AU765882B2 - Antigen presenting system and methods for activation of T-cells - Google Patents

Description

AUSTRALIA

Patents Act 1990 COMPLETE SPECIFICATION STANDARD PATENT Applicant(s): THE SCRIPPS RESEARCH INSTITUTE Invention Title: ANTIGEN PRESENTING SYSTEM AND METHODS FOR ACTIVATION OF T-CELLS The following statement is a full description of this invention, including the best method of performing it known to me/us: 1A- ANTIGEN PRESENTING SYSTEM AND METHODS FOR ACTIVATION OF T-CELLS TECHNICAL FIELD The present invention relates to materials and methods of activating T-cells with specificity for particular antigenic peptides, the use of activated 15 T-cells in vivo for the treatment of a variety of disease conditions, and compositions appropriate for these uses.

BACKGROUND

The efficiency with which the immune system cures or protects individuals from infectious disease has 20 always been intriguing to scientists, as it has been 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 cytotoxic T-cells have been recorded.

Theoretically, cytotoxic T-cells would be the preferable means of treating the types of disease noted above.

However, no procedures have been available to specifically activate cytotoxic T-cells.

Cytotoxic T-cells, or CD8' cells cells expressing the molecule CD8) as they are presently.known, represent the main line of defense against viral:& infections. CD8' lymphocytes specifically recognize and 2 kill cells which are infected by a virus. Thus, the cost of eliminating a viral infection is the accompanying loss of the infected cells. The T-cell receptors on the surface of CD8' cells cannot recognize foreign antigens directly. In contrast to antibodies, antigen must first be presented to the receptors.

The presentation of antigen to CD8' T-cells is accomplished by major histocompatibility complex (MHC) molecules of the Class I type. The major histocompatibility complex (MHC) refers to a large genetic locus encoding an extensive family of glycoproteins which 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 o "non-self". Thus, membrane-bound MHC molecules are intimately involved in recognition of antigens by T-cells.

MHC products are grouped into three major classes, referred to as I, II, and III. T-cells that serve mainly as helper cells express CD4 and primarily interact with Class II molecules, whereas CD8-expressing cells, which mostly represent cytotoxic effector cells, interact with Class I molecules.

Class I molecules are membrane glycoproteins with the ability to bind peptides derived primarily from intracellular degradation of endogenous proteins.

Complexes of MHC 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 molecules with peptides derived from normal cellular products play a role in "teaching" the T-cells to tolerate self-peptides, 3 in the thymus. Class I molecules do not present entire, intact antigens; rather, they present peptide fragments thereof, "loaded" onto their "peptide binding groove".

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" cells by incubating them in vitro with IL-2, a growth factor for T-cells.

However, this protocol (known as LAK cell therapy) will only allow the expansion of those CDS" cells which 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.

Several novel molecules which appear to be involved in the peptide loading process have recently been identified. It has also been noted that Class I molecules without bound peptide "empty" molecules) can be produced under certain restrictive circumstances.

These "empty" molecules are often unable to reach the cell surface, however, as Class I molecules without bound peptide are very thermolabile. Thus, the "empty" Class I molecules disassemble during their transport from the interior of the cell to the cell surface.

4 The presentation of Class I MHC molecules bound to peptide alone has generally ineffective in activating CD8' cells. In nature, the CD8* cells are activated by antigen-presenting cells which present not only a peptide-bound Class I MHC molecule, but also a costimulatory molecule. Such costimulatory.molecules include B7 which is now recognized to be two subgroups designated as B7.1 and B7.2. It has also been found that cell adhesion molecules such as integrins assist in this process.

When the CD8* T-cell interacts with an *9 antigen-presenting cell having the peptide bound by a Class I MHC and costimulatory molecule, the CD8' T-cell is activated to proliferate and becomes an armed effector T-cell. See, generally, Janeway and Travers, Immunobiolov, published by Current Biology Limited, London (1994), incorporated by reference.

Accordingly, what is needed is a means to activate T-cells so that they proliferate and become S 20 cytotoxic. It would be useful if the activation could be done in vitro and the activated cytotoxic T-cells reintroduced into the patient. It would also be desirable if the activation could be done by a synthetic antigen-presenting matrix comprised of a material such as cells which not only presents the selected peptide, but also presents other costimulatory factors which increase the effectiveness of the activation.

It would also be advantageous if it was possible to select the peptide so that substantially only those CD8' cells cytotoxic to cells presenting that peptide would be activated.

BRIEF SUMMARY OF THE INVENTION The present invention relates to a synthetic antigen-presenting system for presenting an MHC molecule 5 complexed to a peptide and assisting molecule to a T-cell to activate the T-cell.

In one aspect, the invention provides a synthetic antigen-presenting matrix comprising a non-cellular support and a group of molecules consisting of an Class I MHC molecule, a microglobulin molecule and an assisting molecule, the group being arranged on the support so that when a peptide is bound to the Class I MHC molecule and the matrix is contacted with a T-cell, the group is capable of activating the T-cell.

In another aspect, the invention provides a synthetic antigen-presenting matrix comprising a support and a group of molecules consisting of an extracellular portion of an Class I MHC molecule, a microglobulin molecule, a B7.1 or B7.2 molecule or combination thereof, and ICAM-1, the group being arranged on the support so that when a peptide is bound to the Class I MHC molecule and the matrix is contacted with a T-cell, the group is capable of activating the T-cell.

In another aspect, the invention provides a method of producing a synthetic T-cell lymphocyte antigen activating matrix comprising inserting genes into a first cell, the genes consisting of: 25 a first gene which encodes a Class I MHC molecule, or an extracellular portion of a Class I MHC molecule, the gene being operably linked to a promoter, (ii) a second gene which encodes a /-2 microglobulin molecule, the gene being operably linked to S 30 a promoter; maintaining the first cell in conditions in which the Class I MHC molecule, or extracellular portion of the Class I MHC molecule, and microglobulin molecule, are expressed in the cell by the first and 35 second genes; harvesting a Class I MHC molecule, or an extracellular portion of a Class I MHC molecule, and a 3-2 5a microglobulin molecule which are expressed in the cell by the first and second genes; inserting a gene into a second cell which encodes an assisting molecule, or an extracellular portion of an assisting molecule; maintaining the second cell in conditions in which the gene encoding the assisting molecule or extracellular portion of an assisting molecule is expressed; harvesting an assisting molecule or extracellular portion of assisting molecule expressed by the gene in the second cell; and arranging the harvested molecules of step (c) and on a support so that the harvested molecules are capable of activating a T-cell when the support contacts the T-cell.

As described herein, the system relates to a synthetic antigen-presenting matrix having a support and at least the extracellular portion of a Class I MHC molecule capable of binding to a selected peptide operably linked to the support. The matrix also includes an assisting molecule operably linked to the support. The assisting molecule acts on a receptor on the CD8 T-cell.

The MHC and assisting molecules are present in sufficient 25 numbers to activate a population of T-cell lymphocytes against the peptide when the peptide is bound to the extracellular portion of the MHC molecule.

It has been found that an antigen-presenting matrix having both an MHC molecule or a portion of a MHC molecule together with an assisting molecule, provides a synergistic reaction in activating T-cell lymphocytes against the peptide. Examples of assisting molecules are costimulatory molecules such as B7.1 and B7.2 or adhesion molecules such as ICAM-1 and LFA-3. The extracellular 35 portion of such costimulatory molecules can also be used.

Another type of costimulatory molecule is one that reacts with the CD28 molecule such as anti-CD28 antibodies or 5b functional portions thereof, e.g. Fab portions.

It has been found that a specifically effective synergistic reaction results from an antigen-presenting matrix having MHC molecules bound with a peptide, a costimulatory molecule, and an adhesion molecule. In particular, a highly effective synergistic generation of cytotoxic T-cell activity results from the combination of B7.1 and ICAM-1.

The support used for the matrix can take several different forms. Examples for the support include solid support such as metals or plastics, porous materials such as resin or modified cellulose columns, -6 microbeads, microtiter plates, red blood cells and liposomes.

Another type of support is a cell fragment, such as a cell membrane fragment or an entire cell. In this embodiment, the matrix is actually cells which have been transfected to present MHC molecules and assisting molecules on the cell surface to create an antigenpresenting cell (APC). This is done by producing a cell line containing at least one expressible Class I MHC nucleotide sequence for the MHC heavy chain, preferably a cDNA sequence, operably linked to a first promoter and an expressible 3-2 microglobulin nucleotide sequence operably linked to a second promoter. The MHC heavy chain and the 0-2 microglobulin associate together form 15 the MHC molecule which binds to the peptide. The MHC protein binds with the antigenic peptide and presents it on the surface of the cell. The cell also includes a gene for a nucleotide sequence of an assisting molecule operably linked to a third promoter. The assisting molecule is also presented on the surface of the cell.

These molecules are presented on the surface of the cell *4 in sufficient numbers to activate a population of T-cell lymphocytes against the peptide when the peptide is bound to the complexes. Other molecules on the surface of a cell or cell fragment such as carbohydrate moieties may also provide some costimulation to the T-cells.

The cell line is synthetic in that at least one of the genes described above is not naturally present in the cells from which the cell line is derived. It is preferable to use a poikilotherm cell line because MHC molecules are thermolabile. A range of species are useful for this purpose. See, for example, U.S. Patent No. 5,314,813 to Petersen et al. which discusses numerous species for this use and.is incorporated by reference.

It is preferred to use eukaryotic cells and insect cells in particular.

In one embodiment, it is particularly preferred to have at least two assisting molecules, one being a costimulatory molecule and the other being an adhesion molecule. It has been found that this combination has a synergistic effect, giving even greater T-cell activation than either of the individual molecules combined. It has also been found to be advantageous and preferable to have at least one of the transfected genes under control of an inducible promoter.

Using the present invention, it is possible to introduce the peptide to the cell while it is producing MHC molecules and allow the peptide to bind the MHC molecules while they are still within the cell.

Alternatively, the MHC molecules can be expressed as empty molecules on the cell surface and the peptide introduced to the cells after the molecules are expressed on the cell surface. In this latter procedure, the use 20 of poikilotherm cells is particularly advantageous because empty MHC molecules, those not yet complexed or bound with peptides, are thermolabile.

Class I MHC molecules have been expressed in insect cells such as Drosophila melanogaster (fruit fly) 25 cells. Since Drosophila does not have all the components of a mammalian immune system, the various proteins involved in the peptide loading machinery should be absent from such cells. The lack of peptide loading machinery allows the Class I molecules to be expressed as empty molecules at the cell surface.

Another advantage of using insect cells such as the Drosophila system is that Drosophila cells prefer a temperature of 28 °C rather than 37 This fact is very important, because empty Class I molecules are thermolabile and tend to disintegrate at 37 By -8incubating the Class I-expressing Drosophila cells with peptides that can bind to the Class I molecule, it is possible to get virtually everyClass I molecule to contain one and the same peptide. The cells are accordingly very different from mammalian cells, where the Class I molecules contain many different types of peptides, most of which are derived from our own, innocuous cellular proteins.

The present invention also relates to methods for producing activated CD8* cells in vitro. One method comprises contacting, in vitro, CD8' cells with one of the antigen-presenting matrices described above for a time period sufficient to activate, in an antigen-specific manner, the CD8* cells. The method may 15 further comprise separating the activated CD8" cells from the antigen-presenting matrix; suspending the activated CD8* cells in an acceptable carrier or excipient; and administering the suspension to an individual in need of treatment. The antigens may comprise native or undegraded proteins or polypeptides, or they may comprise antigenic polypeptides which have been cleaved into peptide fragments comprising at least 8 amino acid residues prior to incubation with the human Class I MHC molecules.

25 In another variation, the invention relates to methods treating conditions in patients and specifically killing target cells in a human patient. The method comprises obtaining a fluid sample containing resting or naive CD8' cells from the patient; contacting, in vitro, the CD8' cells with an antigen-presenting matrix for a time period sufficient to activate, in an antigen-specific manner, the CD8' cells; and (3) administering the activated CD8' cells to the patient.

For example, the use of tumor specific peptides allows for the treatment of tumor related diseases by producing 9 cytotoxic activated CD8' T-cells. The invention also relates to the method of treating a medical condition by administration of an antigen-presenting matrix in a suitable suspension. In various embodiments the condition may comprise cancer, tumors, neoplasia, viral or retroviral infection, autoimmune or autoimmune-type conditions. In one embodiment, the method of administering the matrix comprises intravenous injection.

BRIEF DESCRIPTION OF THE DRAWINGS Figures 1-3 diagram the construction of expression plasmids pRmHa-2 and pRmHa-3. In Figure 1, pRmHa-2 construction is shown; in Figure 2, pRmHa-3 construction is shown; and in Figure 3, the pRmHa-3 vector is illustrated, showing the restriction, 15 polylinker, promoter, and polyadenylation sites, as well as a site at which a nucleotide sequence may be inserted for expression; Figures 4 and 5 show peptide-induced thermostabilization of HLA B27 and HLA A2.1 expressed on 20 the surface of Drosophila cells by HIV peptides. The mean fluorescence of each cell population is shown plotted against the incubation conditions; Figure 6 illustrates data from an experiment designed to determine whether insect cells can process 25 antigen and load it onto the Class I molecules, and whether the latter can present either endogenously or exogenously derived antigen to T-cells. Schneider 2 (SC2) or 3T3 cells transfected with were incubated with ovalbumin protein (OvPro) or ovalbumin peptide, OVA 24 (OvPep) in isotonic (Iso) or hypertonic (Hyp) media.

(Murine cell line BALB/3T3 is available from the ATCC under accession number CCL 163.) After treatment, cells were cocultured with the T-cell hybridoma B3/CD8. B3/CD8 is a T-cell hybridoma between B3 (Carbone, et al., J.

Exo. Med. 169: 603-12 (1989)), cytotoxic T-cell specific 10 for ovalbumin peptide 253-276 presented by H-2Kh Class I molecules, and CD8- bearing IL-2-secreting cell line.

Upon antigenic stimulation, B3/CD8 produces IL-2, measured by 3 H thymidine incorporation in IL-2-dependent cell line CTLL (Gillis, et al., J. Immunol. 120: 2027 91978)). Thus, by measuring the amount of IL-2 produced, one can assay for T-cell recognition. The supernatant from the cocultures were analyzed for IL-2 by 3

H

thymidine incorporation by the IL-2-dependent cell line CTLL (ATCC No. TIB 214). The amount of 3 H thymidine incorporated is plotted against the initial cell treatments; Figure 7 illustrates the expression of B7.1, ICAM-1 and MHC on the surface of transfected Drosophila S 15 (fly) cells according to the present invention; Figure 8 is a graph showing results of a fluorescence-activated cell sorter experiment using recombinant Ld mouse MHC linked to red blood cells; Figure 9 is a graph showing results of a 20 fluorescence-activated cell sorter experiment using recombinant K" mouse MHC linked to red blood cells; Figure 10 is a graph demonstrates binding of .recombinant K' to microtiter plates by use of labeled antibodies; 25 Figure 11 is a series of graphs showing the results from fluorescence-activated cell sorter experiments demonstrating the expression of CD69 and on CD8' 2C cells stimulated with transfected Drosophila cells; Figure 12 is a pair of bar graphs showing IL-2-dependent proliferative responses of CD8' 2C cells elicited by peptides presented by Drosophila cells transfected with L only; Figure 13 is a graph showing the influence of peptide concentration on day 3 proliferative responses of 11 CD8' 2C cells elicited by peptides presented by transfected Drosophila cells; Figure 14 is a pair of graphs showing the influence of antigen-presenting cells dose on day 3 proliferative response and IL-2 production of CD8' 2C cells elicited by peptides presented by transfected Drosophila cells; Figure 15 is a series of graphs showing the influence of peptide concentration on the proliferative response of CD8* 2C cells elicited by Drosophila cells transfected with L'.ICAM and L'.B7.ICAM; Figure 16 is a series of graphs showing the kinetics of the proliferative response of CD8' and CD8- 2C cells elicited by Drosophila cells transfected with Ld, Ld.B7, Ld.ICAM and Ld.B7.ICAM plus QL9 peptide; Figure 17 is a series of graphs showing CTL Sactivity of CD8' 2C cells stimulated by Drosophila cells transfected with Ld.B7.ICAM or L.ICAM antigenpresenting cells plus QL9 peptide (10 pM) in the absence of exogenous cytokines; Figure 18 is a pair of graphs showing CTL activity of CD8' 2C cells stimulated by Drosophila cells transfected with Ld.ICAM antigen-presenting cells plus QL9 peptide (10 pM) in the absence (left) or presence 25 (right) of exogenous IL-2 Figure 19 is a pair of graphs showing the proliferative response of normal (non-transgenic) CD8* T-cells to peptides presented by transfected Drosophila cells (left panel) and the response elicited by graded doses of N B6 and 2C B6 CD8' cells cultured with 5 x 105 B10.D2 spleen cells (2000 cGy) in the absence of peptides for 3 days without the addition of exogenous cytokines (right panel); Figure 20 is a graph showing stimulated mitogenesis of purified 2C+ T-cells cultured (50,000 12 cells per well) in plates coated with immobilized molecules with peptide QL9 (solid line Ld and anti-CD28 antibody, broken line Ld only); Figure 21 is a bar graph showing stimulated mitogenesis of purified 2C+ T-cells at day 5 in culture, with various indicated peptides, cultured in 96-well plates coated with Ld and anti-CD28 antibody (hatched bars no IL-2, black bars IL-2 added); Figure 22 is a graphical representation of the results of cytofluorometric analysis of cells recovered after 12 days of culture in plates coated with Ld and anti-CD28 antibody and exposed to peptide QL9, stained using the 2C T-cell receptor specific antibody 1B2 (M2 positive staining cells, Ml negative staining cells); Figure 23 is a graphical representation of the results of cytofluorometric analysis of cells recovered S after 12 days of culture in plates coated with Ld and anti-CD28 antibody and exposed to peptide p2Ca, stained using the 2C T-cell receptor specific antibody 1B2 (M2 20 positive staining cells, Ml negative staining cells); Figure 24 is a graphical representation of the results of cytofluorometric analysis of cells recovered after 12 days of culture in plates coated with L' and anti-CD28 antibody and exposed to peptide SL9, stained 25 using the 2C T-cell receptor specific antibody 1B2 (M2 positive staining cells, Ml negative staining cells); Figure 25 is a graph showing cytolysis of target cells by activated T-cells (solid line peptide QL9, broken line control peptide LCMV); Figure 26 is a graphical representation of the cytotoxic lysis resulting from activation of human CD8' T-cells with antigen-presenting cells loaded with influenza matrix peptide; Figure 27 is a graphical representation of the cytotoxic lysis resulting from activation of human CD8' 13 T-cells with antigen-presenting cells loaded with HIV-RT peptide; and Figure 28 is a graphical representation of the cytotoxic lysis resulting from activation of human CD8' T-cells with antigen-presenting cells loaded with tyrosinase peptide.

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a synthetic antigen-presenting system which can be used to activate T-cell lymphocytes. The activated CD8* T-cells proliferate, produce cytokines, become cytotoxic or some combination of these results. In one preferred embodiment, the system activates cytotoxic CD8* cells which then proliferate and then are activated to seek out 15 and destroy target cells. The present invention can be used to activate T-cells in vitro and the activated T-cells are then returned to the patient from which they were originally derived or may be used in vivo activation of T-cells.

20 The synthetic antigen-presenting system of the present invention has two major components. The first component is at least the extracellular portion of the Class I MHC molecule which is capable of binding to a selected peptide. The second major component is an assisting molecule which assists in the activation of T-cells. In each case, an extracellular portion of a larger molecule can used, but in certain embodiments, the entire molecules are used.

For ease of description, MHC molecules will be discussed generally, with the understanding that an extracellular portion of the MHC molecule may be used.

The portion of the MHC molecule necessary for the present invention is the part which binds to the selected peptide and presents the peptide to the T-cell.

14 The peptide is selected to activate the appropriate T-cell, depending on the treatment to 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 that will then bind with and destroy the cancer cells.

The present invention allows the MHC molecules to be produced by cells with the peptide already complexed with the MHC molecule or to produce empty MHC molecules which do not yet have a peptide complexed with them. This latter embodiment is particularly useful since it allows the peptide to be chosen after the MHC 15 molecules are prepared.

A Class I MHC molecule includes a heavy chain, sometimes referred to as an alpha chain, and a 0-2 microglobulin. As discussed herein, the extracellular portion of the Class I MHC molecule is made up of an 20 extracellular portion of an MHC heavy chain together with the 3-2 microglobulin.

In preparing the extracellular portions of MHC to be linked to a support, soluble molecules are prepared as discussed below. These molecules generally lack the 25 transmembrane and cytoplasmic domain in the MHC molecule.

The assisting molecule helps facilitate the activation of the T-cell when it is presented with a peptide/MHC molecule complex. The present invention.

includes two major categories of assisting molecules.

The first category is composed of costimulatory molecules such as B7.1 (previously known as B7 and also known as and B7.2 (also known as CD86) which binds to CD28 on T-cells. Other costimulatory molecules are anti-CD28 antibodies or the functional portions of such antibodies, e.g. Fab portions that bind to CD28.

15 The other major category of assisting molecules of the present invention are adhesion molecules. These include the various ICAM molecules, which include ICAM-1, ICAM-2, ICAM-3 and LFA-3. It has been found that the combination of a peptide bound to an MHC molecule used in conjunction with one of these assisting molecules activates the T-cells to an extent previously not seen.

An even greater synergistic reaction has been achieved by using a peptide-bound MHC molecule in conjunction with both a costimulatory molecule and an adhesion molecule. This has been found to be particularly effective in producing cytotoxic CD8' cells.

In accordance with the present invention, the MHC molecule and the assisting molecule are operably 15 linked to a support such that the MHC and assisting molecules are present in sufficient numbers to activate a population of T-cells lymphocytes against the peptide when the peptide is bound to the extracellular portion of the MHC molecule. The peptide can be bound to the MHC 20 molecule before or after the MHC molecule is linked to the support.

The support can take on many different forms.

It can be a .solid support such as a plastic or metal material, it can be a porous material such as commonly 25 used in separation columns, it can be a liposome or red blood cell, or it can even be a cell or cell fragment.

As discussed in more detail below, in the case where a cell serves as a support, the MHC and assisting molecules can be produced by the cell. The MHC molecule is then linked to the cell by at least the transmembrane domain if not also the cytoplasmic domain which would not be present in a soluble form of MHC.

The extracellular portions of MHC molecule and assisting molecule can be linked to a support by providing an epitope which reacts to an antibody 16 immobilized on the support. In addition, the MHC or assisting molecules can be produced with or linked to (His), which would react with nickel in forming part of the support. Other means to immobilize or link MHC molecules to a support are well known in the art.

As discussed above, the support can be a cell membrane or an entire cell. In such a case, an eukaryotic cell line is modified to become a synthetic antigen-presenting cell line for use with T-cell lymphocytes. For ease of description, antigen-presenting cells (APC) will also be called stimulator cells.

Because empty MHC molecules are thermolabile, it is preferred that the cell culture be poikilotherm and various cell lines are discussed in detail below.

A culture of cells is first established. The culture is then transfected with an expressible Class I MHC heavy chain gene which is operably linked to a promoter. The gene is chosen so that it is capable of expressing the Class I MHC heavy chain. The cell line is also transfected with an expressible 0-2 microglobulin gene which is operably linked to a second promoter. The gene is chosen such it is capable of expressing 3-2 microglobulin that forms MHC molecules with the MHC heavy chain. In the case of soluble extracellular portions of 25 MHC molecules to be used with solid supports and the like, a truncated MHC heavy chain gene is used as discussed in more detail below.

The culture is also transfected with an expressible assisting molecule gene operably linked to a third promoter. The assisting molecule gene is capable of being expressed as an assisting molecule which interacts with the molecule on the T-cell lymphocytes.

As discussed below, each of these genes can be transfected using various methods, but the preferred method is to use more than one plasmid.

17 The cell line transfected is chosen because it lacks at least one of the genes being introduced. It has been found that insect cells are advantageous not only because they are poikilothermic, but because they lack these genes and the mechanisms which would otherwise produce MHC molecules bound to peptides. This allows for greater control over the production of peptide-bound MHC molecules, and the production of empty MHC molecules.

The MHC heavy chain is preferably from a different species, more preferably, a homeotherm such as mammals and, optimally, humans.

The preferred cell line is a stable poikilotherm cell line that has the first promoter being inducible to control the expression of the MHC heavy chain. It is preferred that the cell assembles empty MHC molecules and presents them on the cell surface so that the peptides can be chosen as desired.

The resulting MHC molecules bind to the peptide and are present in sufficient numbers with the assisting molecules on the surface of the cell to activate a population of T-cell lymphocytes against the peptide when the peptide is bound to the MHC cells.

In a further embodiment, a second assisting molecule gene is also transfected into the cell culture.

25 In this case, the first assisting molecule gene can be for a costimulatory molecule and the second assisting molecule gene can be for an adhesion molecule.

It is preferred that at least one of the genes and, in particular, the MHC heavy chain gene be linked to an inducible promoter. This allows control over the production of MHC molecules so that they are only produced at a time when the peptide of interest is available and presented in the culture to react with the produced MHC molecules. This minimizes undesirable MHC molecule/peptide complexes.

18 Where the cell line already produces one or more of the desired molecules, it is only necessary to transfect the culture with an expressible gene for the gene which is lacking in the cells. For example, if the cells already present the MHC molecules on their surface, it is only necessary to transfect the culture with an expressible gene for the assisting molecule.

The peptide can be introduced into the cell culture at the time the cells are producing MHC molecules. Through methods such as osmotic shock, the peptides can be introduced in the cell and bind to the produced MHC molecules. Alternatively, partcriilarly -n the case poikilotherm cell lines, the MHC molecules will be presented empty on the cell surface. The peptide can then be added to the culture and bound to the MHC molecules as desired.

After the cells are produced having MHC and assisting molecules on their surfaces, they can be lyophilized and the fragments of the cells used to 20 activate the population of T-cell lymphocytes.

Transfected cultures of cells can be used to produced extracellular portions of. MHC molecules and assisting molecules. The use of extracellular portions in conjunction with supports such as solid supports has 25 certain advantages of production. Where living cells are used to provide a synthetic antigen-presenting cell, at least three genes, two to produce the MHC molecule and one for the assisting molecule must be introduced to the cell. Often, additional genes such as for antibiotic resistance are also transfected.

Where a solid support system is being used, one cell line can produce the extracellular portions of MHC molecules while another cell line produces the extracellular portion of the assisting molecule. The MHC molecule portions and the assisting molecule portions can 19 then be harvested from their respective cultures. The molecules are then linked to an appropriate support in sufficient numbers to activate a population of T-cell lymphocytes against a peptide when the peptide is bound to the extracellular portion of the MHC molecule. From a production standpoint, two different cultures can be used, but it is also possible to use the same culture, however, requiring that the culture be transfected with the additional gene for the extracellular portion of the assisting molecule.

A further modification of this embodiment is to provide a third culture of cells which is transfected with an expressible second assisting molecule gene. In this example, the second culture of cells produces extracellular portions of the costimulatory molecule while the third culture of cells produce an extracellular portion of an adhesion molecule. The adhesion molecule portions are harvested and linked to the support.

The present invention also relates to a method 20 for activating CD8' T-cells against a selected peptide.

The method relates to providing a cell line presenting MHC molecules binding a peptide and assisting molecules on their surfaces. Naive CD8' T-cells can be obtained by removal from a patient to be treated. The cultured cells 25 are then contacted with the CD8' T-cells for a sufficient period of time to activate the CD8' T-cell lymphocytes resulting in proliferation and transforming the T-cells into armed effector cells.

The activated CD8' T-cells can then be separated from the cell line and put into a suspension in an acceptable carrier and administered to the patient.

An alternative method involves the use of the synthetic antigen-presenting matrix to activate the CD8' cells.

It is preferred that human genes are used and, therefore, human molecule analogs are produced. As shown 20 in prior U.S. Patent No. 5,314,813, murine 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).

1. Human Class I MHC Molecules Class I MHC molecules comprise a heavy chain and a /-microglobulin protein. A human Class I MHC heavy chain of the present invention is selected from the group comprising HLA-A, HLA-B, HLA-C, HLA-E, HLA-F, and HLA-G, and more preferably, from the group comprising HLA-A, HLA-B, and HLA-C. The heavy chains are useful in either soluble or insoluble form. In the soluble form, a stop codon is engineered into the nucleotide sequence o e 15 encoding the HLA molecule of choice preceding the e.

transmembrane domain.

While it is possible to isolate nucleotide sequences encoding human Class I MHC heavy chains from known, established cell lines carrying the appropriate 20 variants transformed cell lines JY, BM92, WIN, MOC, and MG it is more practical to synthesize the nucleotide sequence from a portion of the gene via **polymerase chain reaction (PCR), using the appropriate primers. This method has been successfully used to clone 25 full-length HLA cDNA; for example, the sequences for HLA-A25, HLA-A2, HLA-B7, HLA-B57, HLA-B51, and HLA-B37 are deposited in the GenBank database under accession nos. M32321, M32322, M32317, M32318, M32319 and M32320, respectively. Known, partial and putative HLA amino acid and nucleotide sequences, including the consensus sequence, are published (see, Zemmour and Parham, Immunogenetics 33: 310-320 (1991)), and cell lines expressing HLA variants are known and generally available as well, many from the American Type Culture Collection Therefore, using PCR, it is possible to 21 synthesize human Class I MHC-encoding nucleotide sequences which may then be operatively linked to a vector and used to transform an appropriate cell and expressed therein.

Particularly preferred methods for producing the Class I MHC heavy chain, 0-2 microglobulin proteins and assisting molecules of the present invention rely on the use of preselected oligonucleotides as primers in a polymerase chain reaction (PCR) to form PCR reaction products as described herein. Gene preparation is typically accomplished by primer extension, preferably by primer extension in a polymerase chain reaction (PCR) format.

If the genes are to be produced by (PCR) 15 amplification, two primers, a PCR primer pair, must S. be used for each coding strand of nucleic acid to be .amplified. (For the sake of simplicity, synthesis of an exemplary HLA heavy chain variant sequence will be discussed, but it is expressly to be understood that the 20 PCR amplification method described is equally applicable to the synthesis of 0-2 microglobulin, costimulatory molecules, adhesion molecules, and all HLA variants, including those whose complete sequences are presently.

unknown.) 25 The first primer becomes part of the antisense (minus or complementary) strand and hybridizes to a nucleotide sequence conserved among HLA (plus or coding) strands. To produce coding DNA homologs, first primers are therefore chosen to hybridize to be complementary to) conserved regions within the MHC genes, preferably, the consensus sequence or similar, conserved regions within each HLA group consensus sequences within HLA-A, HLA-B, HLA-C, and the less-polymorphic groups, HLA-E, and -G.

22 Second primers become part of the coding (plus) strand and hybridize to a nucleotide sequence conserved among minus strands. To produce the HLA-coding DNA homologs, second primers are therefore chosen to hybridize with a conserved nucleotide sequence at the end of the HLA-coding gene such as in that area coding for the leader or first framework region. In the amplification of the coding DNA homologs the conserved nucleotide sequence of the second primer can be complementary to a sequence exogenously added using terminal deoxynucleotidyl transferase as described by Loh et al., Science 243: 217-220 (1989). One or both of the first and second primers can contain a nucleotide sequence defining an endonuclease recognition site. The 15 site can be heterologous to the immunoglobulin gene being amplified and typically appears at or near the 5' end of the primer.

The high turn over rate of the RNA polymerase amplifies the starting polynucleotide as has been described by Chamberlin et al., The Enzymes, ed. P.

Boyer, PP. 87-108, Academic Press, New York (1982).

Another advantage of T7 RNA polymerase is that mutations can be introduced into the polynucleotide synthesis by replacing a portion of cDNA with one or more mutagenic 25 oligodeoxynucleotides (polynucleotides) and transcribing the partially-mismatched template directly as has been previously described by Joyce et al., Nuc. Acid Res. 17: 711-722 (1989). Amplification systems based on transcription have been described by Gingeras et al., in PCR Protocols. A Guide to Methods and Applications, pp 245-252, Academic Press, Inc., San Diego, CA (1990).

PCR amplification methods are described in detail in U.S. Patent Nos. 4,683,192, 4,683,202, 4,800,159, and 4,965,188, and at least in several texts including "PCR Technology: Principles and Applications 23 for DNA Amplification", H. Erlich, ed., Stockton Press, New York (1989); and "PCR Protocols: A Guide to Methods and Applications", Innis et al., eds., Academic Press, San Diego, California (1990). Various preferred methods and primers used herein are described hereinafter and are also described in Nilsson, et al., Cell 58: 707 (1989), Ennis, et al., PNAS USA 87: 2833-7 (1990), and Zemmour, et al., Immunogenetics 33: 310-20 (1991), for example.

