cell clones (TCC) Th TCR TCC # Well # Phenotype Vbeta Fine specificity TCC-1 17A Th0 Vβ2 VP1.sub.34 18A Th0 Vβ2 VP1.sub.34 TCC-2 16A Th1 Vβ2 VP1.sub.34 28A Th1 Vβ2 VP1.sub.34 18B Th1 Vβ2 VP1.sub.34 TCC-3 29A Th1 Vb18 VP1.sub.54 TCC-4 10A Th0 Vβ5.1 VP1.sub.74-1, VP1.sub.74-2, VP1.sub.74-3 14A Th0 Vβ5.1 VP1.sub.74-1, VP1.sub.74-2, VP1.sub.74-3 27A Th0 Vβ5.1 VP1.sub.74-1, VP1.sub.74-2, VP1.sub.74-3 30A Th0 Vβ5.1 VP1.sub.74-1, VP1.sub.74-2, VP1.sub.74-3 19B Th0 Vβ5.1 VP1.sub.74-1, VP1.sub.74-2, VP1.sub.74-3 TCC-5  3A Th1 Vβ— VP1.sub.91 TCC-6 11B Th1 Vβ— VP1.sub.143 TCC-7 12B Th0 Vβ2 VP1.sub.229 TCC-8 21A Th0 Vβ— VP1.sub.319 25A Th0 Vβ— VP1.sub.319 TCC-9 36A Th1 Vβ— VP1.sub.319  1B Th1 Vβ— VP1.sub.319 TCC-10 19A Th0 Vβ5.3 VP1.sub.335  3B Th0 Vβ5.3 VP1.sub.335 TCC-11 24A Th1 Vβ— VP1.sub.335

(61) Among the CNS-infiltrating T- and B cells, CD4.sup.+ T cells with either Th1- or the above bifunctional Th1-2 phenotype are probably the most critical element based on the following findings. Their parallel secretion of Th1-(IFN-gamma) and Th2 (IL-4) cytokines probably explains the expression of HLA-class II molecules on resident cells such as virus-infected astrocytes and microglia, but also on infiltrating immune cells, since IFN-gamma is the strongest inducer of HLA-class II. Although colocalization studies of HLA-DR with an astrocytic marker such as GFAP could not be performed due the paucity of material, the widespread expression of HLA-DR strongly suggests that these are also positive. In analogy to MS and its animal model experimental autoimmune encephalitis (EAE), where local reactivation of immigrating T cells has been demonstrated, JCV-specific Th1-2 and also Th1 cells are probably locally reactivated by recognition of JCV peptides on JCV-infected, HLA-class II positive astrocytes, microglia/macrophages or recruited dendritic cells (DCs). Furthermore, the secretion of large quantities of IL-4 leads to activation and expansion of memory B cells/plasmablasts in the CNS compartment with the consequence of virus-specific antibody secretion. Locally produced JCV capsid protein (VP1)-specific IgG antibodies may recognize virus-infected oligodendrocytes, which could then be lysed by complement- or antibody-mediated cellular cytotoxicity. The relative increase in the CSF of IgG1 and IgG3 antibodies, which bind complement with high affinity and have been described in the context of other viral infections, supports this notion. Since infected oligodendrocytes do not express HLA-class II, but effectively express HLA-class I, it can be expected that JCV-specific, HLA-A2-restricted CD8.sup.+ cytolytic T cells (Koralnik et al., 2001) (Koralnik et al., 2002) also contribute by killing JCV-infected oligodendrocytes and/or astrocytes. That these previously described cells in the peripheral blood of AIDS patients with PML are probably also participating in the local eradication of JCV in the brain is supported by our observation of CD8.sup.+ T cells specific for JCV VP1.sub.36 and JCV VP.sub.100 as defined by peptide-loaded HLA-A*02:01 tetramers. Infected astrocytes may not only serve as local antigen presenting cells for CD4.sup.+ virus-specific T cells, but may also be killed by Th1-2 cytolytic cells (Hemmer et al., 1997), but this together with the question of DR expression by astrocytes will require further studies.

