WO1996037107A1

WO1996037107A1 - Antigen-specific activation of unprimed cd8+ t cells

Info

Publication number: WO1996037107A1
Application number: PCT/US1996/007436
Authority: WO
Inventors: Marloes L. H. Debruijn; Michael R. Jackson; Per A. Peterson
Original assignee: The Scripps Research Institute
Priority date: 1995-05-23
Filing date: 1996-05-22
Publication date: 1996-11-28
Also published as: AU5801896A

Abstract

An artificial antigen-presenting system provides methods and compositions for the activation of unprimed CD8+ T cells in vitro to form specific activated CD8+ T cells (cytotoxic T lymphocytes, or CTL). CD8+ T cells are activated by contacting them with macrophages that present a MHC molecule complexed to an immunogenic peptide, acquired by phagocytosis of the peptide on an artificial support and with an assisting molecule.

Description

ANTIGEN-SPECIFIC ACTIVATION OF UNPRIMED CD8+ T CELLS

This invention was made with support from the Government of the United States of America under National Institutes of Health Grants GM 49305, CA 27489 and DK

37915, and the Government of the United States of America has certain rights in the invention.

TECHNICAL FIELD The present invention relates to materials and methods of activating T cells with specificity for particular antigenic peptides, the use of activated T cells in vivo for the treatment of a variety of disease conditions, and compositions appropriate for these uses.

BACKGROUND The operation of the immune system and the way it cures or protects individuals from infectious disease continues to be of great interest to scientists. It is believed that it might be possible to activate the immune system to combat other types of diseases. Such diseases include cancer, AIDS, hepatitis and infectious disease in immunosuppressed patients. While various procedures involving the use of antibodies have been applied in those types of diseases, few if any successful attempts using activated CD8+ T cells have been reported. Theoretically, activated CD8+ T cells would be the preferable means of treating such diseases.

Activated CD8+ T cells, also known as cytotoxic T lymphocytes (CTL) , represent the main line of defense against viral infections. Activated CD8+ T cells specifically recognize and kill cells that are infected by a virus. Thus, the cost of eliminating a viral infection is the accompanying loss of the infected cells.

Activated CD8+ T cells are derived from unprimed CD8+ T cells, also called naive, precursor, or resting CD8+ T cells. Unprimed CD8+ T cells are activated by interactions with antigen presenting cells (APC) . The T cell receptors on the surface of CD8+ T cells cannot recognize foreign antigens directly. In contrast to antibodies, which can bind free antigen, T cell receptors must have antigen presented in association with another molecule, the major histocompatibility complex (MHC) molecule.

Major histocompatibility complex (MHC) molecules of the class I type present antigen to unprimed CD8+ T cells. The term "major histocompatibility complex" refers to a large genetic locus encoding an extensive family of integral membrane glycoproteins that play an important role in the immune response. The MHC genes, which are also referred to as the HLA (human leucocyte antigen) complex, are located on chromosome 6 in humans. The molecules encoded by MHC genes are present on cell surfaces and are largely responsible for recognition of tissue transplants as "non-self" as well as the immune response to infection. MHC molecules are grouped into three major classes, referred to as I, II, and III. Unprimed CD4+ T cells interact primarily with Class II molecules, and serve mainly as inflammatory T cells and helper T cells when activated. Unprimed CD8+ T cells interact primarily with Class I molecules.

Class I molecules bind and present peptides derived primarily from intracellular degradation of endogenous proteins. Complexes of MHC class I molecules with peptides derived from viral, bacterial and other foreign proteins comprise the ligand that triggers the antigen responsiveness of T cells. In contrast, complexes of MHC class I molecules with peptides derived from normal cellular products play a role in "teaching" the T cells to tolerate self-peptides, in the thymus. Class I molecules do not present entire, intact antigens; rather, they present pep.ide fragments of antigens, "loaded" onto the "peptide binding groove" of the class I molecule.

When the unprimed CD8+ T cell interacts with an antigen-presenting cell having the peptide bound by a MHC class I molecule and costimulatory molecule, the CD8+ T cell is activated to proliferate and becomes an armed effector T cell. A cos.imulatory molecule can provide a second signal that determines that the presented peptide is recognized by the T cell as "nonself". The second signal thus activates the T cell into an armed effector T cell. The second signal generally provided by antigen presenting cells that activate CD8+ T cells is the B7 molecule, which occurs as B7.1 and B7.2 variants. However, in certain circumstances IL-2 can serve as the second signal in CD8+ T cell activation. See, generally, Janeway and Travers, Immunobiolocrv. published by Current Biology Limited, London (1994) , incorporated by reference. For many years, immunologists have hoped to raise specific cytotoxic cells targeting viruses, retroviruses and cancer cells. While targeting against viral diseases in general may be accomplished in vivo by vaccination with live or attenuated vaccines, no similar success has been achieved with retroviruses or with cancer cells. Moreover, the vaccine approach has not had the desired efficacy in immunosuppressed patients. At least one researcher has taken the rather non-specific approach of "boosting" existing CD8+ T cells by incubating them in vi tro with interleukin-2 (IL-2) .

However, this protocol (known as lymphokine activated killer (LAK) cell therapy) will only^' allow the expansion of those CD8+ T cells that are already activated.

As the immune system is always active for one reason or another, most of the IL-2 stimulated cells will be irrelevant for the purpose of combatting the disease. In fact, it has not been documented that this type of therapy activates any cells with the desired specificity. Thus, the benefits of LAK cell therapy are controversial at best, and the side effects are typically so severe that many studies have been discontinued.

Activated CD8+ T cells generated against specific MHC-peptide complexes are potentially useful in treatment of viral infection or cancer. Such cells have the required specificity for ridding the body completely of diseased cells with minimal side effects. What was needed is a way to activate CD8+ T cells with antigen presenting cells.

The primary physiological role of macrophages is clearing away the debris of microbes and cells. Macrophages engulf and internalize debris by phagocytosis. The membrane-bound internal compartment containing such debris that is formed by phagocytosis is called a phagosome . Phagosomes can fuse with compartments, called lysosomes, that contain degradative enzymes. Generally, lysosomes containing peptide fragments fuse with membrane compartments containing newly translated MHC class II molecules. Peptide fragments are loaded unto the MHC class II molecules and the entire complex is inserted into the plasma membrane of the cell . Thus macrophages can load and present antigen acquired through the phagocytic pathway using MHC class II molecules, which interact primarily with CD4+ T cells. Phagosomes are formed by invagination of the plasma membrane. Using lactoperoxidase covalently coupled to carboxylated latex beads,^' polypeptides of the phago (lyso) somal membrane can be labeled and were shown to be virtually indistinguishable from cell surface polypeptides by gel electrophoresis (Muller, W.A., e_t al. , J. Cell Bid . __ : 292, 1980) . In addition to measuring influx of polypeptides from the plasma membrane to the phago(lyso) somal membrane, this method revealed a rapid flow of polypeptides from the phago(lyso) somal membrane to the plasma membrane (Muller, W.A. , et al. , J. Cell Biol . __ : 303, 1980). A significant portion of membrane proteins introduced into the phago(lyso) somal compartments were sorted and retrieved back to the cell surface. Therefore, the phagocytic pathway is not merely a unidirectional path for molecules destined for digestion.

Recently, macrophages have been found to be able to present antigen acquired through the phagocytic pathway using MHC class I molecules. It has been found that this alternative peptide loading pathway can be used to present antigenic peptides to unprimed CD8+ T cells.

With the discovery that macrophages also produce MHC class I molecules, attempts were made to use macrophages as antigen presenting cells for CD8+ T cells. Peptides obtained from protein fragments were added to cultures of macrophages. However, even at relatively high concentrations of peptide in splenocyte cultures, activation of CD8+ T cells met with only limited success in that the amount of CD8+ T cell activation was usually low.

Other attempts have been made to present proteins to macrophages to produce antigen presenting cells having MHC class I molecules that would serve as antigen presenting cells for CD8+ T cells. In such experiments, ovalbumin proteins were bonded to latex beads which were engulfed by the macrophages. However, the level of desired specificity was^' limited.

Harding and Song attached full-length ovalbumin to microscopic beads, but did not use short synthetic peptides affixed to beads as disclosed in the present invention. Harding, C.Y. and Song, R. J. Immunology 153:4925-4933 (1994). Harding and Song found that bead- attached ovalbumin was more effective in vi tro than soluble ovalbumin in an assay in which macrophages presenting antigen elicited IL-2 secretion from a T cell hybridoma cell line. However, when a T cell mediated cell lysis assay was used, ten times as many macrophages exposed to bead-bound ovalbumin were required to get the same effect as produced by macrophages exposed to soluble ovalbumin, a finding that they considered difficult to interpret. Id. at 4930.

Kovacsovics-Bankowski, M. et al. also disclose the attachment of full-length ovalbumin to microscopic beads, but state that the optimum bead diameter is 2-3 μm and beads 6-10 μm in diameter perform poorly.

Kovacsovics-Bankowski, M. et al. Proc. Natl . Acad. Sci . USA 90: 4942-4946, 4943 (1993).

Another approach has been the presentation of MHC class I binding peptides to macrophages as part of a fusion protein expressed by bacteria. Pfeifer, J.D. et al. Nature 361: 359 (1993). However, this approach required the addition of live bacteria expressing the recombinant fusion protein to the macrophage culture, providing a possible danger of contamination and infection.

Accordingly, what is needed is a means to activate CD8+ T cells specifically so that they proliferate and become cytotoxic. It would be useful if the activation could be done in vi tro to a specific antigen and the activated cytotoxic T cells reintroduced into the patient. It would' also be desirable that the activation could be done by an artificial antigen- presenting system that presents a peptide corresponding to the antigen in a MHC class I molecule so that CD8+ T cells are activated. It would also be advantageous if it was possible to select _he peptide so that substantially only those CD8+ T cells cytotoxic to cells presenting that peptide would be activated. The desired system should be easy to use for the effective treatment of patients. The present system meets these desires.

BRIEF SUMMARY OF THE INVENTION The present invention relates to an artificial antigen-presenting system for activating CD8+ T cells specifically directed to a particular antigen and treatment of a patient using such a system.

The present invention relates to methods for producing activated CD8- T cells. Activation is accomplished using artificial antigen presenting cells, such as transmuted macrophages, to present a peptide corresponding to the antigen to unprimed CD8+ T cells to activate them. The activated CD8+ T cells proliferate and become cytotoxic to cells presenting the antigen. One method comprises affixing peptides to an artificial support such as microscopic beads. The macrophages are then contacted with the peptides for a time period sufficient to allow the macrophages to engulf the peptides and preferably also the support. The macrophages process the peptides and present at least a portion of the peptide en their surface complexed with MHC class I molecules. The macrophages are then contacted with unprimed CD8+ T cells, preferably in vi tro, for a time period sufficient to activate, in an antigen-specific manner, the CD8+ T cells. Preferably a second signal such as interleukin-2 is provided to the CD8+ T cells to assist in priming them.