In particular, it is preferred to design primers from comparison of 5' and 3' untranslated regions of HLA alleles or -G alleles), with selection of conserved sequences. Restriction sites may also be incorporated into the 5' and 3' primers to enable the amplification products to be subcloned into 15 sequencing or expression vectors. It may also be helpful to place a 4-base spacer sequence proximal to the restriction site to improve the efficiency of cutting amplification products with enzymes.

The following primers are preferred for 20 amplification of HLA-A, and -G cDNA, go preferably in separate reactions. Resulting cDNAs may then be cloned and sequenced as described herein. These primers are appropriate for use in amplifying all known and:presently unknown types of HLA.

25 HLA A 5' primer: 5' CC ACC ATG GCC GTC ATG GCG CCC 3' (SEQ ID NO 1) 3' primer: 5' GG TCA CAC TTT ACA AGC TCT GAG 3' (SEQ ID NO 2) HLA B primer: 5' CC ACC ATG CTG GTC ATG GCG CCC 3' (SEQ ID NO 3) 24 HLA C 3' primer: 5' (SEQ ID NO 4) primer: 5' (SEQ ID NO 3' primer: 5' (SEQ ID NO 6) primer: 5' (SEQ ID NO 7) 3' primer: 5' (SEQ ID NO 8) GG ACT CGA TGT GAG AGA CAC ATC CC ACC ATG CGG GTC ATG GCG CCC GG TCA GGC TTT ACA AGC GAT GAG HLA E CC ACC ATG CGG GTA GAT GCC CTC C 0@Oe 0 *000 00

S

S

S

0 GG TTA CAA GCT GTG AGA CTC AGA HLA F 5' primer: 5' CC ACC ATG GCG CCC CGA AGC CTC 20 3' (SEQ ID NO 9) 3' primer: 5' GG TCA CAC TTT ATT AGC TGT GAG A (SEQ ID NO HLA G primer: 5' CC ACC ATG GCG CCC CGA ACC CTC (SEQ ID NO 11) 3' primer: 5' GG TCA CAA TTT ACA AGC CGA GAG 3' (SEQ ID NO 12) In preferred embodiments only one pair of first and second primers is used per amplification reaction.

The amplification reaction products obtained from a 25 plurality of different amplifications, each using a plurality of different primer pairs, are then combined.

However, the present invention also relates to DNA homolog production via co-amplification (using two pairs of primers), and multiplex amplification (using up to about 8, 9 or 10 primer pairs).

In preferred embodiments, the PCR process is used not only to produce a variety of human Class I-encoding DNA molecules, but also to induce mutations which may emulate those observed in the highly-polymorphic HLA loci, or to create diversity from a single parental clone and thereby provide a Class I MHC molecule-encoding DNA "library" having a greater heterogeneity. In addition to the mutation inducing 15 variations described in the above referenced U.S. Patent No. 4,683,195 and such as discussed in U.S. Patent No.

5,314,813.

2. DNA Expression Vectors A vector of the present invention is a nucleic 20 acid (preferably DNA) molecule capable of autonomous replication in a cell and to which a DNA segment, e.g., gene or polynucleotide, can be operatively linked.so as to bring about replication of the attached segment. One 0:00: of the nucleotide segments to be operatively linked to vector sequences encodes at least a portion of a mammalian Class I MHC heavy chain. Preferably, the entire peptide-coding sequence of the MHC heavy chain is inserted into the vector and expressed; however, it is also feasible to construct a vector which also includes some non-coding MHC sequences as well. Preferably, non-coding sequences of MHC are excluded. Alternatively, a nucleotide sequence for a soluble form of an Class I MHC heavy chain may be utilized; the "sol" form differs from the non-sol form in that it contains a "stop" codon inserted at the end of the alpha 3 domain or 26 prior to the transmembrane domain. Another preferred vector includes a nucleotide sequence encoding at least a portion of a mammalian P-2 microglobulin molecule operatively linked to the vector for expression. Still another preferred vector includes a nucleotide sequence encoding at least a portion of a mammalian assisting molecule operably linked to the vector for expression.

It is also feasible to construct a vector including nucleotide sequences encoding a Class I MHC heavy chain and a 3-2 microglobulin and an assisting molecule, or some combination of these.

A preferred vector comprises a cassette that includes one or more translatable DNA sequences .*.operatively linked for expression via a sequence of nucleotides adapted for directional ligation. The cassette preferably includes DNA expression control sequences for expressing the polypeptide or protein that is produced when a translatable DNA sequence is directionally inserted into the cassette via the sequence 20 of nucleotides adapted for directional ligation.. The cassette also preferably includes a promoter sequence upstream from the translatable DNA sequence, and a polyadenylation sequence downstream from the mammalian M:.HC heavy chain sequence. The cassette may also include a selection marker, albeit it is preferred that such a marker be encoded in a nucleotide sequence operatively linked to another expression vector sequence.

The choice of vector to which a cassette of.

this invention is operatively linked depends directly, as is well known in the art, on the functional properties desired, vector replication and protein expression, and the host cell to be transformed, these being limitations inherent in the art of constructing recombinant DNA molecules.

27 In various embodiments, a vector is utilized for the production of polypeptides useful in the present invention, including MHC variants and antigenic peptides.

Exemplary vectors include the plasmids pUC8, pUC9, pUC18, pBR322, and pBR329 available from BioRad Laboratories (Richmond, CA), pPL and pKK223 available from Pharmacia (Piscataway, NJ), and pBS and M13mpl9 (Stratagene, La Jolla, CA). Other exemplary vectors include pCMU (Nilsson, et al., Cell 58: 707 (1989)). Other appropriate vectors may also be synthesized, according to known methods; for example, vectors pCMU/Kh and pCMUII used in various applications herein are modifications of pCMUIV (Nilsson, et al., supra).

In addition, there is preferably a sequence 15 upstream of the translatable nucleotide sequence encoding a promoter sequence. Preferably, the promoter is conditional inducible). A preferred conditional.

promoter used herein is a metallothionein promoter or a heat shock promoter.

20 Vectors may be constructed utilizing any of the well-known vector construction techniques. Those techniques, however, are modified to the extent that the translatable nucleotide sequence to be inserted into the genome of the host cell is flanked "upstream" of the sequence by an appropriate promoter and, in some variations of the present invention, the translatable nucleotide sequence is flanked "downstream" by a polyadenylation site. This is particularly preferred when the "host" cell is an insect cell and the nucleotide sequence is transmitted via transfection. Transfection may be accomplished via numerous methods, including the calcium phosphate method, the DEAE-dextran method, the stable transfer method, electroporation, or via the liposome mediation method. Numerous texts are available which set forth known transfection methods and other 28 procedures for introducing nucleotides into cells; see, Current Protocols in Molecular Biology, John Wiley Sons, NY (1991).

The vector itself may be of any suitable type, such as a viral vector (RNA or DNA), naked straight-chain or circular DNA, or a vesicle or envelope containing the nucleic acid material and any polypeptides that are to be inserted into the cell. With respect to vesicles, techniques for construction of lipid vesicles, such as liposomes, are well known. Such liposomes may be targeted to particular cells using other conventional techniques, such as providing an antibody or other specific binding molecule on the exterior of the liposome. See, A. Huang, et al., J. Biol. Chem.

15 255: 8015-8018 (1980). See, Kaufman, Meth.

Enzymol. 185: 487-511 (1990).

In a preferred embodiment, the vector also contains a selectable marker. After expression, the product of the translatable nucleotide sequence may then S" 20 be purified using antibodies against that sequence. One example of a selectable marker is neomycin resistance. A plasmid encoding neomycin resistance, such as phshsneo, a phsneo, or pcopneo, may be included in each transfection such that a population of cells that express the gene(s) of choice may be ascertained by growing the transfectants in selection medium.

A preferred vector for use according to the present invention is a plasmid; more preferably, it is a high-copy-number plasmid. It is also desirable that the vector contain an inducible promoter sequence, as inducible promoters tend to limit selection pressure against cells into which such vectors (which are often constructed to carry non-native or chimeric nucleotide sequences) have been introduced. It is also preferable that the vector of choice be best suited for expression 29 in the chosen host. If the host cell population is a Drosophila cell culture, then a compatible vector includes vectors functionally equivalent to those such as (see Bello and Couble, Nature 346: 480 (1990)) or pRmHa-1, or -3 (see Bunch, et al., Nucl. Acids Res. 16: 1043-1061 (1988)). In the preferred embodiment, the vector is pRmHa-3, which is shown in Figure 3. This vector includes a metallothionein promoter, which is preferably upstream of the site at which the MHC sequence is inserted, and the polyadenylation site is preferably downstream of said MHC sequence. Insect cells and, in particular, Drosophila cells are preferred hosts according to the present invention. Drosophila cells such as Schneider 2 (S2) cells have the necessary 15 trans-acting factors required for the activation of the promoter and are thus even more preferred.

The expression vector pRmHa-3 is based on the.

bacterial plasmid pRmHa-1 (Figure the latter of which is based on plasmid pUC18 and is deposited with the American Type Culture Collection (ATCC, Rockville, MD), having the accession number 37253. The pRmHa-3 vector contains the promoter, the 5' untranslated-leader sequence of the metallothionein gene (sequences 1-421, S* SEQ ID NO 13) with the Rl and Stu sites removed; see Figure It also contains the 3' portion of the Drosophila ADH gene (sequence #6435-7270, SEQ ID NO 14) including the polyadenylation site. Therefore, cloned DNA will be transcriptionally regulated by the metallothionein promoter and polyadenylated.

Construction of the pRmHa-1 plasmid is described in Bunch, et al., Nucl. Acids Res. 16: 1043-1061 (1988).

Construction of the pRmHa-3 and pRmHa-2 plasmids (the latter of which has a metallothionein promoter sequence that may be removed as an Eco RI fragment) is illustrated in Figures 1, 2, and 3. With regard to pRmHa-3, a 30 preferred plasmid for use according to the present invention, Pst I, Sph I and Hind III are in the promoter fragment and therefore are not unique. Xba is in the ADH fragment (4 bases from its 3' end) and is also not unique. The following restriction sites are, however, unique in pRmHa-3: Eco RI, Sac I, Kpn.I, Sma I, Bam HI, Sal I, Hinc 2, and Acc I.

A cassette in a DNA expression vector of this invention is the region of the vector that forms, upon insertion of a translatable DNA sequence, a sequence of :i nucleotides capable of expressing, in an appropriate host, a fusion protein of this invention. The expression-competent sequence of nucleotides is referred to as a cistron. Thus, the cassette preferably comprises 15 DNA expression control elements operatively linked to one or more translatable DNA sequences. A cistron is formed when a translatable DNA sequence is directionally inserted (directionally ligated) between the control elements via the sequence of nucleotides adapted for that 20 purpose. The resulting translatable DNA sequence, namely the inserted sequence, is, preferably, operatively linked in the appropriate'reading frame.

DNA expression control sequences comprise a set of DNA expression signals for expressing a structural gene product and include both 5' and 3' elements, as is well known, operatively linked to the cistron such that the cistron is able to express a structural gene product.

The 5' control sequences define a promoter for initiating transcription and a ribosome binding site operatively linked at the 5' terminus of the upstream translatable DNA sequence.

Thus, a DNA expression vector of this invention provides a system for cloning translatable DNA sequences into the cassette portion of the vector to produce a 31 cistron capable of expressing a fusion protein of this invention.

3. Cell Lines A preferred cell line of the present invention is capable of continuous growth in culture and capable of expressing mammalian Class -I MHC molecules and assisting molecules on the surface of its cells. Any of a variety of transformed and non-transformed cells or cell lines are appropriate for this purpose, including bacterial, yeast, insect, and mammalian cell lines. (See, e.g., Current Protocols in Molecular Biology, John Wiley Sons, NY (1991), for summaries and procedures for culturing and using a variety of cell lines, E.

coli and S. cerevisiae.) 15 Preferably, the cell line is a eukaryotic cell line. More preferably, the cell line is poikilothermic less sensitive to temperature challenge than *mammalian cell lines). More preferably, it is an insect cell line. Various insect cell lines are available for ~20 use according to the present invention, including moth (ATCC CCL 80), armyworm (ATCC CRL 1711), mosquito larvae (ATCC lines CCL 125, CCL 126, CRL 1660, CRL 1591, CRL 6585, CRL 6586) and silkworm (ATCC CRL 8851). In a preferred embodiment, the cell line is a Drosophila cell line such as a Schneider cell line (see Schneider, J.

Embrvol. Exp. Morph. 27: 353-365 (1972)); preferably, the cell line is a Schneider 2 (S2) cell line (S2/M3) adapted for growth in M3 medium (see Lindquist, et al., Drosophila Information Service 58: 163 (1982)).

Schneider cells may be prepared substantially as follows. Drosophila melanogaster (Oregon-R) eggs are collected over about a 4 hour interval and are dechlorinated in 2.5% aqueous sodium hypochlorite and surface-sterilized by immersion in 70% ethanol for minutes, followed by an additional 20 minutes in 0.05% 32 HgCl. in 70% ethanol. After being rinsed thoroughly in sterile distilled water, the eggs are transferred to petri dishes containing sterile Metricel black filters backed with Millipore prefilters, both previously wetted with culture medium. The eggs are placed overnight in a 22 OC incubator and removed for culturing when 20-24 hours old. The embryos are each cut into halves or thirds, then placed in 0.2% trypsin (1:250, Difco) in Rinaldini's salt solution (Rinaldini, Nature (London) 173: 1134-1135 (1954)) for 20-45 minutes at room temperature. From 100-300 embryos are used to initiate each culture.

SAfter the addition of fetal bovine serum (FBS), the fragments are centrifuged at 100 X g for 2-3 minutes, resuspended in 1.25ml culture medium and seeded into glass T-9 flasks. The cultures are maintained at about E 22-27 oC 0.5 OC, with a gaseous phase of ambient air.

Schneider's culture medium (Schneider, J. Exp. Zool. 156: 91-104 (1964); Schneider, J. Embryol. ExD. Moroh. 20 271-279 (1966)) containing an additional 500mg bacteriological peptone per 100ml medium and supplemented with 15% inactivated FBS is preferably used. The pH (preferably 6.7-6.8) is monitored with 0.01% phenol red.

The cell lines are preferably maintained by subculturing every 3-7 days. The cells readily attach to the glass but not so firmly as to require trypsin treatment; typically, simple pipetting is adequate to flush most of the cells from the bottom of the flasks. The morphological appearance of the cells is described in Schneider, J. Embryol. Exo. Morph. 27: 353-365 (1972) They are essentially epithelial-like in appearance and range from about 5-11bm in diameter and 11-35gm in length. Small pockets containing rounded cells may be dispersed randomly throughout the other cells.

33 Preferably, the Schneider 2 (S2) cells are maintained in Schneider's Drosophila medium plus 10% FBS including penicillin (100 unit/ml) and streptomycin (100mg/ml). It is preferable to keep the cells at a density of more than 0.5 X 10 5 /ml, and to grow them at a 24-30 oC temperature range. The cells tend to double in fewer than 24 hours and grow to high cell density, i.e., about 2 X 10 7 /ml or greater. The cells may also be frozen in 90% FBS and 10% DMSO, for later use or analysis. One may place the cells at -70 OC and then store in liquid nitrogen.

A preferred cell line according to the present invention, identified as Schneider 2 (S2) cells, has been deposited pursuant to Budapest Treaty requirements with the American Type Culture Collection (ATCC), Rockville, MD, on February 18, 1992, and was assigned accession number CRL 10974.

Cells of the present invention are transfected with cDNAs encoding (human) MHC heavy chains, 0-2 20 microglobulin and one or more assisting molecules, which have each been inserted into operatively linked to) an expression vector. In a more preferred embodiment, the vector comprises Drosophila expression plasmid pRmHa-3, into which expressible nucleotide sequences encoding human Class I MHC heavy chains, human 0-2 microglobulin or human assisting molecules have been inserted using techniques disclosed herein. Preferably, the cDNAs encoding MHC heavy chains, those encoding .3-2 microglobulin and those encoding assisting molecules are operatively linked to separate expression plasmids and are cotransfected into the cultured cells.

Alternatively, the cDNAs encoding MHC heavy chains, 0-2 microglobulin and assisting molecules may be operatively linked to the same expression plasmid and cotransfected via that same plasmid. In another variation, cDNAs 34 encoding MHC heavy chains, 0-2 microglobulin, assisting molecules, and a cytokine such as IL-2 are operatively linked to expression plasmids and are cotransfected into a cell line of the present invention. Selection of HLA genes, construction of appropriate vectors and primer selection are described in greater detail above.

Successfully transformed cells, cells that contain an expressible human nucleotide sequence according to the present invention, can be identified via well-known techniques. For example, cells resulting from the introduction of a cDNA or rDNA of the present invention can be cloned to produce monoclonal colonies.

Cells from those colonies can be harvested, lysed, and their DNA content examined for the presence of the rDNA using a method such as that described by Southern, J.

Mol. Biol. 98: 503 (1975). In addition to directly assaying for the presence of rDNA, successful transformation or transfection may be confirmed by well-known immunological methods when the rDNA is capable of directing the expression of a subject chimeric polypeptide. For example, cells successfully transformed with an expression vector may produce proteins displaying particular antigenic properties which are easily determined using the appropriate antibodies. In addition, successful transformation/transfection may be ascertained via the use of an additional vector bearing a marker sequence, such as neomycin resistance, as described hereinabove.

It is also preferable that the culture be stabile and capable of sustained growth at reduced temperatures. For example, it is preferred that the culture be maintained at about room temperature, e.g., about 24-27 OC. In other embodiments, the culture is maintained at higher temperatures, particularly during the process of activating CD8' cells. It is thus 35 preferred that a culture according to the present invention be capable of withstanding a temperature challenge of about 30 oC to about 37 OC. Addition of 0-2 microglobulin to a culture stabilizes the Class I MHC to at least a 30 oC challenge; addition of 0-2 microglobulin and peptides results in greater thermostability at higher temperatures, at 37 oC.

In order to prepare the culture for expression of empty or more preferably, peptide-loaded MHC molecules, the culture may first require stimulation, via CuSO, induction; for a predetermined period of o" time. After a suitable induction period about 12-48 hours, peptides may be added at a predetermined concentration about 100pg/ml). Peptides may be prepared as discussed below. After a further incubation period for about 12 hours at 27 OC the culture is ready for use in the activation of CD8* cells.

While this additional incubation period may be shortened or perhaps omitted, the culture tends to become increasingly stable to temperature challenge if it is allowed to incubate for a time prior to addition of resting or naive CD8' cells. For example, cultures according to the present invention to which peptide has been added are capable of expressing significant amounts of peptide-loaded Class I MHC molecules even when incubated for extended periods of time at 37 OC.

Nutrient media useful in the culturing of transformed host cells are well known in the art and can be obtained from numerous commercial sources. In embodiments wherein the host cell is mammalian, a "serum-free" medium is preferably used.

4. Human 6-2 microqlobulin and Assisting Molecules In order to establish a cell line capable of producing therapeutically useful amounts of 36 surface-expressed human Class I MHC molecules, it is preferable to cotransfect a cell line of the present invention with a vector operably linked to a nucleotide sequence encoding 3-2 microglobulin in order to effect appropriate levels of expression of human MHC molecules in the cell line. While the nucleotide sequence encoding mammalian 3-2 microglobulin such as mouse 3-2 microglobulin increases the stability of the human Class I MHC molecules expressed in the cell lines of the present invention, it is preferable to cotransfect the cell line with a vector operably linked to an expressible nucleotide sequence encoding a human 0-2 microglobulin.

As discussed above, a preferred vector S. according to the present invention includes a nucleotide sequence encoding at least a portion of a mammalian 0-2 microglobulin molecule operatively linked to the vector o. for expression. The gene for the assisting molecules can be linked to the same or another vector. It is also feasible to construct a vector including nucleotide sequences encoding both a Class I MHC heavy chain and a 0-2 microglobulin.

The sequencing and primers used for the assisting molecules are discussed in more detail below.

However, the protocols are similar.

A human 0-2 microglobulin cDNA sequence has been published (see Suggs, et al., PNAS 78: 6613-17, 1981) and the sequence was used as a template for a polymerase chain reaction (PCR) using the following primers: 5' primer:

GCTTGGATCCAGATCTACCATGTCTCGCTCCGTGGCCTTAGCTGTGCT

CGCGCTACTCTC 3' (SEQ ID NO 3' primer 5' GGATCCGGATGGTTACATGTCGCGATCCCACTTAAC 3' 37 (SEQ ID NO 16) The primers are used in a standard PCR reaction (see above and references cited therein). The reaction products are extracted with phenol, purified using a Geneclean kit (Bio 101, San Diego, CA), digested with Bam HI and cloned into the Bam HI site of pBS (Stratagene, La Jolla, CA). After verification of the sequence, this Bam HI fragment is cloned into the Bam HI site of an appropriate expression vector. In the preferred embodiment, human 0-2 microglobulin cDNA is synthesized *44* and operably linked to expression vector pRmHa-3.

5. Peptides Virtually all cellular proteins in addition to viral antigens are capable of being used to generate relevant peptide fragments that serve as potential Class I MHC ligand. In most mammalian cells, then, any particular MHC peptide complex would represent only a small proportion of the total MHC encoded molecules found 20 on the cell surface. Therefore, in order to produce surface-expressed human Class I MHC molecules that have an increased capacity to specifically activate CD8* cells, it is preferable to isolate and load peptide fragments of appropriate size and antigenic characteristics onto Class I molecules.

The peptides of the present invention bind to Class I MHC molecules. The binding occurs under biological conditions which can be created in vivo as well as in vitro. The exact nature of the binding of the peptides need not be known for practice of the invention.

In a preferred embodiment, the peptides to be loaded onto the Class I MHC molecules are antigenic. 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 38 loading onto the MHC molecules be of a single species; 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 may be presented to the cells via various means. Preferably, peptides are presented in a manner which allows them to enter an intracellular pool of peptides. For example, peptides may be presented via osmotic loading. Typically, peptides are added to the culture medium. The peptides may be added to the culture in the form of an intact polypeptide or protein which is subsequently degraded via cellular processes, via 0000 enzymatic degradation. Alternatively, the intact polypeptide or protein may be degraded via some other means such as chemical digestion cyanogen bromide) or proteases chymotrypsin) prior to its addition to 00 the cell culture. In other embodiments, the peptides are presented in smaller segments which may or may not comprise epitopic amino acid sequences.

20 In a preferred embodiment, a sufficient amount of protein(s) or peptide(s) is added to the cell culture to allow'the Class I MHC molecules to bind and subsequently present a large density of the peptide preferably, with the same kind of peptide attached to each MHC on the surface of human Class I MHC-expressing cells of the present invention. It is also preferred to allow the human Class I MHC heavy chains and human 3-2 microglobulin to bind to form heterodimers prior to presenting peptide to the MHC molecules intracellularly.

In another embodiment of the invention, peptides are added to transfected cells of the present invention in order to enhance the thermostability of the MHC molecules expressed by the cells. As noted above, peptides are preferably added to the culture medium.

39 Antigenic peptides that bind to the Class I molecules serve to thermostabilize the MHC molecules and also increase the cell surface expression. Cultures with added peptides which bind to the MHC molecules are thus significantly less susceptible to temperature challenge than cultures without added peptide.

In one embodiment of the present invention, antigenic peptides are presented to the transformed/transfected cell line in various forms. For example, an entire protein or other antigenic polypeptide may be degraded chemically or enzymatically, for example, and added to the cell line in this form. For example, a *°oo protein of interest is degraded with chymotrypsin and the resultant mixture of peptide "fragments" is added to a 15 transformed or transfected cell culture; these cells are then allowed to "choose" the appropriate peptides (which are often smaller peptides, preferably 8mers or 9mers).to load onto the Class I MHC molecules. Alternatively, an entire protein or polypeptide sequence may be cloned into an appropriate vector and inserted into a procaryotic cell, whereby the cell generates significant amounts of the antigenic polypeptide which are then harvested, purified, and digested into peptides which are then added to the transformed/transfected eukaryotic cell culture.

The cells again would be allowed to "choose" the peptides to load onto the expressed MHC.

6. Isolation of Resting or Precursor CD8' cells Resting (or naive or precursor) CD8' cells T-cells that have not been activated to target a specific antigen are preferably extracted from the patient prior to incubation of the CD8' cells with the transformed cultures of the present invention. It is also preferred that precursor CDS' cells be harvested from a patient prior to the initiation of other treatment 40 or therapy which may interfere with the CD8' 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 °C.

Various methods are also available for separating out and/or enriching cultures of precursor 15 CD8' 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* cells, are isolated by Ficoll-Hypaque gradient centrifugation (Pharmacia, Piscataway, NJ). PBL lymphoblasts may be used immediately thereafter or may be stored in liquid nitrogen after freezing in FBS containing 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 precursor 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, Scher and Mage, 41 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' cells from the T- and B-cell population of leucophoresed 15 patients. Depending upon 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., Immunobiolocy, supra.) For example, monoclonal antibodies OKT4 (anti-CD4, 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.

In a further embodiment, CD8' cells can be isolated by combining both negative and positive selection procedures. (See, e.g. Cai and Sprent, J. Exp.

Med. 179: 2005-2015 (1994)).

Preferably, the PBLs are then purified. For example, Ficoll gradients may be utilized for this 42 purpose. The purified PBLs would then be mixed with syngeneic Drosophila cells preincubated with the appropriate antigenic peptides.

7. In Vitro Activation of CD8' Cells In order to optimize the in vitro conditions for the generation of specific cytotoxic T-cells, the culture of antigen-presenting cells is maintained in an appropriate medium. In the preferred embodiment, the antigen-presenting cells are Drosophila cells, which are preferably maintained in serum-free medium Excell 400).

Prior to incubation of the antigen-presenting cells with the cells to be activated, precursor CD8 cells, an amount of antigenic peptide is added to 15 the antigen-presenting cell culture, of sufficient quantity to become loaded onto the human Class I molecules to be expressed on the surface of the antigen-presenting cells. According to the present invention, a sufficient amount of peptide is an amount that will allow about 200 to about 500,000 and preferably about 200 to 1,000-or more, human Class I MHC molecules loaded with peptide to be expressed on the surface of each antigen-presenting cell. Preferably, the antigen-presenting cells are incubated with peptide.

Resting or precursor CD8' cells are then incubated in culture with the appropriate antigen-presenting cells for a time period sufficient to activate and further enrich for a population of CD8' cells. Preferably, the CD8' cells shall thus be activated in an antigen-specific manner. The ratio of resting or precursor CD8' (effector) 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 43 culturing conditions and the nature and severity of the disease condition or other condition for which the within-described treatment modality is used. Preferably, however, the lymphocyte:antigen-presenting cell (e.g.

Drosophila cell) ratio is preferably in the range of about 30:1. to 300:1. For example, in one embodiment, 3 X 107 human PBL and 1 X 106 live Drosophila cells were admixed and maintained in 20 ml of RPMI 1640 culture medium.

The effector/antigen-presenting 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* cells. In general terms, the oo optimum time is between about one and five days, with a 15 "plateau" i.e. a "maximum" specific CD8' activation level generally being observed after five days of culture. In one embodiment of the present invention, in vitro activation of CD8' cells is detected within a brief period of time after transfection of a cell line. In one embodiment, transient expression in a transfected cell line capable of activating CD8' cells is detectable within 48 hours of transfection. This clearly indicates that either stable or transient cultures of transformed cells expressing human Class I MHC molecules are effective in activating CD8' cells.

Preferably, the enrichment and concordant activation of CD8' cells is optimal within one week of exposure to antigen-presenting cells. Thereafter, in a preferred embodiment, the enriched and activated CD8' cells are 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' cell preparation is further expanded by maintenance in culture for a period of time to obtain a population of 44 activated CD8* cells. This period may vary depending on the replication time of the cells but may generally be 14 days. Activation and expansion of CD8' cells has been described by Riddell et al., Curr. Opin. Immunol., 484-491 (1993) 8. Separation of CD8' Cells from Drosophila Cells Activated CD8' cells may be effectively separated from the stimulator Drosophila) cells using one of a variety of known methods. For example, monoclonal antibodies specific for the stimulator cells, for the peptides loaded onto the stimulator cells, or for the CD8' cells (or a segment thereof) may be utilized to bind their appropriate complementary ligand.

15 Antibody-tagged cells may then be extracted from the stimulator-effector cell admixture via appropriate means, via well-known immunoprecipitation or immunoassay methods.

9. Administration of Activated CD8' Cells Effective, cytotoxic amounts of the activated SCD8* cells can vary between in vitro 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 1 X 10' to about 1 X 1012, more preferably about 1 X 1 0 to about 1 X 10", and even more preferably, about 1 X to about 1 X 10 1 activated CD8' cells are utilized for adult humans, compared to about 5 X 10 5 X 10" cells used in mice.

Preferably, as discussed above, the activated CD8' cells are harvested from the Drosophila cell culture prior to administration of the CD8' cells to the individual being treated. It is important to note, 45 however, that unlike other present and proposed treatment modalities, the present method uses a cell culture system Drosophila cells) that are not tumorigenic.

Therefore, if complete separation of Drosophila cells and activated CD8' cells is not achieved, there is no inherent danger known to be associated with the administration of a small number of Drosophila cells, whereas administration of mammalian tumor-promoting cells may be extremely hazardous.

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' cells via intravenous infusion is appropriate.

HLA Typing 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 20 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 Drosophila cells preincubated with the appropriate antigenic peptides in therapeutic applications relating to viral infections, cancers, or malignancies, peptides derived from viral- or cancer-specific proteins.

Continuing to use viral or malignant conditions as an example, 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 46 such antigen, cDNA encoding viral- or cancer-specific proteins is cloned into a bacterial expression plasmid and used to transform bacteria, via methods disclosed herein.

After HLA typing, if Drosophila cells expressing the preferred HLA are not available, cDNAs encoding the preferred HLA may be cloned via use of the polymerase chain reaction. The primers disclosed in section B.1. above (SEQ ID NO 1 through SEQ ID NO 12) may be used to amplify the appropriate HLA-A, -F, or -G cDNAs in separate reactions which may then be cloned and sequenced as described in the methods disclosed for HLA A2.1 below. Stable cell lines expressing the cloned HLA may then be established in the 15 Drosophila cells. Alternatively, a population of insect cells transiently expressing a bulk population of cloned recombinant molecules from the PCR reaction may be used for in vitro CD8' activation.

EXAMPLES

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

Example 1 Expression of Human Class I MHC Molecules A. Preparation of pRmHa-3 Expression Vector The pRmHa-3 expression vector for use in expressing MHC proteins in Drosophila Schneider 2 (S2) cells as described in this invention was constructed by ligating a Sph I linearized pRmHa-1 DNA expression vector with a DNA fragment resulting from a Sph I restriction digest of a pRmHa-2 expression vector as described below.

The ligating of pRmHa-1 with the pRmHa-2 fragment in this manner was performed to remove one of two Eco RI restriction endonuclease cloning sites present in pRmHa-1. Thus, the resultant pRmHa-3 expression vector contained only one Eco RI restriction site in the 47 multiple cloning site (polylinker) into which various MHC-encoding DNA fragments were inserted as described in the Examples.

1. Preparation of pRmHa-1 Expression Vector The pRmHa-1 expression vector, containing a metallothionein promoter, metal response consensus sequences (designated MT) and an alcohol dehydrogenase (ADH) gene containing a polyadenylation signal isolated from Drosophila melanogaster, was constructed as described by Bunch et al., Nucl. Acids Res. 16: 1043-61 (1988). A schematic of the final pRmHa-1 construct is shown in Figure 2. The plasmid expression vector, pUC18, having the ATCC accession number 37253, was used as the source vector from which subsequent vectors described herein were derived. The pUC18 plasmid contains the following restriction sites from 5' to 3' in the multiple cloning site, all of which are not illustrated in the schematic representations of the pUC18-derived vectors in Figure 1: Eco RI; Sac I; Kpn I; Sma I and Sma I located 20 at the same position; Bam HI; Xba I; Sal I, Acc I and Hinc II located at the same position; Pst I; Sph I and Hind III. The pUC18 vector was first digested with Hind III to form a linearized pUC18. Blunt ends were then created by filling in the Hind III ends with DNA polymerase I large fragment as described by Maniatis et al., Molecular Cloning: A Laboratory Manual, eds. Cold Spring Harbor Laboratory, New York (1982).