(62) The above pathogenetic scenario accounts for the effects of IFN-gamma- and IL-4, i.e. the widespread expression of HLA-class II molecules in the brain as well as the strong intrathecal antibody response against JCV, however, it is still puzzling that a large fraction of brain-infiltrating cells show a Th1-2 phenotype. Previously, these cells were referred to as Th0 cells and considered an intermediate differentiation step before naive cells develop into memory cells committed to either Th1 or Th2 lineage (Mosmann and Coffman, 1989). This notion has, however, already been contended early based on following the cytokine patterns of single clones (Kelso, 1995). Today, Th1- and Th2 cells are understood as mutually exclusive fates (Ansel et al., 2006). However, individual TCC with dual cytokine secretion have been described as Th0 cells in measles virus infection (Howe et al., 2005) and among disease-exacerbating autoreactive T cells during altered peptide ligand-based therapy of MS (Bielekova et al., 2000). The inventors' present observation of stable Th1-2 clones based on intracellular cytokine staining, cytokine secretion and transcription factor expression point to a defined T helper cell subpopulation in the CNS rather than an intermediate or transient differentiation stage. Due to the abovementioned ill-defined role of Th0 cells and the prior controversy about their existence as terminally differentiated cells, we propose here to refer to IFN-gamma/IL-4 T helper cells as bifunctional Th1-2 cell. The context and signals that lead to this Th1-2 differentiation need further examination. In a recently published study in a viral infection model, the authors demonstrated that non-protective Th2 cells could be converted to stably IFN-gamma/IL-4-expressing and protective CD4+ cells by concerted action of antigen-specific TCR signal, type I and -II interferons and IL-12 (Hegazy et al., 2010) (Zhu and Paul, 2010). The inventors' findings are the first evidence for the existence of a stable GATA-3+ T-bet+ and IL-4+IFN-gamma+ Th2+1 phenotype in vivo in humans. It is conceivable that these cells were reprogrammed in the brain, and they could well explain the unusually strong immune response and fulminant course of PML-IRIS.

(63) Regarding the fine specificity of brain-infiltrating T cells, the inventors' data are interesting in several aspects. The JCV-specific T cell response is overall broad since peptides from almost all JCV proteins are recognized, which is consistent with the inventors' efforts to map immunodominant epitopes of JCV for peripheral blood-derived CD4.sup.+ T cells in healthy donors and MS patients. However, more than 50% of peptides recognized by brain-derived CD4.sup.+ T cells are part of the major structural protein VP1. Furthermore, VP1-specific T cells dominate with respect to strength of proliferation and precursor frequency. It is intriguing that VP1.sub.34-48 contains not only a major epitope for cytotoxic, HLA-A*02:01-restricted CD8.sup.+ T cells (Du Pasquier et al., 2003b), which the inventors found as well in the brain of the PML-IRIS patients by tetramer staining, but also for HLA-DRB1*15:01/DRB5*01:01-restricted CD4.sup.+ T cells. Furthermore, the recognition of peptide VP1.sub.74 and two variants thereof with single amino acid substitutions indicates that recognition of this epitope may be relevant to protect the host from immune evasion during persistent JCV infection. This has been shown previously for human immunodeficiency—(Borrow et al., 1997) and lymphocytic choriomeningitis virus infections (Ciurea et al., 2001). The vigorous intrathecal antibody response against VP1 further underscores the role of this structural protein. Therefore, the inventors' findings show that VP1 is important for protective immune responses against JCV-infected brain cells and that these are mediated by antibodies, CD4.sup.+ and CD8.sup.+ T cells. The strength of this response is probably in part determined by the HLA type of patient 2, who expresses both the major MS risk allele DRB1*15:01/DRB5*01:01 and A*02:01, which present an identical VP1 epitope to CD4.sup.+ and CD8.sup.+ T cells. He may therefore have experienced a particularly pronounced T cell-mediated immune response in the brain with its immunopathologic consequences of massive PML-IRIS, brain swelling, and neurological worsening. As already pointed out by others (Cinque et al., 2003) the JCV-specific immune response is a double-edged sword. Without a functional immune response brain cells are lysed by uncombated viral infection. On the other hand, if unleashed, the vigorous JCV-specific response during PML-IRIS causes brain inflammation and edema, and while it effectively eliminates JCV from the CNS, it may lead to death of the patient if not at least temporarily attenuated by immunosuppression (Tan et al., 2009b).