The peptides used in the present invention are antigenic polypeptides that are about 8 to about 35 amino acid residues in length and preferably 8 to 11 amino acid residues in length. The peptides correspond to the particular antigen in that they can be peptide fragments of the protein that defines the antigen, or can be peptides having the same amino acid residue sequence as a portion of the particular antigen. Such peptides can also correspond to the particular antigen when the majority of the amino acids in the sequence are the same as those in the sequence of a portion of the particular antigen, the difference being on the end of the peptide to facilitate affixing to the bead (e.g., the addition of a cysteine) or conservative substitutions which have little, if any, effect on activation of the desired CD8+ T cell.

A further method allows for the treatment of a patient. The activated CD8+ T cells are separated from the macrophages and if desired, can be allowed to proliferate for a time period. The activated CD8+ T cells are suspended in an acceptable carrier or excipient. The preparation is then administered to an individual in need of treatment. The CD8+ T cells can be activated using a kit having about 10⁵ macrophages in stasis.

In another variation, the invention relates to methods of treating certain conditions in patients by specifically killing target cells in the patient. A sample containing unprimed CD8+ T cells is removed from the patient. Unprimed CD8+ T cells are contacted in vi tro with antigen-presenting macrophages for a time period sufficient to activate, in an antigen-specific manner, the CD8+ T cells.^' The activated CD8+ cells are then suspended in an acceptable carrier or excipient and administered to the patient^'. In various embodiments the condition may comprise cancer, tumor^'s, neoplasia, viral or retroviral infection, autoimmune or autoimmune-type conditions. The delivery of class I binding peptides by macrophages via the phagocytic pathway can enhance their presentation to unprimed CD8+ T cells in vi tro . Affixing peptides to an artificial support, such as microscopic beads and allowing subsequent internalization by macrophages provides several advantages. The peptides are protected from rapid degradation by serum proteases. Efficient binding and exchange of peptide with MHC class I molecules is promoted by bringing class I molecules expressed on the engulfing membrane and the peptides bound to the artificial support into close contact. Macrophages can specifically activate unprimed CD8+ T cells in vi tro at only nanomolar peptide concentration if artificial supports, such as microscopic beads, are used to deliver antigenic peptide.

Short peptides (8-11 amino acid residues) , administered in the phagocytic pathway at nanomolar concentration, are effective in specifically activating unprimed CD8+ T cells in vi tro . Whole protein antigen coated on beads did render macrophages susceptible to lysis by an antigen-specific activated CD8+ T cells clone, indicating limited proteolytic capability in the phagocytic pathway for class I-restricted presentation. However, the use of the whole protein results in macrophages having insufficient complexes of specific peptides and MHC class I molecules to activate CD8+ T cells.

In a culture containing macrophages, the artificial supports, such as beads, are internalized by macrophages by phagocytosis. Phagocytosis by macrophages is required for optimal stimulation of unprimed CD8+ T cells with bead-affixed peptide in the present invention. The beads themselves were not found to improve the efficiency of primary activated CD8+ T cell response induction of exogenously added peptides provided in the culture medium. Therefore, delivery of peptide via beads is required for efficient presentation by macrophages to unprimed CD8+ T cells.

Bead-affixed peptides of optimal length (8 amino acid residues) or nearly optimal length allow the induction of primary activated CD8+ T cell responses. Long peptide fragments or whole protein are not efficiently presented via phagocytosis to generate these responses . The inability to induce primary activated CD8+ T cell with bead-affixed large peptide fragments or protein is most likely due to insufficient numbers of relevant class I-peptide complexes.

Further attributes and advantages of the present invention are discussed below and would be apparent to one skilled in the art.

BRIEF DESCRIPTION OF THE DRAWINGS FIGURE 1 shows the results of studies demonstrating that the ability to prime CD8+ T cells in vitro with beads coated with OVA-8 peptide is impaired when macrophages are removed;

FIGURE 2 shows the results of studies demonstrating that the ability to prime CD8+ T cells in vi tro with beads coated with VSV-8 peptide is impaired when macrophages are removed;

FIGURE 3 shows the results of studies demonstrating that the ability to prime CD8+ T cells in vi tro with beads coated with SEV-9 peptide is impaired when macrophages are removed; FIGURE 4 shows the results of studies demonstrating that supplementing the culture with macrophages restores the ability to prime CD8+ T cells in vi tro with beads coated with OVA-8 peptide;

FIGURE 5 shows the results of studies demonstrating that supplementing the culture with macrophages restores the ability to prime CD8+ T cells in vi tro with beads coated with VSV-8 peptide;

FIGURE 6 shows the results of studies on the effects of cytochalasin D on the restoration of the ability to prime CD8+ T cells in vi tro with beads coated with OVA-8 or VSV-8 peptides produced by supplementing the culture with either IC-21 or PEM macrophages;

FIGURE 7 shows the results of studies indicating a dependence on haplotype of the stimulating macrophages in activated CD8+ T cell responses;

FIGURE 8 shows the results of studies demonstrating the specificity of priming CD8+ T cells in vi tro with beads coated with OVA-8 peptide or with VSV-8 peptide in the presence of 10 U/ml of interleukin-2 (IL- 2) ;

FIGURE 9 shows the results of studies demonstrating the specificity of priming CD8+ T cells in vi tro with beads coated with OVA-8 peptide or with VSV-8 peptide in the absence of 10 U/ml of interleukin-2 (IL- 2) ;

FIGURE 10 shows the results of studies demonstrating that the ability to produce activated CD8+ T cells in vi tro with beads coated with OVA-8 peptide or VSV-8 peptide is impaired when CD8+ T cells but not CD4+ cells are depleted;

FIGURE 11 shows the results of the titration of class I binding optimal length immunogenic peptides, free in solution, in primary activated CD8+ T cell response induction; FIGURE 12 shows the results of the titration of class I binding optimal length immunogenic peptides, bead-affixed, in primary activated CD8+ T cell response induction; FIGURE 13 shows primary activated CD8+ T cell response induction with bead-affixed peptides, with free peptide or with peptide and beads separately;

FIGURE 14 shows the up-regulation of cell surface K^b expression on RMA-S, EL-4 and IC-21 cells by peptide and/or low temperature;

FIGURE 15 shows the result of a comparison of optimal peptides with 2 amino acid extended peptides in their ability to stabilize cell surface K^b molecules; FIGURE 16 shows the result of a comparison of optimal peptides with 2 amino acid extended peptides in their ability to stabilize cell surface K^b molecules;

FIGURE 17 shows the result of a comparison of optimal peptides with 2 amino acid extended peptides in their ability to induce primary activated CD8+ T cell responses against bead-affixed peptides;

FIGURE 18 shows the result of a comparison of optimal peptides with 2 amino acid extended peptides in their ability to induce primary activated CD8+ T cell responses against bead-affixed peptides;

FIGURE 19 shows the results of a titration of OVA8 peptide for target cell sensitization;

FIGURE 20 shows the results of a titration of VSV8 peptide for target cell sensitization; FIGURE 21 shows the effectiveness of activation of CD8+ T cells with OVA8, long OVA peptides or whole protein, bead-affixed or free;

FIGURE 22 shows the effectiveness of stimulation of unprimed B6 spleen cells with beads coated with OVA8, a 24-mer peptide (OVA24) or a 35-mer peptide (OVA35) containing the minimal activated CD8+ T cell epitope, the OVA8 sequence, or with the ovalbumin protein;

FIGURE 23 shows the effectiveness of recognition of IC-21 cells by the OVA specific activated CD8+ T cell clone after incubation of the target cells with OVA8 or ovalbumin protein in unassociated form or bead-affixed form;

FIGURE 24 shows the effect of brefeldin A and cytochalasin D on the presentation of free OVA8 peptide by macrophages;

FIGURE 25 shows the effect of brefeldin A and cytochalasin D on the presentation of free VSV8 peptide by macrophages; FIGURE 26 shows the effect of brefeldin A and cytochalasin D on the presentation of bead-affixed OVA8 peptide by macrophages;

FIGURE 27 shows the effect of brefeldin A and cytochalasin D on the presentation of bead-affixed VSV8 peptide by macrophages;

FIGURE 28 shows the recovery of target cell sensitivity after pronase treatment of IC-21 cells pre-treated with VSV8 peptide;

FIGURE 29 shows the recovery of target cell sensitivity after pronase treatment of IC-21 cells pre-treated with bead-affixed VSV8 peptide;

FIGURE 30 shows the recovery of target cell sensitivity after pronase treatment of IC-21 cells pre-treated with OVA8 peptide, which -is prevented by brefeldin A; and

FIGURE 31 shows the recovery of target cell sensitivity after pronase treatment of IC-21 cells pre-treated with bead-bound OVA8 peptide, which is prevented by brefeldin A. DETAILED DESCRIPTION OF THE INVENTION

The present invention relates to a artificial antigen-presenting system which can be used to activate CD8+ T cell lymphocytes. The system produces activated CD8+ T cells which then proliferate, seek out and destroy target cells. The present invention can be used to activate CD8+ T cells in vi tro and the activated CD8+ T cells can then be returned to the patient from which they were originally derived. Alternatively, the CD8+ T cells can be contacted in vivo. The peptide is selected to activate the appropriate T cell, depending on the treatment co be conducted. For example, in the treatment of particular cancers, certain antigenic peptides are presented on the surface of the cancer cells which will react with activated T cells. Thus, it is appropriate to use a peptide selected to activate the appropriate T cells which will then bind with and destroy the cancer cells.

The efficiency of providing the peptide to the antigen presenting cell is substantially improved by using macrophages as antigen presenting cells and affixing the peptide to an artificial support, such as a microscopic bead. The macrophages engulf the peptide, and at least some portion of the support, by phagocytosis. Peptide is loaded onto newly synthesized MHC class I molecules within the phagocytic pathway. Peptide - MHC class I complexes are inserted into the membrane and presented on the surface of the macrophage. The present invention relates to a method for activating CD8+ T cells against a selected peptide, and antigen presenting cells presenting the peptide. Unprimed CD8+ T cells are isolated from a patient to be treated. The antigen presenting cells are then contacted with the CD8+ T cells for a sufficient period of time to activate the CD8+ T cells resulting in proliferation and transforming the T cells into armed effector cells, also known as activated CD8+ T cells. The activated CD8+ T cells can then be separated from the^' antigen presenting cells, suspended in an acceptable carrier and administered to the patient. The present invention provides also a kit to be used to activate the patient's CD8+ T cells. It is preferred that human macrophage cells are used and, therefore, human MHC class I molecules complexed to peptides are produced. Macrophages may also be obtained from the patient to be treated, or, alternatively, from donors or from storage in stasis, such as frozen in liquid nitrogen in an suitable medium (such as culture medium comprising from about 5 to about 10% dimethylsulfoxide (DMSO) (Sigma Chemical Co., St. Louis, MO) , or alternatively, fetal bovine serum (FBS) containing about 10% DMSO) .