The resultant linearized blunt-ended pUC18 vector was ligated with a 740 base pair (bp) Hinf I fragment from the Drosophila melanogaster ADH gene containing a polyadenylation signal. The ligated ADH allele was first isolated from the plasmid pSACI, described by Goldberg et al., PNAS USA 77: 5794-5798 (1980), by digestion with Hinf I followed by blunt ending with Klenow resulting in the nucleotide sequence listed 48 in SEQ ID NO 14. The pSACI vector containing the ADH allele was constructed by subcloning into pBR322 (ATCC accession number 31344) a 4.7 kilobase (kb) Eco RI fragment of Drosophila DNA selected from a bacteriophage lambda library containing random, high molecular weight (greater than 15 kb). The 5' Hinf I restriction site occurred naturally in the ADH gene at position 1770 as described by Kreitman, Nature 304: 412-417 (1983). The 3' Hinf I site was derived from the pUC18 vector into which the ADH gene had been cloned. This position was four bases 3' to the Xba I site at position 2500 of the ADH gene. The ADH segment extended from the 35 bp o*o upstream of the polyadenylation/cleavage sequence in the 3' untranslated portion of the ADH mRNA to 700 bp 15 downstream of the polyadenylation signal. The resultant pUC18-derived vector containing the ADH gene fragment was designated pHA-1 as shown in Figure 1.

The 421 bp Eco RI/Stu I MT gene fragment was obtained from a clone containing DNA of approximately 15.3 kb in a Drosophila melanogaster genomic DNA library.

The library, prepared with a Mbo I partial digestion of imaginal DNA, was cloned in the lambda derivative EMBL4.

The fragment contained the MT promoter and metal response consensus elements of the Drosophila MT gene (Maroni et al., Genetics 112: 493-504 (1986)). This region, containing the promoter and transcription start site at nucleotide corresponded to position -370 to nucleotide position +54 of the MT gene (SEQ ID NO 13).

The resultant fragment was then ligated into pHA-1 expression vector prepared above that was previously linearized with Eco RI and Sma I. The 3' blunt end in MT created by the Stu I digest was compatible with the blunt end in pHA-1 created by the Sma I digest. The resultant pUC18-derived vector containing a 5' Drosophila MT gene fragment and a 3' ADH gene fragment was designated 49 pRmHa-1. The pRmHa-1 expression vector, shown in Figure 2, contained the origin of replication (ori) and the beta-lactamase gene conferring resistance to ampicillin (Amp') from pUC18 as shown in Figure 1 on the pHa-1 vector. The diagram of pRmHa-1 also shows the to 3' contiguous positions of the MT gene fragment, the multiple cloning site and the ADH gene fragment. The pRmHa-1 vector was used as described in c. below in the construction of the pRmHa-3 expression vector.

2. Preparation of DRmHa-2 Expression Vector The construction of pRmHa-2 is shown in Figure 1. For constructing the pRmHa-2 expression vector, the MT fragment prepared above was inserted into the pUC18-derived vector pHA-1 as described for constructing pRmHa-1 above with a few modifications.

An Eco RI linker was added to the Stu I site of the Eco RI/Stu I-isolated MT gene fragment prepared above to form a metallothionein fragment having Eco RI restriction sites on both ends. The resultant fragment was then 20 ligated into the ADH fragment-containing pUC18 expression vector that was previously linearized with Eco RI. The resultant pUC18-derived vector containing a 5' Drosophila MT gene fragment and a 3' ADH gene fragment having two Eco RI restriction sites 5' to the multiple cloning site was designated pRmHa-2. The pRmHa-2 expression vector, shown in Figure 1, contained the origin of replication (ori) and the beta-lactamase gene conferring resistance to ampicillin (Amp') from pUC18. The diagram of pRmHa-2 also shows the 5' to 3' contiguous positions of the MT gene fragment, the multiple cloning site and the ADH gene fragment. The pRmHa-2 vector was used along with pRmHa-1 as described in c. below in the construction of the pRmHa-3 expression vector.

50 3. Preparation of pRmHa-3 Expression Vector To prepare the pRmHa-3 expression vector that had only one Eco RI restriction site, a fragment from pRmHa-2 was ligated into pRmHa-1. For this construction, pRmHa-2, prepared in b. above, was first digested with Sph I. The resultant Sph I fragment beginning in the middle of the MT gene and extending to the Sph I site in the multiple cloning site was first isolated from the pRmHa-2 vector and then ligated into pRmHa-1 prepared in A.1. above. The pRmHa-1 vector was previously modified to remove the Eco RI restriction site 5' to the MT gene fragment then linearized with Sph I. This process is o •'co schematically illustrated in Figure 2. To remove the Eco RI site in pRmHa-1, the vector was first digested with S 15 Eco RI to form a linearized vector, then blunt ended with Mung Bean nuclease and religated.

The pRmHa-1 vector lacking an Eco RI site was then digested with Sph I to remove the region corresponding to the Sph I fragment insert from pRmHa-2 20 and form a linearized pRmHa-1 vector. The Sph I fragment from-pRmHa-2 was then ligated into the Sph I linearized pRmHa-1 to form the pRmHa-3 expression vector. A schematic of the pRmHa-3 vector is shown in Figure 3.

The relative positions of the various restriction sites from the pUC18 vector from which pRmHa-3 was derived are indicated on the figure. In addition, the relative positions and lengths of the MT and ADH gene fragments separated by the multiple cloning site (polylinker).into which the MHC gene of interest is cloned are indicated on the figure. The pRmHa-3 vector, being derived from pUC18, contains the pUC18 origin of replication and beta-lactamase gene conferring ampicillin resistance.

Thus, MHC encoding DNA fragments as prepared in this invention and cloned into the multiple cloning site of 51 pRmHa-3 were transcriptionally regulated by the MT promoter and polyadenylated via the ADH gene.

B. cDNA Synthesis Detailed descriptions of the cDNA of Class I MHC molecules of various HLA groups can be found in U.S.

Patent No. 5,314,813 to Peterson et al. which has been incorporated by reference.

cDNAs encoding any preferred HLA may be cloned via use of the polymerase chain reaction. The primers disclosed in section B.1. above (SEQ ID NO 1 through SEQ ID NO 12) may be used to amplify the appropriate HLA-A, or -G cDNAs in separate reactions which e may then be cloned and sequenced as described in the methods disclosed for HLA A2.1 above. Preparation of cDNA from human cells is carried out as described in Ennis, et al., PNAS USA 87: 2833-2837 (1990). Briefly, a blood sample is obtained from the individual and cells are collected after centrifugation and used to prepare total RNA. First strand cDNA is synthesized by using 20 oligo(dT) and avian myeloblastosis virus reverse transcriptase. The resulting cDNA is used in a PCR amplification reaction utilizing the appropriate primer(s) as noted in section B.1. above, and a GeneAmp kit and thermal cycler (Perkin-Elmer/Cetus). Reaction conditions are preferably as follows. 100ng cDNA template and 50 picomoles of each oligonucleotide primer are used. Thirty cycles are run as follows: one minute at 94 oC; one minute at 60 oC; and one minute, 30 seconds at 72 oC. The PCR reaction is then heated to 100 OC for 10 minutes to kill the Taq polymerase and the ends of the DNA made blunt by T4 polymerase (Stratagene, San Diego, CA).

To synthesize HLA A2.2, cDNA encoding a complete A2.2 (see Holmes, et al., J. Immunol. 139: 936-41 (1987), for the published sequence) is cloned into 52 an M13mpl9 plasmid, a commercially available bacteriophage vector (Stratagene, La Jolla, CA). cDNA is synthesized by PCR using primers derived from the published sequence of A2. The cDNA is released from an M13mpl9 clone as a Not I (overhang filled with Klenow)/Eco RI fragment. (Klenow fragments are part of the E. coli DNA polymerase I molecule, produced by the treatment of E. coli DNA pol I with subtilisin. They are used to "fill out" 5' or 3' overhangs at the ends of DNA molecules produced by restriction nucleases.) The Not I/Eco RI fragment is inserted into pSP64T digested with SBg III (ends filled with Klenow) and Eco RI. pSP64T is an SP6 cloning vector designed to provide 5' and 3' o flanking regions from an mRNA which is efficiently 15 translated (0-globin) to any cDNA which contains its own initiation codon. This translation SP6 vector was constructed by digesting pSP64-X3m with Bal I and Bst EII, filling in the staggered ends with T4 DNA polymerase and adding a Bgl II linker by ligation. Bal I cuts the 20 3-globin cDNA two bases upstream of the ATG (start codon) and Bst EII cuts eight bases upstream of the TAA (stop codon). There is only one Bgl II si-te in pSP64T so that restriction enzymes cutting in the polylinker fragment, from Pst I to Eco RI can still be used to linearize the plasmid for transcription. (See Kreig and Melton, Nucleic Acid Res. 12: 7057-7070, (1984), which also describes the construction of the plasmid pSP64-Xom.) The resulting plasmid is cleaved with Eco RI (end filled with Klenow) and Hind III which is cloned into the pCMUII polylinker between Hind III and Stu I (See Paabo, et al., EMBO J. 5: 1921-1927 (1986).) The entire cDNA is removed as a Hind III (end filled with Klenow) Bam HI fragment which is cloned into pRmHa-3 cleaved with Sma I and Bam HI.

53 HLA A2.2 soluble form was prepared by engineering a stop codon into the above-described A2.2 cDNA immediately preceding the transmembrane domain. The modification is achieved by cleaving the A2.2 cDNA cloned in the eukaryotic expression vector pCMUII between Hind III 5' and Stu I 3' (see above) with Mbo II and Bam HI inserting the following oligonucleotides: primer: 5' GGAGCCGTGACTGACTGAG 3' (SEQ ID NO 17) 3' primer: 5' CCCTCGGCACTGACTGACTCCTAG 3' (SEQ ID NO 18) The resulting recombinant plasmid is cleaved with Hind III, the overhanging end filled with Klenow, then cut with Bam HI releasing a restriction fragment which is cloned into pRmHa-3 in the same way as A2.2 full length.

1. Construction of Murine ICAM-1 Expression Vector Spleen cells were isolated from Balb/c mice.

The spleen cells were stimulated with conA; mRNA was 20 isolated using the FastTrack kit (Invitrogen, San Diego, according to the manufacturers' instructions. cDNA was synthesized from the mRNA using AMV reverse transcriptase kit (Promega, Madison, WI) according to the manufacturers' instructions. Based on the published cDNA nucleotide sequence (Siu, G. et al., J. Immunol. 143, 3813-3820 (1989) the following oligonucleotides were synthesized as PCR primers: TTTAGAATTCAC CATGGCTTCA ACCCGTGCCA AG TTTAGTCGACTC AGGGAGGTGG GGCTTGTCC The cDNA synthesized was subjected to PCR using these primers. The product was cleaved with the restriction enzymes Eco RI and Sal I and ligated into pRmHa-3, which had been digested with the restriction enzymes Eco RI. and Sal I.

54

SSS@

0 000* 0 09 *0 6055 2. Construction of Murine B7.1 Expression Vector Spleen calls were isolated from Balb/c mice and stimulated with conA. Messenger RNA was isolated using the FastTrack kit (Invitrogen, San Diego, CA) according to the manufacturer's instructions. cDNA was synthesized from the mRNA using AMV reverse transcriptase kit (Promega, Madison, WI) according to the manufacturer's instructions.

Based on the published cDNA nucleotide sequence (Freeman, et al., J. Exp. Med. 174: 625-631 (1991)) the following oligonucleotides were synthesized as PCR primers: TTTAGAATTCAC CATGGCTTGC AATTGTCAGT TG TTTAGTCGACCT AAAGGAAGAC GGTCTGTTC 15 The cDNA synthesized was subjected to PCR using these primers. The product was cleaved with the restriction enzymes Eco RI and Sal I and ligated into pRmHa-3, which had been digested with the restriction enzymes Eco RI and Sal I.

3. Construction of Murine B7.2 Expression Vector IC-21 cells (obtained from ATCC) were propagated in RPMI 1640 medium containing 10% Fetal Calf Serum. mRNA was isolated from these cells using the FastTrack kit (Invitrogen, San Diego, CA) according to the manufacturer's instructions. cDNA was synthesized from the mRNA using AMV reverse transcriptase kit (Promega, Madison., WI) according to the manufacturer's instructions. Based on the published cDNA nucleotide sequence (Freeman, et al., J. Exp. Med. 178: 2185-2192 (1993)) the following oligonucleotides were synthesized as PCR primers: TTTAGAATTCAC CATGGACCCC AGATGCACCA TGGG TTTAGTCGACTC ACTCTGCATT TGGTTTTGCT GA The cDNA synthesized was subjected to PCR using these primers. The product was cleaved with the restriction 55 enzymes Eco RI and Sal I and ligated into pRmHa-3, which had been digested with the restriction enzymes Eco RI and Sal I.

The above expression constructs were transfected into Drosophila S2 cells using the calcium phosphate method as listed in Table 1. Stable cell lines were selected by including 500 yg/ml Geneticin in the cell culture medium.

TABLE 1 Transfected MHC I f/2 B7.1 B7.2 ICAM-1 phsneo Cells (CD (CD54) pg _g pg 80) pg pg pg 1 A 12 12 1 2 B 8 8 8 1 3 C 8 8 8 1 4 C 8 8 8 1 D 6 6 6 6 1 6 E 6 6 6 6 1 7 F 6 6 6 6 1 8 G 4.8 4.8 4.8 4.8 4.8 1 Human accessory and costimulatory molecules were cloned from human cell lines demonstrated to express these proteins by FACS analysis with monoclonal antibodies specific for the particular proteins.

Adhesion molecules belonging to the integrin family, ICAM-I (CD54) and LFA-3 (CD58), were cloned from human cell lines K562 and HL60, respectively. The K562 cells, originated from human chronic myelogenous leukemia, were obtained from ATCC (CCL-243) and cultured under conditions recommended RPMI with 10% fetal calf serum at 37 degrees C with 5% CO,) HL60 cells, originated from a human promyelocytic leukemia, and were obtained from ATCC (CCL-240) and were cultured according 56 to ATCC's recommendations. Costimulatory molecules B7.1 and B7.2 were also cloned from K562 and HL60 cells respectively.

4. cDNA Messenger RNA samples were prepared from each cell line from RNA isolated by the modified guanidinium thiocyanate method (Chromczynski, et al. Anal. Biochem.

162: 156-159, 1987) followed by poly A+ RNA selection on oligo(dt)-cellulose columns (Sambrook, et al, Molecular Cloning: A Laboratory Manual, Second Edition, 6.22-6.34, Cold Spring Harbor laboratory, CSH, NY), Induction of HL60 cells with vitamin D3 (usually required to express some cell surface molecules) was not required to obtain the B7.2 and LFA-3 molecules, the proteins were 15 expressed in the absence of induction. cDNA was prepared using AMV reverse transcriptase kit according to the manufacturers' instructions (Promega, Madison WI).

S5. PCR Primers PCR primers were designed and synthesized after obtaining copies of the known sequences from the GENEWORKS database (Intelligenetics) and considering the ends needed to clone into the appropriate vectors. They are as follows with the top sequence of each protein the 5'primer and the bottom one the 3'primer: B7.1 5'-ACCCTTGAAT CCATGGGCCA CACACGGAGG CAG-3' CCTTATACAG GGCGTACACT TTCCCTTCT-3' B7.2 5'-ACCCTTGAGC TCATGGATCC CCAGTGCACT ATG-3' GGTTAAAAAC ATGTATCACT TTTGTCGCAT GA-3' LFA-3 5'-ACCCTTGAGC TCATGGTTGC TGGGAGCGAC GCGGGG-3' 3'-ATTACCGGAT CCTTAAAGAA CATTCATATA CAGCACAATA CA-3' ICAM-1 5'-ACCCTTGAAT TCATGGCTCC CAGCAGCCCC CGGCCC-3' 3'-ATTACCGGAT CCTCAGGGAG GCGTGGCTTG TGTGTTCGG-3' 57 6. Expression of DNA Fragment The cDNA preparations from each of the cell lines was used to clone the desired proteins. The polymerase chain reaction was used to generate cDNA fragments utilizing the appropriate PCR primer (see above). The appropriate DNA fragments.were cloned into the Drosophila fly vector pRMHA-3. Plasmid preparations have been prepared from all of the preparations and are now ready for transfection into the fly cells.

Human 3-2 microglobulin cDNA is prepared using a published partial cDNA sequence (see Suggs, et al., PNAS 78: 6613-17, 1981) is used as a template for a polymerase chain reaction (PCR) with the following primers: 5' primer

GCTTGGATCCAGATCTACCATGTCTCGCTCCGTGGCCTTAGCTGTGCTCGC

GCTACTCTC 3' (SEQ ID NO 3' primer 20 5' GGATCCGGATGGTTACATGTCGCGATCCCACTTAAC 3' (SEQ ID NO 16) The primers are used in a standard PCR reaction (see Nilsson, et al., Cell 58: 707 (1989)). The reaction products are extracted with phenol, purified using a Geneclean kit (Bio 101, San Diego, CA), digested with Bam HI and cloned into the Bam HI site of pBS (Stratagene, La Jolla, CA). After verification of the sequence, this Bam HI fragment is cloned into the Bam HI site of pRmHa-3.

As noted in the Examples, murine Class I cDNA was utilized in various instances. Murine Class I cDNA was prepared as follows.

H-2Kb: cDNA encoding a complete K h molecule is obtained from an expression plasmid pCMU/K h constructed as follows. A partial H-2K cDNA missing the leader 58 sequence and most of the alpha I domain is prepared according to the method of Reyes, et al., PNAS 79: 3270-74 (1982), producing pH202. This cDNA is used to generate a full-length molecule. The missing sequence is provided using a genomic clone encoding H-2K" (Caligan, et al., Nature 291: 35-39, 1981) as a template in a PCR reaction, using a 5' primer flanked by a Not I site, followed by 21 nucleotides encoding the last seven amino acids of the leader sequence and 18 nucleotides complementary to the beginning of the alpha I domain and a 3' primer complementary to the region encompassing the Sty I site. The resulting fragment is ligated with pH202 at the Sty I site. The 5' sequence encoding the remainder of the signal sequence is obtained form the Db cDNA (see below) as a Bam HI/Not I fragment. The entire coding sequence is cleaved from the expression plasmid as a Bam HI fragment and cloned into pRmHa-3 cleaved with Bam HI.

H-2Ld: cDNA encoding a complete L' molecule is 20 obtained from an expression plasmid pCMUIV/Ld (see Joly and Oldstone, Gene 97: 213, 1991). The complete cDNA is cleaved from a eukaryotic expression vector pCMU IV/Ld as a Bar HI fragment and cloned into pRmHa-3 as Kb.

As noted previously, the pCMU vector (pCMUIV) is derived from eukaryotic expression vector pC81G as described in Nilsson, et al., supra. Vector pC81G, in turn, is derived from pA81G (Paabo, et al., Cell 33: 445-453 (1983)) according to the method disclosed in Paabo, et al., EMBO J. 5: 1921-7 (1986).

H-2D': cDNA encoding a complete D" molecule is obtained from expression plasmid pCMUIV/D' (see Joly and Oldstone, Science 253: 1283-85, 1991). The complete cDNA is cleaved from a eukaryotic expression vector pCMUIV/Dh as a Bam HI fragment and cloned into pRmHa-3 as K".

59 Murine 0-2 microglobulin: full-length murine 0-2 microglobulin cDNA is obtained as a Hind III (filled with Klenow)/Bgl II fragment from pSV2neo (ATCC No. 37149) mouse 0-2 microglobulin cDNA and cloned into pRmHa-3 cleaved with Sma I and Bam HI.

Vector phshsneo confers neomycin (G418) resistance and is a derivative of phsneo (pUChsneo) with an additional heat-shock promoter (hs) sequence, which may be synthesized from commercially-available pUC8 as described in Steller, et al., EMBO J. 4: 167 (1985). The heat shock promoter contained in these vectors is the hsp70 promoter. Other useful vectors conferring neomycin resistance (G418 resistance) include cosmid vector smart2 (ATCC 37588), which is expressed under the control of Drosophila hsp70 promoter, and plasmid vector pcopneo (ATCC 37409).

C. Insertion of Genes into Expression Vectors The restriction products are subjected to electrophoresis on a 1% agarose gel (Maniatis, et al., Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory (1982)). The restriction fragments encoding the cDNAs are excised from the gel and purified away from the agarose using "Geneclean", according to manufacturers' directions (Bio 101, San Diego, CA). The expression plasmid pRmHa-3 (Figure 3) is cleaved with the appropriate restriction enzymes in One Phor All buffer according to the manufacturer's directions (Pharmacia, Piscataway, NJ) and treated with alkaline phosphatase as described in the manufacturer's literature (Boehringer Mannheim, Indianapolis, IN). One hundred ng of cleaved and phosphatased pRmHa-3 vector is mixed with 300ng of agarose gel purified Class I MHC heavy chain cDNA or 0-2 microglobulin cDNA and ligated using T4 DNA ligase and One Phor all buffer as described in the manufacturers' literature. After incubation at 16 OC for five hours, 60 the ligation mixture is used to transform competent E.

coli JM83 (Maniatis, et al., supra (1982)).

Methods disclosed in Maniatis, et al., supra are used to prepare the cDNA needed. The presence of the MHC heavy chain cDNA and its orientation in the vector is determined by restriction mapping. Bacteria containing the vector with the cDNA in the correct orientation relative to the metallothionein promoter are used for large scale preparation of DNA using the alkaline lysis method and cesium chloride gradient purification. The amount of DNA obtained is determined spectrophotometrically.

D. Transfection and Labeling of S2 Cells e SS2 cells are grown in Schneider medium 15 (Gibco/BRL, Grand Island, NY) supplemented with 10% fetal calf serum (heat treated for one hour at 55 OC), 100 units/ml penicillin, 100mg/ml streptomycin, and ImM o glutamine. (For convenience, this supplemented medium is hereinafter referred to as Schneider medium.) Cells are grown at 27 OC and typically passaged every seven days by diluting 1:17 in fresh medium. Cells are-converted to growth in serum free media '(Excell 400 or 401 supplemented with 100 units/ml penicillin, 100 mg/ml streptomycin, ImM glutamine, and 500pg/ml G418 (JRH Biosciences, Lenexa, KS) by initial dilution at Excell 401. One week later, cells may be passaged into 10% Schneider medium/90% Excell 401 and one week later into 100% Excell 401.' Cells are maintained in this medium and passaged every seven days by diluting 2:17 in fresh medium.

X 10' S2 cells at a concentration of cells per ml are plated out in 85mm petri dishes. Twelve hours later, calcium phosphate/DNA precipitates, prepared as described bel9w (1 ml) are added dropwise to the cells. After 48 hours, the supernatant is carefully 61 removed and the cells transferred to a 175cm flask in a total volume of 50 ml in Schneider medium containing 500Sg/ml Geneticin (G418) (Gibco/BRL, Grand Island, NY) After 21 days, 20 ml of the culture is removed to a fresh flask containing 30 ml of Schneider medium containing 500jg/ml G418. Ten days later, a stable population of cells that weakly adhered to the flask and grew with a doubling time of approximately 24 hours is obtained and these cells are subsequently cultured and passaged in the selection media as described above. Frozen aliquots of these cells are prepared by collecting 5-20 X 106 cells by centrifugation and resuspending them in 1 ml of cell freezing media (93% fetal calf serum/7% dimethylsulfoxide). Aliquots are then placed at -70 °C for one week and subsequently transferred to liquid nitrogen storage.

Calcium phosphate precipitates are prepared as described by Paabo, et al. (EMBO J. 5: 1921-27 (1986)), except that 25ig of DNA is used per transfection. The following combinations of DNA are used to prepare the indicated transfectant: MHC Class I heavy chain alone: 23pg heavy chain expression vector DNA 2Ag of phshsneo DNA.

MHC Class I heavy chain 0-2 microglobulin: 11.5/g heavy chain expression vbctor DNA 11.5Ag of 3-2 microglobulin (human or mouse) expression vector DNA 2Ag of phshsneo DNA.

Other combinations of mouse genes are presented in Table 1.

Twenty-four hours prior to metabolic labeling, cells are plated out at a cell density of 3-5 X 106 cells/ml (10 ml/85mm petri dish) in Schneider medium containing ImM CuSO,. Thirty minutes prior to labelling the medium is aspirated from the dishes and the cells are washed with 2 X 10 ml of PBS and then incubated in Graces 62 insect medium minus methionine and cysteine (special order from Gibco/BRL, Grand Island, NY) for 20 minutes, and then in 1 ml of this medium containing 0.1mCi 35

S

Trans label (New England Nuclear; duPont, Boston, MA) After the labelling period, the labelling solution is aspirated and the cells are either lysed immediately on ice, with ice cold PBS/1% Triton X100 (1 ml) or after a chase period in the presence of methionine containing Schneider or Excell 400 medium (5 ml) (JRH Biosciences).

The chase medium is collected if soluble Class I MHC molecules are being analyzed.

:i"The following operations are all carried out with the lysates kept cold (less than 8 oC). The lysates were collected into Eppendorf tubes, centrifuged in a 15 microfuge tube for 15 minutes at 13,000 X g, transferred to a fresh tube containing 100pl of a 10% slurry of protein A sepharose and placed on an end-over-end rotator for two hours. Following a further centrifugation in the microfuge for 15 minutes, the cell lysates are ready for analysis.

In experiments utilizing murine MHC, S2 cells were transfected with the murine MHC recombinants described above using the CaPO, precipitation method; each heavy chain is transfected either alone or as a 50:50 mix with the vector encoding 0-2 microglobulin. A plasmid encoding neomycin resistance, phshsneo DNA, is included in each transfection such that a population of cells that stably expressed MHC Class I could be obtained by growing the transfectants in selection medium (Geneticin G418-sulphate, Gibco/BRL, Grand Island, NY).

E. Peptide Generation Antigenic peptides according to the present invention may be obtained from naturally-occurring sources or may be synthesized using known methods. In various examples disclosed herein, peptides are 63 synthesized on an Applied Biosystems synthesizer, ABI 431A (Foster City, CA) and subsequently purified by HPLC.

Isolation or synthesis of "random" peptides may also be appropriate, particularly when one is attempting to ascertain a particular epitope in order to load an empty MHC molecule with a peptide most likely to stimulate precursor CD8 cells. One may produce a mixture of "random" peptides via use of proteasomes (see, Example 2.B.6) or by subjecting a protein or polypeptide to a degradative process digestion with chymotrypsin or peptides may be synthesized.

While we have observed that the cell lines of the present invention are able to degrade proteins and polypeptides Sinto smaller peptides capable of being loaded onto human Class I MHC molecules, it is preferable to introduce smaller peptides 8-mers and 9-mers directly into the cell culture to facilitate a more rapid loading and expression process.

If one is synthesizing peptides, random 20 9- and 18-amino acid peptides, all varieties of amino acids are preferably incorporated during each cycle of the synthesis. It should be noted, however, that various parameters solvent incompatibility of certain amino acids may result in a mixture which contains peptides lacking certain amino acids. The process should thus be adjusted as needed by altering solvents and reaction conditions to produce the greatest variety of peptides.

As noted hereinabove, murine heavy chains complexed with human 0-2 microglobulin were stable at temperatures approximately 6-8 degrees higher than if complexed with murine It was also observed that the stabilities imparted by peptide and xenogeneic 0-2 microglobulin are additive. A large increase in the thermostability of the Class I molecules occurs if 8-9 64 mers are used, as compared to 12-25 mers; indeed, the difference between the stabilization imparted by the 8-9 mers compared with the larger peptides might be even greater than what was observed previously, for even though the peptides have been purified by HPLC, it is likely that there is some contamination of the larger peptides by 8-9 mers.

The thermostability of a Class I molecule is apparently dependent on: the origin of P-2 microglobulin; the presence of peptide; and the length and sequence of this peptide.

Previous work Patent No. 5,314,813 to Peterson et al.; Jackson et al., PNAS USA 89: 12117-12121 (1992)) has shown that Class I MHC heavy chains can bind S" 15 peptide either alone or when they are associated with 3-2 microglobulin. Surface expression of peptide-loaded human Class I MHC, however, appears to be best facilitated by loading the molecules with peptide after the heavy chains have complexed with 3-2 microglobulin.

20 1. Expression of Human MHC Once we determined that the thermostability of a Class I molecules is dependent on the origin of 3-2 microglobulin, the presence of peptide, and the length and sequence of this peptide, we utilized this information in the creation of cell lines capable of specifically activating CD8' cells via the expression of peptide-loaded human Class I MHC molecules.

Thermolability appears to be an inherent property of Class I molecules; it has presumably evolved to ensure that Class I molecules which contain either no peptide or a peptide of poor binding properties (that confers little thermostability) self-destruct. In this way, the cell minimizes the number of empty Class I molecules on its surface, for such a situation would presumably be dangerous in that exogenously derived 65 peptides could be bound and presented. Human Class I molecules expressed in insect cells with human /2 are not stable to extended incubation at 37 OC; neither are human Class I molecules expressed in the mutant cell line T2 which has been shown to be deficient in peptide loading onto the Class I molecules (Hosken and Bevan, Science 248: 367-70 (1990); Cerundolo, et al., Nature 345: 449-452 (1990)). Thus, it seems that the affinity between the heavy chain and P-2 microglobulin has been carefully conserved through co-evolution of the molecules such that empty Class I molecules, or those carrying poorly-binding peptides, self-destruct at the body temperature of the "host" organism.

SHuman Class I MHC molecules were expressed in 15 S2 cells. Cell lines co-expressing human 0-2 microglobulin and HLA A2.2Y, HLA A2.1, HLA B7, or HLA B27 were established using previously-described methods.

Briefly, cDNAs encoding the above proteins were cloned into the Drosophila expression vector pRmHa-3 and 20 cotransfected with a human 3-2 microglobulin-containing plasmid and phshsneo plasmid into S2 cells via methods disclosed herein. Three to four weeks later, the population of G418-resistanT-cells was diluted 1:5 with fresh selection media. Once a healthy growing population of cells was obtained, CuSO, was added to an aliquot of cells and 24 hours later, cells were analyzed via flow cytometry using a monoclonal antibody W6/32 (ATCC Bethesda, MD) which recognizes a monomorphic determinant of human Class I heavy chains when they are in association with P-2 microglobulin. (See Barnstable, et al., Cell 14: 9 (1978).) High levels of surface expression of each of the human Class I molecules were induced by the addition of CuSO, (data not shown). These stable populations were sorted for high expressing cells using cytofluorimetry as described below. It is these 66 sorted populations of cells which were used for all subsequent experiments.

Twenty-four hours prior to FACS analysis, CuSO 4 is added to the stably transfected S2 cells (3-4 X 106 cells/ml) to a final concentration of imM, thereby "switching on" expression from-the transfected genes.

Cells are plated out in 24-well cluster dishes (2 ml per well). Eight hours prior to FACS analysis, the CuSO, medium is replaced with fresh medium (1 ml) with or without peptide at a concentration of 50g/ml. 37 °C temperature challenges are carried out by transferring the dishes onto a flat surface in a 37 OC room at various time intervals prior to harvesting the cells for analysis.

15 To analyze surface expression of Class I MHC on the S2 cells, aliquots of cells (5 X 1 0

S

are transferred into tubes on ice, collected by centrifugation (1,000 X g for 4 minutes), resuspended in 3 ml of PBS/1% BSA, 0.02% sodium azide, collected by centrifugation and resuspended 20 in PBS/BSA ml) containing the appropriate primary antibody (ascites fluids Y3, 28:14:8S, 30.5.7, W6/32, diluted 1:200). Rabbit antisera are diluted 1:500 and B22.293 hybridoma supernatant is used directly. After a one hour incubation on ice, cells are washed twice in 3 ml of PBS/BSA and resuspended in 0.5 ml of PBS/BSA containing FITC labelled secondary antibody (Cappell, Durham, NC) and 1 ng/ml propidium iodide. After a minute incubation on ice, cells are washed once with PBS/BSA and resuspended in this buffer at a concentration of 1 X 10'/ml. Samples are then analyzed by FACS 440 (Becton Dickinson). Dead cells stained with propidium iodide, are excluded by including a live gate in the analysis.

For cell sorting, the same procedure outlined above is used, except that all staining operations are 67 carried out in a sterile hood. Solutions, including antibodies, are filter-sterilized, and Schneider media or Excell 400 is used in place of PBS/BSA. Cells that specifically bound the primary antibody are sorted using a Becton Dickinson cell sorter. Sorted cells (2-8 X 10 5 are washed once in medium before plating out at a concentration of 2 X 10 5 cells/ml.

F. Loading of Membrane-Bound Empty MHC Molecules by in vitro Incubation with Peptides In order to demonstrate that the human Class I molecules expressed on the surface of the Drosophila cells were empty, the cells were incubated at 37 OC for two hours and the cell surface expression was analyzed by 15 cytofluorimetry. The surface expression of both HLA B27 and A2.1 is greatly reduced if cells are incubated at 37 oC for 2 hours; however, preincubating the cells in HIV peptides known to bind to the Class I molecules 00 affords significant thermal stability to the Class I,.

while peptides that do not bind have little effect (see Figure (A 9-amino acid peptide ILKEPVHGV (SEQ ID NO 42) from the POL protein of HIV binds and stabilizes HLA A2.1. A nine-amino-acid peptide from the Vpr protein of HIV binds and stabilizes B27 (FRIGCRHSR; SEQ ID S 25 NO 41). These data show that the human Class I molecules expressed on the surface of Drosophila cells are empty and can be stabilized by binding specific HIV peptides.