(64) The cellular and humoral JCV-specific immune response in the brain during PML-IRIS not only complicates the treatment, but may also cloud the diagnosis of PML in the first place. Different from current routine, which relies on CSF JCV viral load and, if a biopsy is performed, on immunohistochemistry and in situ hybridization for JCV antigen and DNA respectively, the intrathecal antibody response against VP1 appears more robust and should be examined. In both PML-IRIS patients of this study intrathecal VP1-specific antibody titers were extremely high despite almost undetectable JCV DNA by PCR and in situ hybridization. The important role of JCV antibody testing is supported by prior observations of high antibody titers in AIDS patients with PML (Weber et al., 1997), but also recent data in natalizumab-treated MS patients (Gorelik et al., 2010).

(65) Another important and unexpected observation of this study is that, different from the JCV-specific antibody response, pathogenetically relevant T cells are confined to the CNS parenchyma itself, and that the CSF is of little use for investigating T cell specificity and function. This finding is probably highly relevant not only to PML-IRIS, but also to MS, where most studies have focused on CSF as a surrogate for the responses within the CNS from obvious reasons, i.e. because CNS tissue is rarely available to investigators. Future research should therefore make every possible effort to examine biopsy or autopsy tissue if it can be acquired. When studying the brain-infiltrating CD4+ T cells of this MS patient with PML-IRIS, the inventors were further surprised to see that none of the peptides from three major myelin proteins were recognized, suggesting that bystander activation or—recruitment of myelin-specific T cells during massive brain inflammation does not occur, but that cells are exquisitely specific for the causal agent.

EXAMPLE 2—IMMUNISATION TO JCV

(66) An individual healing attempt was performed in a patient with idiopathic CD4+ lymphopenia, a rare constitutive immunodeficiency, who developed PML at the age of 64 years (referred to as “patient Hamburg” in FIG. 11). The male patient had been healthy, i.e. not experienced unusual or frequent infections, throughout his life, and in February 2010 was hospitalized with signs of an encephalitis of unknown origin. He was thoroughly worked up, and a diagnosis of suspected EBV-related encephalitis was made. Following transient improvement during antiviral therapy, he deteriorated further albeit slowly. During 2010, two brain biopsies were performed, and in the second one at the end of 2010, a diagnosis of progressive multifocal leukoencephalopathy (PML) was made based on positive JCV viral load in the CSF and on demonstration of JCV-infected oligodendrocytes and astrocytes in the brain. In parallel, the inventors found a low CD4+ T cell count (around 300/microliter) as well as a CD4+/CD8+ ratio of 0.5 or less, which are both consistent with the diagnosis of idiopathic CD4+ lymphopenia. In addition, in vitro experiments in the laboratory documented an absent T cell response to JCV virus VP1 in peripheral blood mononuclear cells and an almost complete absence of naïve CD4+ T cells. Following these observations, the inventors reasoned that the patient probably had developed PML based on a pre-existing and probably genetically determined low CD4+ number, which became further accentuated by entirely physiological immune involution, which sets in and increases above 50 years of age.

(67) The inventors wanted to test if vaccination with VP1, and, in this case of CD4+ lymphopenia, preferably combined with recombinant IL-7, would increase the number of JCV-specific T cells that the patients must have had, since he is JCV-positive. Further, if this were to occur, the inventors hoped that the vaccine-induced or -boosted JCV VP1-specific T cell response would lead to these cells' migrating to the CNS and elimination of virus and virus-infected cells from the CNS compartment. The inventors therefore applied for an “individual healing attempt”, discussed this option and its potential risks with the patient and obtained his consent. The use of IL-7 (Cytheris) was further supported by a recent publication in another case of CD4+ lymphopenia (Patel et al., 2010), in whom recombinant IL-7 together with antiviral drugs had led to substantial improvement of the patient, however, in that patient, no immunological studies were performed, and therefore nothing was known about improvement of antigen-specific immune responses.

(68) The vaccination approach in the above patient included the following steps (for timing of vaccinations and tests see scheme below): Subcutaneous injection with the entire recombinant major capsid protein VP1 (provided by the Life Science Inkubator, Bonn) in combination with a dermally applied TLR7 agonist (imiqimod, Aldara; commercially available) and i.v. recombinant IL-7 (Cytheris). The VP1 protein was administered in the form of virus-like particles (VLP), as the recombinantly expressed VP1 protein associated to such particles under the conditions used herein. As shown below, the patient not only showed an in vitro proliferative response against JCV VP1 after only two vaccinations, but also reduced the JCV viral load to 0 and finally began to show slight contrast enhancement around the PML lesions by brain MRI imaging, which all support that the vaccination worked in vivo. He also showed clinical improvement with slight delay after developing a JCV-specific immune response. Furthermore, since the inventors' data from the brain-infiltrating T cells in the PML-IRIS patient described in Example I suggested that JCV-specific CD4+ T cells with a T helper 1-2 phenotype are probably crucial for elimination of JCV virus from the brain, the inventors also stained for Th1-2 CD4+ T cells in the cerebrospinal fluid of the patient, and could demonstrate that these cells are indeed present.