As shown in prior U.S. Patent No. 5,314,813, urine systems provide particularly useful models for testing the operation of T cell activation and demonstrate the applicability of the process for human systems. See also Sykulev et al . , Immunity 1 : 15-22 (1994) . Peptides Previous attempts to activate CD8+ T cell in vi tro by addition of high levels of peptide to splenocyte cultures have met with limited success. While this approach was somewhat successful for the K^b-binding OVA8 peptide, SIINFEKL, CD8+ T cell response induction with other class I-binding peptides was relatively ineffective. Greater activation of CD8+ T cells can be obtained with cell types that express a large number of empty cell surface class I molecules. Thus, the cell surface density of specific MHC/peptide complexes and therefore the efficiency of loading MHC class I molecules with peptide appear to be important features for the antigen-presenting ability to unprimed CD8+ T cells.

Synthetic peptides that constitute a activated CD8+ T cell epitope can bind directly to cell surface MHC class I molecules. No intracellular events are apparently involved. Cell surface class I - peptide association depends primarily on the presence of empty class I molecules or class I molecules that carry poorly bound endogenous peptides. TAP-mutated cell types, which are deficient in loading peptides onto class I molecules (e.g. RMA-S cells and T2 cells), express significant amounts of empty class I molecules at their cell surface, which can be loaded with a single peptide species. On most cell types however, only a small fraction of the total cell surface class I molecules is empty. Therefore, loading cell surface class I of 'normal' cells with exogenous class I-binding peptides is poor.

Bead-affixed antigenic peptide is believed to be superior to free peptide for the following reasons: a) peptide present in phagosomes is protected from serum proteases, b) a high concentration of peptide at the contact site of cell surface and bead may improve peptide loading on surface class I molecules, and/or c) the small space between bead and membrane in phago(lyso) somes may promote peptide loading on class I molecules.

Virtually all cellular proteins in addition to viral antigens are capable of being used to generate peptides that serve as potential MHC class I ligands. It is preferable to isolate and load peptide fragments of appropriate size and antigenic characteristics onto MHC class I molecules.

The peptides used in the present invention are antigenic polypeptides which are about 8 to about 35 amino acid residues in length. The peptides correspond to the particular antigen in that they can be peptide fragments of the protein that defines the antigen. Alternatively, the peptides can be peptides having the same amino acid residue sequence as a portion of the particular antigen. Such peptides can also correspond to the particular antigen when the majority of the amino acids in the sequence are the same as those in the sequence of a portion of the particular antigen, the difference being on the end of the peptide to facilitate affixing to the bead (e.g., the addition of a cysteine) or conservative substitutions which have little, if any, effect on activation of the desired CD8+ T cell. It is also preferred that the peptides be of a uniform size, preferably 8-mers or 9-mers, and most preferably, 8-mers. It is also preferable that the peptides prepared for loading onto the MHC molecules be of a single species; i.e., that all peptides loaded onto the MHC be identical in size and sequence. In this manner, it is possible to produce monoantigenic peptide-loaded MHC molecules.

Peptides were presented to the cells attached to a support. Preferably, peptides are affixed to an artificial support in a manner which allows them to enter an intracellular pool of peptides through the phagocytic pathway. Preferably, the peptides to be presented are affixed to beads. The beads are typically from about 0.05 μm to about 10 μm in diameter. One preferred size is 6.76 μm in diameter.

The beads, commonly called "latex" beads, may be made of polystyrene, polyacrolein, poly(methyl methyacrylate) , a mixture of polystyrene and glycidyl methacrylate, or an iron-containing magnetic polymer. Functional groups, such as carboxyl, sulfate, amino, acrylic ester, or chloromethylstyrene may be provided on the surface of the bead for sites at which peptide may be affixed. Most preferably, however, peptides are non- covalently affixed to the beads. The amount of peptide affixed to the bead is preferably from about 0.05 nmol/10⁶ beads to about 0.5 nmol/10⁶ beads.

Typically, peptides affixed to an artificial support are added to the culture medium that is accessible to the macrophages. The peptides may be in the form of an intact pclypeptide or protein which may be subsequently degraded via cellular processes, e.g., via enzymatic degradation. Alternatively, the intact polypeptide or protein may be degraded via some other means such as chemical digestion (e.g. cyanogen bromide) or proteases (e.g. chymotrypsin) prior to its addition to the cell culture. In other embodiments, the peptides are presented in smaller segments which may or may not comprise epitopic amino acid sequences. The peptides may be synthesized instead of being derived from a larger peptide by degradation.

In a preferred embodiment, a sufficient amount of artificial support with affixed protein(s) or peptide (s) is added to the cell culture to allow the MHC class I molecules to bind and subsequently present a high density of the peptide on the antigen presenting cells of the present invention. Preferably the same kind of peptide is presented by each human MHC class I expressing cells of the present invention. Isolation of Unprimed CD8+ T cells

Unprimed (or resting or naive or precursor) CD8+ T cells -- i.e., T cells that have not been activated to target a specific antigen -- are preferably extracted from the patient prior to incubation of the CD8+ T cells with the antigen presenting cells of the present invention. It is also preferred that unprimed CD8+ T cells be harvested from a patient prior to the initiation of other treatment or therapy which may interfere with the CD8+ T cells' ability to be specifically activated. For example, if one is intending to treat an individual with' a neoplasia or tumor, it is preferable to obtain a sample of cells and culture same prior to the initiation of chemotherapy or radiation treatment. Methods of extracting and culturing lymphocytes are well known. For example, U.S. Patent No. 4,690,915 to Rosenberg describes a method of obtaining large numbers of lymphocytes via lymphocytopheresis. Appropriate culturing conditions used are for mammalian cells, which are typically carried out at 37 degrees C.

Various methods are also available for separating out and/or enriching cultures of unprimed CD8+ T cells. Some examples of general methods for cell separation include indirect binding of cells to specifically-coated surfaces. In another example, human peripheral blood lymphocytes (PBL) , which include CD8+ T cells, are isolated by Ficoll-Hypaque gradient centrifugation (Pharmacia, Piscataway, NJ) . PBL lymphoblasts may be used immediately thereafter or placed in stasis. Stasis may be achieved by storage in liquid nitrogen after freezing in a suitable medium, such as fetal bovine serum (FBS) containing about 10% DMSO (Sigma Chemical Co., St. Louis, MO) , which conserves cell viability and lymphocyte functions.

Alternative methods of separating out and/or enriching cultures of unprimed CD8+ T cells include both positive and negative selection procedures. For positive selection, after lymphocyte-enriched PBL populations are prepared from whole blood, sub-populations of CD8+ lymphocytes are isolated therefrom by affinity-based separation techniques directed at the presence of the CD8+ receptor antigen. These affinity-based techniques include flow microfluorimetry, including fluorescence-activated cell sorting (FACS) , cell adhesion, and like methods. (See, e.g., Scher and Mage, in Fundamental Immunology, W.E. Paul, ed., pp. 767-780, River Press, NY (1984).) Affinity methods may utilize anti-CD8+ receptor antibodies as the source of affinity reagent. Alternatively, the natural ligand, or ligand analogs, of CD8+ receptor may be used as the affinity reagent. Various anti-T cell and anti-CD8+ monoclonal antibodies for use in these methods are generally available from a variety of commercial sources, including the American Type Culture Collection (Rockville, MD) and Pharmingen (San Diego, CA) .

Negative selection procedures are utilized to effect the removal of non-CD8+ from the CD8+ population. This technique results in the enrichment of CD8+ T cells from the T- and B-cell population of leucophoresed patients. Depending upcn the antigen designation, different antibodies may be appropriate. (For a discussion and review of nomenclature, antigen designation, and assigned antibodies for human leucocytes, including T cells, see Knapp, et al. , Immunology Today 10: 253-258 (1989) and Janeway et al. , Immunobiology. supra. ) For example, monoclonal antibodies OKT4 (anti-CD., ATCC No. CRL 8002) OKT 5 (ATCC Nos. CRL 8013 and 8016), OKT 8 (anti-CD8, ATCC No. CRL 8014), and OKT 9 (ATCC No. CRL 8021) are identified in the ATCC Catalogue of Cell Lines and Hybridomas (ATCC, Rockville, MD) as being reactive with human T lymphocytes, human T cell subsets, and activated T cells, respectively. Various other antibodies .are available for identifying and isolating T cell species. CD8+ T cells can also be isolated by combining both negative and positive selection procedures. (See, e.g. Cai and Sprent, J. EXP. Med. 179: 2005-2015 (1994)). Prior to incubation of the antigen-presenting cells with the unprimed CD8+ T cells, an amount of antigenic peptide affixed to the artificial support is added to the macrophage cell culture, of sufficient quantity to become loaded onto the MHC class I molecules to be expressed on the surface of the antigen-presenting cells. The macrophages may be used immediately thereafter or placed in stasis. Stasis may be achieved by storage in liquid nitrogen after freezing in a suitable medium, such as FBS containing 10% DMSO (Sigma Chemical Co., St. Louis, MO).

Unprimed CD8+ T cells are then incubated in culture with the appropriate antigen-presenting cells for a time period sufficient to activate and further enrich fcr a population of CD8+ T cells. The ratio of unprimed CD8+ T cells to antigen-presenting cells may vary from individual to individual and may further depend upon variables such as the amenability of an individual's lymphocytes to culturing conditions and the nature and severity of the disease condition. The effector cell / antigen presenting cell culture may be maintained for as long a time as is necessary to activate and enrich for a population of a therapeutically useable or effective number of CD8+ T cells. In general terms, the optimum time is between about one and about five days, with a "plateau" -- i.e. a "maximum" specific CD8+ activation level -- generally being observed after five days of culture. Thereafter, the enriched and activated CD8+ T cells can be further purified by isolation procedures including site restriction, rosetting with antibody-red blood cell preparations, column chromatography and the like.