Figures 4 and 5 show peptide-induced thermostabilization of HLA B27 and HLA A2.1 expressed on the surface of Drosophila cells by HIV peptides.

Drosophila cells expressing either HLA B27 or A2.1 were incubated with peptides where indicated and then either maintained at 28 OC or incubated at 37 OC for two hours prior to analysis of the surface expression of the Class I molecules by use of the antibody W6/32 (from ATCC and cytofluorimetry. The mean fluorescence of each 68 cell population is shown plotted against the incubation conditions The HIV POL peptide (ILKEPVHGV, SEQ ID NO 42) stabilizes A2.1 but not B27 (Figure while the HIV Vpr peptide (FRIGCRHSR, SEQ ID NO 41) stabilizes B27, but not A2.1 (Figure Example 2 Preparation of Synthetic Antigen-Presenting Cells A. Osmotic Loading Osmotic loading of SC2 and 3T3 cells with ovalbumin protein was carried out as described by Moore, et al., Cell 54: 777-785 (1988). The assay procedure is as follows. In a 96-well dish, 1 X 10' Drosophila cells (with or without peptide/protein loaded) or 3T3 cells were cocultured with 1 X 10 5 B3/CD8 T-cell hybridoma cells in 2001p of RPMI media supplemented with 10% fetal bovine serum. After 24 hours of incubation, 100il of the supernatant from these cultures was added to 100l of RPMI containing 5,000 CTLL cells. The cells were cocultured for 24 hours at 37 OC when 1Ci of 3

H

20 thymidine (Amersham) was added. After a further incubation of 15 hours at 37 oC, the incorporation of radiolabel into the CTLL cells was determined by scintillation counting.

Assays conducted with murine MHC also verified that the insect cells are capable of loading peptide onto the Class I molecules. Cells expressing as few as 200-500 MHC molecules containing a particular antigen can be detected by a T-cell. As the Drosophila cells do not accumulate chromium, an antigen presentation assay based on B3/CD8, a T-cell hybridoma, was used. B3/CD8 is a hybridoma between B3, cytotoxic T-cell specific for ovalbumin peptide 253-276 presented by H-2K h Class I molecules, and CD8- bearing IL-2-secreting cell line (see Carbone, et al., supra, 1989). Upon antigenic stimulation, B3/CD8 produces IL-2, measured by 'H 69 thymidine incorporation in IL-2-dependent T-cell line CTLL (Gillis, et al., J. Immunol. 120: 2027 91978)).

Thus, by measuring the amount of IL-2 produced, one can assay for T-cell recognition.

In order to provide an intracellular pool of ovalbumin protein from which OVA peptides can be derived, ovalbumin (Sigma Chem. Co., MO) was osmotically loaded into the cells as described by Moore, et al, supra (1988). Immediately after loading, the cells were mixed with the T-cell hybridoma. After two days' incubation, the medium was removed and assayed for IL-2. The amount of IL-2 was determined by the ability of the medium to support the growth of the IL-2-dependenT-cell line CTLL (Gillis, et al., supra, 1978), and growth was quantitated 15 by the amount of radioactive thymidine incorporated into the cells.

S2 or 3T3 cells transfected with K'/02 were incubated with ovalbumin protein (OvPro) or ovalbumin peptide, OVA 24 (OvPep) in isotonic (Iso) or hypertonic 20 (Hyp) media. (Murine cell line BALB/3T3 is available from the ATCC under accession number CCL 163.) After treatment, cells were cocultured with the T-cell hybridoma B3/CD8. .B3/CD8 is a T-cell hybridoma between B3 (Carbone, et al., J. Exp. Med. 169: 603-12 (1989)), cytotoxic T-cell specific for ovalbumin peptide 253-276 presented by H-2Kh Class I molecules, and CD8- bearing IL-2-secreting cell line. Upon antigenic stimulation, B3/CD8 produces IL-2, measured by 3 H thymidine incorporation in IL-2-dependent cell line CTLL (Gillis, et al., J. Immunol. 120: 2027 91978)). Thus, by measuring the amount of IL-2 produced, one can assay for T-cell recognition. The supernatant from the cocultures were analyzed for IL-2 by 'H thymidine incorporation by the IL-2-dependent cell line CTLL (ATCC No. TIB 214).

70 The amount of 3 H thymidine incorporated is plotted against the initial cell treatments.

It can be seen in Figure 6 that the T-cells responded well to the Drosophila cells if the ovalbumin peptide was added to the culture medium, but no recognition occurred if the cells were loaded with the ovalbumin protein. The MHC Class I molecules expressed on the cell surface of the insect cell are fully functional in that they can bind peptide if it is added to the culture medium and can present it in the correct context for it to be recognized by a T-cell.

o*o B. Optimization of In Vitro Conditions For the optimization of in vitro conditions for the generation of specific cytotoxic T-cells, the culture of Drosophila cell stimulator cells is preferably maintained in serum-free medium Excell 400).

SDrosophila cell stimulator cells are preferably incubated with >20pg/ml peptide. The effector:stimulator ratio (lymphocyte:Drosophila cell ratio) is preferably in the 20 range of about 30:1 to 300:1. The maximum specific CD8' is generally observed after five days of culture. The culture of target cells for killing-assay is preferably maintained in a serum-free medium.

Example 3 Stimulation of Proliferation and Differentiation of Armed Effector T-Cells We have found that Drosophila S2 cells transfected with MHC class I molecules and specific assisting molecules are able to stimulate primary responses from T-cells in vitro. We present data below in this example from a mouse model system. In this example, constructs coding for mouse MHC class I 32 microglobulin, specific assisting molecules were used and tested with CD8* cells from lymph nodes of T-cell receptor transgenic mice.

71 The data in Figure 7 provides evidence that the transfected Drosophila S2 cells express the protein products of the transfected murine genes. Flow cytometry using a fluorescence-activated cell sorter (FACS) and fluorescently labelled antibodies were used to demonstrate the expression of Ld (MHC molecule which includes heavy chain and P2) and the specific assisting molecules B7.1 (CD80) and ICAM-1 (CD54) molecules by transfected Drosophila S2 cells. Transfected cells were separated with a FACS to obtain cells expressing Ld molecules and were then maintained in vitro.

The transfection of Drosophila S2 cells is summarized in Table 2. The data show Ld, B7.1 and ICAM-1 expression measured by flow cytometry on the cell lines 15 after induction with CuS04 It is apparent that, relative to the control antibody (ctr Ab), all of the *transfectants express Ld molecules on the cell surface.

Likewise, cells cotransfected with Ld and B7.1 (Ld.B7) express B7.1 but not ICAM-1, whereas cells cotransfected 20 with Ld and ICAM-1 (Ld.ICAM) express ICAM-1 but not B7.1; triple transfection with Ld, B7.1 and ICAM-1 (Ld.B7.ICAM) led to expression of all three molecules.

Using a standard tissue culture system (Cai, Z.

and Sprent, J. (1994) J. Exp. Med. 179: 2005-2015), doses of 5 x 10" purified CD8* 2C lymph node (LN) cells were cultured at 37 oC with doses of 3 x 10 5 transfected fly cells peptides (10 pM final concentration). Peptides were synthesized by R. W. Johnson Pharmaceutical Research Institute (Sykulev, et al.(1994) Immunity 1: 15-22.

Proliferative responses were measured by adding 'HTdR (1 pCi/well) 8 hours prior to harvest. IL-2 production was measured by removing supernatants from the cultures at 48 hours and adding 50 Al supernatant to an IL-2 responsive indicator cell line (CTLL); proliferation of the indicator line was measured by addition of 'HTdR.

72 The data shown in Table 2 are the means of triplicate cultures. The transfected Drosophila S2 cells die rapidly at 37 oC and fail to incorporate 3 HTdR at this temperature.

The data in Table 2 demonstrate that the transfectants are able to stimulate primary responses of mouse T-cells.

Table 2. Capacity of transfected fly cells to stimulate primary proliferative responses and IL-2 production by CD8' lymph node cells from 2C T-cell receptor transgenic mice.

TABLE 2 3 HTdR incorporation (cpm x 10 3 with transfected fly cells expressing: 15 Assay Peptides Ld Ld Ld Ld Ld B7.1 added B7.1 ICAM-1 B7.1 combined with *9*e ICAM-1 L d ICAM-1 ICAM-1 Ld ICAM-1 Proliferation 0.2 0.1 0.3 0.2 (Day 3) p2Ca 0.2 0.3 1.5 142.0 QL9 0.2 60.9 73.9 263.7 132.9 IL-2 Production 0.3 0.2 0.1 1.2 (Day 2) p2Ca 0.2 0.2 0.1 64.6 0.3 QL9 0.1 -0.4 0.2 158.6 The 2C T-cell receptor (TCR) is strongly reactive to Ld molecules complexed with certain peptides, e.g. p2Ca (SEQ ID NO 46) or QL9 (SEQ ID NO 47). These two peptides have moderate to high affinity for soluble Ld molecules, 4 x M- for p2Ca, and 4 x 10' M- for QL9 (Sykulev. et When complexed to soluble L' molecules, the two peptides also have high binding affinity for soluble 2C TCR molecules. However, in both TCR binding and L'' binding, the QL9 peptide clearly has a higher affinity than the p2Ca peptide.

73 Table 2 shows that proliferative responses and IL-2 production by the responder 2C cells to the weaker peptide, p2Ca, requires that the stimulator Ld-transfected cells coexpress both B7.1 and ICAM-1; a mixture of cells expressing either Ld B7.1 or Ld ICAM-1 is nonstimulatory. By contrast, with the stronger peptide, QL9, Ld.fly cells expressing either B7 or ICAM elicit clearly-significant responses, although combined expression of B7 and ICAM generates much higher responses. In contrast to these findings on T-cell proliferation, IL-2 production in response to the QL9 peptide requires joint expression of B7 and ICAM; expression of these molecules on separate cells is .*ineffective.

The results show that Drosophila cells transfected with murine class I molecules and costimulatory molecules induce murine T-cells to mount.

primary proliferative responses and lymphokine (IL-2) production in response to peptide antigens. The system 20 is also applicable to human, T-cells and could be used to stimulate unprimed (or primed) T-cells specific for tumor-specific antigens in vitro; in vivo infusion of clonally-expanded T-cells specific for tumor-specific antigens might be therapeutic for patients with cancer.

Infusion of T-cells specific for viral antigens would be useful in patients with viral infections, e.g. HIV.

Example 4 Immobilization of Biotinylated MHC Molecules on Avidin-Coated Red Blood Cells NHS-LC-biotin, neutravidin and biotin-BMCC were purchased from Pierce (Rockford, IL). Sheep red blood cells were obtained from the Colorado Serum Company (Denver, CO). Drosophila S2 cells expressing L" and recombinant L d were prepared as described in Examples 1 and 2. Monoclonal antibodies 30.5.7 (anti-Ld) and 1B2 74 (anti-clonotypic antibody to the 2C T-cell receptor) were used as hybridoma cell culture supernatants.

The protocol used is described by Muzykantov and Taylor (Anal. Biochem. (1994) 223, 142-148). Briefly, SRBC were washed 4 times in phosphate buffered saline (PBS), biotinylated using NHS-LC-biotin, washed again 4 times in PBS, incubated with neutavidin, and finally washed 4 times and stored at 4 OC in PBS containing 3% fetal calf serum and 0.02% sodium azide.

Recombinant Ld was biotinylated using biotin-BMCC, a maleimide-coupled biotin which reacts with thiol groups. Ld displays a free thiol group, the side chain of cystein 121, which is not in the peptide binding site. Biotinylation was performed as recommended by the.

manufacturer. Unreacted biotin was removed using Centricon Biotinylated L' was immobilized by incubation.

at a final concentration of 0.2 mg/ml with avidin-coated SRBC for 30 minutes followed by washing in DMEM 20 containing 10% fetal calf serum. SRBC with attached Ld e.9 were used immediately.

T-cells expressing the 2C TCR transgene from lymph nodes of mice were purified by magnetic depletion.

Purified T-cells were consistently 97-98% positive for staining in flow cytofluorometry using the anti-clonotypic antibody 1B2.

Immobilization of biotinylated Ld on avidin-coated SRBC was done as indicated above.

Attachment was assessed using flow cytofluorometry using anti-Ld antibody 30.5.7.

A typical experiment is represented in Figure 8. The negative control (cells minus antibody) is shown in dotted lines. The filled peak comprises cells labeled with fluorescent antibody. 99.78% of the cells were labeled. Fluorescence intensity was in the same 75 range than the highest levels of intensity that we observed for Ld on synthetic antigen-presenting cells.

Kb (MHC molecule which includes heavy chain and g2) was also biotinylated using the same procedure. We could immobilize biotinylated Kb on avidin-coated SRBC as assessed by flow cytofluorometry (Figure 99.88% of the cells were labeled.

Rosetting experiments verified that the attached MHC molecules interacted functionally with T-cells. Drosophila S2 cells expressing Ld, Ld-coated SRBC were incubated with QL9 peptide (0.02 mM) or an irrelevant peptide (MCMV, 0.02mM) for 30 minutes on ice; 2C+ T-cells were then added, the proportion being 10 2C+ T-cells for 1 Drosophila S2 cell, or 10 SRBC for 1 2C+ 15 T-cell; the mixture was pelleted and kept on ice for at least 30 min. Cells were then carefully resuspended and rosettes were counted, a rosette being a Drosophila S2 cell bound to at least 3 2C+ T-cells, or a 2C+ T-cell bound to at least 3 SRBC. Rosettes were observed in all 20 cases. Typically, 30-40% of the lymphocytes were included in rosettes when QL9 peptide was added. No rosette was observed in the presence of the irrelevant peptide, although occasional attachment of a few single cells was observed.

These examples describe a new method to immobilize high amounts of MHC class I molecules on various surfaces (fly cells, red blood cells, latex beads) in native conformation as judged by monoclonal antibody binding and rosetting experiments (T-cell receptor binding). This method can be extended to other synthetic surfaces including artificial phospholipid membranes. Phosphatidylethanolamine as well as avidin-coupled phospholipids are particularly relevant to our studies. These phospholipids are commercially 76 available from Lipex Biomembrane Inc., Vancouver, BC, Canada.

Example Immobilization of Biotinylated MHC Molecules on Avidin-Coated Latex Beads Six micron diameter latex sulfate beads were purchased from Interfacial Dynamics Corporation (Portland, OR) and biotinylated according to the protocol described in Example 4.

Avidin-coated latex beads were prepared using a 1% suspension of the latex beads incubated in PBS containing Img/ml of neutravidin for one hour at room temperature. An equal volume of PBS containing 10% fetal calf serum was then added. After one hour of incubation S 15 at room temperature, the beads were washed 3 times and used for binding of recombinant biotinylated Ld.

Recombinant biotinylated L was immobilized by 4 incubation at a final concentration of 0.2 mg/ml with avidin-coated latex beads for 30 minutes followed by 20 washing in DMEM containing 10% fetal calf serum. SRBC with attached Ld were used immediately.

Rosetting experiments'verified that the attached MHC molecules on latex beads interacted i functionally with T-cells. Drosophila S2 cells expressing recombinant L d and Ld-coated latex beads were incubated with QL9 peptide (0.02 mM) or an irrelevant peptide (MCMV, 0.02mM) for 30 minutes on ice; 2C+ T-cells were then added, the proportion being 10 2C+ T-cells for 1 Drosophila S2 cell, or L"-coated latex beads for 1 2C+ T-cell; the mixture was pelleted and kept on ice for at least 30 min. Cells were then carefully resuspended and rosettes were counted, a rosette being a Drosophila S2 cell bound to at least 3 2C+ T-cells, or a 2C+ T-cell bound to at least 3 latex beads. Rosettes were observed in all cases. Typically, 30-40% of the lymphocytes were 77 included in rosettes when QL9 peptide was added. No rosette was observed in the presence of the irrelevant peptide, although occasional attachment of a few single cells was observed.

Example 6 Immobilization and Detection of Recombinant Protein Bound to Various Solid Supports Such as Plastic Microwell Plates The MHC molecules were immobilized by direct binding to microtiter plates (Corning) and detected as follows: MHC Kb molecules were diluted to desired concentration in PBS, e.g. 0.001mg/ml for 100 ng /well.

100ti1 of diluted Kh was added to each well on the plastic 15 microtiter plate. The plate was incubated for 1 hour at room temperature. After incubation, the plate was washed once with PBS and 2001 2% bovine serum albumin (BSA) in PBS and Tween (PBST) was added, and incubated for another hour at room temperature. The plate was 20 washed three times with PBST and biotinylated anti-K h mAb was added (1:2500) -in 2% BSA in PBS. The plate was incubated another hour at room temperature and washed three times with PBST. Avidin conjugated HRP was added (1:2500) in 2% BSA in PBS. Following another hour of incubation at room temperature, the plate was washed three times with PBST and H202 or thophenyldiamine was added. The reaction was stopped with H2SO,. Reaction product was detected colorimetrically at 490 nm.

Figure 10 shows the results of detecting the presence of MHC K" molecules using three different monoclonal antibodies.

Recombinant MHC K' molecules can alternatively be bound through biotin-avidin linked interactions with the substrate. In this embodiment, the microwell plates were coated with 100l avidin diluted in PBS to a 78 concentration of 0.001mg/ml. Excess avidin was removed by a PBS wash. The above procedure for presenting and detecting Kb binding followed.

Recombinant MHC molecules may alternatively be immobilized by a linkage based on a poly-histidine tag added to the-MHC interacting with the nickel bound to the substrate.

The above procedure for binding and detection is followed using nickel chelate coated microwell plates (Xenopore) and recombinant MHC molecules with a poly-histidine tag expressed using vector pRmHa/His 6 described above.

Example 7 Direct Binding of Peptide to Soluble, 15 Empty Class I MHC Molecules In Vitro A. Procedures H-2Kb: prepared as described above in Example 1.B.

H-2Kb Sol: Kb sol cDNA is a derivative of Kb, 20 encoding the extracellular portion of the Class I MHC molecule. Kb sol cDNA may be produced by PCR according to known methods, such as those described in Ennis, et al., PNAS USA 87: 2833-7 (1990) and Zemmour, et al., Immunoqenetics 33: 310-20 (1991). Specifically, cDNA encoding a truncated Kb molecule with a stop codon inserted at the end of the alpha 3 domain at amino acid position +275 is excised from the pCMU expression plasmid as a Bam HI fragment and cloned into pRmHa-3 as Kb cDNA.

The K' sol cDNA is a derivative of the complete Kb cDNA (see above) which is used as a template in a PCR reaction using a 5' oligonucleotide that encompassed the Sty I site, and the following 3' oligonucleotide: ATATGGATCCTCACCATCTCAGGGTGAGGGGC 3' (SEQ ID NO 43) 79 The resulting PCR fragment is blunt-end cloned into the Sma I site of pBS (Stratagene, La Jolla, CA), sequenced, and the remaining 5' sequence of Kb cloned into the Sty I site. A cDNA encoding the complete Khsol protein could be obtained as a Bam HI restriction fragment.

H-2D b and H-2Ld are prepared as discussed in Example 1.B. above.

The cDNAs encoding Kb aloa2a3 domains (274 residues) and murine 0-2 microglobulin (99 residues) were respectively cloned into the unique Bam HI site of an expression vector harboring the metallothionein promoter pRMHa-3 (Bunch, et al., Nucleic Acid Res. 16: 1043-1061 (1988)). Drosophila S2/M3 cells were transformed with these recombinant plasmids in addition to plasmid 15 phshsneo (containing a neomycin-resistance gene) by the calcium-phosphate precipitation method described previously. The transformed cells selected against neomycin-analog antibiotics G418 were grown at 27 OC in serum-free medium and soluble heavy-chain Kb and P-2 microglobulin were co-expressed by the addition of 0.7mM CuSO,.

The soluble, assembled heterodimer of Kb was purified from the culture supernatants by affinity chromatography using anti-Kb monoclonal antibody Y3, followed by ion-exchange chromatography on a Pharmacia Mono Q FPLC column according to the instructions of the manufacturer (Pharmacia, Piscataway, NJ).

SDS-polyacrylamide gel electrophoresis (SDS-PAGE) of the Kh preparation followed by staining with Coomassie blue showed only one band of relative molecular mass (Mr) at about 32,000 and one band of Mr at about 12,000 with no detectable impurities. The highly-purified K" was dialyzed against phosphate-buffered saline (PBS), filter-sterilized, and used for further study.

80 Extinction coefficient of the soluble K h ("KbSol") protein (43.2 kDa) is 69,200 M'cm at 280 nm.

The purified Kb sol (0.3AM) in PBS with or without 1% TX-100 were exposed to varying temperatures 40, 230, 320, 370, 420, and 47 0 C) for one hour.

The proteins were then immunoprecipitated by incubating with the monoclonal antibody Y3 and protein A sepharose beads (Pharmacia, Piscataway, NJ) at 4 OC for two hours, respectively. The samples were analyzed by 12.5% SDS-PAGE, followed by staining with Coomassie blue. The two thick bands on the gel are heavy and light chains of antibody Y3. In another procedure, Kbsol (0.31iM) were incubated with 50M of peptides in PBS at 23 OC for two hours to allow for Kbsol-peptide complex formation.

15 After the addition of 1% TX-100, the samples were exposed to 12 oC, 37 OC, or 47 OC temperatures for one hour. The complexes were immunoprecipitated and analyzed by SDS-PAGE as described above. In a third procedure, Khsol (2.7AM) were incubated with 5OpM of OVA-8, VSV-8 or SEV-9 20 peptides, respectively, at 23 OC for two hours. The samples were applied on a 5% polyacrylamide IEF gel. IEF was run from pH 5-7 and the gel was stained with silver.

Next, VSV-8 peptide was radioiodinated using the chloramine-T method (Hunter, et al., Nature 194: 495-6 (1962)) and free 12'I was removed by C1 column (OPC cartridge, Applied Biosystems, Foster City, CA). The labelled peptide was further purified by C, 1 reverse-phase HPLC. After elution, the labelled peptide was lyophilized and resuspended in PBS.

The specific activity of 1 I]VSV-8 (about 250 Ci/mmole) was determined spectrophotometrically by using extinction coefficient of tyrosine at 274nm (1420 M-cm') First, K"sol (0.5M) was mixed with [(12I]VSV-8 and unlabelled VSV-8 (50nM) at 23 OC for 16 hours to allow for complex formation. A portion of the sample 81 was analyzed by gel filtration (Superose 12, Pharmacia, Piscataway, NJ) in PBS. After elution, radioactivity contained in each fraction (0.05ml) was measured.

Protein was monitored by absorbance at 280nm.

In a second procedure, [1 25 I] VSV-8 (0.39nM) was mixed with various concentrations of Kbsol in PBS containing 1% bovine serum albumin (BSA). After incubation at 23 OC for 2-16 hours, Kbsol-peptide complexes were separated from free peptide by small gel filtration (Bio-Gel P30, BioRad, Richmond, CA) in PBS.

gel filtration permitted over 95% separation of bound and free peptide within about 5 minutes. Radioactivity of bound and free peptides was measured and the data were analyzed by linear regression. At maximal levels of Kbsol offered, about 65% of the total labelled peptides were bound. This maximal binding capacity of labelled peptide to Kbsol protein deteriorated over time, presumably due to radiation by 125I bound to VSV-8.

In a third procedure, each sample contained 20 0.39nM of [12 5 I]VSV-8 (about 18,000 cpm), unlabelled peptides at the indicated concentration, and 30nM of 'Ksol that gives about 50% of the SI]VSV-8 binding in the absence of unlabelled peptide at a final volume of 72Al. All components were dissolved and diluted in PBS containing 1% BSA. After incubation for 2-16 hours at 23 oC, 50Al samples were analyzed by P30 gel filtration as described above. The dissociation constants for unlabelled peptides were determined from molar concentrations of [12 5 I]VSV-8 and unlabelled peptides giving 50% inhibition of 12 SI]VSV-8 binding to Kbsol as described. (See Muller, et al., Meth. Enzvmol. 92: 589-601 (1983).) K'sol (0.3jiM) and [1 25 I]VSV-8 (0.39nM) were then incubated at 40 C, 23 OC, and 37 oC, and the association was determined at various times by P30 gel filtration.

82 Murine P-2 microglobulin was added, when necessary, before the incubation at the indicated concentration.

The murine 0-2 microglobulin was prepared by affinity chromatography using anti-3-2 microglobulin polyclonal antibody K355 from culture supernatants of the recombinant Drosophila cells. (See also Logdberg, et al., Molec. Immun. 14: 577-587 (1979).) In another experiment, Kbsol (0.3AM or 1.8AM) and [12sI]VSV-8 (2.4nM) were incubated at 23 OC for two hours, and the peptide-Kbsol complexes were isolated by P30 gel filtration. The samples contained very small amounts of 5 2]VSV-8 and Ksol complexes (at the maximum, 2.4nM) and empty Kbsol at final concentration of about 50 to 300nM. To some samples, 3yM of P-2 microglobulin, 3AM of 15 0-2 microglobulin plus 20yM of unlabelled VSV-8, 20pM of unlabelled VSV-8, or 1% TX-100 were added. The samples were incubated for various times at 37 oC and the degree of dissociation was determined by passage over columns.

20 B. Discussion Class I MHC molecules present antigenic peptides to cytotoxic T lymphocytes. Direct binding of peptide to Class I molecules in vitro has been hampered -by either the presence of previously bound peptides at the binding site (Chen and Perham, Nature 337: 743-5 (1989)) or the lack of binding specificity. (See, e.g., Frelinger, et al., J. ExD. Med. 172: 827-34 (1990); Choppin, et al., J. Exp. Med. 172: 889-99 (1990); Chen, et al., J. Exp. Med. 172: 931-6 (1990).) In vitro analysis of peptide binding to soluble, empty Class I molecules purified from Drosophila cells transformed with truncated H-2Khsol and murine 3-2 microglobulin genes is disclosed herein. The results demonstrate that peptide binding is very rapid and naturally processed peptides (octapeptides; see, Van Bleek, et al., Nature 348: 83 213-6 (1990); Falk, et al., Nature 351: 290-6 (1991)) have the highest affinities to Kbsol of the nanomolar range and indicate that K'sol complexed with octapeptides are stable, whereas those complexed with slightly shorter or longer peptides are short-lived. Interactions between free heavy chain and 0-2 microglobulin is basically reversible in the absence of detergent. Peptides spontaneously bind to empty Class I molecules without dissolution of 0-2 microglobulin. However, excess 0-2 microglobulin apparently promotes the binding of peptide to empty Class I as a consequence of reassociation of free heavy chain with 0-2 microglobulin under conditions where the heterodimers are unstable.

Soluble H-2Kh molecules (composed of the ala2a3 15 domain of heavy chain) and murine 0-2 microglobulin, were purified from the culture supernatants of Drosophila cells which were concomitantly transformed with the truncated heavy chain and 3-2 microglobulin genes.

Preliminary examinations suggested that Drosophila cells 20 express Class I MHC molecules devoid of endogenous peptides on the cell surface. Some of the properties of empty Class I molecules include the-observation that they are less stable at 37 OC and their structure is •stabilized by the binding of peptide. (See, e.g., Schumacher, et al., Cell 62: 563-7 (1990); Ljunggren, et al., Nature 346: 476-80 (1990).) To confirm that ourified soluble Kh are also empty, their thermal stability in detergent-free solution was examined.

Surprisingly, the proteins heated for one hour at 47 °C were well recovered by immunoprecipitation using a conformational antibody, Y3. This unexpected result led us to add detergent, 1% Triton X-100 (polyoxyethylene (9) octyl phenyl ether), to the protein solution, since similar experiments to test the stability of Class I molecules have always been conducted in detergent lysates 84 (See Schumacher, et al., cited supra). The results obtained in the presence of detergent show that the purified K'sol is now unstable at 37 OC. This and other lines of evidence suggest that K"sol heterodimer disassembles into the heavy chain and microglobulin at elevated temperatures and that detergent may prevent 0-2 microglobulin from reassociating with dissociated free heavy chain (see below). Second, the possibility of stabilizing purified Kbsol with peptides was studied.

The results of the first-described examination demonstrated that the proteins can be stabilized only when they are mixed with octapeptide (vesicular stomatitis virus nucleocapsid protein [VSV-8], see Table 3 below) which is shown to be a naturally processed 15 peptide (see Van Bleek, et al., cited supra). These observations are consistent with the characteristics of o empty Class I molecules mentioned above.

Independent support that the purified Kbsol molecules are empty is provided by isoelectric focusing (IEF) under native conditions (data not shown). The I soluble Kb purified from Drosophila cells exhibited a much simpler pattern than HLA-A2 molecules purified from human lymphoblastoid cell lines (see Figure 3 in Silver, et al., Nature 350: 619-22 (1991)). The complicated pattern of HLA-A2 on IEF is presumed to be the result of the presence of heterogeneous peptides bound to the molecules. The simple band of purified Kbsol indicates the absence of endogenous peptides. In addition, the incubation of Kbsol with antigenic peptides caused the distinct shifts of band on IEF gel, reflecting the change in isoelectric point of K'sol due to the peptide binding.

It should be noted that such band-shifting was not observed in HLA-A2 molecules when they were simply mixed with peptides, unless HLA-A2 are incubated with peptides in "reconstituting conditions" after removal of 85 previously bound endogenous peptides. Taken together, these observations on native IEF also indicate that soluble Kb purified from Drosophila cells are empty.

The association of 2 SI-labelled VSV-8 with Kbsol was demonstrated by gel filtration (not shown).

The radioactivity of high molecular weight materials corresponds to peptide-Kbsol complexes, while that of low molecular weight materials represents free peptides.

Unlabelled VSV and ovalbumin (OVA) peptides could compete with the labelled VSV-8 (see below), arguing that [12 5 I]VSV-8 is bound specifically to Kbsol molecules.

Reversed-phase HPLC revealed that Kb-bound 12 I]VSV-8 has the identical retention time to the input peptide. The binding to Kbsol of the labelled VSV-8 was saturable, 15 exhibiting a dissociation constant (KD) of about 33nM (not shown). From the x-axis of the Scatchard plot, it was noted that about 65% of the labelled VSV-8 is able .to bind to K'.

To determine affinities of various peptides to 20 Kb, competitive radioimmunoassays (RIA) using 12 5 I]VSV-8 were carried out (data not shown). The inhibitory peptides used for the RIA are listed in Table 3. KD for each peptide is summarized in Table 3 as well.

TABLE 3 Various Antiqenic Peptides* Used in Present Studies Code Sequence K (M VSV-7 GYVYQGL 5.3 X 10- 8 VSV-8 RGYVYQGL 3.7 X VSV-9N LRGYVYQGL 7.3 X DLRGYVYQGL 3.9 X 10- 7 VSV-9C RGYVYQGLK 6.9 X RGYVYQGLKS 2.1 X OVA-8 SIINFEKL 4.1 X 10- 9 OVA-9N ESIINFEKL 8.9 X 86 OVA- 9C OVA-24 SEV- 9 LES IINFEKL SI INFEKLT SI INFEKLTE EQLESI INFEKLTEWTSSNVMEER

FAPGNYPAL

2.8 X 1.1 X l0-1 1.4 X 7. 1 X 2.7 X 10-9 VSV-8: Vesicular stomatitis virus nucleocapsid protein 52-59 (Van Bleek, et al., Nature 348: 213-216 (1990)) OVA-8: Ovalbumin 257-264 (Carbone, et al., J. ExID.Med.

169: 603-12 (1989)); SEV-9: Sendai virus nucleoprotein 324-332 (Schumacher, et al., Nature 350: 703-706 (1991)) All peptides were purified by C 1 reversed-phase HPLC to exclude contaminating shorter peptides with different binding properties. The 3-letter code designations and SEQ ID NO for each peptide are given below.