(69) Another individual healing attempt was performed in a patient with breast cancer who received chemotherapy and developed acute myeloid leukaemia (AML) as a side effect of the chemotherapy. The patient then received an autologous and allogeneic hematopoietic stem cell transplant as treatment of the AML and subsequently developed acute graft-versus-host disease (grade IV). The immunodeficiency acquired as a consequence of these treatments resulted in PML. This case is referred to as “patient Zurich” in FIG. 11. Both patients were treated by recombinant IL-7 (s.c.). 2 days later, the first dose of VP1 (s.c. 1 mg, imiquimod cream on the skin) was administered. 2 days after the first dose of VP1, the second dose of rIL-7 was given.

(70) On day 12 after the first IL-7 dose, the patients received a second vaccination with VP1 s.c. plus imiquimod, and in this case simultaneously rIL-7. At week 6, a third dose of VP1/imiquimod and a fourth dose of rIL-7 were administered. FIG. 11 shows the results from the below described proliferation assay performed at the time points indicated in the graph with peripheral blood mononuclear cells (PBMC) obtained from both patients. A stimulation index of >2 is considered a positive response. It can be seen that both patients showed a positive immune response against JCV after vaccination.

(71) Material and Methods

(72) Blood and CSF Samples

(73) Biological samples were obtained after informed written consent. Peripheral blood mononuclear cells (PBMCs) were separated from EDTA-blood by Ficoll (PAA, Pasching, Austria) density centrifugation.

(74) Cerebrospinal fluid (CSF)-derived mononuclear cells were expanded by seeding 2000 cells/well plus 2×10.sup.5 irradiated (45 Gy) allogeneic feeder cells. 1 μg/ml PHA-L (Sigma-Aldrich, Munich, Germany) and 500 IU/ml IL-2 (kindly provided by Federica Sallusto, Institute for Research in Biomedicine, Bellinzona, CH) was added. The addition of IL-2 was repeated every 3-4 days until day 14.

(75) Proliferation Assays

(76) The proliferation response of PBMC to VP1 (kindly provided by Viktorya Demina, Life Science Inkubator, Bonn, Germany) and Tetanus toxoid (TTx, Novartis, Marburg, Germany) was tested by seeding 2×10.sup.5 cells in a 96-well U-bottom microtiter plates. VP1 was used at 2 μg/ml and TTx at 5 μg/ml. After 7 days incubation, incorporation of .sup.3H-thymidine (Hartmann Analytic, Braunschweig, Germany) was measured. Stimulatory indices (SI) were calculated by dividing the mean CPM (counts per minute) of the wells plus antigen by the mean CPM of the wells without antigen.

(77) To measure proliferative responses to VP1 and TTx by flow cytometry the CellTrace™ CFSE Cell Proliferation Kit (Invitrogen, Darmstadt, Germany) was used. Therefore, cells were seeded as described above and restimulated with antigen after six days and treated with CFSE following the manufacturer's instruction. After five days cells were analysed by flow cytometry.

(78) Flow Cytometry Analysis

(79) Whole blood stainings were performed by adding the appropriate antibody cocktail in a volume of 50 μl to 100 μl blood. The mixture was incubated for 30 minutes at room temperature, followed by 10 minutes of red blood cell lysis with FACS Lysing Solution (BD PharMingen). After washing, the cells were analysed by flow cytometry in a LSR II (BD). Following antibodies were used: CD4 (APC, RPA-T4, eBioscience), CD8 (PB, DK25, Dako, Glostrup, Denmark), CD45RO (FITC, UCHL1, eBioscience), CD25 (PE-Cy7, eBioscience), CD3 (PE, DakoCytomation, Denmark), CD8 (PB, DakoCytomation, Denmark), IFN-gamma (FITC, BDPharmingen), IL-4 (PE-Cy7, eBioscience).