Following the purification, the resulting CD8+ T cell preparation can be further expanded by proliferation in culture for a period of time to obtain a population of 10^s activated CD8+ T cells. This period may vary depending on the replication time of the cells but may generally be 14 days. Activation and proliferation of CD8+ T cells has been described by Riddell et al. , Curr. Opin. Immunol. , 5: 484-491 (1993). Separation of CD8+ T cells from macrophages If desired, activated CD8+ T cells can be effectively separated from the antigen-presenting cells using one of a variety cf known methods. For example, monoclonal antibodies specific for the stimulator cells, for the peptides loaded onto the stimulator cells, or for the CD8+ T cells (or a segment thereof) may be utilized to bind their appropriate complementary ligand. Antibody-tagged cells may then be extracted from the stimulator-effector cell admixture via appropriate means, e.g., via well-known i munoprecipitation, FACS or immunoassay methods. Administration of Activated CD8+ T cells

Effective, cytotoxic amounts of the activated CD8+ T cells can vary between in vi tro and in vivo uses, as well as with the amount and type of cells that are the ultimate target of these killer cells. The amount will also vary depending on the condition of the patient and should be determined via consideration of all appropriate factors by the practitioner. Preferably, however, about I X 10⁶ to about 1 X 10², more preferably about 1 X 10⁸ to about 1 X 10¹¹, and even more preferably, about 1 X 10⁹ to about 1 X 10¹⁰ activated CD8+ T cells are utilized for adult humans, compared to about 5 X 10⁶ - 5 X 10⁷ cells used in mice. Preferably, as discussed above, the activated

CD8+ T cells are harvested from the macrophage cell culture prior to administration of the CD8+ T cells to the individual being treated. Methods of re-introducing cellular components are known in the art and include procedures such as those exemplified in U.S. Patent No.

4,844,893 to Honsik, et al.' and U.S. Patent No. 4,690,915 to Rosenberg. For example, administration of activated CD8+ T cells via intravenous infusion is appropriate.

Preferably, as discussed above, the activated CD8+ T cells are harvested from the macrophage cell culture prior to administration of the CD8+ T cells to the individual being treated. Methods of re-introducing cellular components are known in the art and include procedures such as those exemplified in U.S. Patent No. 4,844,893 to Honsik, et al. and U.S. Patent No. 4,690,915 to Rosenberg. For example, administration of activated CD8+ T cells via intravenous infusion is appropriate.

Alternatively, macrophages may be introduced into the patient to activate unprimed CD8+ T cells in vivo. Monocytes from the blood of the patient are cultured in vi tro together with bead-affixed peptide corresponding to the particular antigen. Macrophages differentiate from the monocytes in culture in the presence of the bead-affixed peptide. Macrophages are contacted with the bead-affixed peptides for a sufficient period of time for the macrophages to engulf the bead- affixed peptide and to present at least a portion of the peptide on the surface cf the macrophage. The macrophages are then preferably separated from the remaining beads and introduced into the patient to activate CD8+ T cells in vivo .

As noted previously, HLA haplotypes/allotypes vary from individual to individual and, while it is not essential to the practice of the present invention, it is often helpful to determine the individual's HLA type. The HLA type may be determined via standard typing procedures and the PBLs purified by Ficoll gradients. The purified PBLs would then be mixed with syngeneic macrophages preincubated with the appropriate antigenic peptides -- e.g., in therapeutic applications relating to viral infections, cancers, or malignancies, peptides derived from viral- or cancer-specific proteins.

In those instances in which specific peptides of a particular viral- or cancer-specific antigen have been characterized, the synthesized peptides encoding these epitopes will preferably be used. In cases in which the preferred antigenic peptides have not been precisely determined, protease digests of viral- or cancer-specific proteins may be used. As a source for such antigen, cDNA encoding viral- or cancer-specific proteins is cloned into a bacterial expression plasmid and used to transform bacteria, e.g., via methods disclosed herein.

EXAMPLES The following examples are intended to illustrate, but not limit, the present invention.

Example 1: The role of macrophages in the presentation of bead-affixed peptide The use of macrophages that had engulfed peptides affixed to a support as antigen presenting cells was demonstrated using spleen cell primary cultures as a source of unprimed CD8+ T cells. Cell culture C57BL/6 (B6, H-2^b) mice were bred under specific pathogen-free conditions. Unprimed CD8+ T Cells

Unfractionated spleen cells (SC) from unprimed B6 mice were used in assays of CD8+ T cell activation. Alternatively, adherent cells, including macrophages and dendritic cells, were removed by passing SC over nylon wool and incubating the remaining nonadherent SC in plastic petri dishes for 90 minutes at 37 degrees C.

Quadruple cultures consisted of 5 x 10⁶ SC or nonadherent SC of B6 unprimed mice and 1-2 x 10⁶ peptide-coated beads in 12-well plates (Corning, New York, NY) in 2 ml of RPMI-1640 with 10% FCS, penicillin, streptomycin, glutamine, 2-mercaptoethanol, non-essential amino acids, sodium pyruvate, HEPES. The primary activated CD8+ T cell cultures were incubated for 5 days at 37 degrees C, 5% CO-. Where applicable, 2 x 10⁵ macrophages per well were added to nonadherent SC cultures. On day 1, IL-2 was added all cultures to a final concentration of 10 U/ml. Low concentrations of IL-2 were included in the cultures to improve the magnitude and reproducibility of the induced responses. However, IL-2 was not required for specific priming of CDS- T cells, as shown in Example 2 and FIGURE 9, below. The use of IL-2 in final concentrations of about 2 U/ml to about 75 U/ml in cultures of CD8+ T cells is known in the art. See, for example, Riddell, S. et al. Science, 257:238-241 (1992) . A final concentration of 10 U/ml IL-2 is preferred. Macrophages Peritoneal exudate macrophage cells (PEM) were collected by rinsing the peritoneal cavity of the mouse twice with 2.5 ml ice-cold PBS. These cells were allowed to adhere to plastic culture plates for 90 minutes at 37 degrees C in RPMI-1640 with 8% FCS. Non-adherent cells were discarded. The adherent cells were detached with calcium-free PBS and resuspended in ice-cold RPMI-1640 with 1% BSA, collected by centrifugation and incubated with 10 μg/ml 33D1 mAb for 30 minutes at 4 degrees C. Low-tox-M rabbit complement, or EL-4 cells transfected with the nucleoprotein gene of VSV (Nl) , and 10 U/ml IL-2 in RPMI-1640 with 10% FCS, penicillin, streptomycin, glutamine, 2-mercaptoethanol (5.5 x 10^"5 M) , sodium pyruvate (1 mM) , non-essential amino acids (100 μM) , HEPES (10 mM) . EG70VA and Nl cells were maintained in RPMI-1640 with 10% FCS, penicillin, streptomycin, glutamine and 400 μg/ml G418 (Geneticin) . Peptides

Three K^b-restricted peptide antigens, VSV8 (an octamer activated CD8+ T cell epitope from vesicular stomatitis virus) , SEV9 (a nonamer activated CD8+ T cell epitope from Sendai virus) and 0VA8 (an octamer activated CD8+ T cell epitope from ovalbumin) were used to demonstrate peptide delivery through the phagocytic pathway. Peptides of optimal length are chosen preferably for binding class I in order to avoid the need for antigen processing and thereby optimizing loading and presentation processes.

Peptides were synthesized on an Applied Biosystems 431A synthesizer with amino- and carboxyl- ends. All peptides were purified with C18 reversed-phase HPLC.

The sequences of the peptides are: OVA8 (SEQ ID NO:l) :

Ser lie lie Asn Phe Glu Lys Leu 1 5

OVA10N (SEQ ID NO: 2) :

Leu Glu Ser lie lie Asn Phe Glu Lys Leu 1 5 10

OVA10C (SEQ ID NO: 3) : Ser lie lie Asn Phe Glu Lys Leu Thr Glu

1 5 10

OVA24 (SEQ ID NO: 4) :

Glu Gin Leu Glu Ser lie lie Asn Phe Glu Lys Leu Thr Glu Trp Thr 1 5 10 15 Ser Ser Asn Val Met Glu Glu Arg

20 OVA35 (SEQ ID NO: 5) :

Leu Val Leu Leu Pro Asp Glu Val Ser Gly Leu Glu Gin Leu Glu Ser 1 5 10 15 lie He Asn Phe Glu Lys Leu Thr Glu Trp Thr Ser Ser Asn Val Met

20 25 30

Glu Glu Arg 35 VSV8 (SEQ ID NO:6) : Arg Gly Tyr Val Tyr Gin Gly Leu 1 5

SEV9 (SEQ ID NO:7) : Phe Ala Pro Gly Asn Tyr Pro Ala Leu 1 5

SEV11N (SEQ ID NO: 8) :

Gly Glu Phe Ala Pro Gly Asn Tyr Pro Ala Leu 1 5 10

SEV11C (SEQ ID NO: 9) :

Phe Ala Pro Gly Asn Tyr Pro Ala Leu Trp Ser 1 5 10

About 100 x 10^s sulfate latex beads with a diameter of 6.76 μm (IDC, Portland, Oregon) were incubated for 18 hours at 4 degrees C with 1 mM or 10 μM peptide for OVA8 , VSV8 or SEV9 or 100 μM of the peptide for the other peptides in 1 ml PBS under continuous rotation. After 3 washes with PBS, the pellets containing the peptide-coated beads were placed in an acid-washed ignition tube (10 x 7 mm) and 200 μl of 6N HCl and 2 μl phenol was added. The tubes were alternately purged with argon and evacuated three times to remove dissolved oxygen. The tubes were then sealed under a vacuum of 20-50 microns and placed in a heating block at 110 degrees C for 20-24 h. The sealed tubes were opened and the sample dried under vacuum. The hydrolyzed samples were reconstituted in 250 μl of sodium citrate dilution buffer (pH 2.2, 0.2 N, Beckman) . A 50 μl aliquot was analyzed on a Beckman Model 6300 High Performance

Analyzer according to the methods of Spackman et al . , Anal . Chem . 30: 1190 (1958) . The instrument has been modified with a 25 cm long cation exchange column (P/N 38049) for improved resolution of amino acids. The binding of OVA8, SEV9 or VSV8 was comparable: after incubating 100 x 10⁶ beads with 1 ml containing 1 mM, 100 μM or 10 μM peptide, 50 nmol, 10 nmol and 5 nmol respectively were bound. This was confirmed by binding studies with radioactively labeled peptide, as described. The amount of ovalbumin protein bound to beads was determined by the difference in optical density between the solution before and after incubation with beads. Cell lysis assay Varying numbers of effector cells were added to

5,000 Na₂ ⁵¹Cr0₄ (⁵¹Cr) labeled target cells and incubated for 5 hours at 37 degrees C. RMA-S target cells were labeled for 1.5 hours and IC-21 target cells for 18 hours at 37 degrees C with 100 μCi ⁵¹Cr per 1 x 10⁶ cells. The percentage of specific ^≡1Cr release was calculated as:

Specific lysis = 100 x \^E~B\

( T-B)

where E is ⁵¹Cr release in counts per minute (cpm) in the experimental well, 23 is background ⁵¹Cr release in cpm, and T is ⁵¹Cr release in cpm after 2% Triton X-100 treatment. Background (medium) release was always less than 20% of maximal (2% triton X-100) release. The standard error of the mean of triplicate cultures was less than 5% of the specific ⁵¹Cr release.