VSV-7 VSV- 8 VSV- 9N VSV- 9C VSV- lOC OVA- 8 OVA- 9N OVA- 9C GlyTyrValTyrGlnGlyLeu (SEQ ID NO 40, residue nos. 4-10) ArgGlyTyrValTyrGlnGlyLeu (SEQ ID NO 40, residue nos. 3-10) LeuArgGlyTyrValTyrGlnGlyLeu (SEQ ID NO residue nos. 2-10) AspLeuArgGlyTyrValTyrGlnGlyLeu (SEQ ID NO ArgGlyTyrValTyrGlnGlyLeuLys (SEQ ID NO 44, residue nos. 1-9) ArgGlyTyrValTyrGlnGlyLeuLysSer (SEQ ID NO 44) SerllelleAsnPheGluLysLeu (SEQ ID NO 39, residue nos. 5-12) GluSerllelleAsnPheGluLysLeu (SEQ ID NO 39, residue nos. 4-12) LeuGluSerllelleAsnPheGluLysLeu (SEQ ID NO 39, residue nos. 3-12) SerllelleAsnPheGluLysLeuThr (SEQ ID NO 39, residue nos. 5-13) 87 SerIleIleAsnPheGluLysLeuThrGlu (SEQ ID NO 39, residue nos. 5-14) OVA-24 GluGlnLeuGluSerIleIleAsnPheGluLysLeuThrGlu-TrpT hrSerSerAsnValMetGluGluArg (SEQ ID NO 39) SEV-9 PheAlaProGlyAsnTyrProAlaLeu (SEQ ID NO The peptides of naturally processed size (8mer for VSV and OVA, and 9mer for sendai virus nucleoprotein [SEV]) had the highest and remarkably similar affinities from the range of 2.7 to 4.1 nM. this exceedingly high affinity of the natural peptides is consistent with recent observations. (See, Schumacher, et al., Nature 350: 703-6 (1991); Christnick, et al., Nature 352: 67-70 (1991).) However, peptides that were shorter or 15 longer by as little as one or two residues lowered the affinity by a factor of from 2 to 100. This reduction of the affinity is even more drastic for a much longer peptide; the affinity of 24mer peptide (OVA-24) is more than 10,000-fold lower than that of OVA-8. These 20 results help to explain why earlier reports using longer peptides claim the affinity of micromolar range. (See, Frelinger, et al. and Choppin, et al., both cited supra.) It is of particular interest that the extension of peptides at the carboxyl terminus is much less destructive of the affinity than extension at the amino terminus. According to the three-dimensional structure of HLA-A2, the peptide-binding groove is formed by two long a helices on the antiparallel 0 strands, and the cleft is about 25 angstroms long, which is proposed to accommodate an extended peptide chain of about eight residues (see, Bjorkman, et al., Nature 329: 506-12 (1987)). At one end of the cleft, the al and a2 helices come close together tightly, while at the other end, the cleft is fairly open. It is now speculated that both VSV and OVA peptide bind to the cleft in the same orientation 88 *and the carboxyl terminus of the peptides might interact with the relatively open end of the cleft so that the extension of peptide at the carboxyl terminus does not cause severe steric hindrance.

Examinations were then performed to determine the rate of peptide binding to Kb at 4 oC and 23 OC, respectively (not shown). Binding was very rapid, especially at 23 OC, with a half-time of about 5 minutes even in extremely low concentrations of labelled peptides (about 0.4nM). This contrasts with previous observations, which show a half-time of association of about two hours. (See, Choppin, et al., cited supra.) Again, only 65% of the total labelled peptide was able to bind. The addition of excess 0-2 15 microglobulin did not affect the peptide-binding kinetics at such low temperatures that Kb heterodimer is stable (remained to be assembled). This implies that exchange of 0-2 microglobulin is not a prerequisite for peptide binding; peptides can spontaneously bind to empty 20 Class I molecules without dissociation of 0-2 microglobulin. In contrast, excess free 0-2 microglobulin apparently promotes peptide binding at 37 OC (data not shown). As the concentration of added S-2 microglobulin increased, more peptides bound to Kb molecules. Since empty Kb are unstable at 37 oC, a fraction of heterodimers must be dissociated to the heavy chain and 0-2 microglobulin and thereby, the heterodimer must be in equilibrium with free heavy chain and free 0-2 microglobulin. Then, the addition of 0-2 microglobulin should shift the equilibrium toward the formation of heterodimer that can bind peptides. This view is supported by recent observations that there are substantial numbers of Class I free heavy chains on the normal cell surface and exogenously added 0-2 microglobulin facilitates peptide binding to empty 89 Class I molecules on cells as a consequence of the reassociation of 0-2 microglobulin with free heavy chain.

(See, Rock, et al., Cell 65: 611-620 (1991); Kozlowski, et al., Nature 349: 74-77 (1991); Vitiello, et al., Science 250: 1423-6 (1990).) The dissociation kinetics of peptide at 37 °C were then observed. Immediately after isolating 2 I]VSV-8 and Kb complexes by gel filtration, the samples containing either 50 or 300 nM Kb were exposed to 37 oC temperatures. Some samples were supplemented with 3pM 0-2 microglobulin and/or 20AM unlabelled VSV-8, or 1% TX-100. The dissociation of labelled peptides from Kb was measured at various times (not shown). In the presence of a large excess of unlabelled peptides, the 15 dissociation rate of peptide followed first-order kinetics with a half-time dissociation of about 36 minutes (a dissociation rate constant of 3.2 X 10' 4 This unexpected, relatively rapid dissociation of labelled peptide does not agree with some current views 20 of stable peptide-Class I complexes. In fact, the results ascertained (not shown) demonstrated that Kb and VSV-8 complexes are stable. This discrepancy must arise from the 10-fold lower affinity of radiolabelled VSV-8 (33 nM) compared with that of unlabelled VSV-8 (3.7 nM) The first-order kinetics were also observed when the detergent was added instead of unlabelled peptide, indicating that the detergent makes the peptide dissociation process irreversible. In contrast, the peptide dissociation profile did not follow the first-order kinetics in the absence of unlabelled peptide or the detergent. This suggests that the peptide association/dissociation is reversible and the binding of peptide is dependent on the concentration of heterodimer (compare the kinetics between 50 nM and 300 nM of Kb).

This became more evident when excess 0-2 microglobulin 90 was added. These results support the previous argument that interaction between the heavy chain, 0-2 microglobulin and peptide are basically reversible at 37 OC, if not entirely, in the absence of detergent. It is probable that a detergent such as TX-100 may prevent 0-2 microglobulin from reassociating with free heavy chain at 37 OC. This could reasonably explain why Kb once heated to elevated temperatures in the absence of detergent can be efficiently immunoprecipitated by conformational antibody (not shown). Interestingly, the addition of 3-2 microglobulin did not suppress the peptide dissociation in the presence of excess unlabelled peptides, indicating that labelled-peptides are released from the complexes without dissociation of 0-2 15 microglobulin. It should be remembered, however, that the affinity of [1 25 I]VSV-8 is about 10-fold lower than that of the natural peptides. Therefore, this is not necessarily the case for the natural peptides.

The study using in vitro peptide-binding assay 20 systems suggests that peptide binding to Class I molecules is a simple mass action and a ligand-receptor interaction. The approach used herein allows characterization of the peptide binding specificity to Class I molecules and of the interaction of peptide-Class I complexes with the T-cell receptor.

Example 8 Therapeutic Applications A. Class I Molecule Bank A reservoir or "bank" of insect cell lines may be established and maintained, with each cell line expressing one of the 50 to 100 most common Class I MHC heavy chain, a P-microglobulin molecule, as well as at least one assisting molecule. cDNAs encoding these proteins may be cloned based on HLA variants obtained from cell lines containing same via the 91 polymerase chain reaction (see Ennis, et al., PNAS USA 87: 2833-7 (1990)) and inserted into the appropriate vector, such as an insect expression vector, to generate cell lines expressing each HLA variant.

Testing according to the following protocol, for example, can be used to determine which peptides derived from the virus of choice bind the best to the different Class I MHC molecules. The various cultures may appropriately be labeled or catalogued to indicate which Class I MHC molecules are best for use with particular peptides. Alternatively, transient cultures may be established as needed. As discussed herein, after approximately 48 hours' incubation of a culture of insect cells with a vector, that culture is apparently capable 15 of expressing empty MHC molecules which may be loaded with the peptide(s) of choice for the purpose of activating CD8* cells.

B. Preparation of "Special" Cell Lines After HLA typing, if insect cell lines expressing the preferred HLA are not available, cDNAs encoding the preferred HLA and assisting molecules may be cloned via use of the polymerase chain reaction. The primers disclosed in section B.l. above (SEQ ID NO i i through SEQ ID NO 12) may be used to amplify the appropriate HLA-A, or -G cDNAs in separate reactions which may then be cloned and sequenced as described in the methods disclosed in Example 1, section 1 above. DNA is then purified from the PCR reaction using a Gene Clean kit (Bio 101, San Diego, CA) and ligated directly into the Sma I site of pRmHa-3.

Individual clones are isolated, the sequences verified, and stable Drosophila cell lines expressing the HLA established. Alternatively, a bulk population of recombinant plasmids may be grown in large scale and DNA purified by cesium chloride gradients. The purified DNA 92 is then used to transfect S2 cells using calcium phosphate precipitation techniques. After 24 hours, the precipitate is washed off the cells and replaced with fresh Schneider media containing ImM CuSO 4 Forty-eight hours later, the bulk population of transiently transfected cells is used for in vitro-activation of CD8' after incubation with syngeneic peptides or protease digests of specific proteins.

Stable cell lines expressing the cloned HLA may then be established. Alternatively, a population of -insect cells transiently expressing a bulk population of cloned recombinant molecules from the PCR reaction may be used for in vitro CD8* activation.

0* It is also possible to activate 15 haplotype-specific CD8s in vitro using insect cells expressing Class I MHC incubated with peptides where the cell line-expressed MHC is not the expressed element in vivo. This provides a unique opportunity to proliferate CD8' cells which recognize a specific antigen associated 20 with a particular MHC which would not be possible in vivo due to allelic restriction. For example, a peptide (NP) from'the nuclear protein of Influenza virus is ordinarily **b "4 restricted to the Db molecule; however, we have found that such a peptide can bind to Kb (albeit more weakly than-to Db) and can generate a degree of thermal stability to the Kb (see Figure Furthermore, Kb-expressing Drosophila cells preincubated with the NP peptide and cocultured with splenocytes from a B6 mouse results in the in vitro activation of CD8* which specifically recognize the Kb molecule associated with the NP peptide. In addition, the reciprocal experiment using a Kb-restricted peptide (OVA) derived from ovalbumin and Db-expressing Drosophila cells results in the proliferation of CD8* which specifically recognize Db containing the OVA peptide. Such CD8s are able to kill 93 cells (EL4 OVA) transfected with cDNA encoding the ovalbumin protein, indicating that in vivo, some Db molecules are loaded with the OVA peptide.

This system therefore provides a unique opportunity to proliferate CD8 against specific antigens presented by a Class I molecule which, in vivo, is not the restriction element for that peptide. Although enough antigen is presented in vivo by said Class I for the cell to be recognized by CD8 and killed, it is not enough to proliferate such CD8s in vivo. By loading empty Class I molecules expressed by Drosophila cells with peptide, we are able to override the in vivo restriction by providing an excess of antigenic peptide to the Class I molecule in a non-competitive environment 15 such that enough antigen is presented by the Class I to activate the specific CD8 recognizing this complex.

C. AIDS Treatment In vitro activated cells may be administered to patients for in vivo therapy. Preferably, the Class I 20 MHC genotype (haplotype) of the individual is first determined. Conventional tissue typing is appropriate for this purpose and may be performed at the treatment center or by some appropriate commercial operation. Once the individual's HLA type(s) is (are) determined, the best combination of peptides and Class I MHC molecules suitable for the individual patient is ascertained and prepared as noted above and the appropriate insect cell lines and peptides are provided. Resting or precursor CD8* T-cells from the blood of the patient are then stimulated with the appropriate peptide-loaded MHC produced by the insect cell culture. After activation, the CD8 cells are reintroduced into the patient's bloodstream, and the disease process in the patient continues to be monitored. Methods of removing and re-introducing cellular components are known in the art 94 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.

Additional treatments may be administered as necessary until the disease is sufficiently remediated.

Similar treatment protocols are appropriate for use with other immunosuppressed individuals, including transplant patients, elderly patients, and the like.

D. Cancer Treatment In cancer patients, a treatment procedure similar to that described above is utilized. However, in such patients, it should be anticipated that conventional *o therapy to reduce the tumor mass may precede the immune therapy described herein. Therefore, it is preferred 15 that blood samples from the putative patient be obtained and stored via freezing) prior to the commencement of conventional therapy such as radiation or chemotherapy, which tends to destroy immune cells. Since few, if any, forms of cancer arise in direct response to viral infection, target peptides for immune treatment are less readily observed. However, recent studies indicate that mutations in the oncogenes ras, neu, and p53o contribute to cancer in as much as 50% of all cancer cases. Thus, peptides derived from these mutated: regions of the molecules are prime candidates as targets for the present therapy. Pursuant to the protocols disclosed herein, the best combination of peptides and Class I molecules for the individual patient may be determined and administered.

For example, many tumors express antigens that are recognized in vitro by CD8 cells derived from the affected individual. Such antigens which are not expressed in normal cells may thus be identified, as well as the HLA type that presents them to the CD8" cells, for precisely targeted immunotherapy using the methods of the 95 present invention. For example, van der Bruggen, et al.

have described an antigen whose expression is directed by a specific gene and which antigen appears to be presented by HLA Al (Science 254: 1643-1647 (1991)). As various human tumor antigens are isolated and described, they become good candidates for immunotherapeutic applications as described herein.

In another, alternative therapeutic mode, it may be feasible to administer the in vitro activated CD8' cells of the present invention in conjunction with other immunogens. For example, the Large Multivalent Immunogen disclosed in U.S. Patent No. 5,045,320 may be administered in conjunction with activated CD8* cells.

It is also possible that cytokines such as IL-2 15 and IL-4, which mediate differentiation and activation of T-cells, may be administered as well, as cytokines are able to stimulate the T-cell response against tumor cells in vivo. It is believed that IL-2 plays a major role in the growth and differentiation of CD8* precursors and in 20 CD8* proliferation. The administration of IL-2 to cancer patients is frequently associated with an improved anti-tumor response which is likely-related to induction of tumor-specific T-cells. However, the best therapeutic effects of IL-2 might be obtained by continuous local rather than systemic administration of IL-2, thus minimizing the IL-2 toxicity and prolonging its biological activity. One may achieve local delivery via transfecting tumor cells with an IL-2 gene construct.

IL-2 cDNA is constructed as described by Karasuyama and Melchers in Eur. J. Immunol. 18: 97-104 (1988). The complete cDNA sequence of IL-2 is obtained as an Xho I fragment from the plasmid pBMGneo IL-2 (see Karasuyama and Melchers, supra) and directly ligated into the Sal I site in pRmHa-3. Recombinant pRmHa-3 plasmid with the insert in the correct orientation (determined 96 via restriction mapping with Hind III) is purified by cesium gradients and used to cotransfect S2 cells using the calcium phosphate technique. (A mixture of plasmid DNA was prepared for this purpose: 10pg pRmHa-3 containing IL-2 cDNA, 6ig each of pRmHa-3 plasmid containing MHC Class I heavy chain or 0-2 microglobulin and 2pg of phshsneo DNA.) Stable cell lines which are inducible via CuSO, to express heavy chain, 0-2 microglobulin and IL-2 were obtained by growing the transfectants in G418 medium. These stable cell lines were coated with peptide and used in the in vitro assay o as described above. Tumor cells transfected with IL-2 are observed to enhance the CTL (CD8) activity against the parental tumor cells and bypass CD4 and T helper 15 function in the induction of an antitumor or cytotoxic response in vivo. Therefore, increasing the potential of the Drosophila system via cotransfection with the IL-2 gene is suggested herein.

Example 9 20 Dose Dependence of the Production of Activated T-Cells Using the Antigen-Presenting System Antigen-presenting cells (APC) were produced by transfecting Drosophila cells as described in Example 3 and then tested for their capacity to present antigen to T-cells from the 2C line of transgenic mice. With mouse cells as antigen-presenting cells, this line displays strong alloreactivity for L d molecules complexed with an endogenous 8-mer peptide, p2Ca (Leu-Ser-Pro-Phe-Pro-Phe- Asp-Leu, SEQ ID NO 46), derived from a Krebs cycle enzyme, 2-oxoglutarate dehydrogenase (OGDH). The p2Ca peptide is exposed naturally bound to L on the surface of H-2d cells such as B10.D2 cells. The p2Ca peptide has intermediate binding affinity for soluble Ld molecules (4 x 106 M and high affinity for 2C TCR molecules (2 x 106

M-

1 to 1 x 10 97 A closely-related 9-mer peptide, QL9 (Gln-Leu- Ser-Pro-Phe-Pro-Phe-Asp-Leu, SEQ ID NO 47), has even higher affinity for these molecules (2 x 108 for Ld and 2 x 10 7

M

1 for 2C TCR). Except for one additional amino acid (glutamine) at the N terminus, QL9 has an identical sequence to p2Ca and, like p2Ca, forms part of the native sequence of OGDH.

With p2Ca and QL9 peptides (prepared in synthetic form), antigen-presenting cell requirements for mature unprimed 2C CD8* cells were studied in vitro. The responder CD8* cells were purified from pooled lymph *nodes (LN) of 2C mice on a C57BL/6 (B6, H-2 b background by first removing CD4' cells, class II-positive cells and B cells by mAb complement treatment followed by 15 positive panning.

Cell suspensions were prepared from pooled cervical, axillary, inguinal and mesenteric LN of young adult mice using a tissue grinder. For cell purification, LN cells were first treated with a cocktail 20 of mAbs (anti-CD4, anti-HSA, anti-I-Ab) plus complement for 45 minutes at 37 OC. The surviving cells were further separated into CD8' and CD8-- (CD4-) cells by panning at 4 OC for 60-90 minutes on petri dishes coated with anti-CD8 mAb. The attached CD8* cells were recovered by incubation at 37 OC for 5 minutes followed by vigorous pipetting. Non-attached cells were eluted and treated with anti-CD8 mAb and complement to obtain CD8- 1B2* 2C cells. More than 95% of the CD8* cells obtained were clonotype-positive (IB2*) and 98% of these cells displayed a naive (CD44 1 phenotype.

TCR stimulation elicited a complex pattern of intracellular events which lead to early up-regulation of CD69 and CD25 (IL-2 receptor or IL-2R) on the cell surface. These changes are apparent within a few hours of stimulation and are followed by cytokine synthesis and 98 cell proliferation. Drosophila cells were transfected with genes for Ld, Ld and B7.1 Ld and ICAM-1 (Ld.ICAM) or Ld, B7.1 and ICAM-1 B7.ICAM). Figure 11 shows CD69 and CD25 expression on purified naive CDS- 2C cells stimulated for 12 hours with transfected Drosophila cells and p2Ca versus QL9 peptide at a concentration of iM. Purified CD8' 2C cells were incubated with various Drosophila cells plus a peptide (either p2Ca or QL9, 10 AM) in bulk (2 ml) culture for 12 hours and then stained for CD69 or CD25. Either p2Ca or QL9 presented by Drosophila cells transfected with Ld.B7, L .ICAM or L".B7.ICAM were effective in stimulating the upregulation of CD69 and CD25. However, non cultured 2C cells without either peptide or Drosophila cells did not 15 show up-regulation of CD69 and CD25 (top panel).

In the presence of Drosophila cells expressing L4 alone, induction of CD69 and CD25 expression on CD8* 2C cells was low, but significant, with the strong QL9 peptide but barely detectable with the weaker p2Ca peptide. With L.B7 or L=.ICAM antigen-presenting cells, both peptides elicited marked up-regulation of CD69 and However, B7.ICAM antigen-presenting cells induced even higher expression of these molecules.

Drosophila cells transfected with Ld alone failed to cause proliferation of 2C CD8' cells to either p2Ca or QL9 peptide in the absence of exogenous lymphokines, consistent with the minimal capacity of these cells as antigen-presenting cells to induce CD69 and CD25 expression. The results are shown in Figure 12.

Responses to p2Ca (above) and QL9 (below) were measured by culturing 5 x 10' purified CD8' 2C cells with 2 x 10 s Drosophila cells in the presence or absence of the indicated concentrations of peptides for 3 days. 3

H]

TdR was added during the last 8 hours of culture; rIL-2 was added at a final concentration of 20 units/ml. The 99 data are the mean of triplicate cultures. When supplemented with exogenous IL-2, however, both peptides stimulated significant proliferative responses at high doses (10 pM). The proliferative responses elicited by QL9 were far stronger than for p2Ca (note the large difference between the scales of x-axes of the upper and the lower panel).

Dose-response relationships for proliferation on day 3 elicited by Drosophila cells transfected with L6.B7.ICAM and presenting p2Ca peptide were compared to those elicited by Drosophila cells transfected with LC.B7, L'.ICAM or Ld.B7.ICAM presenting QL9 peptide. The results (means of triplicate cultures) are shown in Figure 13. Drosophila APC (2 x 10 5 cells) were cultured 15 with 5 x 104 CD8' 2C cells for three days in the presence or absence of peptides. 3 H] TdR was added during the last 8 hours of culture; no IL-2 was added to the cultures.

Optimal proliferative responses elicited by L .B7.ICAM antigen-presenting cells on day 3 required high concentrations of p2Ca peptide, e.g. 10 iM (10 5

M),

(Figure 13). The results for QL9 peptide were different.

The does-response curves for L°.B7 antigen-presenting cells plus QL9 and Ld.ICAM antigen-presenting cells plus d 25 QL9 approximated:the results for Ld.B7.ICAM antigenpresenting cells-plus p2Ca (Figure 13). However, with L^.B7.ICAM antigen-presenting cells, proliferative responses to QL9 on day 3 were maximal with 100 nM (10 M) and were clearly apparent with doses as low as 10 pM (10 1 M) (Figure 13). At high doses, e.g. 10 iM (10" 5

M),

QL9 inhibited the proliferative response on day 3 (Figure 13, note the log scale).

Inhibition of proliferation by L.B7.ICAM antigen-presenting cells and QL9 peptide did not apply to IL-2 production (Figure 14, right), and was only seen 100 with high doses of antigen-presenting cells (Figure 14, left). Ld.B7.ICAM antigen-presenting cells plus QL9 peptide was again found to be more effective in eliciting IL-2 production than L.3B7.ICAM antigen-presenting cells plus p2Ca peptide (Figure 14, right). The data indicates that antigen-presenting cells without peptides were ineffective in eliciting either response over the 100fold range of antigen-presenting cells density tested (Figure 14, open squares).

Changes in the dose-response relationships of the responses of CD8* 2C cells on Day 3, Day 4 and Day were examined using QL9 peptide with Ld.B7, Ld.ICAM or L.B7.ICAM antigen-presenting cells. The results are shown in Figure 15. Responses were measured with 5 x 10 4 15 CDS8 2C cells and 3 x 0 l antigen-presenting cells at the 0.

indicated peptide concentrations. The data are the mean results of triplicate cultures.

With an intermediate dose of 100 nM (10 7

M)

QL9 peptide, proliferative responses to L'.B7.ICAN antigen-presenting cells were high on Day 3 (Figure left), reached a peak on Day 4 (Figure 15, center) and then declined to low levels on Day 5 (Figure 15, right).

With a high dose of 10 pM M) QL9 peptide, however, the response was low on Day 3 (Figure 15, left), but then 25 increased markedly to reach a high peak on Day 5 (Figure right) Transient inhibition of proliferation induced by QL9 on Day 3 was only seen when the avidity of T-cell/APC interaction was very high. Decreasing the avidity of T-cell/APC interaction by using lower doses of APC (Figure 14, left) or lower doses of QL9 peptide (Figure 15, left) augmented the Day 3 proliferative response. Reducing the avidity of T-cell/APC interaction enhanced the early (Day 3) proliferative response but reduced the late (Day 5) response and also reduced IL-2 101 production (Figure 16, right). In Figure 16 the results obtained using CD8* 2C and CD8- 2C cells on Days 2, 3, 4 and 5 are comoared.

The observations apply with the highly immunogenic LZ.B7.ICAM antigen-presenting cells.

However, with the less immunogenic Ld.B7 or Ld.ICAM antigen-presenting cells, proliferative responses to QL9 peptide required high doses of peptide (Figure 13, Figure and were crucially dependent upon CD8 expression by the responder cells (Figure 16). These responses with L.B7 and Ld.ICAM antigen-presenting cells reached a peak on Days 3 or 4 (rather than Day 5) and were far lower i" than with Ld.B7.ICAM antigen-presenting cells. With low doses of QL9, e.g. 1 nM (10-9 proliferative responses 15 with Ld.B7 and LC.ICAM antigen-presenting cells were completely undetectable 100 cpm) (Figure 13). This was in striking contrast to the results seen using L.B7.ICAM antigen-presenting cells, where 1 nM QL9 led to high responses 10,000 cpm) (Figure 13). In 20 contrast to the results with high doses of QL9 peptide iM, Table the synergistic interaction between B7 and ICAM for proliferative responses became pronounced at low peptide doses.

As can be seen, Ld.B7.ICAM cells act as 25 extremely potent antigen-presenting cells for naive 2C cells. Raising the avidity of T-cell/APC interaction to a high level inhibits the early proliferative response but potentiates the late response and IL-2 production is enhanced. For the high-affinity peptides such as QL9, the synergy between B7 and ICAM is pronounced at low doses of antigen.

102 ExamDle Production of Cytotoxic T-Cells Usina the Antiaen Presenting System Antigen-presenting cells (APC) were produced by transfecting Drosophila cells as described in Example 3 and then tested for their ability to induce CTL activity.

.CTL activity was tested on "Cr-labelled RMA.S-Ld targets sensitized with QL9 peptide, no peptide or an irrelevant peptide (MCMV). CD8 2C cells (5 x 10 s were cultured with 2 x 106 transfected Drosophila cells in a volume of 2 ml in a 24-well culture plate. Cai, and J. Sprent, (1994). Peptides were present during the culture at a :concentration of 10 mM. After 4 days, the cells were pooled and adjusted to the required number. To prepare 15 targets, RMA-S.L 0 cells were labeled with 5 1 Cr (100 mCi/l-2 x 106 cells) at 37 oC for 90 minutes in the presence or absence of peptides. After labeling, the cells were thoroughly washed and resuspended in medium with or without peptides. Specific 5 Cr release was 20 calculated according to established procedure. Id.

The capacity of Ld.B7, Lc.B7.ICAM and L.ICAM antigen-presenting cells to induce CTL activity to 10 M QL9 peptide in bulk cultures is shown in Figure 17.

Strongly immunogenic LA.37.ICAM antigen-presenting cells 25 were efficient in generating QL9-specific CTL (Figure 17, center). Significantly, Ld.B7 antigen-presenting cells were also effective in generating CTL to QL9 (Figure 17, left). The surprising finding, however, was that L .ICAM antigen-presenting cells were totally unable to stimulate CTL generation (Figure 17, right). This result (representative of three different experiments) was unexpected, because L6.ICAM antigen-presenting cells were no less efficient than LA.B7 antigen-presenting cells-at inducing proliferative responses to QL9 (Figures 13 and 15). The L. ICAM-stimulated cultures in Figure 17 103 contained large numbers of blast cells, and total cell yields were about 3-fold higher than the input number.

The surprising finding that L.ICAM antigenpresenting cells were totally unable to stimulate CTL generation applied to cultures not supplemented with exogenous lymphokines. However, when exogenous IL-2 was added to the cultures, LI.ICAM antigen-presenting cells induced strong CTL activity to QL9 (Figure 18, right; exogenous IL-2).

Example 11 Proliferation of Normal T-Cells Induced by the Antigen-Presenting System Antigen-presenting cells (APC) were produced by 15 transfecting Drosophila cells as described in Example 3 and then tested for their ability to induce proliferation in normal (nontransgenic) murine T-cells.

The capacity of Ld.B7.ICAM Drosophila cells to induce a strong primary response of 2C TCR transgenic 20 CD8- cells raised the question whether Drosophila cells could act as antigen-presenting cells for normal (nontransgenic) CD8- cells. Since the 2C mice were on a B6 (H-2 b background, the response of normal B6 CD8* cells was tested. Graded doses of CD8' cells from normal 25 B6 mice were cultured with 10 pM QL9 peptide presented by Ld.B7.ICAM Drosophila antigen-presenting cells, i.e. a situation where a diverse repertoire of T-cells was exposed to only a single alloantigen (L d QL9) but at high concentration. As shown in Figure 19 (left), presentation of QL9 by L=.B7.ICAM antigen-presenting cells was indeed immunogenic for normal B6 CD8' cells and led to appreciable proliferative responses on Day 3 (80,000 cpm) with large doses of responder cells (1 x 106) in the absence of added cytokines; responses to an unrelated peptide, MCMV, were much lower (though 104 significant) and no response occurred in the absence of peptide. As expected, the response of normal B6 CD8' cells to QL9 plus Ld.B7.ICAM antigen-presenting cells was substantially lower than with normal B10.D2 spleen cells as antigen-presenting cells (where the allostimulus was provided by a multiplicity of self peptides bound to L, K' and Dc, but at low concentration) (Figure 19, right.

Note the difference in the Y axis scales).

ExamPle 12 Activation Of Cvtotoxic T-Cells Using Immobilized Purified Recombinant MHC Class I Molecules And Assistina Molecules Except as noted, cells, materials and reagents 15 were prepared as described in Example 4. Biotinylated anti-mouse CD28 monoclonal antibody (clone 37.51) was purchased from Pharmingen (San Diego, CA). This antibody is also available from Caltag (South San Francisco, CA).

o This antibody, which binds to CD28 co-stimulatory receptor, a ligand of B7.1 and B7.2 on the surface of T-cells, augments the proliferation of T-cells (Gross et al. J. Immunol. 149, 380-388, 1992). IL-2 was used as a 0oD -concanavalin A supernatant (10% final concentration).

Substrates were prepared as follows: Fifty 25 microliters of PBS containing 1 microgram/ml of neutravidin were added to each well of a 96 well cell culture plate (Corning cat.#25860). After 2 hours at room temperature, the wells were washed 3 times with PBS prior to incubation with biotinylated molecules.

Avidin-coated mouse red blood cells were prepared as described for avidin-coated sheep red blood cells in Example 4. Avidin-coated latex beads were prepared as described Example 4.

Biotinylation of recombinant Ld was performed as described in Example 4. Biotinylated recombinant Ld 105 was immobilized on the substrate together with biotinylated anti-CD28 antibody. Avidin-coated red blood cells or latex beads were incubated with 0.2 mg/ml biotinylated LC, 0.025 mg/ml biotinylated anti-CD28, or a mixture thereof for 30 minutes at room temperature, then washed 3 times in DMEM containing 10% FCS and used immediately. Avidin-coated 96 well plates were incubated with 50 microliters per well of 2 microgram/ml biotinylated Ld, 0.25 microgram/ml anti-CD28, or mixtures thereof for 30 minutes at room temperature, then washed 3 times using DMEM containing 10% FCS and used immediately.

2C+ T-cells were prepared as described in Example 4. CD8' cells from C57BL/6 mice were prepared from the lymph nodes of these mice by magnetic depletion.

*.0 15 Purified cells were consistently 90-92% positive for CD8 expression, as assessed by flow cytofluorometry.

T-cell activation was performed in culture Plates coated with purified Ld molecules and anti-CD28 antibody. T-cells and peptide (0.02 mM final concentration) were added to 96 well plates coated with L and/or anti-CD28, in a final volume of 0.2ml/well and cultured for the appropriate time at 37 OC in humid atmosphere containing 5% CO T-cell activation using red blood cells or 25 latex beads coated with purified L' molecules and anti- CD28 antibody was performed in uncoated culture plates.

T-cells and peptide (0.02mM) were added to each well, together with 100,000 red blood cells or latex beads.

Final volume was 0.2ml. Culture conditions were as described above.

T-cell mitogenesis was assayed by incorporation of a pulse of 1 iCi of tritiated thymidine per well by triplicate cell cultures. Cells were harvested 8 hours later and thymidine incorporation was determined by counting the filters in a scintillation counter.

106 RMA.S cells (target cells) expressing LA were incubated with radiolabeled chromium for 90 minutes at 37 washed 3 times and distributed in 96 well U-bottom plate (Costar cat#3799) (5000-10000 cells per well) in the presence of the appropriate peptide at 0.01m.M.

Various amounts of activated T-cells (effector cells) were added to reach effector/target ratios (E/T ratios) ranging between 150 and 1. After 5 hours of incubation at 37 0.lml of supernatant was collected from each well and counted in a gamma counter. Percent lysis was determined by a standard method (Coligan et al., Current Protocols in Immunology, section 3.11, Wiley, New York (1991)) ee** Immobilized purified Ld and anti-CD28 antibody 15 were mitogenic for 2C+ T-cells. When QL9 peptide was .4* "used, a thymidine incorporation above 100,000 czm per well was consistently measured by day 3-5 of culture (Figure 20). Maximum thymidine incorporation in that oooo same range was obtained using any of the three methods of activation (molecules immobilized on plastic, on red blood cells or on latex beads). When activation molecules were immobilized on plastic, L d alone **oo immobilized on plastic induced a transient mitogenesis (Figure 20, broken line) whereas Ld plus anti-CD28 25 antibody induced-a higher and more sustained mitogenesis

S

(Figure 20, solid line). Anti-CD28 antibody in addition to L° was required for the induction of any mitogenesis in the model using red blood cells. However, anti-CD28 antibody alone did not induce any mitogenesis.