REFERENCES

(80) Anonymous. http://tysabri.de/index.php?inhalt=tysabri.pmlinzidenz 2011. Astrom K E, Mancall E L, Richardson E P, Jr. Progressive multifocal leuko-encephalopathy; a hitherto unrecognized complication of chronic lymphatic leukaemia and Hodgkin's disease. Brain 1958; 81: 93-111. Carreno B M, Turner R V, Biddison W E, Coligan J E. Overlapping epitopes that are recognized by CD8+ HLA class I-restricted and CD4+ class II-restricted cytotoxic T lymphocytes are contained within an influenza nucleoprotein peptide. J Immunol 1992; 148: 894-9. Cavacini L A, Kuhrt D, Duval M, Mayer K, Posner M R. Binding and neutralization activity of human IgG1 and IgG3 from serum of HIV-infected individuals. AIDS Res Hum Retroviruses 2003; 19: 785-92. Cinque P, Bossolasco S, Brambilla A M, Boschini A, Mussini C, Pierotti C, et al. The effect of highly active antiretroviral therapy-induced immune reconstitution on development and outcome of progressive multifocal leukoencephalopathy: study of 43 cases with review of the literature. J Neurovirol 2003; 9 Suppl 1: 73-80. Du Pasquier R A, Clark K W, Smith P S, Joseph J T, Mazullo J M, De Girolami U, et al. JCV-specific cellular immune response correlates with a favorable clinical outcome in HIV-infected individuals with progressive multifocal leukoencephalopathy. J Neurovirol 2001; 7: 318-22. Du Pasquier R A, Corey S, Margolin D H, Williams K, Pfister L A, De Girolami U, et al. Productive infection of cerebellar granule cell neurons by JC virus in an HIV+ individual. Neurology 2003a; 61: 775-82. Du Pasquier R A, Kuroda M J, Schmitz J E, Zheng Y, Martin K, Peyerl F W, et al. Low frequency of cytotoxic T lymphocytes against the novel HLA-A*0201-restricted JC virus epitope VP1(p36) in patients with proven or possible progressive multifocal leukoencephalopathy. J Virol 2003b; 77: 11918-26. Du Pasquier R A, Kuroda M J, Zheng Y, Jean-Jacques J, Letvin N L, Koralnik I J. A prospective study demonstrates an association between JC virus-specific cytotoxic T lymphocytes and the early control of progressive multifocal leukoencephalopathy. Brain 2004a; 127: 1970-8. Du Pasquier R A, Schmitz J E, Jean-Jacques J, Zheng Y, Gordon J, Khalili K, et al. Detection of JC virus-specific cytotoxic T lymphocytes in healthy individuals. J Virol 2004b; 78: 10206-10. Du Pasquier R A, Stein M C, Lima M A, Dang X, Jean-Jacques J, Zheng Y, et al. JC virus induces a vigorous CD8+ cytotoxic T cell response in multiple sclerosis patients. J Neuroimmunol 2006; 176: 181-6. Egli A, Infanti L, Dumoulin A, Buser A, Samaridis J, Stebler C, et al. Prevalence of polyomavirus BK and JC infection and replication in 400 healthy blood donors. J Infect Dis 2009; 199: 837-46. Gillespie S M, Chang Y, Lemp G, Arthur R, Buchbinder S, Steimle A, et al. Progressive multifocal leukoencephalopathy in persons infected with human immunodeficiency virus, San Francisco, 1981-1989. Ann Neurol 1991; 30: 597-604. Goldmann C, Petry H, Frye S, Ast O, Ebitsch S, Jentsch K D, et al. Molecular cloning and expression of major structural protein VP1 of the human polyomavirus JC virus: formation of virus-like particles useful for immunological and therapeutic studies. J Virol 1999; 73: 4465-9. Hegazy A N, Peine M, Helmstetter C, Panse I, Frohlich A, Bergthaler A, et al. Interferons direct Th2 cell reprogramming to generate a stable GATA-3(+)T-bet(+) cell subset with combined Th2 and Th1 cell functions. Immunity 2010; 32: 116-28. Houff S A, Major E O, Katz D A, Kufta C V, Sever J L, Pittaluga S, et al. Involvement of JC virus-infected mononuclear cells from the bone marrow and spleen in the pathogenesis of progressive multifocal