The number of effector cells giving half- maximal specific lysis was determined from a plot of effector/target ratio v. percent specific lysis and divided into 10⁶ to yield the number of lytic units per 10⁶ effector cells. Inhibitors and enzymes

Brefeldin A (Epicentre Technologies) was added at 4 μg/ml to cells in culture medium at 37 degrees C for 30 minutes before and during pulsing, with the antigen or during a recovery phase after pronase treatment. Cytochalasin D (Sigma) was added at 10 μg/ml to cells in culture medium at 37 degrees C for 30 minutes before and during pulsing with antigen. Cytochalasin D-treated macrophages that were used as stimulator cells in primary activated CD8+ T cell responses were fixed with 1% paraformaldehyde. Macrophages that served as targets for activated CD8+ T cell clones were not fixed, because no ^Ξ1Cr is released by fixed cells. Brefeldin A (at 4 μg/ml) was present during the "Cr-release assay, however cytochalasin D was omitted at this stage because it inhibits the lytic machinery of activated CD8+ T cell. In order to strip glycoproteins from the cell surface of lymphocytes, pronase treatment was used. Briefly, cells were washed and resuspended in DMEM with 20 mM HEPES. One tenth of volume of a 50 mg/ml solution of pronase (Calbiochem-Behring, La Jolla, CA) was added and incubated for 1 hour at 8 degrees C. Digestion was terminated by addition of 25% FCS and 20 mM EDTA, followed by three washes in medium with 10% FCS. Recovery of cell surface expression was carried out in RMPI-1640 with 10% FCS, penicillin, streptomycin and glutamine for 3 hours at 37 degrees C. The efficacy of removal and recovery of cell surface K^b molecules was checked by immunofluorescence. After pronase treatment the mean fluorescence of cell surface K^b on IC-21 cells decreased from 32 ± 5 to 9 ± 3 and increased to 25 ± 4 after 3 hours recovery at 37 degrees C (the mean of the negative control was 6 ± 2) . Detection of empty class I molecules

Cells were cultured for 18 hours at 37 degrees C or 22 degrees C in HL-1 medium in the absence or presence of 100 μM 0VA8 peptide. Then, they were washed in FACS-buffer and spun down at 4 degrees C for FACScan analysis.

Macrophages were capable of producing primary in vi tro activated CD8+ T cell responses against specific peptides presented as peptide-coated beads. B6 unprimed SC or SC without adherent cells were cultured with beads coated with OVA8, VSV8 or SEV9 peptide, corresponding to 1 μM in culture, and 10 U/ml IL-2. After 5 days the effectors were tested for lysis of RMA-S targets without or with different peptides (FIGURES 1-3) .

Unfractionated SC or nonadherent SC were cultured without or with supplementation with macrophages from the peritoneal cavity (PEM) and bead-affixed peptide, corresponding to 100 nM in culture, and 10 U/ml IL-2 for five days (FIGURES 4,5). These primary activated CD8+ T cell cultures were tested by lysis of RMA-S targets with the relevant peptide, OVA8 and VSV8 respectively.

Macrophages, 1 x 10^s either PEM or IC-21 cells were treated with cytochalasin D (cyto D) before and during exposure to 2 nmol of peptide coated on 10 x 10⁶ beads in 2 ml and fixed prior to culture with 20 x 10⁶ nonadherent spleen cells in 8 ml. Cytochalasin D blocks polymerization of actin filaments and thereby blocks phagocytosis, The cytolytic activity was tested on RMA-S targets loaded with the inducing peptide (FIGURE 6) . Depleting adherent cells (macrophages and dendritic cells) from the spleen cell cultures decreased the lytic capacity four- to five-fold (FIGURES 1 - 5) . FIGURES 1-3 also show the peptide specificity of the primary activated CD8+ T cells induced using bead-affixed peptide. The activated CD8+ T cell activity of cultures depleted of adherent cells can be restored by addition of either purified macrophages from a peritoneal lavage of B6 mice (PEM, FIGURES 4-6) or IC-21 cells, a B6 macrophage cell line (FIGURE 6) .

Introduction of PEM from Balb/c mice or J774, a Balb/c macrophage cell line, did not restore response induction in cultures of unprimed B6 lymphocytes deficient of adherent cells (FIGURE 7) . The unprimed CD8+ T cells were obtained from nylon-wool-passed SC cells from unprimed B6 mice. The antigen was VSV8 peptide affixed to beads and presented at a final concentration in culture of 100 nM. The target cells were RMA-S cells loaded with VSV8 peptide.

Supplementation of nylon-wool-passed SC cells with a different haplotype, either PEM from Balb/c mice or J774, a Balb/c macrophage cell line, did not activate unprimed CD8+ T cells. Therefore, macrophages are directly stimulating unprimed CD8+ T cells.

When the B6 macrophages were treated with cytochalasin D (a potent inhibitor of phagocytosis) before and during incubation with peptide-coated beads, activated CD8+ T cell response induction was greatly inhibited (FIGURE 6) . It is believed that the presentation of the peptides to unprimed CD8+ T cells is mediated by macrophages in a process that requires phagocytosis of the beads and loading of MHC class I in the phago(lyso) somal pathway. By depleting adherent cells from the splenic cell (SC) responder population, we removed not only macrophages but also dendritic cells. We believe that involvement of dendritic cells in the antigen-presenting system of the present invention is unlikely because treatment of the adherent cell fraction of peritoneal exudate cells with 33D1 antibody plus complement treatment (in order to deplete dendritic cells) did not reduce primary activated CD8+ T cell response induction in vi tro against bead-affixed peptide. Furthermore, activation of purified unprimed CD8+ T cells against bead-affixed peptide could be established with a macrophage cell line, and optimal presentation of bead-affixed peptide required phagocytosis, a characteristic closely associated with macrophages but not with cultured dendritic cells. Together, these observations lead us to conclude that macrophages can efficiently present bead-affixed peptide to unprimed CD8+ T cells.

Example 2: Effect of Interleukin-2 Primary spleen cell cultures from unprimed B6 mice were depleted of macrophages by passing the cells over nylon wool as described in Example 1. Similarly, macrophages (PEM) were purified from a peritoneal lavage of B6 mice. OVA8 and VSV8 peptides coated on beads were presented in culture at a concentration of 100 nM. Culture conditions, treatment protocols, lysis assay and analysis were the same as used in Example 1, except that interleukin-2 was not added to the media used in studies shown in FIGURE 9. Comparison of FIGURES 8 and 9 reveals that activated CD8+ T cell responses primed in absence of IL-2 are specific and dose-dependent, albeit somewhat reduced in magnitude. The presence of IL-2 is thus not required for the specificity of the primed activated CD8+ T cell response, but does increase the magnitude of the response.

Example 3: Effect of depletion of CD8+ T cells

T cell depletion studies investigated the role of various cell types in the presentation of bead-affixed peptide in splenocyte cultures. Depletion of CD8+ T cells, but not CD4+ T cells, abolished activated CD8+ T cell activity (FIGURE 10) , consistent with the fact that class I-restricted peptides and IL-2 were used in these primary activated CD8+ T cell cultures. Cell types and culture conditions were the same as used in Example 2. The macrophage-induced activated CD8+ T cell responses were dependent on the presence of CD8+ T cells, consistent with the fact that^' the peptides on the beads contain activated CD8+ T cell epitopes. Example 4 : Effect of Beads in Primary activated CD8+ T cell Response Induction in Spleen Cell Cultures Cell types and culture conditions were the same as used in Example 1. The amount of peptide bound to beads was determined by amino acid analysis after acid hydrolysis. FIGURES 11 and 12 show a comparison of the use of bead-affixed versus free peptide for activate activated CD8+ T cell in vi tro in splenocyte (SC) cultures from unprimed C57BL/6 (B6) mice in the presence of 10 U/ml IL-2. Low concentrations of IL-2 were included in the cultures to improve the magnitude and reproducibility of the induced responses. However, as shown in FIGURE 10 and above, specific responses were obtained in studies in which IL-2 was not included. Half-maximal VSV8- and SEV9-specific activated

CD8+ T cell responses required one hundredfold less bead-affixed than free peptide: 1 μM of free peptide (FIGURE 11) compared to 10 nM of bead-affixed peptide (FIGURE 12) . Bead delivery was less important for stimulating OVA8-specific activated CD8+ T cell, improving the efficiency of this process only tenfold (from 100 nM free peptide to 10 nM bead-affixed peptide required for half-maximal activated CD8+ T cell response induction) , consistent with the unusual potency of free OVA peptide to stimulate unprimed CD8+ T cells in vi tro . Taken together, these results show that bead-affixed peptides are presented more efficiently than free peptides to unprimed CD8+ T cells in splenocyte cultures. After harvesting the effector activated CD8+ T cell from the 5 day splenic cultures with peptide-coated beads, the presence of adhered, bead-filled macrophages was apparent, suggesting that bead-affixed peptide may be preferentially delivered to macrophages.

To determine whether improved presentation efficiency of bead-affixed peptide was a result of delivery of peptide via beads to APC or a consequence of the use of latex beads per se, activated CD8+ T cell response induction by bead-associated peptide was compared with activated CD8+ T cell response induction by free peptide plus uncoated beads. Peptide concentrations were used that allowed activated CD8+ T cell response induction with bead-affixed peptide but not or εuboptimally with free peptide, i.e. 100 nM of OVA8 or VSV8 or SEV9. FIGURE 13 shows that uncoated beads did not improve primary activated CD8+ T cell responses against free peptide, demonstrating the requirement for delivery of peptide via beads to increase activated CD8+ T cell response induction at low antigen concentration. 1 nmol of peptide coated on 4 x 10⁶ beads (which corresponds to 100 nM in culture) or 100 nM free peptide was added to a culture consisting of 20 x 10⁶ SC of B6 unprimed mice in 10 ml of medium with 10 U/ml IL-2. In addition, 100 nM of free peptide was added to a culture with 20 x 10⁶ SC of B6 unprimed mice containing 10 U/ml IL-2 after 4 x 10⁶ beads were allowed to be internalized by macrophages. The effector cells were harvested after 5 days and tested on RMA-S targets loaded with the relevant peptide. Cytolytic activity is expressed as lytic units per 10⁶ effector cells, calculated from individual dose-response curves. Example 5: The macrophage cell line IC-21 expresses few empty cell surface MHC class I K^b molecules

MHC class I molecules survive longer at the cell surface when bound to high affinity class I-binding peptides. Similarly, the lifespan of unstable MHC class I molecules that are empty or with low affinity peptides can be increased by incubation at low temperature (22 degrees C) . Thus, increases in the level of class I expression at the cell surface resulting from incubation of cells with high affinity peptide and/or culturing cells at reduced temperature can be used as a measurement of the number of MHC class I molecules that arrive at the surface either empty or filled with low affinity peptides.