QL9-L d complexes are recognized by the 2C Tcell receptor with a high affinity. Complexes of other peptides and Ld recognized by this receptor with a lower affinity were able to activate 2C+ T-cells; these included peptides p2Ca and SL9 (Figure 21). However, IL- 2 added at day 2 of culture was required in order to 107 observe mitogenesis using these peptides. Activation was peptide specific since LCMV peptide, a control peptide that is not recognized by the 2C T-cell receptor, did not induce activation. Mitogenesis was measurable starting with as little as 700 T-cells per well (96 plate well), demonstrating that the method activated a very small number of T-cells in a peptide specific manner.

2C+ T-cells were mixed with total CD8- cells from C57BL/6 mice at a 1/99 ratio and cells were cultured with immobilized L d and anti-CD28 antibody in the presence of peptide QL9. IL-2 was added at day 2 of culture, and subsequently every other day. Cell activation and proliferation was noted, as evidenced by the formation of clumps, presence of enlarged cells that 15 progressively spread in the well. At day 12 of culture, 0 cells were harvested and percentage of cells expressing the 2C T-cell receptor was assessed by staining with the anticlonotypic antibody 1B2 and analysis by flow cytofluorometry: 43% of the cells cultured initially in the presence of peptide QL9 were 2C+, (Figure 22) whereas 49% of the cells expressed 2C after stimulation with peptide p2Ca (Figure 23), and 22% with peptide SL9 (Figure 24). This shows a considerable enrichment (43, 49 and 22 times) in specific T-cells after 12 days of culture. Enrichment was observed even with SL9, a peptide that makes a low affinity complex with the 2C Tcell receptor. This method thus can specifically activate and enrich a small subpopulation of antigen specific T-cells out of a heterogenous mixture of Tcells.

2C+ T-cells, activated using L d and anti-CD28 antibody in the presence of QL9 peptide, were cultured for 5 days, then their cytotoxic capacity was assessed.

The results, shown in Figure 25, demonstrated that immobilized activation molecules induced the cells to 108 differentiate into effector cytotoxic T-cells. Lysis was specific, since it was observed in the presence of QL9 peptide but not in the presence of a control peptide (LCMV). Resting T-cells were thus activated using immobilized molecules to differentiate into cytotoxic T-cells able to specifically kill targets.

As can be seen, immobilized purified MHC Class I and assisting molecules can be used to specifically activate naive resting T-cells into cytotoxic T-cells. MHC class I and assisting molecules are sufficient for activation; no additional signal :i originating from antigen-presenting cells is necessary.

MHC class I and assisting molecules immobilized on substrates, such as cell culture plates, provide an 15 appropriate tool for T-cell activation. Several advantages are offered by such coated substrates. They are easy to prepare and to manipulate, they can be coated with well-controlled amounts of molecules, ensuring reproducible activation conditions.

Examole 13 o-0, Activation of Human Cytotoxic T-Cells A unit of human blood (450 mls) collected in o* heparin (10 4/ml) was obtained through the General 25 Clinical Research Center (GCRC) at Scripps Clinic, La Jolla, CA. The blood was first processed in a Ficoll-Hypaque density gradient in a 50 cc conical centrifuge (Histopaque 1077, Sigma) according to the manufacturer's specifications. Once the buffy coat was obtained, the sample was washed in buffer 1 (D-PBS without Ca" or then buffer 2 (RPMI with 4% fetal calf serum (FCS)) and a final wash in buffer 3 (D-PBS 1% human serum albumin (HSA, 25% Buminate/Baxter-Hyland) and 0.2% sodium citrate 109 The total peripheral blood mononuclear cell (PBMCs) preparation was counted from the washed cells.

This preparation was then taken through a MaxSepT isolation procedure (Baxter) where CD8' cells were selected by negative selection. A cocktail of mAbs to cells targeted for removal (CD19-PharMingen, CD4-Ortho-mune, CD15-PharMingen, CD56-PharMingen, CD14-PRI) was prepared at 2 gg/ml of cells. The total P3MC cell count was diluted to 20 x 106/ml in buffer 3 and the antibodies were added.

The mixture (approximately 40 mls) was rotated (4 rpm) on a rotary shaker at 4 so that the tube mixed the sample end over end, for a total of 30 minutes.

After the sensitization phase, the cells were washed with 15 buffer 3 and resuspended in the same buffer with magnetic Dynal beads coated with sheep-anti-mouse IgG (SAM) (Dynabeads M450 #110.02). The stock of beads is usually 4 x 108 beads/ml. The final bead:target cell ratio is 10:1.

20 To determine the final volume of beads to use, *o 1 the formulas below were followed: (Total cells sensitized) x of population of target cells) Total theoretical target cell number.

Total theoretical target cell number x Total beads required.

Total beads required/Bead concentration of stock Volume of beads required.

The volume of the final sensitized cells:beads mixture was apprcximately 50 mls. The mixture was put into a 150 ml Fenwal Transfer Pack (#4R2001), air added with a needle and the mixture rotating under similar conditions as above (30 minutes, 4 4 rpm, agitated end-over-end). At the end of the incubation period, the target cells were removed with a MaxSep separation device according to the manufacture's specifications.

110 The separated cells were transferred from the bag and counted to determine recovery. FACS analysis was performed to determine the purity of the sample.

The resulting separated human CD8' cells were stimulated with Drosophila (fly) antigen-presenting cells which were transfected and shown to express human HLA A2.1, B7.1 and/or LFA-3. The fly cells were diluted to in Schneider's medium with 10% FCS serum. On the following day CuS04 was added to the cultured cells which were incubated for 24 hours at 27 'C and harvested. The harvested cells were washed and suspended in Insect .X-oress medium with a final peptide concentration of 100 jg/ml and incubated for 3 hours at 27 'C.

The peptides used were HIV-RT (ILKEPVHGV) (SEQ 15 ID NO 48), Tyrosinase (YMNGTMSQV) (SEQ ID NO 49), and Influenza Matrix (GILGFVFTL) (SEQ ID NO The control peptide was derived from the core of the hepatitis virus and had the sequence FLPSDFFPSV (SEQ ID NO 51).

The-fly antigen-presenting cells were added to the CD8' cells at a ratio of 1:10 (APC:CD8* T-cell). The cells were incubated in flat bottom wells (48 wells) for four days at 37 At day 5, IL-2 (10 y/ml) and IL-7 S* (10 p/ml) (Genzyme) were added with a media change. On 25 day 11 a CTL assay was performed.

JY, an EBV-transformed human B cell line expressing HLA 2.1, B7 was used as the target cell in the chromium release assay. Vissereu, M.J.W. et al. J.

Immunol., 154:3991-3998 (1995). JY cells were seeded at 3 x 105 cells/ml in RPMI with 10% FCS, 24 hours prior to the assay. The JY cells were counted and washed once in RPMI wash solution Rehautin FCS, 1% HEPES, 0.025% gentamycin in RPMI) to provide 5 x 106 cells per sample.

The cell pellet was gently resuspended in 100 pl of s Chromium stock (0.1 mCi, NEN) and placed in a 37 OC water 111 bath for 1 hour with agitation every fifteen minutes.

Labelled JY cells were washed four times for 6 minutes each at 1400 rpm in 10 mis of RPMI wash solution. Cells were adjusted to 1 x 10 cells/ml in RPMI/10% Hyclone.

Efficiency of target cell labelling is confirmed by standard gamma counter techniques.

For peptide loading of target cells, two mis of labelled JY cells (2 x 10 5 cells) were incubated with pg/ml peptide for 30 minutes at room temperature.

Peptide stock solutions (1mg/ml) were stored at -70 'C.

In a 96 well round bottom plate, 100 pl effector cells and 100 yl peptide loaded target cells were combined at ratios of 84:1, 17:1 and 3.4:1 (effector:target).

Controls to measure spontaneous release and maximum 15 release of 5 Cr were included in duplicate. Samples were incubated at 37 "C for 6 hours.

K562 cells at a concentration of 107/ml in RPMI with 10% FCS were added at a ratio of 20:1 (unlabelled K562:labelled JY). This erythroleukemic cell line was used to reduce NK background cell lysis in the chromium release assay. Plates were centrifuged at 1000 rpm for minutes and 160 1p supernatant from each sample transferred to 96 tube collection tubes. Analysis of cell Slysis is determined by standard gamma counting techniques 25 (Gammacell 1000, Nordion).

CTL activity produced by activating human CD8' T-cells with antigen-presenting cells loaded with influenza matrix peptide (SEQ ID NO 50) is shown in Figure 26. Prior exposure to influenza virus indicates that this CTL activity was a secondary response. Data points are the mean of values from triplicate cultures.

Antigen-presenting cells expressing A2.1, B7.1 and ICAM-1 were more effective than antigen-presenting cells expressing A2.1 and B7.1 or antigen-presenting cells expressing A2.1, B7.1 and LFA-3.

112 The results of activating human CD8' T-cells with antigen-presenting cells loaded with HIV-RT peptide (SEQ ID NO 48) are presented in Figure 27. Since screening of the blood of this patient had indicated no prior exposure to HIV, these results indicate the CTL activity was based on a primary response. In this case, only antigen-presenting cells expressing A2.1, B7.1 and ICAM-1 produced CTL activity that was significantly greater than control levels.

Figure 28 shows the CTL activity of human CD8' T-cells activated by antigen-presenting cells loaded with tyrosinase peptide (SEQ ID NO 49). Tyrosinase is a normally occurring enzyme that is over-expressed in melanoma tumor cells. Again, the antigen-presenting 15 cells expressing all three molecules, ICAM-1 as well as A2.1 and B7.1, were most effective, especially at the lowest effector:target ratio tested.

**w These results show that the antigen presenting system is capable of producing effective CTL activity in human CD8' T-cells directed against a peptide that is derived from an endogencus protein that is over-expressed in tumor cells. This also demonstrates that this in vitro stimulation of CDS5 T-cells using both B7 and ICAM will generate cytotoxic CD8' T-cells even against S 25 peptides that would otherwise be recognized as self.

This is contrary to present knowledge that such "self" peptides could not be used to create cytotoxic T-cells.

This method greatly expands the number of possible tumor specific antigens that can be used to activate CD8 T-cells against the tumor.

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.

112a In the claims which follow and in the preceding description of the invention, except where the context requires otherwise due to express language or necessary implication, the word "comprise" or variations such as "comprises" or "comprising" is used in an inclusive sense, i.e. to specify the presence of the stated features but not to preclude the presence or addition of further features in various embodiments of the invention.

It is to be understood that a reference herein to a prior art document does not constitute an admission that the document forms part of the common general knowledge in the art in Australia or in any other country.

:ooo o oo *o* 113 SEQUENCE LISTING GENERAL INFORMATION: APPLICANT: Cai, Zeling Sprent, Jonathan Brunmark, Anders Jackson, Michael Peterson, Per A Luxembourg, Alain Leturcq, Didier Jean Moriarty, Ann M.

(ii) TITLE OF INVENTION: ANTIGEN PRESENTING SYSTEM AND METHODS FOR ACTIVATION OF T-CELLS (iii) NUMBER OF SEQUENCES: 51 (iv) CORRESPONDENCE ADDRESS: ADDRESSEE: Olson Hierl, Ltd.

STREET: 20 North Wacker Drive, Suite 3600 CITY: Chicago STATE: Illinois COUNTRY: USA ZIP: 60606 COMPUTER READABLE FORM: MEDIUM TYPE: Floppy disk COMPUTER: IBM PC compatible OPERATING SYSTEM: PC-DOS/MS-DOS SOFTWARE: PatentIn Release Version #1.25 (vi) CURRENT APPLICATION DATA: APPLICATION NUMBER: FILING DATE: 8-MAR-1996

CLASSIFICATION:

(vii) PRIOR APPLICATION DATA: APPLICATION NUMBER: US 08/400,338 FILING DATE: 8-MAR-1995

CLASSIFICATION:

(viii) ATTORNEY/AGENT INFORMATION: NAME: Olson, Arne M.

REGISTRATION NUMBER: 30,203 REFERENCE/DOCKET NUMBER: TSRI4711 (ix) TELECOMMUNICATION INFORMATION: TELEPHONE: (312) 580-1180 TELEFAX: (312) 580-1189 INFORMATION FOR SEQ ID NO:1: SEQUENCE CHARACTERISTICS: LENGTH: 23 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) 114 (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO (xi) SEQUENCE DESCRIPTION: SEQ ID NO:1: CCACCATGGC CGTCATGGCG CCC 23 INFORMATION FOR SEQ ID NO:2: SEQUENCE CH-ARACTERISTICS: LENGTH: 23 base Dairs TYPE: nucleic acid STRANDEDNESS: sinzle TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO e i" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:2: GGTCACACTT TACAAGCTCT GAG 23 INFORMATION FOR SEQ ID NO:3: SEQUENCE CHARACTERISTICS: LENGTH: 23 base pairs TYPE: nucleic acid STRANDEDNESS: sincle TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO (xi) SEQUENCE DESCRIPTION: SEQ ID NO:3: CCACCATGCT GGTCATGGCG CCC 23 INFORMATION FOR SEQ ID NO:4: SEQUENCE CHARACTERISTICS: LENGTH: 23 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) 115 (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO (xi) SEQUENCE DESCRIPTION: SEQ ID NO:4: GGACTCGATG TGAGAGACAC ATC 23 INFORMATION FOR SEQ ID SEQUENCE CHARACTERISTICS: LENGTH: 23 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO (xi) SEQUENCE DESCRIPTION: SEQ ID CCACCATGCG GGTCATGGCG CCC 23 INFORMATION FOR SEQ ID NO:6: SEQUENCE CHARACTERISTICS: LENGTH: 23 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO (xi) SEQUENCE DESCRIPTION: SEQ ID NO:6: GGTCAGGCTT TACAAGCGAT GAG 23 INFORMATION FOR SEQ ID NO:7: SEQUENCE CHARACTERISTICS: LENGTH: 23 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) 116 (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO (xi) SEQUENCE DESCRIPTION: SEQ ID NO:7: CCACCATGCG GGTAGATGCC CTC 23 INFORMATION FOR SEQ ID NO:8: SEQUENCE CHARACTERISTICS: LENGTH: 23 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO e (xi) SEQUENCE DESCRIPTION: SEQ ID NO:8: GGTTACAAGC TGTGAGACTC AGA 23 INFORMATION FOR SEQ ID NO:9: SEQUENCE CHARACTERISTICS: LENGTH: 23 base cairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (gencmic) (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO (xi) SEQUENCE DESCRIPTION: SEQ ID NO:9: CCACCATGGC GCCCCGAAGC CTC 23 INFORMATION FOR SEQ ID SEQUENCE CHARACTERISTICS: LENGTH: 23 base pairs TYPE: nucleic acid STRANDEDNESS: sincle TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) 117 (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO (xi) SEQUENCE DESCRIPTION: SEQ ID GGTCACACTT TATTAGCTGT GAG 23 INFORMATION FOR SEQ ID NO:11: SEQUENCE CHARACTERISTICS: LENGTH: 23 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear MOLECULE TYPE: DNA (genomic) (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO (xi) SEQUENCE DESCRIPTION: SEQ ID NO:l1: CCACCATGGC GCCCCGAACC CTC 23 INFORMATION FOR SEQ ID NO:12: SEQUENCE CHARACTERISTICS: LENGTH: 23 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO (xi) SEQUENCE DESCRIPTION: SEQ ID NO:12: GGTCACAATT TACAAGCCGA GAG 23 INFORMATION FOR SEQ ID NO:13: SEQUENCE CHARACTERISTICS: LENGTH: 427 base pairs TYPE: nucleic acid STRANDEDNESS: double TOPOLOGY: linear 118 (ii) MOLECULE TYPE: DNA (ger~rnic) (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO (xi) SEQUENCE DESCRIPTION: SEQ ID NO:13: AATTCGTTGC AGGACAGC-AT GTGGTGCCCG ATGTGACTAG CTCTTTC-CTG TATCCTCTGG TTCCGA-TA.AG AGACCCAGAA1 CTCCGGCCCC CCACCGCCCA 120 CATACATATG TGGTACGCA.A GTAAGAC-TGC CTGCGCATGC CCCATGTGCC 180 TTTTGCATCC CATACAAGTC CCCAAAGTGC- AGAACCGAAC CAATTCTTCG 240 AAAGCTTCT GCACACGTCT CCACTCGAAT TTGGAGCCGG CCGGCGTGTG 300 GATTCGAACG A.AAGACCCGT GTC-TAAAGCC GCGTTTCCAA AATGTATAAA 360 TCTGGCCAAkT GTGCATCAGT TGTGGTCAGC A'GCAAAATCA AGTGAATCAT 420

CTAAAGG

427 INFORMATION FOR SEQ ID NO:14: (iJ) SEQUENCE CHARACTERISTICS: LENGTH: 740 base zairs TYPE: nucleic acid STPRANDEDNESS: double TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (geromic) (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO

CAGGCCGTCC

CCGCCACCCC

CCACC.AA!GAG

CGGGCAGAAC

CAAAAGAGGT

ACCGAGAGCA

CTCAGTGA

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 14: ATTCGATGCA CACTCACATT CTTCTCCTAA.. TACGATA.ATA AAACTTTCCA TGAAAAATAT GGAAkAAATAT ATGAAAATTG AGAAATCC7A. KAAACTGATA AACGCTCTAC TTAATTA.AAAL 120 TAGATAAATG GGAGCGGCTG GAATGGCGGA GCATGACCAA GTTCCTCCGC CAATCAGTCG 180 TAAACAGAA GTCGTGGAA GCGGATA.GA AGAATGTTCG ATTTGACGGG CAAGCATGTC 240 119 0* TGCTATGTGG CGGATTGCGG AGGAATTGC! CTGGAGACCA 300 AATATAGCGG TGTGAGTGAG CGGGAAGCTC GGTTTCTGTC 360 TCCAGCCAGT CGCTGTCC-AA ACTA.TTAAG TTAATGAGTT 420 GAGCAACAAT TAAGTTTATG TTTCAGTTCG GCTTAGATTT 480 TCAATCAATA CTTTAGAACA AAATCA.AAC TCATTCTAAT 540 TTTTA-ATGAT AAkGCATTTTG TCGTTTATAC TTTTTATATT 600 AAGTTCATTT TAATCGCCAG ATAAGCAJATA TATTGTGTAA.

660 GAAATTCAGG GAGACGGGGA AGTTACTATC TACTAAAAGC 720 TACTCTCTCT ACTCTAGAGT 740 INFORMATION FOR SEQ ID Ci) SEQUENCE C.HARACTERISTICS: LENGTH: 60 base pairs TYPE: nucleic acid STRANDEDNESS: sincle TOPOLOGY: linear (iJi) MOLECULE TYPE: DNZA (genornic) (iii) HYPOTHETICAL: NO ANTI-SENSE: NO GCAAkGGTTCT

CAGATCGAAC

TTTCATGTTA

CGCTGAAGGA

AGCTTGGTGT

TCGATATTAA

ATATTTGTAT

CAAACAATTT

CATGACCAAG

TCAAAACTAG

GTTTCGCGCT

CTTGCCACTT

TCATCTTTTT

ACCACCTATG

TCTTTATCAG

CTTACAGTTT

(xi) SEQUENCE DESCRIPTION: SEQ ID GCTTGGATCC AGATCTACCA TC-TCTCGCTC CGTGGCCTTA GCTC-TGCTCG CGCTACTCTC INFORMATION FOR SEQ ID NO:lE: SEQUENCE CHARACTERISTICS: LENGTH: 36 base zairs TYPE: nucleic ac-4 STRANDEDNE--SS: sincle TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genornic) (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO 120 (xi) SEQUENCE DESCRIPTION: SEQ ID NO:16: GGATCCGGAT GGTTACATGT CGCGATCCCA CTTAAC 36 INFORMATION FOR SEQ ID NO:17: SEQUENCE CHARACTERISTICS: LENGTH: 19 base pairs TYPE: nucleic acid STRANDEDNESS: sincle TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (gencmic) (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO (xi) SEQUENCE DESCRIPTION: SEQ ID NO:17: GGAGCCGTGA CTGACTGAG :..19 INFORMATION FOR SEQ ID NO:18: SEQUENCE CHARACTERISTICS: LENGTH: 24 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO (xi) SEQUENCE DESCRIPTION: SEQ ID NO:18: CCCTCGGCAC TGACTGACTC CTAG 24 INFORMATION FOR SEQ ID NO:19: SEQUENCE CHARACTERISTICS: LENGTH: 38 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO 121 a. a (xi) SEQUENCE DESCRIPTION: SEQ ID NO:19: CATCCTTATT AGATCTCACC ATCACCATCA CCATTGAG 38 IN'FORMATION FOR SEQ ID SEQUENCE CHARACTERISTICS: LENGTH: 38 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genornic) (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO SEQUENCE DESCRIPTION: SEQ ID TCGACTCAAT GGTGATGGTG ATGGTGAGAT CTAA TAA.G 38 INFORMATION FOR SEQ ID NO:21: Ui) SEQUENCE CHARACTERISTICS: LENGTH: 3875 base =airs TYPE: nucleic acid STRANDEDNESS: double TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO (xi) SEQUENCE DESCRIPTION: SEQ ID NO:21: TTGCAGGACA GGATGTGGTG CCCGATGTGA CT-AGCTCTTT GCTGCAGGCC CTGGTTCCGA TAAGAGACCC AG.AACTCCGG CCCCCCACCG CCCACCGCCA TATGTGGTAC GCAAGTAAGA GTGCCTGCGC ATGCCCCATG TGCCCCACCA ATCCCATACA AGTCCCCAAA GTGGAGAACC GAACCAATTC TTCGCGGGCA TTCTGCACAC GTCTCCACTC GAATTTGGAG CCGGCCGGCG TGTGCAAAAG A.ACGAAAGAC CCGTGTGTAA AGCCGCGTTT CCAAAATGTA TAAA.ACCGAG CAATGTGCAT CAGTTGTGGT CAGCAGCAAA ATCAAGTGAA TCATCTCAGT

GTCCTATCCT

CCCCCATACA

AGAGcTTTGC

GAACAA.AGC

AGGTGAATCG

AGCATCTGGC

GCAACTAAAG

120 180 240 300 360 420 122 0 to *:to to..

S S so.&

GGGPAATTCGA

AGTCGACCTG

AATA.AAACTT

GATA;-.ACGCT

CCAA-TTCCT

T TCC-.AT 7TG A

ACCAGCAAGG

TCCAC-ATCGA

TTTTTCATGT

TTCGCTGA.AG

ATAC-TGGT

TTTCCATATT

AAATATTTGT

CC7-AU.CAATT TTTTAkCA.ACG

ATCCCCCTTT

AGTTGCGCAG

GC GG T ATT TC TA-AG CCAGC C

CGGCATCCGC

CACCGTCATC

TTAAkTGTCAT

GCGC-AACCCC

AAkTAACCCTG

TCCGTGTCGC

AA-ACGCTGGT

AACTGGM.TCT

TGATGAGCAC

AAGAGC.AACT

TCACAGAAA

CCATGAGTGA

GCTCGGTACC

CAGGCATGCA

TCCATGAA

CTACTTAATT

CCGCCAATCA

CGGGCPAAGCA

TTCTCATGAC

ACT CAA AACT

TAGTTTCGCG

GACTTGCCAC

GTTCATCTTT

AAACCACCTA

ATTCTTTATC

TCTTACAkGTT

TCGTGACTGG

CGCCAGCTGG

CC-GA.ATGGC

ACACCGCATA

CCGACACCCG

TTACAGACAAk

ACCGAAACGC

GATAATA-ATG

TATTTGTTTA

ATAAATGCTT

CCTTATTCCC

G.AAAGTAAA

CAACAGCGGT

TTTTAAAGTT

CGGTCGCCGC

GCATCTTACG

TAACACTGCG

CGGGGATCCT

AGCTATTCGA

ATATGGA.:-AA--

AAAATAGTh'

GTCGTAXL-.AC

TGTCTGCTAT

C AAkG A-kTA TA.

MLGTCCAGCCOA

CTGAC-CAACA

TTTC.AATCAA

TTTTTTA-AT G TGAAGTctLT

AGC-AAATTCA.

TTACTCTCTC

GAAA-ACCCTG

CCT-ATAGCG

G-ATGGCGC!C

TGIGTGC A C:-C

CC;ACACCCG

GCTGTGACCG

GCC-AGACGA-A

GTTTCTTA

TTTTTCTAA-A

CPAATA'7

TTTTTTGCGG

GATGCTGA.f-2 A-AGATCCTT27

CTGCTATGTG

ATACACTAT"A

GATGGC-AT.GA

GCCAACTTAC

TATTAGATCT

TGCACACTCA

ATATATGAAA

AATGGGAGCG

7.GAAGTCGTG G-3TGGCGGATT

C-CGCTGAGTG

G-TCGCTGTCG

7.TTAAC-TTTA

ACTTTAGAA

ATAAGCATTT

TTTAATCGCC

GGGAGACGGG

-ACTCTAGAG

GCGTTACCCA

AAGAGGCCCG

-GCATGCGGTA

T CAGTACAAkT

CTGACGCGCC

TCTCCGGGAG

AGGGCCTCGT

CC-TCAGGTGG

-ACPATTCAAA

G.A-A.-AGGAA

CATTTTGCCT

ATCAGTTGGG

A 'GAGTTTTCG

GCGCGGTATT

CTCAGAATGA

CAGTAAGAGA

TTCTGACAAC

CACCATCACC

CPATTCTTCTC

ATTGAGAAAT

GCAGGAATGG

C-AAAGCGGAT

GC:2GAGC-AAT

AGCGGGAAGC

AAA-CTAATTA

TGTTTCAC-TT

C;AA:AATCAAA

TGTCGTTTAT

AGATA-AGCA,

PAGTTACTAT

TAGCTTGGCA

ACTTAATCGC

CLCCGA-TCGC

T-TTCTCCTT

CTGCTCTGAT

CT-7ACGGGCT

CTGCATGTGT

GtAACGCCTA

CACTTTTCGG

TATGTATCCG

GAGTATGAGT

TCCTIGTTTTT

TGCACGAGTG

CCCCGAAGAA

ATCCCGTATT

CTTGGTTGAG

ATTATGCAGT

GATCGGAGGA

ATCACCATTG

CTA.TACGAT

CCAA-AACT

CGGAGCITGG

A G AGzA T G

TGCACTGG

-TCGC-TT-TCTG

AGTtAA'TGAG

CGGCTTAGAT

ACTClATTCTA

ACTTTT-TATA

TATATTGTGT

CTACT.1AI-AG

CTGGCCGTCG

CTTGCAGCAC

CCTTCCCAAC

ACGC1TCTC-T

GCCGCA-.TAC-T

TGTCTGCTCC

CAGAGGTTTT

TTTTTATAGG

GGAA!ATGTGC

CTCATGAGAC

ATTCAACATT

GCTCACCCAG

GGTTACATCG

CGTTTTCCAA

GACGCCGGGC

TACTCACCAG

GCTGCCATAA

CCGA-AGGAGC

480 540 600 660 720 780 840 900 960 1020 1080 1140 1200 1260 1320 1380 1440 1500 1560 1620 1680 1740 1800 1860 1920 1980 2040 2100 2160 2 22 0 2280 2340 TAACCGCTTT TTTGCACAAC ATGGGGGATC ATGTAACTCG CCTTGATCGT TGGGAACCGG 123 AGCTGAATGA AGCCATACCA AACGACGA.GC

CAACGTTGCG

TAGACTGGAT

GCTGGTTTAT

CACTGGGGCC

CAACTATGGA

GGTAACTGTC

AATTTAAA-AG

GTGAGTTTTC

ATCCTTTTTT

TGGTTTGTTT

GAGCGCAGAT

CAAACTATTA

GGAGGCGGAT

TGCTGATAA-A

AGATGGTAAG

TGAACGA.AAT

AGACCAAGTT

GATCTAGGTG

GTTCCACTGA

TCTGCGCGTA

GCCGGATCAA

ACCAAATACT

S

0000 0005 0506 0S

OSSS

00 0 0 00

S

OOSO

OS@t ACTCTC-TAGC ACCGCCTACA

S.

0500

S

0 0006

S.

50 0

S

*500*S 5

GTGGCGATAA

AGCGC-TCGGG

CCGAACTGAG

AGGCGGACAG

CAGGGGGAAA

GTCGATTTTT

CCT.TTTTACG

CCTGATTCTG

CGAACcGACC

ACCGCCTCTC

CTGGAAAGCG

CCAGGCTTTA

ATTTCACACA

GTCGTGTCTT

CTGA7ACGGGG

ATACCTACAG

GTATCCGGTA

CGCCTGGTAT

GTGATGCTCG

GTCCTGGCCI

TGGATAACCC

GAGCGCAGCC

CCCGCGCGTI

GGCAGTGAGC

CACTTTATGC

GGWACAGC"

ACTGGCGAAC

AAAGTTGCAG

TCTGGAGCCG

CCCTCCCGT

AGACAC-ATCG

TACTCATATA

A-AGATCCTTT

GCC-TCAGACC

ATCTGCTGCT

GAGCTACCAA

GTCCTTCTAG

TACCTCGCTC

ACCGGGTTGG

GGTTCGTGCA

CGTGAGCATT

AGCGGCAGG

CTTTATAGTC

TCAGGGGGGC

TTTGCTGGCC

TATTACCGCC

AGTCAGTGAC

IGGCCGATTC

GCAACGCAJ

TTCCGGCTC(

C ATGACAT-A'

CTGACACCAC

TACTTACTCT

GACCACTTCT

GTGAGCGTGG

TCGTAGTTAT

CTGAGATAGG-

TACTTTAGPAT

TTC-ATAATCT

CCGTAGAAA

TGCAAACAA-3

CTCTTTTTCC

TGTAGCCGTA

TGCTAATCCT

ACTCAAGACG

CACAGCCCAG

GAGAAAGC-

TCGGAACAGG

*CTGTCGGGTT

*GGAGCCTATG

*TTTTGCTCAC

*TTTGAGTGAG

CGAGGAAGCG

STTAATGCAGC

*TAATGTGAGI

3TATGTTGTGI r TACCG AGCTTCCCGG C GCGCTCGGCC C GTCTCGCGGT CTACACGACG C

TGCCTCACTG

TC-TTTAAAA

CATGACCAA

GATCAAAGGA

AAAACCACCG

GAA~GGTAACT

GTTAGGCCAC

GTTACCAGTG

ATAGTTACCG

CTTGGAGCGA

CACGCTTCCC

AGAGCGCACG

TCGCCACCTC

GAAACGCC

ATGTCTTTCC

CTGATACCGC

GL-AGAGCGCC

TGGCACGACA

TAGCTCACTC

GGAATTGTGA

C-ATGCCTGTA GCA.ATGGCA-A

:AACAATTAA

TTCCGGCTG

.TCATTGCAG

;GGAGTCAGG

'TTAAGCATT

:TTCATTTTT

%TCCCTTAAC

TCTTCTTGAG

CTACCAGCGG

GGCTTCAGCA

CACTTCAAGA

GCTGCTGCCA

GATAAGGCGC

ACGACCTACA

GAAGGGAGAA

AGGGAGCTTC

TGACTTGAGC

AGCAACGCGG

TGCGTTATCC

TCGCCGCAGC

CAATACGCAA

GGTTTCCCGA

ATTAGGCACC

GCGGATAACA

2400 2460 2520 2580 2640 2700 2760 2820 2880 2940 3000 3060 3120 3180 3240 3300 3360 3420 3480 3540 3600 3660 3720 3780 3840 3875 INFORMATION FOR SEQ ID NO: 22: SEQUENCE CHARACTERISTICS: LENGTH: 71 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) 124 (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO (xi) SEQUENCE DESCRIPTION: SEQ ID NO:22: GATCCTTATT AGATCTTACC CATACGACGT CCCAGATTAC GCTCGATCTC ACCATCACCA TCACCATTGA G 71 INFORMATION FOR SEQ ID NO:23: SEQUENCE CHARACTERISTICS: LENGTH: 71 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO (xi) SEQUENCE DESCRIPTION: SEQ ID NO:23: TCGACTCAAT GGTGATGGTG ATGGTGAGAT CGAGCGTAAT CTGGGACGTC GTATGGGTAA GATCTAATAA G 71 INFORMATION FOR SEQ ID NO:24: SEQUENCE CHARACTERISTICS: LENGTH: 3908 base pairs TYPE: nucleic acid STRANDEDNESS: double TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO (xi) SEQUENCE DESCRIPTION: SEQ ID NO:24: TTGCAGGACA GGATGTGGTG CCCGATGTGA CTAGCTCTTT GCTGCAGGCC GTCCTATCCT CTGGTTCCGA TAAGAGACCC AGAACTCCGG CCCCCCACCG CCCACCGCCA CCCCCATACA 120 TATGTGGTAC GCAAGTAAGA GTGCCTGCGC ATGCCCCATG TGCCCCACCA AGAGcTTTGC 180 ATCCCATACA AGTCCCCAAA GTGGAGAACC GAACCAATTC TTCGCGGGCA GAACAAAAGC 240 TTCTGCACAC GTCTCCACTC GAATTTGGAG CCGGCCGGCG TGTGCAAAAG AGGTGAATCG 300 125

AACGAAZAGAC

CAATGTGCAT

GGGAATTCGA

TTACGCTCGA

CGATGCACAC

ArAATATATG

ATA.AATGGGA

AACAGAAGTC

TATGTGGCGG

ATAGCGGTGA

CCAGTCGCTG

ACAATTAAGT

CAATACTTTA

ATGATAAGCA

tATTTTAATC

TCAGGGAGAC

CTCTACTCTA

CTGGCGTTAC

GCGAAGAGGC

GCCTGA.TGCG

CTCTCAGTAC

CCGCTGACGC

CCGTCTCCGG

GAAZAGGGCCT

AGACC-TCAGG

AAATACATTC

ATTGAAAAAG

CGGCATTTTG

AAGATCAGTI

TTGAGAGTT'I

GTGGCGCGGI

CCGTGTGTAA

CAGTTGTGGT

GCTCGGTACC

TCTCACCATC

TCACATTCTT

AAA.ATTGAGA

GCGGCAGGMA

GTGGAAAGCG

ATTGCGGAGG

GTGAGCGGGA

TCGAAACTAA

TTATGTTTCA

GAACAAAATC

TTTTGTCGTT

GCCAGATAAG

GGGAAGTTAC

GAGTAGCTTG

CCAACTTAAT

CCGCACCGAT

GTATTTTCTC

PAATCTGCTCT

GCCCTGACGG

GAGCTGCATG

CGTGATACGC

TGGCACTTTT

AA.ATATGTAT

G.AAGAGTATG

CCTTCCTGTT

GGGTGCACGA

TCGCCCCGA-P

ATTATCCCGI

AGCCGCGTTT

CAGCAGCA-AA

CGGGGATCCT

ACCATCACCA

CTCCTAATAC

AATCCA.AAkA

TGGCGGAGCA

GATAGAAAC-A

AATTGCACTG

AGCTCGGTTT

TTAAGTtAAT

GTTCGGCTTA

AAAACTCATT

TATACTTTTT

CAATATATTG

TATCTACTAA

GCACTGGCCG

CGCCTTGCAG

CGCCCTTCCC

CTTACGCATC

GATGCCGCAT

GCTTGTCTGC

TGTCAGAGGT

CTATTTTTAT

CGGGG~kA-TG

CCGCTCATGA

AGTATTCA.AC

TTTGCTCACC

GTGGGTTAC;

GAACGTTTTC

ATTGACGCCC

CCAAAATGTA

ATCAAGTGAA

TATTAGATCT

TTGAGTCGAC

GATAATAAA

ACTGATAAAC

TGGCCAAGTT

ATGTTCGATT

GAGACCAGCA

CTGTCCAGAT

GAGTTTTTCA

GATTTCGCTG

CTAATAGCTT

ATATTTCGAT

TGTAAATATT

AAGCCAAACA

TCGTTTTACA

CACATCCCCC

AACAGTTGCG

TGTGCGGTAT

AGjTTAAGCCA

TCCCGGCATC

TTTCACCGTC

AGGTTAATGT

TGCGCGGAAC

GACAATA3ACC

*ATTTCCGTGT

*CAGAAACGCT

TCGAACTGGA

*CAATGATGAG

GGCAAGAGCP

TCATCTCAGT

TACCCATACG

CTGCAGGCAT

CTTTCCATGA

CCTCTACTTA

CCTCCGCCAA

TGACGGGCAA

AGGTTCTCAT

CGAACTCAAA

TGTTAGTTTC

AAGGACTTGC

GGTGTTCATC

ATTAAACCAC

TGTA.TTCTTT

ATTTCTTACA

ACGTCGTGAC

TTTCGCCAGC

CAGCCTGAAT

TTCACACCGC

GCCCCGACAC

CGCTTACAGA

ATCACCGAAA

CATGATAATA

CCCTATTTGT

CTGATAAATG-

CGCCCTTAT'I

GGTGAA.AGTP.