leukoencephalopathy. N Engl J Med 1988; 318: 301-5. Jilek S, Jaquiery E, Hirsch H H, Lysandropoulos A, Canales M, Guignard L, et al Immune responses to JC virus in patients with multiple sclerosis treated with natalizumab: a cross-sectional and longitudinal study. Lancet Neurol 2010; 9: 264-72. Kleinschmidt-DeMasters B K, Tyler K L. Progressive multifocal leukoencephalopathy complicating treatment with natalizumab and interferon beta-1a for multiple sclerosis. N Engl J Med 2005; 353: 369-74. Koralnik I J. Progressive multifocal leukoencephalopathy revisited: Has the disease outgrown its name? Ann Neurol 2006; 60: 162-73. Koralnik I J, Du Pasquier R A, Kuroda M J, Schmitz J E, Dang X, Zheng Y, et al. Association of prolonged survival in HLA-A2+ progressive multifocal leukoencephalopathy patients with a CTL response specific for a commonly recognized JC virus epitope. J Immunol 2002; 168: 499-504. Koralnik I J, Du Pasquier R A, Letvin N L. JC virus-specific cytotoxic T lymphocytes in individuals with progressive multifocal leukoencephalopathy. J Virol 2001; 75: 3483-7. Langer-Gould A, Atlas S W, Green A J, Bollen A W, Pelletier D. Progressive multifocal leukoencephalopathy in a patient treated with natalizumab. N Engl J Med 2005; 353: 375-81. Major E O. Progressive multifocal leukoencephalopathy in patients on immunomodulatory therapies. Annu Rev Med 2010; 61: 35-47. Ogg G S, Jin X, Bonhoeffer S, Dunbar P R, Nowak M A, Monard S, et al. Quantitation of HIV-1-specific cytotoxic T lymphocytes and plasma load of viral RNA. Science 1998; 279: 2103-6. Padgett B L, Walker D L, ZuRhein G M, Eckroade R J, Dessel B H. Cultivation of papova-like virus from human brain with progressive multifocal leucoencephalopathy. Lancet 1971; 1: 1257-60. Patel A, Patel J, Ikwuagwu J. A case of progressive multifocal leukoencephalopathy and idiopathic CD4+ lymphocytopenia. J. Antimicrob. Chemother. 2010; 65:2489. Ransohoff R M. Natalizumab and PML. Nat Neurosci 2005; 8: 1275. Stoner G L, Ryschkewitsch C F, Walker D L, Webster H D. JC papovavirus large tumor (T)-antigen expression in brain tissue of acquired immune deficiency syndrome (AIDS) and non-AIDS patients with progressive multifocal leukoencephalopathy. Proc Natl Acad Sci USA 1986; 83: 2271-5. Stuve O, Marra C M, Jerome K R, Cook L, Cravens P D, Cepok S, et al Immune surveillance in multiple sclerosis patients treated with natalizumab. Ann Neurol 2006; 59: 743-7. Tan C S, Dezube B J, Bhargava P, Autissier P, Wuthrich C, Miller J, et al. Detection of JC virus DNA and proteins in the bone marrow of HIV-positive and HIV-negative patients: implications for viral latency and neurotropic transformation. J Infect Dis 2009a; 199: 881-8. Tan K, Roda R, Ostrow L, McArthur J, Nath A. PML-IRIS in patients with HIV infection: clinical manifestations and treatment with steroids. Neurology 2009b; 72: 1458-64. Weber T, Trebst C, Frye S, Cinque P, Vago L, Sindic C J, et al. Analysis of the systemic and intrathecal humoral immune response in progressive multifocal leukoencephalopathy. J Infect Dis 1997; 176: 250-4. Zhu J, Paul W E. CD4+ T cell plasticity-Th2 cells join the crowd. Immunity 2010; 32: 11-3. Zhu J, Yamane H, Paul W E. Differentiation of effector CD4 T cell populations (*). Annu Rev Immunol 2010; 28: 445-89. Zonios D I, Falloon J, Bennett J E, Shaw P A, Chaitt D, Baseler M W, et al. Idiopathic CD4+ lymphocytopenia: natural history and prognostic factors. Blood 2008; 112: 287-94. Zurhein G, Chou S M. Particles Resembling Papova Viruses in Human Cerebral Demyelinating Disease. Science 1965; 148: 1477-9. DE 195 43 553 Goldmann et al., 1999, Journal of Virology 73, p. 4465-4469