K^b stabilization assay RMA-S cells were cultured for 24 hours at 22 degrees C in HL-1 medium (Hycor Bio edical Inc, Portland, Maine) at 10⁷ cells per ml and subsequently incubated with 100 μM peptide in KL-1 medium with 4 μg/ml brefeldin A at 22 degrees C for 30 min. The cells were washed three times with ice-cold HL-1 medium, resuspended in HL-1 medium with 4 μg/ml brefeldin A and transferred to 37 degrees C for different time periods. At the end of each incubation period, 10-fold excess of ice-cold FACS-buffer (PBS with 0.5% BSA and 0.03% NaN₃) was added and the cells were spun down at 4 degrees C for FACscan analysis.

Using this method we compared IC-21 macrophages, and the TAP-mutated cell line, RMA-S and the cell line EL-4. FIGURE 14 shows that incubating IC-21 macrophage cells for 18 hours in media containing high levels of OVA8 peptide did not result in a significant increase in cell surface expression levels of K^b. RMA-S, EL-4 and IC-21 cells were cultured for 18 hours at 37 degrees C or 22 degrees C in the absence or presence of 100 μM OVA8 peptide. The cells were then stained with anti-K^b (Y-3) and FITC-labeled F(ab')₂ goat anti-mouse Ig and analyzed on a FACScan flow cytometer. The same treatment resulted in a modest increase in K^b surface expression on EL-4 and a relatively dramatic increase on RMA-S cells, as predicted. These data show that IC-21 macrophage cells express few rescuable, and presumably empty, K^b molecules at their cell surface. The observations that 10-100 fold less peptide is required for primary activated CD8+ T cell response induction with bead-affixed peptide than with free peptide plus the requirement for phagocytosis in this process, indicates that exchange of peptide from the bead to cell surface class I molecules without an internalization event is unlikely to play a significant role in generating MHC/peptide complexes on IC-21 cells that stimulate unprimed CD8+ T cells.

Example 6: Effect of peptide length on the off-rate from cell surface K^b molecules and presentation when delivered in the phagocytic pathway Small extensions to optimal peptides have been shown to affect the affinity of the peptide for soluble class I molecules in vi tro. For instance, a peptide extended with 2 amino acids at the amino-terminus (OVA10N) has a 76-fold lower affinity for K^b than the optimal peptide (OVA8) , but the affinity of a peptide 2 amino acids longer at the carboxy-terminus (OVA10C) is only 4-fold decreased compared to the optimal peptide (OVA8) . In general, it has been proposed that peptides with a less than optimal fit for MHC class I molecules, e.g. that are too long, are less efficient than optimal peptide at inducing in vi tro activated CD8+ T cell responses. Below are the studies for the likely half-lives of various peptide/class I complexes at the cell surface and the requirement for optimal length peptides in phagocytic loading of class I. FACS analysis

For staining, 2 x 10⁵ cells were washed in FACS-buffer and incubated with 1:100 diluted anti-K^b (Y-3, ATCC HB176) ascites on ice. After 30 minutes the cells were washed and incubated with 10 μg/ml FITC-labeled goat anti-mouse Ig (Cappel, Durham, NC) for 30 minutes on ice. Control samples were stained with the secondary reagent only. Cells were washed and passed through a FACScan flow cytometer (Becton Dickinson, Mountain View, CA) . To measure peptide off-rates, the half-lives of homogeneous class I-peptide populations are first measured at 37 degrees C. The RMA-S cell line which expresses at its cell surface large amounts of empty class I which are relatively stable at reduced temperature. RMA-S cells were cultured at reduced temperature for 16 hours prior to a 30 minutes exposure to optimal and extended OVA and SEV peptides. After washing, the cell were transferred to 37 degrees C in serum-free medium for different time periods. Brefeldin A, which effectively blocks transport from the ER of newly synthesized proteins, was included in this step to minimize any contribution of nascent K molecules. Cells were then stained for surface expression of K^b molecules. Optimal length peptides (OVA8 and SEV9) fully stabilized the K^b molecules for at least 30 min, while the empty MHC class I molecules disintegrated within 5 minutes. FIGURE 15 shows the stabilization of cell surface K^b on RMA-S by a peptide of optimal length (OVA8) and peptides with 2 amino acid additions at the amino- or carboxy-terminus (OVA10N or OVA10C) . FIGURE 16 shows the stabilization of cell surface K^b on RMA-S cells by a peptide of optimal length (SEV9) and peptides with 2 amino acid additions at the amino- or carboxy-terminus (SEV11N or SEV11C) .

RMA-S cells were cultured at 22 degrees C for 24 hours and subsequently incubated with 100 μM peptide for 30 minutes at 22 degrees C in the presence of brefeldin A. The cells were washed at 4 degrees C and transferred to 37 degrees C in the presence of brefeldin

A for different time periods, before washing and staining with anti-K^b Ab (Y-3) and FITC-labeled F(ab')₂goat anti-mouse Ig. Immunofluorescence was determined on a FACScan flow cytometer. RMA-S cells were cultured at 22 degrees C for 24 hours and subsequently incubated with 100 μM peptide for 30 minutes at 22 degrees C in the presence of brefeldin A. The cells were washed at 4 degrees C and transferred to 37 degrees C in the presence of brefeldin A for different time periods, before washing and staining with anti-K^b Ab (Y-3) and FITC-labeled F(ab')₂goat anti-mouse Ig. Immunofluorescence was determined on a FACScan flow cytometer. Interestingly, the half-life of the extended

OVA peptides was about 20 minutes, whether the peptide was 2 amino acids longer at the amino- or at the carboxy-terminus. Extended SEV peptides stabilized the K^b molecules only slightly, with a half-life between 5 and 10 minutes. Clearly, there is a significant discrepancy between the peptide affinity measurement carried out in vi tro and the off-rate measurement described here. These differences most likely reflect the fact that the in vi tro studies were carried out under equilibrium conditions.

Conditions at the cell surface and the phagosome are also quite different, both in terms of antigen processing capabilities and concentrations of peptide and β₂microglobulin. It is reasonable to assume that the rules for peptide loading in these two locations may also be different. Delivery of peptide via microscopic beads was studied to determine if it has any effect on the ability of a sub-optimal peptide to be presented by class I and to stimulate primary activated CD8+ T cell responses.

The abilities of OVA (FIGURE 17) and SEV (FIGURE 18) peptides, free or bead-affixed, to activate CD8+ T cells were compared with those of peptides with 2 more amino acids at the carboxy terminus (OVA10C and SEV11C) at high concentrations of peptide (1 μM and 10 μM) that allowed maximal activation with optimal length peptide. Spleen cells of unprimed B6 mice were cultured for 5 days with either 10 μM free peptide or 2 nmol of bead-affixed peptide in 8 ml (corresponding to 250 nM) in the presence of 10 U/ml IL-2. The effector cells were tested on RMA-S cells coated with OVA8 or SEV9.

Peptides that are two amino acids longer (OVA10C and SEV11C) induced higher activated CD8+ T cell responses in bead-affixed form than unbound at a 40-fold lower peptide dose (250 nM of bead-affixed peptide versus 10 μM free peptide) . The most plausible explanation for these findings is that the phagosome has antigen processing capabilities and that these slightly too long peptides are trimmed into optimal peptides. In addition, binding of peptides to K^b molecules in the phago(lyso) some may be more efficient than at the cell surface. The small volume between bead and phagolysosomal membrane is likely to promote peptide exchange from the bead onto the class I molecule. We conclude that there are significant differences in peptide loading at the cell surface (with free peptide) versus peptide loading on class I in phagocytic compartments (with bead-affixed peptide) , both in terms of processing and capacity to stimulate unprimed CD8+ T cells. Example 7: Primed activated CD8+ T cell clones are at least 10, 000-fold more sensitive to peptide than unprimed

T lymphocytes Several studies have indicated the importance of class I-peptide density for activation of CD8+ T cells. CD8+ T cells that meet antigen on the APC for the first time can only effectively respond at high doses of the antigen, whereas primed T lymphocytes require much lower doses of antigen for activation. Free peptide was titrated onto RMA-S cells or IC-21 macrophages and tested for recognition by an OVA-specific activated CD8+ T cell clone at E/T ratio of 5. 10 fM of OVA8 on RMA-S or IC-21 targets was required for -50% of target cell lysis by the OVA activated CD8+ T cell clone (FIGURE 19) . Similarly, half-maximal lysis by the VSV specific activated CD8+ T cell clone was obtained at 10 pM VSV8 on RMA-S and 200 pM VSV8 on IC-21 target cells (FIGURE 20) . In comparison to unprimed CD8+ T cells, activated CD8+ T cell clones are at least 10, 000-fold more sensitive to peptide (compare FIGURES 4 and 5 to 19 and 20) . Delivery of peptide via beads to IC-21 cells improved the presentation to the VSV-specific activated CD8+ T cell 5-10 fold. In contrast, delivery via beads did not further improve the presentation of OVA8 peptide by IC-21 to the OVA-specific activated CD8+ T cells. Previous work has shown that a density of relevant class I-peρtide on the target cells of 200-500 complexes is sufficient for activation of primed activated CD8+ T cell. The difference between the various peptides we have analyzed for phagocytic peptide loading is presumably a reflection of quantitative differences in the density of specific class I-peptide complexes needed for activated CD8+ T cells relative to unprimed CD8+ T cells. As a general rule, activated CD8+ T cell clones are several orders of magnitude more sensitive than unprimed CD8+ T cells.

Example 8: Limited antigen processing occurs in the phagocytic pathway In order to analyze antigen processing in the phagocytic pathway for antigen presentation by MHC class I we compared primary activated CD8+ T cell response induction by beads coated with the optimal peptide (OVA8) , coated with longer peptides that encompass the OVA8 sequence (OVA24, OVA35) or coated with ovalbumin protein (FIGURES 21-23) . SC of unprimed B6 mice were stimulated for 5 days with three different doses of 0VA8 or ovalbumin that were administered to the culture either unassociated or associated with beads in the presence of 10 U/ml IL-2 in 8 ml. The effectors were harvested and tested for lysis of ⁵¹Cr-labeled RMA-S cells loaded with OVA8 peptide at E/T ratio 50. In FIGURE 22, stimulation of unprimed B6 spleen cells with beads coated with 0VA8, a 24-mer peptide (OVA24) or a 35-mer peptide (OVA35) containing the minimal activated CD8+ T cell epitope, the OVA8 sequence, or with the ovalbumin protein. 25 x 10⁶ beads were coated in 1 ml containing 100 μM peptide or protein overnight and washed three times before use. Quadruple cultures of 2 x 10^δ coated beads and 5 x 10^s unprimed B6 SC were incubated at 37 degrees C for 5 days in the presence of 10 U/ml IL-2. The effectors were harvested and tested for cytotoxicity on ⁵¹Cr-labeled RMA-S cells loaded with OVA8 peptide at different E/T ratios. In FIGURE 23, 25 x 10^€ beads were coated with three different doses of antigen, as indicated in the figure and incubated with 2.5 x 10^s IC-21 cells in 2.5 ml for 4 hours at 37 degrees C. In addition, free OVA8 peptide or ovalbumin protein were incubated with 2.5 x 10⁶ IC-21 cells at the same concentrations as bead-affixed antigen for the same time period. These cells were used in a ⁵¹Cr release assay as targets for the OVA specific activated CD8+ T cell clone at E/T ratio 10.