TCTCAACAGC

CACTTTTAV

ACTCGGTCGC

GCAACTALAAG

ACGTCCCAGA

GCAAGCTATT

AAAATATGGA

ATTA.AAATAG

TCAGTCGTAA

GCATGTCTGC

GACCAAGAAT

ACTAGTCCAG

GCGCTGAGCA

CACTTTCAAT

TTTTTTTTTA

CTATGAAGTc

ATCAGGAAAT

GTTTTA.CTCT

TGGGAT-A-2kCC

TGGCGTPAATA

GGCGAATGGC

ATATGGTGCA

CCGCCA-ACAC

CAAGCTGTGA

CGCGCGAGAC

ATGGTTTCTT

TTATTTTTCT

CTTCM.TPAT

CCCTTTTTTG

LAAAGATGCTG

GGTAAGATCC

LGTTCTGCTAT

CGCATACACT

TA.AAACCGAG AGCATCTGGC 360 420 480 540 600 660 720 780 840 900 960 1020 1080 1140 1200 1260 1320 1380 1440 1500 1560 1620 1680 1740 1800 1860 1920 1980 2040 2100 2160 2220

Q

ATTCTCAGAA TGACTTGGTT GAGTACTCAC CAGTCACAGA AAAGCATCTT ACGGATGGCA 126 .t.

TGACAGTAAG

TACTTCTGAC

ATCATGTAAC

AGCGTGACAC

A.ACTACTTAC

CAGGACCACT

CCGGTGAGCG

GTATCGTAGT

TCGCTGAGAT

ATATACTTTA

TTTTTGATAA

ACCCCGTAGA

GCTTGCAAAC

CAACTCTTTT

TAGTGTAGCC

CTCTGCTAAT

TGGACTCAAG

GCACACAGCC

ATTGAGAAAG

GGGTCGGAAC

GTCCTGTCGG

GGC GGAGCCT

GCC-TTTTGCT

GCCrTTGAGT GAGCGAGGA7A

TCATTAATGC

AATTAATGTG

TCGTATGTTG

GATTACCG

AGAATTATC

AACGATCGGA

TCGCCTTGAT

CACGATGCCT

TCTAGCTTCC

TCTGCGCTCG

TGGGTCTCGC

TATCTACACG

AGGTGCCTCA

GATTGATTTA

TCTCATGACC

AAAGATCAA

AAAAAAACCA

TCCGAAGGTA

GTAGTTAGGC

CCTGTTACCA

ACGATAC-TTA

CAGCTTGGAG

CGCCACGCTT

AGGAC-AGCGC

GTTTCGCCAC

ATGGAAAAAC

CACATGTCTT

GAGCTGATAC

GCGGAAGAGC

AGCTGGCACG

AGTTAGCTCA

TGTGGAATTG

AGTGCTGCCA:

GGACCGAAGG

CGTTGGGAAC

GTAGCAATGG

CGGCAACAAT

GCCCTTCCGG

GGTATCATTG

ACGGGGAGTC

CTGATTAAGC

AAACTTCATT

A-AAATCCCTT

GGATCTTCTT

CCGCTACCAG

ACTGGCTTCA

CACCACTTC-k

GTGGCTGCTC-

CCGGATAAGG

CGAACGACCT-

CCCGAAGGGA

ACGAGGGAGC

CTCTGACTTG

GCCAGCAACG

TCCTGCGTTA

CGCTCGCCGC

GCCCAATACG

ACAGGTTTCC

CTCATTAGGC

TGAGCGGATA

TAACCATGAG

AG CTAAC CG C

CGGAGCTGAA

CAACAACGTT

TAATAGACTG

CTGGCTGGTT

CAGCACTGGG

AGGCAACTAT

ATTGGTAACT

TTTAATTTAA

AA.CGTGAGTT

GAGATCCTTT

CGGTGGTTTG

GCAGAGCGCA

AGAACTCTGT

CCAGTGGCGA

CGCAGCGGTC

ACACCGAACT

GAALAGGCGGA

TTCCAGGGGG

AGCGTCGATT

CGGCCTTTTT

TCCCCTGATT

AGCCGAACCG

CAAACCGCCT

CGACTGGAAA

ACCCCAGGCT

ACAATTTCAC

TGATAACACT

TTTTTTGCAC

TGAAGCCATA

GCGCAAACTA

C-ATGGAGGCG

TATTGCTGAT

GCCAGATGGT

GGATGAACGA

GTCAGACCAA

AAGGATCTAG

TTCGTTCCAC

TTTTCTGCGC

TTTGCCGGAT

GATACCAAAT

AGCACCGCCT

TAAGTCGTGT

GGGCTGAACG

GAGATACCTA

CAGGTATCCG

AAACGCCTGG

TTTGTGATGC

ACGGTCCTGG

CTGTGGATAA.

ACCGAGCGCA

CTCCCCGCGC

GCGGGCAGTG

TTACACTTTA

ACAGGAAACA

GCGGCCAACT

AACATGGGGG

CCAAACGACG

TTAACTGGCG

GATAAAGTTG

AAATCTGGAG

AAGCCCTCCC

AATAGACAGA

GTTTACTCAT

GTGAAGATCC

TGAGCGTCAG

GTAATCTGCT

CA-AGAGCTAC

ACTGTCCTTC

ACATACCTCG

CTTACCGGGT

GGGGGTTCGT

CAGCGTGAGC

GTAAGCGGCA

TATCTTTATA

TCGTCAGGGG

CCTTTTGCTG

CCGTATTACC

GCGAGTCAGT

GTTGGCCGAT

AG CGCAACG C

TGCTTCCGGC

GCTATGACAT

2280 2340 2400 2460 2520 2580 2640 2700 2760 2820 2880 2940 3000 3060 3120 3180 3240 3300 3360 3420 3480 3540 3600 3660 3720 3780 3840 3900 3908

S

S

INFORMATION FOR SEQ ID SEQUENCE CHARACTERISTICS: LENGTH: 41 base pairs TYPE: nucleic acid 127 STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO (xi) SEQUENCE DESCRIPTION: SEQ ID GATCCTTATT AGATCTCACC ATCACCATCA CCATTGTTGA G 41 INFORMATION FOR SEQ ID NO:26: SEQUENCE CHARACTERISTICS: LENGTH: 41 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO (xi) SEQUENCE DESCRIPTION: SEQ ID NO:26: TCGACTCAAC AATGGTGATG GTGATGGTGA GATCTAATAA G 41 INFORMATION FOR SEQ ID NO:27: SEQUENCE CHARACTERISTICS: LENGTH: 3878 base pairs TYPE: nucleic acid STRANDEDNESS: double TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO (xi) SEQUENCE DESCRIPTION: SEQ ID NO:27: TTGCAGGACA GGATGTGGTG CCCGATGTGA CTAGCTCTTT GCTGCAGGCC GTCCTATCCT CTGGTTCCGA TAAGAGACCC AGAACTCCGG CCCCCCACCG CCCACCGCCA CCCCCATACA 120 TATGTGGTAC GCAAGTAAGA GTGCCTGCGC ATGCCCCATG TGCCCCACCA AGAGcTTTGC 180 ATCCCATACA AGTCCCCAAA GTGGAGAACC GAACCAATTC TTCGCGGGCA GAACAAAAGC 240 TTCTGCACAC GTCTCCACTC GAATTTGGAG CCGGCCGGCG TGTGCAAAAG AGGTGAATCG 300 128

AACGAAGAC

CAATGTGCAT

GGGPAATTCGA

TTGAGTCGAC

GATPAAT-AAAA

ACTGATAATAC

TGGCCAAGTT

ATGTTCGATT

GAGACCAGCA

CTGTCCAGAT

GAGTTTTTCA

GATTTCGCTG

CTAATAGCTT

ATATTTCGAT

TGTAAkATATT AAGCCkAACA

TCGTTTTACA

CACATCCCCC

AACAGTTGCG

TGTGCGGTAT

AGTTAAkGCCA

TCCCGGCATC

TTTCACCGTC

AGGTTA-ATGT

TGCGCGG;AC

GACAATAACC

ATTTCCGTGT

CAG-AA-ACGCT

TCGA.ACTGGA

CAATGATGAG

GGCAAGAGCA

CCGTGTGTA-k

CAGTTGTGGT

GCTCGGTACC

CTGCAGGCAT

CTTTCCATGAk

GCTCTACTTA

CCTCCGCCZA

TGACGGGCAA

AGGTTCTCAT

CGAACTCAAA

TGTTAGTTTC

AAGGACTTGC

GGTGTTCATC

ATTAAACCAC

TGTATTCTTT

ATTTCTTACA

ACGTCGTGAC

-ITTCGCCAGC

CAGCCTGJAAT

TTCACACCGC

GCCCCGACAC

CGCTTACAGA

ATCACCGAAA

CATGATAATA

CCCTATTTGT

CTGATAAATG

CGCCCTTATT

GGTGAAAGTA

TCTCAACAGC

CACTTTTAAA

ACTCGGTCGC

AGCCGCGTTT

CAGCAGCAAA

CGGGGATCCT

GCAAGCTATT

AAAATATGGA

ATTAAAATAG

TCAGTCGTA-k

GCATGTCTGC

GACCAAGAAT

ACTAGTCCAG

GCGCTGAGC-A

CACTTTCAAT

TTTTTTTTTA

CTATGAAGTc

ATCAGGAKAT

GTTTTACTCT

TGGGAAAACC

TGGCGTAATA

GG-CGAATGGC

ATATGGTGCA

CCGCCAACAC

CAAGCTGTGA

CGCGCGAGAC

ATiGGTTTCTT

TTATTTTTCT

CTTCAkATAAkT

CCCTTTTTTG

AAAGATGCTG

GGTAAGATCC

GTTCTGCTAT

CGCATACACT

CCAA.AATGTA

ATCAAGTGAA

TATTAGATCT

CGATGCACAC

AAAAZTATATG

ATA-ATGGC-A

AACAGAAGTC

TATGTGGCGG

ATAGCGGTGA

CCAGTCGCTG

ACAATTAAGT

CAATACTTTA

ATGATA-AGCA

tATTTTAATC

TCAGGGAGAC

CTCTACTCTA

CTGGCGTTAC

GCGAAGAGGC

GCCTGATGCG

CTCTCAGTAC

CCGCTGACGC

CCGTCTCCGG

GAAAGGGCCT

AGACC-TCAGG

AAATIACATTC

ATTGAAAAAG

CGGCATTTTG

AAGATCAGTT

TTGAGAGTTT

GTGGCGCGGT

ATTCTCAGAA

TAAAACCGAG

TCATCTCAGT

CACCATCACC

TCACATTCTT

AAAkATTGAGA

GCGGCAGGAA

GTGGAAAGCG

ATTGCGGAGG

GTGAGCGGGA

TCGAAACTAA

TTATGTTTCA

GAACAAAATC

TTTTGTCG-T

GCCAGATA.AG

GGGAAGTTAC

GA'GTAGCTTG

CCAAL~CTTP-A-T

CCGCACCGAT

GTATTTTCTC

AATCTGCTCT

GCCCTGACGG

G-AGCTGCATG

CC-TGATACGC

TGG-CACTTTT

AAATATGTAT

GAP.GAGTATG

CCTTCCTGTT

GGGTGCACGA

TCGCCCCGAA

ATTATCCCGT

TGACTTGGTT

AGCATCTGGC

GCAI4CTA-AAG

ATCACCATTG

CTCCTAATAC

AATCCAAA

TGGCGGAGCA

GATAC-AA-AGA

AATTGCACTG

AGCTCGC-TTT

TTAAGTtAAT

GTTCGGCTTA

AAAACTCATT

TATACTTTTT

CAATATATTG

TATCTACTAA

GCACTGGCCG

CGCCTTGCAG

CGCCCTTCCC

CTTACGCATC

GATGCCGCAT

GCTTdTCTGC

TGTCAG-AGGT

CTATTTTTAT

CGGGC-A.AATG

CCGCTCATGA

AG-TATTCAAC

TTTGCTCACC

GTGGGTTACA

GAACGTTTTC

ATTGACGCCG

GAGTACTCAC

360 420 480 540 600 660 720 780 840 900 960 1020 1080 1140 1200 1260 1320 1380 1440 1500 1560 1620 1680 1740 1800 1860 1920 1980 2040 2100 2160 2220 CAGTCACAGA AAAGCATCTT ACGGATGGCA TGACAGTAAG AGAATTATGC AGTGCTGCCA 129

TAACCATGAG

AGCTAACCGC

CGGAGCTGAA

CAACAACGTT

TAATAGACTG

CTGGCTGGTT

CAC-CACTCGGG

AGGCAACTAT

ATTGGTAACT

TTTAATTTAA

AACGTGAGTT

GAGATCCTTT

CGGTGGTTTG

GCAGAGCGCA

AGAACTCTGT

CCAGTGGCGA

CGCAGCGGTC

ACACCGAACT

GA.AAGGCGGA

TTCCAGGGGG

AGCGTCGATT

CGGCCTTTTT

TCCCCTGATT

AGCCGAACcG

CAAACCGCCT

CGACTGGAA

ACCCCAGGCT

TGATAACACT

TTTTTTGCAC

TGAAGCCATA

GCGCAAACTA

GATGGAGGCG

TATTGCTGAT

GCCAGATGGT

GGATGAACGA

GTCAGACCAA

AAGGATCTAG

TTCGTTCCAC

TTTTCTGCGC

TTTGCCGGAT

GATACCAAA.T

AGCACCGCCT

TAAGTCGTGT

GGGCTGAACG

GAGATACCTA

CAGGTATCCG

AAACGCCTGG

TTTGTGATGC

ACGGTCCTGG

CTGTGGATAA

ACCGAGCGCA

CTCCCCGCGC

GCGGGCAGTG,

TTACACTTTA

GCGGCCAACT

AACATGGGGG

CCAAACGACG

TTAACTGGCG

GATAAAGTTG

AAATCTGGAG

AAGCCCTCCC

AATAGACAGA

GTTTACTCAT

GTGAAGATCC

TGAGCGTCAG

GTAAtCTGCT

CAAGAGCTAC

ACTGTCCTTC

ACATACCTCG

CTTACCGGGT

GGGGGTTCGT

CAGCGTGAGC

C'T-AGCGGCA

TATCTTTATA-

TCGTtAGGGG

CCTTTTGCTG

CCGTATTACC

GCGAGTCAGT

GTTGGCCGAT

AGCGCAACGC

TGCTTCCGGC

TACTTCTGAC AACGATCGGA ATCATGTAAC TCGCCTTGAT AGCGTGACAC CACGATGCCT AACTACTTAC TCTAGCTTCC CAGGACCACT TCTGCGCTCG CCGGTGAGCG TGGGTCTCGC GTATCGTAGT TATCTACACG TCGCTGAGAT AGGTGCCTCA ATATACTTTA GATTGATTTA TTTTTGATAA TCTCATGACC ACCCCGTAGA AA.AGATCAA GCTTGCA7LAC AAAAAAACCA CAACTCTTTT TCCGAAGGTA TAGTGTAGCC GTAGTTAGGC CTCTGCTAAT CCTGTTACCA TGGACTCAAG ACGATAGTTA GCACACAGCC CAGCTTGGAG ATTGAGAAAG CGCCACGCTT GGGTCGGAAC AGGAGAGCGC GTCCTGTCGG -GTTTCGCCAC GGCGGAGCCT ATGGAAAAAC GCCTTTTGCT CACATGTCTT GCCTTTGAGT GAGCTGATAC GAGCGAGC-AA GCGGAAGAGC TCATTAATGC AGCTGGCACG AATTA-ATGTG AGTTAGCTCA TCGTATGTTG TGTGGAATTG

GATTACCG

GGACCGAAGG

CGTTGGGAAc

GTAGCAATGG

CGGCAACAAT

GCCCTTCCGG

GGTATCATTG

ACGGGGAGTC

CTGATTAAGC

AAACTTCATT

AAAATCCCTT

GGATCTTCTT

CCGCTACCAG

ACTGGCTTCA

CACCACTTCA

GTGGCTGCTG

CCGGATAAGG

CGAACGACCT

CCCGAAGGGA

ACGAGGGAGC

CTCTGACTTG

GCCAGCAACG

TCCTGCGTTA

CGCTCGCCGC

GCCCAATACG

ACAGGTTTCC

CTCATTAGGC

TGAGCGGATA

2280 2340 2400 2460 2520 2580 2640 2700 2760 2820 2880 2940 3000 3060 3120 3180 3240 3300 3360 3420 3480 3540 3600 3660 3720 3780 3840 3878

S

ACAJ-'TTTCAC ACAGGAAACA GCTATGACAT INFORMATION FOR SEQ ID NO:28: SEQUENCE CHARACTERISTICS: LENGTH: 47 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear 130 9 (ii) MOLECULE TYPE: DNA (genomic) (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO (XiJ) SEQUENCE DESCRIPTION: SEQ ID NO:28: GATCCTTATT AGATCTGCTT GGCGCCATCC TCAATTTGGG GGTTGAG INFORPTION FOR SEQ ID NO:29: Ci) SEQUENCE CHARACTERISTICS: LENGTH: 47 base pairs TYPE: nucleic aci' d STRANDEDNESS: sincale TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (iii) HYPOTHETICA.L: NO (iv) ANTI-SENSE: NO (xi) SEQUENCE DESCRIPTION: SEQ ID NO:29: TCGACTCAAC CCCCAAATTG AGGATGGCGC CAAGCAGATC T~kAAG INFOR!MATION FOR SEQ ID (iJ) SEQU ENCE CHARACTERISTICS: LENGTH: 3883 base pairs TYPE: nucleic acid STRANDEDNESS: double TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA (iiiJ) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO (xi) SEQUENCE DESCRIPTION: SEQ ID, TTGCAGGACA GGATGTGGTG CCCGATGTGA CTAGCTCTTT GCTGCAGGCC TGGTTCCGAT AAGAGACCCA GAACTCCGGC CCCCCACCGC CCACCGCCAC ATGTGGTACG CAAGTAAGAG TGCCTGCGCA TGCCCCATGT GCCCCACCAA TCCCATACAA. GTCCCCAI.AG TGGAGAACCG AACCAATTCT TCGCGGGCAG TCTGCACACG TCTCCACTCG AATTTGGAGC CGGCCGGCGT GTGCAAAAGA ACGAAAGACC CGTGTGTAAA. GCCGCGTTTC CAAAATGTAT AAA-ACCGAGA

GTCCTATCCT

CCCCATACAT

GAGcTTTGCA

AACAA.AAGCT

GGTGAATCGA

GCATCTGGCC

120 180 240 300 360 131 AATGTGCATC AGTTGTGGTC AGCAGCAAA-k TCAAGTGAAT CATCTCAGTG CAACTAAAGG 0

GGAATTCGAG

GGGGGTTGAG

AATACGATAA\

PJ.AkAACTGA

GAGCATGGCC

AAkAGAATGTT

CACTGGAGAC

GGTTTCTGTC

TtAATGAGTT

GCTTAGATTT

TCATTCTAAT

TTTTTATATT

TATTGTGTAA

ACTAAAAGCC

GGCCGTCGTT

TGCAGCACAT

TTCCCAACAG

GCATCTGTGC

CGCATAGTTA

TCTGCTCCCG

GAGGTTTTCA

TTTATAGGTT

AAAkTGTGCGC

CATGAGACAA

TCAACATTTC

TCACCCAGAA

TTACATCGA-A

TTTTCCAATC

CGCCGGGCA.P

CTCACCAGTC

TGCCATAACC

CTCGGTACCC G TCGACCTGCA C TAkA.ACTT-TC C TAAACGCTCT AAGTTCCTCC C

CGATTTGACG

CAGCAAGGTT

CAGATCGAAkC

TTTCATGTTA

CGCTGAAGGA

AGCTTGGTGT

TCGATATTAA

ATATTTGTAT

AAACAATTTC

TTACAACGTC

CCCCCTTTCG

TTGCGCAGCC

GGTATTTCAC

AGC CAGC CC C

GCATCCGCTT

CCGTCATCAC

AATGTCATGA

GGAACCCCTA

TAACCCTGAT

CGTGtCGCCC

ACGCTGGTGA

CTGGATCTCA

IATGAGCACTT

GAGCAACTCG

ACAGAAAAGC

ATGAGTGATA

;GGGATCCTT

GCATGCAAG

:ATGAAA.AAT

CTTAATTAA

;CCAATCAC-T

'GCAAGCATG

:TCATGACCA

E'CAAAACTAG

GTTTCGCGCT

CTTGCCACTT

TCATCTTTTT

ACCACCTATG

TCTTTATCAG

TTACAGTTTT

GTGACTGGGA

CCAGCTGGCG

TGAATGGCGA.

ACCGCATATG

GACACCCGCC

ACAG-ACAAGC

CGAAACGCGC

TAATAATGGT

TTTGTTTATT

AAATGCTTCA~

TTATTCCCT7I

AAGTAAAG.

ACAGCGGTAZ

TTAAAGTTCI

GTCGCCGCAI

ATCTTACGG,

ACACTGCGG(

ATTAGATCTG

CTATTCGATG

ATGGAAAAAT

AATAGATAAA

CGTAAAACAG

TCTGCTATGT

AGAPTATAGC

TCCAGCCAGT

GAGCAACAAT

TCAATCAATA

TTTTAATGAT

AAGTctATTT

GAAATTCAGG

ACTCTCTCTA

AAACCCTGGC

TAATAGCGAA

ATGGCGCCTG

GTGCACTCTC

AACACCCGCT

TGTGACCGTC

GAGACGA-AAG

TTCTTAGACG

TTTCTAAAT;

.ATAATATTG)!

TTTTGCGGC;

6TGCTGAAGAI

LGATCCTTGAC

GCTATGTGG(

ACACTATTC'

TGGCATGACJ

CAACTTACT'

CTTGGCGCCA

CACACTCACA

ATATGAAAAT

TGGGAGCGGC

AAGTCGTGGA

GGCGGATTGC

GGTGAGTGAG

CGCTGTCGAA~

TAAGTTTATG

CTTTAGAACA

AAGCATTTTG

TAATCGCCAG

GAGACGGGAA

CTCTAGAGTA

GTTACCCAAC

GAGGCCCGCA

ATGCGGTATT

AGTACAATCT

GACGCGCCCT

TCCGGGAGC'

GGCCTCGTG-A

ITCAGGTGGCP

CATTCAAATP

L At-tAGGAAG; L TTTTGCCTTC

"CAGTTGGGTC

AGTTTTCGCC

GCGGTATTA'

r' CAGAATGAC'.

GTAAGAGAK

r CTGACAACG

TCCTCAATTT

TTCTTCTCCT

TGAGAAATCC

AGGAATGGCG

AAGCGGATAG

GGAGGPATTG

CGGGAAGCTC

ACTA?.TTAAG

TTTCAGTTCG

AAATCAA.AAC

TCGTT-TATAC

ATAAGCAATA

GTTACTATCT

GCTTGGCACT

TTAATCGCCT

CC-GATCGCCC

TTCTCCT.TAC

GCTCTGATGC

GACGGC-CTTG

GCATGTGTCA

TACGCCTATT

CTTTTCGGGG

LTGTATCCGCT

GTATGAGTAT

CTGTTTTTGC

CACC-AGTGGG

CCGAAGA.ACG

r' CCCGTATTGA C' TGGTTGAGTA r' TATGCAGTGC k. TCGGAGGACC 420 480 540 600 660 720 780 840 900 960 1020 1080 1140 1200 1260 1320 1380 1440 1500 1560 1620 1680 1740 1800 1860 1920 1980 2040 2100 2160 2220 2280 132 9

GAAGGAGCTA

GGAACCGGAG

AATGGCAACA

ACAATTAATA

TCCGGCTGGC

CATTGCAGCA

GAGTCAGG-CA

TAAGCATTGG

TCATTTTTAA

CCCTTAACGT

TTCTTGAGAT

ACCAGCGGTG

CTTCAGCAGA

CTTCAAGAAC

TGCTGCCA.GT

TAAGGCGCA G

GACCTACACC

AGGGAGAG

GGAGCTTCCA

ACTTGAGCGT

CAACGCGGCC

CGTTATCCCC

GCCGCAGCCG

ATACGCAAAC

TTTCCCGACT

TAGGCACCCC

GGATAACAAT

ACCGCTTTTT

CTGAATGAAG

ACGTTGCGCA

GACTGGATGG

TGGTTTATTG

CTGGGGCCAG

ACTATGGATG

TAACTGTCAG

TTTAAAAGGA

GAGTTTTCGT

CCTTTTTTTC

GTTTGTTTGC

GCGCAGATAC

TCTGTAGCAC

GGCGATAAGT

CGGTCGGGCT

G.AACTGAGAT

GCGGACAG'T

GGGGGAAACG

CGATTTTTGT

TTTTTACGGT

TGATTCTGTG

AACCGACCGA

CGCCTCTCCC

GGAAAGCGGG

AGGCTTTACA

TTCACACAGG

TGCACAACAT

CCATACCAAA

A.ACTATTAAC

AGGCGGATAA:-

CTGATAAATC

ATGGTAAGCC

AACGAAATAG

ACCAAGTTTA

TCTAGGTGAA

TCCACTGAGC

TGCGCGTAAT

CGGATCAAGA

CAAATACTGT

CGCCTACATA

CGTGTCTTAC

GAACGGGGGG

ACCTACAGCG

ATCCGGTAAG

CCTGGTATCT

GATG'CTCGTC

CCTGGCCTTT

GATAACCGTA

GCGCAGCGAG

CGCGCGTTGG

CAGTGAGCGC

CTTTATGCTT

AAACAGCTAT

GGGGGATCAT

CGACGAGCGT

TGGCGAACTA

AGTTGCAGGA

TGGAGCCGGT

CTCCCGTATC

ACAGATCGCT

CTCATATATA

GATCCTTTTT

GTCAGACCCC

CTGCTGCTTG

GCTACCAACT

CCTTCTAGTG

CCTCGCTCTG

CGGGTTGGAC

TTCGTGCACA

TGAGCATTGA

CGGCAGGGTC

TTATAGTCCT

AGGGGGGCGG

TGCTGGCCTT

TTACCGCCTT

TCAGTGAGCG

CCGATTCATT

AACGCAATTA

CCGGCTCGTA

GTAACTCGCC

GACACCACGA

CTTACTCTAG

CCACTTCTGC

GAGCGTGGGT

GTAGTTATCT

G'AGATAGGTG

CTTTAGATTG

GATA.ATCTCA

GTAGAAAAGA

CAAACAAAA

CTTTTTCCGA

TAGCCGTAGT

CTAATCCTGT

TCAAGACGAT

CAGCCCAGCT

GAAAL-GCGCCA

G!A-CAGGAG

GTCGGGTTTC

AGCCTATGGA

TTGCTCACAT

TGAGTGAGCT

AGGAAGCGGA

AATGCAGCTG

ATGTGAGTTA

TTGATCGTTG

TGCCTGTAGC

CTTCCCGGCA

GCTCGGCCCT

CTCGCGGTAT

ACACGACGGG

CCTCACTGAT

ATTTAALAACT

TGACCAAAAT

TCAAAGGATC

AACCACCGCT

AGGTAACTGG

TAGGCCACCA

TACCAGTGGC

AGTTACCGGA

TGGAGCGAAC

CGCTTCCCGA

AGCGCACGAG

GCCACCTCTG

AAA.ACGCCAG

GTCTTTCCTG

GATACCGCTC

AGAGCGCCCA

GCACGACAGG

GCTCACTCAT

2340 2400 2460 2520 2580 2640 2700 2760 2820 2880 2940 3000 3060 3120 3180 3240 3300 3360 3420 3480 3540 3600 3660 3720 3780 3840 3883 9 *99*4* TGTTGTGTGG AATTGTGAGC GACATGATTA CCG INFORMATION FOR SEQ ID NO:31: SEQUENCE CH.ARACTERISTICS: LENGTH: 879 base pairs TYPE: nucleic acid STRANDEDNESS: double TOPOLOGY: linear 133 (ii) MOLECULE TYPE: cDNA (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO (xi) SEQUENCE DESCRIPTION: SEQ ID NO:31: GAATTCATGG GCCACACACG GAGGCAGGGA ACATCACCAT CCAAGTGTCC ATACCTCAAT 0*St

TTCTTTCAGC

ACCAAGGAAG

CTGGCACAAA

GGGGACATGA

CTCTCCATTG

CTGAAGTATG

AAAGCTGACT

AGGATAATTT

GGAGAAGAAT

GCTGTTAGCA

AAGTATGGAC

TTTCCTGATA

GTGATATGCT

AGATTGAGA-4

TCTTGGTGCT

TGAA.AGAAGT

CTCGCATCTA

ATATATGGCC

TGATCCTGGC

AAAA-AGACGC

TCCCTACACC

GCTCAACCTC

TAAATGCCAT

GCAAACTGGA

ATTTAAGAGT

ACCTGCTCCC

GCCTGACCTA

GGGAAAGTGT

GGCTGGTCTT

GGCAACGCTG

CTGGCAAAAG

CGAGTACAAG

TCTGCGCCCA

TTTCAAGCGG

TAGTATATCT

TGGAGGTTTT

CAACAkCAACA

TTTCAATATG

GAATCAGACC

ATCCTGGGCC

CTGCTTTGCC

ACGCCCTGTA.