125 nM of free OVA8 (equivalent to 1 nmol affixed to beads) allowed half-maximal primary activated CD8+ T cell response induction (FIGURE 21) . The same OVA8 dose administered via beads to the culture induced a maximal activated CD8+ T cell response. On the other hand, introduction of free ovalbumin to the cultures did not generate primary antigen-specific activated CD8+ T cell responses (FIGURE 21) . Indeed, addition of up to 10 μM of free ovalbumin in the culture did not result in a primary antigen specific activated CD8+ T cell response. Even at the highest dose of ovalbumin protein possible, 1 nmol (a dose limited by the binding capacity of the beads and number of beads that could be added to the culture) , we were only able to obtain a slight OVA8-specific activated CD8+ T cell response (FIGURE 21) . Similarly, coating beads with long peptides, OVA24 and OVA35 sequences, resulted in very poor primary activated CD8+ T cell responses (FIGURE 22) . As can be seen, when bead-affixed whole protein was incubated with IC-21 cells, recognition did take place by the 0VA8-specific activated CD8+ T cell clone (Fig 22) . The association of whole protein with the beads greatly increased target cell sensitization. After incubation with free ovalbumin protein in concentrations up to 50 μM ovalbumin protein, recognition was poor, whereas bead-affixed ovalbumin sensitized IC-21 cells at nanomolar concentration (FIGURE 23) . Therefore, it appears that limited processing of antigen within the phago(lyso)some can occur. Sufficient levels of peptide-class I complexes are generated at the cell surface for recognition by activated CD8+ T cells, but this level is insufficient to activate unprimed CD8+ T cells. Presentation of bead-affixed ovalbumin protein was not detected by in vi tro activation assay, whereas the presentation of bead-affixed peptide was detected. This finding suggests that the present invention which allows for circumventing the requirement for antigen processing results in more efficient loading of MHC class I molecules with phagocytosed antigen.

Maximum T cell activation requires a combination of signals, partly from the T cell receptor and partly from ligands of costimulatory/adhesion molecules. One study has shown that activation of macrophages, e.g. by ingestion of microbial debris, induced costimulatory activity. However, as shown, the ingestion of uncoated beads did not significantly improve the presentation of free peptide, suggesting that uncoated beads were not as effective as microbial debris in inducing costimulatory activity.

Incubating macrophages in a high concentration of MHC class I binding peptide that is an optimal length has little effect on the steady state expression level of MHC class I surface expression. These data suggest that cell surface class I loading of peptide is inefficient on macrophages. Interestingly, loading MHC class I in the phagocytic pathway also had little measurable effect on the overall surface MHC class I expression levels, although it was apparently efficient compared to cell surface loading in that it required 10-100 fold less peptide to induce primary activated CD8+ T cell responses.

Example 9: Optimal presentation of bead-affixed peptide by macrophages requires phagocytosis and nascent class I molecules To investigate pathways involved in the presentation of bead-affixed peptide, the macrophage cell line IC-21 was treated with brefeldin A or cytochalasin D and used these cells as a target for the OVA- and VSV-specific activated CD8+ T cell. Brefeldin A effectively blocks transport from the ER of newly synthesized proteins. Under these conditions, the concentration used completely prevented presentation of VSV antigen in infected cells to the VSV8 K^b-restricted activated CD8+ T cell clone.

Inhibition of phagocytosis by cytochalasin D was scored microscopically by the absence of internalized beads. Recognition and lysis of IC-21 cells that had been incubated with bead-affixed peptides and were treated with cytochalasin D or brefeldin A was diminished by 50% or more (FIGURES 26 and 27) . In these studies, 2 x 10^s IC-21 cells were cultured in 2 ml for 18 hours at 37 degrees C with 20 x LO⁶ beads coated with 2 nmol of OVA8 peptide (corresponding to 1 μM in culture) in the presence or absence of 4 μg/ml brefeldin A (bref A) or 10 μg/ml cytochalasin D (cyto D) , all in the presence of ⁵¹Cr. The adherent cells (i.e. viable cells) were harvested, extensively washed and incubated with the OVA specific activated CD8+ T cell clone in a ⁵¹Cr release assay for 5,hours at 37 degrees C.

These results demonstrate a role for both phagocytosis and nascent MHC class I molecules in the presentation of bead-affixed peptide by macrophages. Presentation of free peptide by IC-21 cells to the activated CD8+ T cell clones was not influenced by cytochalasin D or brefeldin A (FIGURES 24 and 25) .

In these studies, IC-21 cells were cultured (2 x 10^s in 2 ml dishes for 18 hours at 37 degrees C in the presence of ⁵¹Cr) with 1 μM OVA8 in the presence or absence of 4 μg/ml brefeldin A (bref A) or 10 μg/ml cytochalasin D (cyto D) . The viable cells (i.e. those cells adhering) were harvested, extensively washed and incubated with the OVA specific activated CD8+ T cell clone in a ⁵¹Cr release assay for 5 hours at 37 degrees

C. In the studies shown in FIGURE 25, 2 x 10⁶ IC-21 cells were cultured in 2 ml for 18 hours at 37 degrees C with 1 μM VSV8 in the presence or absence of 4 μg/ml brefeldin A

(bref A) or 10 μg/ml cytochalasin D (cyto D) , all in the presence of ^E1Cr. The adherent cells (i.e. viable cells) were harvested, extensively washed and incubated with the VSV specific activated CD8+ T cell clone in a ⁵¹Cr release assay for 5 hours at 37 degrees C.

Cytochalasin D could not be included in the 4 hour cytotoxicity assay, because this reagent interferes with activated CD8+ T cell effector function. Thus it is possible that some of the remaining activated CD8+ T cell activity observed against IC-21 cells incubated with bead-affixed peptide and treated with cytochalasin D was due to recovery of the phagocytic machinery. Furthermore, it is possible that in both cytochalasin D and brefeldin A-treated cells, MHC class I molecules were loaded at the cell surface with peptide that detached from the beads extracellularly, an inefficient process that could still be sufficient for detection by the highly sensitive activated CD8+ T cell clones. Example 10: Class I-peptide complexes reappear at the cell surface of macrophages containing bead-affixed peptide after stripping cell surface class I molecules Intracellular trafficking of class I complexed with bead-delivered peptide was investigated by removing cell surface class I molecules and following re-expression of class I-peptide complexes at the cell surface. IC-21 cells were incubated with free peptide (FIGURES 26 and 30) or with bead-affixed peptide (FIGURE 29 and 31) , prior to stripping the cell surface of IC-21 cells with pronase. Thereafter, the cells were washed, allowed to recover for three hours and used as targets for the peptide-specific activated CD8+ T cell clones. There were three phases of treatment. In phase

1, 2 x 10⁶ IC-21 cells were labeled overnight in the presence of 1 μM peptide in 2 ml (1st) . VSV8 was used in FIGURE 28 and OVA8 was used in FIGURE 30. In the second phase, on the next day the cells were washed or pronase-treated (2nd) . In the third 'recovery phase'

(3rd) , medium only or medium with 10 μM peptide was added before using these cells as targets for the VSV-specific activated CD8+ T cell line at an E/T ratio of 5. In FIGURE 30, in the 'recovery phase', medium only or medium with 10 μM peptide was added with or without 4 μg/ml brefeldin A before using these cells as targets for the OVA-specific activated CD8+ T cell clone.

Similar studies (shown in FIGURES 29 and 31) demonstrated the effect of presenting peptides affixed to beads. In phase l, 2 x 10⁶ IC-21 cells were labeled overnight in the presence of 2 nmol of peptide coated on 20 x 10⁶ beads in 2 ml (1st) . VSV8 was used in FIGURE 30 and OVA8 was used in FIGURE 31. On the next day the cells were washed or pronase-treated (2nd) . In the 'recovery phase' (3rd), medium only or medium with 10 μM peptide was added before using these cells as targets for the VSV-specific activated CD8+ T cell line at an E/T ratio of 5. In FIGURE 31, in the 'recovery phase', medium only or medium with 10 μM peptide was added with or without 4 μg/ml brefeldin A before using these cells as targets for the OVA-specific activated CD8+ T cell clone.

After pronase treatment and recovery in medium, half-maximal activated CD8+ T cell recognition was observed when cells were pre-incubated with bead-affixed peptide (FIGURE 29 and 31, filled squares) . In contrast, cells pre-incubated with free peptide were not recognized (FIGURE 28 and 30, filled squares) . This indicates that an intracellular pool of peptide is present in macrophages after ingesting peptide affixed to beads.

When free peptide was given to the cells during the recovery phase, the targets were optimally recognized, irrespective of the primary peptide administration route (filled triangles) . This demonstrates that new MHC class I molecules were expressed on the cell surface. Furthermore, this process was completely blocked when brefeldin A was included during the recovery stage (open triangles) . The inhibitory effect of brefeldin A was also observed on presentation of bead-affixed peptide after pronase 96/37107 PCΪ7US96/07436

- 47 -

treatment (FIGURE 31, open squares) . These results indicate a role for newly synthesized class I molecules in presenting peptides affixed to beads that have been internalized into phagosomes. The present invention discloses that short synthetic peptides affixed to beads are more than ten times more effective than ovalbumin attached to beads in eliciting T cell responses. In fact, the peptide of the present invention is short, preferably from about eight to about thirty-five amino acid residues long.

The present invention differs from the report by Harding and Song, Id. , in the amount of peptide that could be loaded onto microscopic beads. Harding and Song report that typically 0.2 to 1 ng of ovalbumin was attached to 10⁶ beads that were 1 μm in diameter. The present invention discloses that 5 - 50 nmol of OVA8, VSV8 or SEV9 was typically affixed to 100 x 10⁶ beads that were 6.76 μm in diameter. This corresponds to about 54 to 540 ng per 10⁶ beads for the octapeptides and 55 to 550 ng per 10⁶ beads for the nonapeptide. The present invention thus loads about 250 fold to as much as 2750 fold more antigen per bead. The difference in the size of the beads would predict only a 45.7 fold increase, and does not account for the superior loading of antigen on beads in the present invention.