TCTCACTTCT

TCCTGTGGTC

GAGAAGAAAAk

AACCGGACCA

TCTGACGAGG

GAACACCTGG

GACTTTGAAA

CCAGAGCCTC

GTTTCCCAAG

ACAACCAACC

TTCAACTGGA

ATTACCTTPA

CCAAGATGCA

TAAGGATTC

C-TTCAGGTC-T

ACAATGTTT C

TGGTGCTGAC

TCTTTGATAT

GCACATACGA

CTGAAGTGAC

TTCCAACTTC

ACCTCTCCTG

ATCCTGAAAZ!C

ACAGCTTCAT

ATACAACCAA

TCTCAGTAAA

GAGAGAGAAG

TATCCACGTG

TGTTGA-GAG

TATGATC-TCT

CACTAATAAC

GTGTGTTGTT

GTTATCAGTC

TA.ATATTAGA

GTTGGAAAAT

TGAGCTCTAT

GTGTCTCATC

GCAAGAGCAT

TGGA.ATTTTT

GAGGAJ.TGAG

120 180 240 300 360 420 480 540 600 660 720 *78 0 840 879 INFORMATION FOR SEQ ID NO:32: SEQUENCE CHARACTERISTICS: LENGTH: 738 base pairs TYPE: nucleic acid STRANDEDNESS: double TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA (iii) HYPOTHETICAL: NO Civ) ANTI-SENSE: NO (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 32: GAATTCATGG GCCACACACG GAGGCAGGGA ACATCACCAT CCAAGTGTCC ATACCTCAAT TTCTTTCAGC TCTTGGTGCT GGCTGGTCTT TCTCACTTCT GTTCAGGTGT TATCCACGTG ACCAAGGAAG TGAAAGAAGT GGCAACGCTG TCCTGTGGTC ACA.ATGTTTC TGTTGAAGAG 120 180 134 CTGGC-ACAAA CTCGCATCTA CTGGCAAAAG GAGAAGAAAA GGGGACATGA ATATATGGCC CGAGTACAAG AACCGGACCA CTCTCCATTG TGATCCTGGC TCTGCGCCCA TCTGACGAGG CTGAALGTATG AA-AAAGACGC TTTCAAGCGG GAACACCTGG AAkAGCTGACT TCCCTACACC TAGTATATCT GACTTTGAAA AGGATAATTT GCTCAACCTC TGGAGGTTTT CCAGAGC-CTC GGAG;AAIGAAT TAAATGCCAT CAACACAACA GTTTCCCAAG GCTC-TTAGCA GCAAACTGG7A TTTCATATATG ACAACCAACC AAGTATGGAC ATTTAAGAGT GAATCAGACC TTCAACTGGA TTTCCTGATA ACGGATTC INFORMATION FOR SEQ ID NO:33: Ci) SEQUENCE CHARACTERISTICS: LENGTH: 1002 base pairs TYPE: nucleic acid STRANDEDNESS: double TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA Ciii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO

TGGTGCTGAC

TCTTTGATAT

GCACATACGA

CTGAAGTGAC

TTCCAACTTC

ACCTCTCCTG

ATCCTGAAAC

ACAGCTTCAT

ATACAACCAA

TATC-ATGTCT

CACTAATAAC

GTGTGTTGTT

GTTATCAGTC

TAATATTAGA

GTTGG:A-AT

TGAGCTCTAT

GTGTCTCATC

GCAAGAGCAT

240 300 360 420 480 540 600 660 720 738 (xi) SEQUENCE DESCRIPTION: SEQ ID NO:33: GAGCTCA-TGG ATCCCCAGTG CACTATGGGA CTGAGTAACA TTCTCTTTGT GATGGCCTTC

C

CTGCTCTCTG

CCATGCCAAT

GACCAGGAAA

CATTCCAAGT

AATCTTCAGA

GGAATGATTC

CCTGAAATAG

TCTATACACG

ACTATCGAGT

GTTTCCATCA

ATTCTGGAAA

CCTCAGCCTC

GTGCTGCTCC

TTGCAAACTC

ACTTGGTTCT

ATATGGGCCG

TCAAGGACAA

GCATCCACCA

TACCAATTTC

GTTACCCAGA

ATGATGGTAT

GCTTGTCTGT

CTGACAAGAC

CCCCAGACCA

TCTGAAGATT

TCAAAACCAA

GAA.TGAGGTA

CACAAGTTTT

GGGCTTGTAT

GATGAATTCT

TAATATAACA

ACCTAAGAAG

TATGCAGAAA

TTCATTCCCT

GCGGCTTTTA

CATTCCTTGG

CAAGCTTATT

AGCCTGAGTG

TACTTAGGCA

GATTCGGACA

CAATC-TATCA

GAACTGTCAG

Ck-A ATGTGT

ATGAGTGTTT

TCTCAAGATA

GATGTTACGA

TCTTCACCTT

ATTACAGCTG

TCAATGAGAC

AGCTLAGTAGT

AAGAGAAATT

GTTGGACCCT

TCCATCACAA.

TGCTTGCTAA

ACATAAATTT

TGCTAAGAAC

ATGTCACAGA

GCAATATGAC

TCTCTATAGA

TACTTCCAAC

TGCAGACCTG

ATTTTGGCAG

TGACAGTGTT

C-AGACTTCAC

A7AAGCCCACA

CTTCAGTCAA

GACCTGCTCA.

CAAGAATTCA

ACTGTACGAC

CATCTTCTGT

GCTTGAGGAC

AGTTATTATA

120 180 240 300 360 420 480 540 600 660 720 780 135

TGTGTGATGG

TCTTATAA.AT

GATAAAAATCC

ACATCTTCAT

TTTTCTGTCT AATTCTATGG A.AATGGAAGA AGAAGAAGCG GCCTCGCAAc GTGGAACCAA CACAATGGAG AGGGAAGAGA GTGAACAGAC CAAGAAP-AAGA ATATACCTGA AAGATCTGAT GAAGCCCAGC GTGTTTTTAAk AAGTTCGALAG GCGACA.A',AG TGATACATGT TTTTA.AGGGC CC 840 900 960 1002 INFORMATION FOR SEQ ID NO:34: SEQUENCE CH-ARACTERISTICS: LENGTH: 751 base pDairs TYPE: nucleic acid STRANDEDNESS: double TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA (iii) HYPOTHETICAL: NO (iv) AN'TI-SENSE: NO (xi) SEQUENCE DESCRIPTION: SEQ ID NO:34: *5

GAGCTCATGG

CTGCTCTCTG

CCATGCCAAT

GACCAGGAAA

CATTCCAAGT

AA.TCTTCAGA

GGAAkTGATTC!

CCTGMA.ATAG

TCTATACACG

ACTATCGAGT

CTTTCCATCA

ATTCTGGAAA

CCTCAGCCTC

ATCCCCAGTG

GTGCTGCTCC

TTGCAAACTC

ACTTGGTTCT

ATATGGGCCG

TCAAGGACAA

GCATCCACCA

TACCAATTTC

GTTACCCAGA

ATGATGGTAT

GCTTGTCTGT

CTGACAAGAC

CCCCAGACCA

CACTATGGGA

TCTGAAGATT

TCAA.AACCAA

GAATGAGGTA

CACA.AGTTTT

GGGCTTGTAT

GATGAATTCT

TAA.TATAACA

ACCTAAGAAG

TATGCAGAAA

TTCATTCCCT

GCGGCTTTT.A

CATTGGGGCC

CTGAGTAACA

CAAGCTTATT

AGCCTGAGTG

TACTTAGGCA

GATTCGGACA

CAATGTATCA

GAACTGTCAG

GAAAATGTGT

ATGAGTGTTT

TCTCAAGATA

GATGTTACGA

TCTTCACCTT

TTC!TCTTTGT

TCAATGAGAC

AGCTAGTAGT

AAGAGAAATT

GTTGGACCCT

TCCATCACAA

TGCTTGCTAA

AC.ATAAATTT

TGCTAAGAAC

ATGTCACAGA

GCAATATGAC

TCT.?CTATAGA

GATGGCCTTC

TGCAGACCTG

ATTTTGGCAG

TGACAGTGTT

GAGACTTCAC

AAAGCCCACA

CTTC.AGTCAA.

GACCTGCTCA

CAAGAATTCA

ACTGTACGAC

CATCTTCTC-T

GCTTGAGGAC

120 180 240 300 360 420 480 540 600 660 720 751 INFORMATION FOR SEQ ID SEQUENCE CHARACTERISTICS: LENGTH: 1611 base pairs TYPE: nucleic acid STRA NDEDNESS: double TOPOLOGY: linear (ii) MOLECUE TYPE: cDNA 136 (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO (xi) SEQUENCE DESCRIPTION: SEQ ID *5

GAATTCATGG

GCTCTGTTCC

CCCCGGGGAG

GGCATAGAGA

TATGAACTGA

GGGCAGTCAA

GCACCCCTCC

GGTGGGGCAC

CGGGAGCCAG

CACCATGGAG

CTGTTTGAGA

CCACAACTTG

CTGGACGGGC

TTGAACCCC;

ACCGCAGAGC

CAGGAGACAC

AAGCCAGAG(

AAGGTGACG(

AAGGCCACC,

GGCCAGCTT

CAGGACCTGG C GCTCCGTGCT G CCCCGTTGCC

TI

GCAATIGTGCA CAGCTAAAAC C CCTCTTGGCA

C

CCCGGGCCAA

CTGTGGGGGA

CCAATTTCTC

ACACCTCGGC

TCAGCCCCCG

TGTTCCCAGT

SCAGTCACCTA

ACGAGGGCAC

TGCAGACAGT

3 TCTCAGAAGG 2 TGAATGGGGT C CAGAGGACAA A TAcACAAGAA AATGCCCAG ACATCTGTGT ,GTGACATGC

AGCACCTCCT

~AAAAAGGAG

LGAAGATAGC

TTCCTCACC

CCAGTGGGC

:CTCACCC-TG

GCCCGCTC-AG

GTGCCGCACT

CCCCTACCAG

GGTCCTAGAG

CTCGGAGGCC

TGGC.AACGAC

CCAGCGGCTC

GACCATCTAC

GACCGAGGT(

TCCAGCCCA(

CGGGCGCAGi

CCAGACCCG

AAACTGGAC

ATTGCCCGA

GACTGTCAC

GGTCACCCC

TGTGGTAGC

CCAGCGGA;

GAACACACJ

TTGCTCCTGC

CAACCAATGT

GTGTACTGGA

AAGAACCTTA

GTGCTGCTCC

GTCACGACCA

GAACTGGACC

CTCCAGACCI

GTGGAC.ACGC

CAGGTCCACC

TCCTTCTCGC

;ACGTGTGCAC

AGCTTTCCG(

3 ACAGTGAAG' CTCCCAGCAG CCCCCGGCCC GCGCTGCCCG

CACTCCTGGT

CCCCCTCAA

GTGACCAGCC

CTGGGAACAA

GCTATTCAAA

CTCCAGAACG

CCCTACGCTG

GTGGGGAGAA

CGGTGCTGGT

TGCGGCCCCA

TTGTCCTGCC

AGGGGACCG'I

TGGCACTGGG

CCAAGGCCTC

TAATACTGGC

3CGCCCAACG'.

r GTGAGGCCC

CCTGCTCGGG

AGTCATCCTG

C-ILCGTTGT TG

CCGGAGGTG

CTGCCCTGAT

GGTGC-CTG

CCAGGTGGAG

GGAGCTGAAA

GAGGAGAGAT

AGGGCTGGAG

AGCGACTCCC

GGTCTGTTCC

GGACCAGAGG

-AGTCAGTGTG

GAACCAGAGC

C GATTCTGACG

SCCCTAGAGCC

120 180 240 300 360 420 480 540 600 660 720 780 840 900 960 1020 1080 1140 1200 1260 1320 1380 1440 1500 1560 1611 .555

S

S.

S

CCACTGGGCC CGAGGGCCCA GCTCCTGCTG GACGAGAGGG

ATTGTCCGGG

TGCCAGGCTT

CTGCCCATCG

GCCAGGAGCA

TATGAGATTG

AGCACGTACC

AAAGGGACCC

GGGGGAACCC

GGGAATCAGT

CTCAAGGGGA

TCATCATCAC

TCTATAACCG

CCATGAAACC

TTCTCCTGCT

GAGCTTCGTG

TGGCCAGAAA

CTCAAGTGTC

CGAGATCTTG

GAGGTGACCG

GCCGCAGTCA

ATCAAGAAAT

GCCACGCCTC

CTGCAACCCT GGAGGTGGCC

TCCTGTATGG

ATTCCCAGCA

TAAAGGATGG

AGGGCACCTA

TGAATGTGCT

TAAkTGGGCAC

ACAGACTACA

CCTGAGGATC

CCCCCGACTG

GACTCCAATG

CACTTTCCCA

CCTCTGTCGG

CTCCCCCCGG

TGCAGGCCTC

ACAGGCCCAA

C

137 INFORMATION FOR SEQ ID NO:36: SEQUENCE CHAPACTERISTICS: LENGTH: 1452 base pairs TYPE: nucleic acid STRANDEDNESS: double TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO (xi) SEQUENCE DESCRIPTION: SE-;Q ID NO:36: GAATTCATGG CTCCCAGCAG CCCCCGGCCC GCGCTGCCCG CACTCCTGGT CCTGCTCGGG

S.

S

S

SS*~

S S S.

S

GCTCTGTTCC

CCCCGGGGAG

GGCATAGAGA

TATGAACTGA

GGGCAGTCPA

GCACCCCTCC

GGTGGGGCAC

CGGGAGCCAG

CACCATGGAG

CTGTTTGAGA

CCAC!.A CTTG

CTGGACGGGC

TTGAACCCCA

ACCGCAGAGG

CAGGAGACAC

AAGCCAGAGG

AAGGTGACGC

AAGGCCACCC

GGCCAGCTTA

GACGAGAGGG

TGCCAGGCTT

CAGGACCTGG

GCTCCGTGCT

CCCCGTTGCC

GCAATGTGCA

CAGCTAAAAC

CCTCTTGGCA

CCCGGGCCAA

CTGTGGGGGA

CCAATTTCTC

ACACCTCGGC

TCAGCCCCCG

TGTTCCCAGT

CAGTCACCTA

ACGAGGGCAC

TGCAGACAGT

TCTCAGAAGG

TGAATGGGGT

CAGAGGACAA

TACACAAGAA

ATTGTCCGGG

GGGGGAACCC

CAATGCCCAG

GGTGACATGC

TAAAAAGGAG

AGAAGATAGC

CTTCCTCACC

GCCAGTGGGC

CCTCACCGTG

GCCCGCTGAG

GTGCCGCACT

CCCCTACCAG

GGTCCTAGAG

CTCGGAGGCC

TGGCAACGAC

CCAGCGGCTG

GACCATCTAC

GACCGAGGTG

TCCAGCCCAG

CGGGCGCAGC

CCAGACCCGG

AAACTGGACG

ATTGCCCGAG

ACATCTGTGT

AGCACCTCCT

TTGCTCCTGC

CAACCAATGT

GTGTACTGGA

AAGAACCTTA

GTGCTGCTCC

GTCACGACCA

GAACTGGACC

CTCCAGACCT

GTGGACACGC

CAGGTCCACC

TCCTTCTCGG

ACGTGTGCAG

AGCTTTCCGG

ACAGTGAAGT

CCACTGGGCC

TTCTCCTGCT

GAGCTTCGTG

TGGCCAGAAA

CTCAAGTGTC

CCCCCTCAAA

GTGACCAGCC

CTGGGAACAA

GCTATTCAAA

CTCCAGAACG

CCCTACGCTG

GTGGGGAGAA

CGGTGCTGGT

TGCGGCCCCA

TTGTCCTGCC

AGGGGACCGT

TGGCACTGGG

CCAAGGCCTC

TAATACTGGG

CGCCCAACGT

GTGAGGCCCA

CGAGGGCCCA

CTGCAACCCT

TCCTGTATGG,

ATTCCCAGCA

TA-AAGGATGG

AGTCATCCTG,

CAAGTTGTTG

CCGGAAGGTG

CTGCCCTGAT

GGTGGAACTG.

CCAGGTGGAG

GGAGCTGAAA

GAGGAGAGAT

AGGGCTGGAG

AGCGACTCCC

GGTCTGTTCC

GGACCAGAGG

AGTCAGTGTG

GAACCAGAGC

GATTCTGACG

CCCTAGAGCC

GCTCCTGCTG

GGAGC-TGGCC

CCCCCGACTG

GACTCCAATG

CACTTTCCCA

120 180 240 300 360 420 480 540 600 660 720 780 840 900 960 1020 1080 1140 1200 1260 1320 1380 CTGCCCATCG, GGGAATCAGT GACTGTCACT CGAGATCTTG AGGGCACCTA CCTCTGTCGG 138 GCCAGGAGCA CTCAAGGG-A GGTCACCCGC TATGAGGGAT CC INFORMATION FOR SEQ ID NO:37: SEQUENCE CHARACTERISTICS LENGTH: 726 base pa TYPE: nucleic acid STRANDEDNESS: doubl TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO GAGGTGACCG TGAATGTGCT CTCCCCCCGG 1440 1452 (xi) SEQUENCE DESCRIPTION: SEQ ID NO:37: GAGCTCATGG TTGCTGGGAG CGACGCGGGG CGGGCCCTGG GGGTCCTCAG CGTGGTCTGC a a a a.

a a S.

CTGCTGCACT

GGGAATGTAA

CAAAAGGATA

AATAGGGTTT

GATGAAGATG

TATGTGCTTG

GAAGTCCAAT

TGGGATTGTC

AATGATCTTC

TC.AATCATTT

ATACCCATAC

GGATCC

GCTTTGGTTT

CTTTCCATGT

AAGTTGCAGA

ATTTAGACAC

AGTATGAAAT

AGTCTCTTCC

GCATGATACC

CTATGGAGCA

CAC.A-A-AAAT

TGACAACCTG

CATTAGCAGT

CATCAGCTGT

ACCAAGCAAT

ACTGGAAAAT

TGTGTCAGGT

GGAATCGCCA

ATCTCCCACA

AGAGCATTAC

ATGTAAACGT

ACAGTGTACT

TATCCCAAGC

A.ATTACAACA

TTTTCCCAAC

GTGCCTTTAA

TCTGAATTCA

AGCCTCACTA

AATATTACTG

CTAACTTGTG

AACAGCCATC

AACTCAACCA

CTTAGCAATC

AGCGGTCATT

TGTATTGTGC

AAATATATGG

AAGAGGTCCT

GAGCTTTCTC

TCTACAACTT

ATACCATGAA

CATTGACTAA

GAGGACTTAT

GTATATATTT

CATTATTTAA

CA.AGACACAG

TGTATATGAA

TGTTGTG TAT

ATGGA.AAAAA

ATCTTTTAAA

AACATCATCA

GTTCTTTCTT

TGGAAGCATT

AATGTACTCA

TAAGATGGAA

TACAACATCA

ATATGCACTT

TGTTCTTTAA

120 180 240 300 360 420 480 540 600 660 720 INFORMATION FOR SEQ ID NO:38: SEQUENCE CHARACTERISTICS: LENGTH: 657 base pairs TYPE: nucleic acid STRANDEDNESS: double TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA (iii) HYPOTHETICAL: NO (iJv) ANTI-SENSE: NO (xi) SEQUENCE DESCRIPTION: SEQ ID NO:38: 1~39

GAGCTCATGG

CTGCTGCACT

GGG.ZATGTAA

CAAAAGGATA

AATAGGGTTT

GATGAAGATG

TATGTGCTTG

GAAGTCCAAT

TGGGATTGTC

AATGATCTTC

TCAATCATTT

TTGCTGGGAG

GCTTTGGTTT

CTTTCCATGT

AAGTTGCAGA

ATTTAGACAC

AGTATGAAAT

AGTCTCTTCC

GCATGATACC

CTATGGAGCA

CACAAAAAAT

TGACAACCTG

CGACGCGGGG

CATCAGCTGT

ACCAAGCAAT

ACTGGAAAAT

TGTGTCAGC-T

GGAATCGCCA

ATCTCCCACA

AGAGCATTAC

ATGTAAACGT

ACAGTGTACT

TATCCCAAGC

CGGGCCCTGG

TTTTCCCAAC

GTGCCTTTA.A

TCTGAATTCA

AGCCTCACTA

AATATTACTG

CTAACTTGTG

AACAGCCATC

AACTCAACCA.

CTTAGCAATC

AGCGGTCATT

GGGTCCTCAG

AAATATATGG

AAGAGGTCCT

GAGCTTTCTC

TCTACAACTT

ATACCATCA.A

CATTGACTAA

GAGGACTTAT

GTATATATTT

CATTATTTAA

CAAGACACAG

CGTGGTCTGC

TGTTGTGTAT

ATGGAAAA

ATCTTTTAAA

AACATCATCA

GTTCTTTCTT

TGGAAGCATT

AATGTACTCA

TAAGATGGAA

TACAACATCA

AGGATCC

120 180 240 300 360 420 480 540 600 657 9S INFORMATION FOR SEQ ID NO:39: SEQUENCE CHARACTERISTICS: LENGTH: 24 amino acids TYPE: amino acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULhE TYPE: peptide (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO FRAGMENT TYPE: internal (xi) Glu SEQUENCE DESCRIPTION: SEQ ID NO:39: Gin Leu Glu Ser Ile Ile Asn Phe Glu Lys Leu Thr Glu Trp Thr 5 10 Ser Ser Asn Val Met Glu Glu Arg INFORMATION FOR SEQ ID SEQUENCE CHARACTERISTICS: LENGTH: 10 amino acids TYPE: amino acid STR.ANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO 140 *50 4555

OS..

o oooo oooo

S

oooo 5oo

*OO

*o o *Ooo

S

5000

OSL

0605 oe 0

S

FRAGMENT TYPE: internal (xi) SEQUENCE DESCRIPTION: SEQ ID Asp Leu Arg Gly Tyr Val Tyr Gin Gly Leu 1 5 INFORMATION FOR SEQ ID NO:41: SEQUENCE CHARACTERISTICS: LENGTH: 9 amino acids TYPE: amino acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO FRAGMENT TYPE: internal (xi) SEQUENCE DESCRIPTION: SEQ ID NO:41: Phe Arg Ile Gly Cys Arg His Ser Arg 1 INFORMATION FOR SEQ ID NO:42: SEQUENCE CHARACTERISTICS: LENGTH: 9 amino acids TYPE: amino acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO FRAGMENT TYPE: internal (xi) Ile 1 SEQUENCE DESCRIPTION: SEQ.ID NO:42: Leu Lys Glu Pro Val His Gly Val INFORMATION FOR SEQ ID NO:43: SEQUENCE CHARACTERISTICS: LENGTH: 32 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear 141 (ii) MOLECULE TYPE: DNA (genomic) (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO (xi) SEQUENCE DESCRIPTION: SEQ ID NO:43: ATATGGATCC TCACCATCTC AGGGTGAGGG GC 32 INFORMATION FOR SEQ ID NO:44: SEQUENCE CHARACTERISTICS: LENGTH: 10 amino acids TYPE: amino acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO FRAGMENT TYPE: internal (xi) SEQUENCE DESCRIPTION: SEQ ID NO:44: Arg Gly Tyr Val Tyr Gln Gly Leu Lys Ser 1 5 INFORMATION FOR SEQ ID SEQUENCE CHARACTERISTICS: LENGTH: 9 amino acids TYPE: amino acid S* STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO FRAGMENT TYPE: internal (xi) SEQUENCE DESCRIPTION: SEQ ID Phe Ala Pro Gly Asn Tyr Pro Ala Leu 1 INFORMATION FOR SEQ ID NO:46: SEQUENCE CHARACTERISTICS: 142 LENGTH: 8 amino acids TYPE: amino acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO FRAGMENT TYPE: internal (xi) SEQUENCE DESCRIPTION: SEQ ID NO:46: Leu Ser Pro Phe Pro Phe Aso Leu 1 INFORMATION FOR SEQ ID NO:47: SEQUENCE CHARACTERISTICS: LENGTH: 9 amino acids TYPE: amino acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO FRAGMENT TYPE: internal (xi) SEQUENCE DESCRIPTION: SEQ ID NO:47: Gln Leu Ser Pro Phe Pro Phe Asp Leu 1 143 INFORMATION FOR SEQ ID NO:48: SEQUENCE CHARACTERISTICS: LENGTH: 9 amino acids TYPE: amino acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO FRAGMENT TYPE: internal (xi) SEQUENCE DESCRIPTION: SEQ ID NO:48: Ile Leu Lys Glu Pro Val His Gly Val INFORMATION FOR SEQ ID NO:49: SEQUENCE CHARACTERISTICS: LENGTH: 9 amino acids TYPE: amino acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO FRAGMENT TYPE: internal (xi) SEQUENCE DESCRIPTION: SEQ ID NO:49: Tyr Met Asn Gly Thr Met Ser Gln Val 1 INFORMATION FOR SEQ ID SEQUENCE CHARACTERISTICS: LENGTH: 9 amino acids TYPE: amino acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO FRAGMENT TYPE: internal (xi) SEQUENCE DESCRIPTION: SEQ ID 144 Gly Ile Leu Gly Phe Val Phe Thr Leu 1. INFORMATION FOR SEQ ID SEQUENCE CHARACTERISTICS: LENGTi: 10 amino acids TYPE: amino acid CC) STRJANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO FRAGMENT TYPE: internal (xi) SEQUENCE DESCRIPTION: SEQ ID NO:51: Phe Leu Pro Ser Asp Phe Phe Pro Ser Val.

1 5

Claims (31)

1. A synthetic antigen-presenting matrix comprising a non-cellular support and a group of molecules consisting of an Class I MHC molecule, a P-2 microglobulin molecule and an assisting molecule, the group being arranged on the support so that when a peptide is bound to the Class I MHC molecule and the matrix is contacted with a T-cell, the group is capable of activating the T-cell.

2. A matrix according to claim I wherein the Class I MHC molecule is an extracellular portion of an Class I MHC molecule which is capable of binding the peptide.

3. A matrix according to claim 1 or 2 wherein the support is a cell fragment.

4- A matrix according to claim 1 or 2 wherein the support is a liposome, A matrix according to claim 1 or 2 wherein the support is a solid surface.

6. A matrix according to claim 2 wherein the extraceflular portion V of the Class I MHC molecule is linked to an epitope which is capable of reacting with an antibody so that the portion may be linked to the support, when the epitope is reacted with the antibody.

7. A matrix according to claim 2 wherein the extracellular portion of the Class I MHC molecule is linked to (His) 6 which is capable of reacting with nickel so that the portion may be linked to the support, when (His), is reacted with nickel. 146

8. A porous material. matrix according to claim 1 or 2 wherein the support is a r r r I r r r r r r r

9. A matrix according to claim 1 or 2 wherein the assisting molecule is a costimulatory molecule. A matrix according to claim 9 wherein the costimulatory molecule is B7.1 or B7.2.

11. A matrix according to claim 9 wherein the costimulatory molecule is an anti-CD80 antibody, or a portion of an anti-CD80 antibody which is capable of costimulating the T-cell.

12. A matrix according to claim 1 or 2 wherein the assisting molecule is an adhesion molecule.

13. A matrix according to claim 12 wherein the adhesion molecule is ICAM-1, ICAM-2, ICAM-3 or LFA-3.

14. A matrix according to claim 1 or 2 comprising an adhesion molecule and a costimulatory molecule. The matrix of claim 7 wherein the extracellular portion of the MHC molecule is empty.

16. A synthetic antigen-presenting matrix comprising a support and a group of molecules consisting of an extracellular portion of an Class I MHC molecule, a f-2 microglobulin molecule, a B7.1 or 87.2 molecule or combination thereof, and ICAM-1, the group being arranged on the support so that when a peptide is bound to the Class I MHC molecule and the matrix is contacted with a T-cell, the group is capable of activating the T-cell.

17. A method of producing a synthetic T-cell lymphocyte antigen activating matrix comprising inserting genes into a first cell, the genes consisting of: a first gene which encodes a Class I MHC molecule, 147 or an extracellular portion of a Class I MHC molecule, the gene being operably linked to a promoter; (ii) a second gene which encodes a f-2 microglobulin molecule, the gene being operably linked to a promoter; maintaining the first cell in conditions in which the Class I MHC molecule, or extracellular portion of the Class I MHC molecule, and g-2 microglobulin molecule, are expressed in the cell by the first and second genes; harvesting a Class I MHC molecule, or an extracellular portion of a Class I MHC molecule, and a 0-2 microglobulin molecule which are expressed in the cell by the first and second genes; inserting a gene into -a second cell which encodes an assisting molecule, or an extracellular portion of an assisting molecule; maintaining the second cell in conditions in which the gene encoding the assisting molecule or extracellular portion of an assisting molecule is expressed; harvesting an assisting molecule or extracellular portion of assisting molecule expressed by the gene in the second cell; and arranging the harvested molecules of step and on a support so that the harvested molecules are capable of activating a T-cell when the support contacts the T-cell.

18. .A method according to claim 17 wherein the first and second e cells are identical.

19. A method according to claim 17 further comprising inserting a gene into a third cell which encodes a further assisting molecule, or an extracellular portion of a further assisting molecule; maintaining the third cell in conditions in which the gene 148 encoding the further assisting molecule or extracellular portion of the further assisting molecule is expressed; harvesting the further assisting molecule or extracellular portion of the further assisting molecule expressed by the gene in the third cell; and arranging the harvested molecules of steps and on a support so that the harvested molecules are capable of activating a T-cell when the support contacts the T-cell. A method according to claim 17 wherein the assisting molecule is a costimulatory molecule.

21. A method according to claim 20 wherein the costimulatory molecule is B7.1 or B7.2.

22. A method according to claim 17 wherein the assisting molecule is an adhesion molecule.

23. A method according to claim 22 wherein the adhesion molecule :is ICAM-1, ICAM-2, ICAM-3 or LFA-3.

24. A method according to claim 17 wherein the peptide is bound to .the Class I MHC molecule, or extracellular portion of the Class I MHC molecule, before the Class I MHC molecule, or extracellular portion of the .Class I MHC molecule, is arranged on the support. A method according to claim 17 wherein the peptide is bound to the Class I MHC molecule, or extracellular portion of the Class I MHC molecule, after the Class I MHC molecule, or extracellular portion of the Class I MHC molecule, is arranged on the support.

26. A method for activating a T-cell comprising contacting a T-cell with a matrix according to any one of claims 1 to 16 so that the T-cell is activated by the group of molecules arranged on the support of the matrix. 149

27. A method for generating a cytotoxic T-cell comprising contacting a T-cell with a matrix according to any one of claims 1 to 16 so that the T-cell is made cytotoxic by the group of molecules arranged on the support of the matrix.

28. A method according to claim 27 wherein the support is a cell,

29. A method according to claim 27 wherein the costimulatory molecule is B7.1 or B7.2. A method according to claim 27 wherein the costimulatory molecule is at least a portion of an anti-CD28 antibody which is capable of costimulating the T-cell clone.

31. A method according to claim 27 wherein the adhesion molecule is ICAM-1, ICAM-2, ICAM-3 or LFA-3.

32. A method of treating a tumor in a patient comprising constructing a T-cell isolated from the patient with a '*matrix according to any one of claims 1 to 16 so that the T-cell is activated by the group of molecules arranged on the support of the matrix; and administering the activated T-cell to the patient. S33. A method according to claim 32 wherein the Class I MHC molecule, or the extracellular portion of the Class I MHC molecule which is arranged on the support of the matrix is bound to a peptide derived from the tumor cell.

34. A method according to claim 33 wherein the peptide is a self antigen of the patient. A method according to claim 34 wherein the assisting molecule is B7.1 or B7.2 and the adhesion molecule is ICAM-1, ICAM-2, ICAM-3 or LFA-3. 150

36. A method according to claim 33 wherein the assisting molecule is B7.1 and the adhesion molecule is ICAM-1.

37. A matrix according to claim 1 or claim 16, the matrix being substantially as described herein with references to the accompanying Examples.

38. A synthetic antigen-presenting matrix substantially as hereinbefore described with reference to the accompanying Examples.

39. A method of producing a synthetic T-cell lymphocyte antigen activating matrix substantially as hereinbefore described with reference to the accompanying Examples. Dated this 2 2 n d day of July 2003 THE SCRIPPS RESEARCH INSTITUTE By their Patent Attorneys GRIFFITH HACK *oo e