The present invention does not require transfer of substantial amounts of peptide from the bead surface to nascent class I molecules. These studies delineate the pathway by which MHC class I are loaded with peptide coated onto beads and highlight the importance of phagocytosis and nascent MHC class I molecules for this process. Taken together with previous data on the antigen routes from the phagosome to the cell surface, it is believed that the pathway starts with nascent MHC class I molecules loaded with either high and low affinity peptides exiting the endoplasmic reticulum ^■ (ER) . After passage through the Golgi complex, the MHC class I molecules loaded with peptides are targeted equally efficiently to the cell surface and the newly created phagosomal membrane. Those MHC class I molecules complexed with low affinity peptide and arriving directly at the cell surface have a short half-life. They rapidly disintegrate because β₂ icroglobulin and peptide concentrations are low. In contrast, MHC class I molecules complexed to low affinity peptides that are delivered to the bead-induced phagosome may be rescued by high affinity peptide released from the bead. This peptide exchange process may be aided by relatively high concentrations of β₂ microglobulin and peptide in the narrow space between the bead and the phagosomal membrane. Once formed, these rescued complexes are recovered to the cell surface via the membrane recycling route.

It was observed that IC-21 macrophages can internalize on average 12 latex beads, each with a diameter of 6.76 μm. Consequently, as much as the equivalent of the entire plasma membrane can be internalized as phagosomal membrane (and rapidly replenished by massive synthesis) , providing a huge area for potential peptide acquisition and MHC class I rescue.

The phagocytic process is rapid, essentially complete within 30 minutes. Such rapid redistribution of so much membrane most likely results in inefficient maturation of phagosomes into lysosomes, with much retrieval and recycling of cell surface proteins. These conditions greatly increase the time available for MHC class I molecules to form complexes with bead-delivered peptide and thus the efficiency with which they are subsequently retrieved to the cell surface. The foregoing is intended to be illustrative of the present invention, but not limiting. Numerous variations and modifications may be effected without departing from the true spirit and scope of the invention.

SEQUENCE LISTING

(1) GENERAL INFORMATION:

(i) APPLICANT: De Bruijn, Marloes L. H. Jackson, Michael R. Peterson, Per A.

(ii) TITLE OF INVENTION: ANTIGEN-SPECIFIC ACTIVATION OF UNPRIMED

CD8+ T CELLS

(iii) NUMBER OF SEQUENCES: 9

(iv) CORRESPONDENCE ADDRESS:

(A) ADDRESSEE: Olson S. Hierl, Ltd.

(B) STREET: 20 North Wacker Drive, 36th Floor

(C) CITY: Chicago

(D) STATE: Illinois

(E) COUNTRY: USA

(F) ZIP: 60606

(v) COMPUTER READABLE FORM:

(A) MEDIUM TYPE: Floppy disk

(B) COMPUTER: IBM PC compatible

(C) OPERATING SYSTEM: PC-DOS/MS-DOS

(D) SOFTWARE: Patentin Release #1.0, Version #1.30

(vi) CURRENT APPLICATION DATA:

(A) APPLICATION NUMBER:

(B) FILING DATE: 22-MAY-1996

(C) CLASSIFICATION:

(vii) PRIOR APPLICATION DATA:

(A) APPLICATION NUMBER: US 08/447,761

(B) FILING DATE: 23-MAY-1995

(C) CLASSIFICATION:

(viii) ATTORNEY/AGENT INFORMATION:

(A) NAME: Olson, Arne M

(B) REGISTRATION NUMBER: 30,203

(C) REFERENCE/DOCKET NUMBER: TSRI 475.0 PCT

(ix) TELECOMMUNICATION INFORMATION:

(A) TELEPHONE: 312-580-1180

(B) TELEFAX: 312-580-1189

(2) INFORMATION FOR SEQ ID NO:l:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 8 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS: not relevant

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: peptide (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO (xi) SEQUENCE DESCRIPTION: SEQ ID NO:1:

Ser He He Asn Phe Glu Lys Leu 1 5

(2) INFORMATION FOR SEQ ID NO:2 :

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 10 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS: not relevant

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: peptide

(iii) HYPOTHETICAL: NO

(iv) ANTI-SENSE: NO

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:2:

Leu Glu Ser He He Asn Phe Glu Lys Leu 1 5 10

(2) INFORMATION FOR SEQ ID NO:3:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 10 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS: not relevant

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: peptide

(iii) HYPOTHETICAL: NO

(iv) ANTI-SENSE: NO

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:3 :

Ser He He Asn Phe Glu Lys Leu Thr Glu 1 5 10

(2) INFORMATION FOR SEQ ID NO:4:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 24 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS: not relevant

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: peptide (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO (xi) SEQUENCE DESCRIPTION: SEQ ID NO:4 :

Glu Gin Leu Glu Ser He He Asn Phe Glu Lys Leu Thr Glu Trp Thr 1 5 10 15

Ser Ser Asn Val Met Glu Glu Arg 20

(2) INFORMATION FOR SEQ ID NO:5:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 35 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS: not relevant

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: peptide

(iii) HYPOTHETICAL: NO

(iv) ANTI-SENSE: NO

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:5:

Leu Val Leu Leu Pro Asp Glu Val Ser Gly Leu Glu Gin Leu Glu Ser 1 5 10 15

He He Asn Phe Glu Lys Leu Thr Glu Trp Thr Ser Ser Asn Val Met 20 25 30

Glu Glu Arg 35

(2) INFORMATION FOR SEQ ID NO:6:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 8 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS: not relevant

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: peptide

(iii) HYPOTHETICAL: NO

(iv) ANTt-SENSE: NO

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:6:

Arg Gly Tyr Val Tyr Gin Gly Leu 1 5

(2) INFORMATION FOR SEQ ID NO:7:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 9 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS: not relevant

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: peptide (iii) HYPOTHETICAL: NO ( iv) ANTI -SENSE : NO

(xi) SEQUENCE DESCRIPTION: SΞQ ID NO:7:

Phe Ala Pro Gly Asn Tyr Pro Ala Leu 1 5

(2) INFORMATION FOR SEQ ID NO: 8 :

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 11 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS: not relevant

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: peptide

(iii) HYPOTHETICAL: NO

(iv) ANTI-SENSE: NO

(xi) SEQUENCE DESCRIPTION: SΞQ ID NO: 8:

Gly Glu Phe Ala Pro Gly Asn Tyr Pro Ala Leu 1 5 10

(2) INFORMATION FOR SEQ ID NO:9:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 11 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS: not relevant

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: peptide

(iii) HYPOTHETICAL: NO

(iv) ANTI-SENSE: NO

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 9:

Phe Ala Pro Gly Asn Tyr Pro Ala Leu Trp Ser 1 5 10

Claims

WE CLAIM :

1. A method for production of activated CD8+ T cells specifically directed towards a particular antigen, comprising the steps of providing peptides corresponding to the particular antigen; affixing the peptides to an artificial support; contacting macrophages with the affixed peptides for a sufficient period of time for the macrophages to engulf the affixed peptides and to present at least a portion of the peptides on the surface of the macrophage; and contacting unprimed CD8+ T cells with the peptide-presenting macrophages for a sufficient period of time to activate the unprimed CD8+ T cells.

2. The method of claim 1 wherein the peptides comprise from about eight to about thirty-five amino acid residues.

3. The method of claim 1 wherein the support is a bead from about 0.05 μm to about 10 μm in diameter.

4. The method of claim 1 wherein the support is engulfed by the macrophage.

5. The method of claim 1 wherein the period of time to activate the unprimed CD8+ T cells is about one to about five days.

6. The method of claim 1 further comprising the step of providing a second signal effective in priming unprimed CD8+ T cells.

7. The method of claim 6 wherein the second signal is about 2 to about 75 international units/ml of interleukin-2.

8. The method of claim 3 wherein the amount of peptide affixed to the beads is about 0.05 nmol/10⁶ beads to about 0.5 nmol/10⁶ beads.

9. A macrophage having a peptide corresponding to the particular antigen presented on the surface of the macrophage and having at least a portion of an artificial support in the interior of the macrophage.

10. A method for production of activated CD8+ T cells specifically directed towards a particular antigen, comprising the steps of providing macrophages of claim 9; and contacting unprimed CD8+ T cells with the macrophages for a sufficient period of time to activate the unprimed CD8+ T cells.

11. The method of claim 10 wherein the CD8+ cells are activated in vi tro .

12. The method of claim 10 wherein the CD8+ cells are activated in vivo.

13. A kit for the production of activated CD8+ T cells comprising at least 10^s macrophages of claim 9 stored in stasis.

14. An artificial antigen presenting cell having a peptide corresponding to the particular antigen presented on the surface of the artificial antigen presenting cell and having at least a portion of an artificial support in the interior of the artificial antigen presenting cell.

15. The artificial antigen presenting cell of claim 14 wherein the artificial antigen presenting cell is a transmuted macrophage.

16. A method for production of artificial antigen presenting cells, comprising the steps of providing peptides corresponding to the particular antigen; affixing the peptides to an artificial support; and contacting macrophages with the affixed peptides for a sufficient period of time for the macrophages to engulf the affixed peptides and to present at least a portion of the peptide on the surface of the macrophage.

17. The method of claim 16 wherein the peptides comprise from about eight to about thirty-five amino acid residues.

18. The method of claim 16 wherein the support is a bead from about 0.05 μm to about 10 μm in diameter.

19. The method of claim 16 wherein the support is engulfed by the macrophage.

20. The method of claim 16 wherein the period of time to activate the unprimed CD8+ T cells is about one to about five days.

21. The method of claim 16 further comprising the step of providing a second signal effective in priming unprimed CD8+ T cells.

22. The method of claim 21 wherein the second signal is about 2 to about 75 international units/ml of interleukin-2.

23. A macrophage presenting a particular antigen produced by the steps of providing peptides corresponding to the particular antigen; affixing the peptides to an artificial support; and contacting macrophages with the affixed peptides for a sufficient period of time for the macrophages to engulf the affixed peptides and to present at least a portion of the peptide on the surface of the macrophage.

24. A method of treatment comprising the steps of removing CD8+ T cells from a patient; contacting the removed CD8+ T cells with the macrophages having a peptide corresponding to the particular antigen presented on the surface of the macrophage for a sufficient period of time to activate the unprimed CD8+ T cells; suspending the activated CD8+ T cells in an acceptable carrier or excipient; and administering the suspension to the patient.

25. The method of claim 24 wherein the macrophages originate from the patient.

26. The method of claim 24 wherein the macrophages have at least a portion of artificial support in the interior of the artificial antigen presenting cell.

27. The method of claim 24 further comprising the step of proliferating the activated CD8+ T cells before administering the activated CD8+ T cells to the patient.

28. A system for transmuting macrophages to artificial antigen presenting cells comprising: a) beads having a diameter of about 0.05 μm to about 10 μm in diameter; and b) peptides comprising about 8 to about 35 amino acid residues and corresponding to the particular antigen, the peptides being affixed to the beads.

29. The system of claim 28 wherein the amount of peptides affixed to the beads is about 0.05 nmol/10⁶ beads to about 0.5 nmol/10⁶ beads.