WO1995009642A1

WO1995009642A1 - Method for provoking immunity by peptides labelled with a photoactivatable group which binds to mhc molecules

Info

Publication number: WO1995009642A1
Application number: PCT/US1994/010897
Authority: WO
Inventors: Pedro Romero; Jean-Charles Cerottini; Immanuel Leuscher
Original assignee: Ludwig Institute For Cancer Research
Priority date: 1993-10-05
Filing date: 1994-09-27
Publication date: 1995-04-13
Also published as: AU7958994A

Abstract

The invention involves the cross-linking of presented peptides to their corresponding MHC molecule or a soluble MHC derivative which has similar antigen binding properties, to form covalent complexes of the two molecules. The resulting materials are useful in provoking an immune response as well as in the identification of cytolytic T lymphocytes specific for the peptide of interest. The peptides are cross-linked to their partner molecule via a photoactivatable material which is attached to the peptide. Upon exposure to light at an appropriate frequency, the peptide cross-links to the MHC molecule.

Description

METHOD FOR PROVOKING IMMUNITY BY PEPTIDES LABELLED WITH PHOTOACTIVATABLE GROUP WHICH BINDS TO MHC MOLECULES

RELATED APPLICATION This application is a continuation-in-part of Serial

Number 08/133,407, filed on October 5, 1993, and incorporated by reference in its entirety.

FIELD OF THE INVENTION This invention relates to methodologies for provoking an immune response, such as the production of cytolytic T lymphocytes ("CTLs" hereafter) directed against specific covalent complexes of presenting major histocompatibility complex (MHC) molecule or MHC derivative and peptide. The methodology is applicable in the context of both .in vitro and in vivo methodologies. The invention also relates to a diagnostic method for determining whether CTLs of a particular, restricted type are present in a sample, such as a subject's blood or serum. This methodology acts as an aid in diagnosis. In the context of improving CTL production, it is well known that the CTLs are useful as both diagnostic and therapeutic reagents, as well as raw materials for, e.g., generation of antibodies, as sources of DNA, proteins, and so forth. Thus, the invention has applicability in a number of specific contexts. The complexes referred to supra may be cell bound, or soluble. BACKGROUND AND PRIOR ART

The T lymphocyte population of a mammal is divided into several different types. Of these, the so-called "CD8⁺" CTLs are well known as playing an important role in cell mediated immune responses against viruses, bacteria, parasites, and transformed cells. Exemplary of the vast literature on this point are Townsend et al. Prog. Allergy 36:10 (1985) ; Palmer et al., Nature 353: 852 (1991) ; Romero et al. , Nature 341: 332 (1989) ; Van der Bruggen et al. Curr. Opinion. Immunol. 4: 608 (1992) .

The mechanism by which the subpopulation of these cells, the cytolytic T lymphocytes or "CTLs" act is known in some detail, although it would be incorrect to say that it is understood completely. In general, antigenic peptides are generated by intracellular degradation of endogenously produced proteins. The peptides bind to nascent major histocompatibility complex molecules of type I ("MHC-I") in the endoplasmic reticulum. This binding promotes assembly and transport to the cell surface. Townsend et al. , supra, summarize this quite well . The binding of the antigenic peptides to MHC molecules is restricted, meaning that certain peptides bind only to certain MHC molecules. Once the complexes of antigenic peptide and MHC molecule is present on a cell surface, the complex is "recognized" by CTLs specific for it. Binding between the CTL, via the T cell receptor or "TCR", and the antigenic presenting cell, or "APC" , results in a cascade of events which includes the proliferation of the CTLs specific for the complex without across the board expansion of the CTL population. Observations have revealed that while the majority of assembled MHC-I molecules are occupied by endogenous peptides, there is a small proportion which have empty binding sites, and are therefore available for binding exogenous peptides. The antigenic peptides which bind to the MHC generally are restricted in length to 8-10 amino acids, and have MHC allele specific binding motifs, in this regard see Romero et al . , J. Exp. Med. 174: 603 (1991); Rammensee et al. , Curr. Opinion Immunol. 5:35 (1993) . Endogenously produced peptides have been eluted from their partner MHC-I molecules, and identified by sequence analysis. Exemplary of this work are the papers by Rδtvschke, Nature 348: 252 (1990) and Van Bleek et al. , Nature 348: 231 (1990) . In experiments involving systematic truncation of synthetic peptides in the context of CTL recognition, optimal length CTL epitopes have been identified. See Romero et al . , supra.

The work described supra has led to identification of a rapidly growing number of CTL epitopes, and to searches for efficient procedures to induce antigen specific CTL responses. This work, performed in vivo, used synthetic CTL epitopes as immunogens. Parenteral administration of antigenic peptides in aqueous solution generally failed to elicit an in vivo CTL response, as reported by Deres et al . , Nature 342: 561 (1989) ; Kast et al. Proc. Natl. Acad. Sci. USA 88: 2283 (1991) ; Rock et al . , J. Immunol. 150: 1244 (1993) . It has also been found that parental administration of conjugates of antigenic peptide and lipids does result in efficient CTL response induction, as per Deres et al. , supra; Romero et al. , J. Immunol. 148: 1871 (1992) Martinon et al., J. Immunol 149: 3416 (1992) . Subcutaneous administration of peptides emulsified in Freund's incomplete adjuvant has also succeeded in generating antigen specific CTLs (Aichele et al. , J. Exp. Med. 171: 1815 (1990) ; Kast et al. , supra; Fayolle et al. , J. Immunol. 147: 4069 (1991) ; Gao et al. , J. Immunol. 147: 3268 (1991) ; Romero et al. supra, Zhou et al. , J. Immunol. Meth 153: 1992)) . Immunization with adjuvants such as Freund's incomplete adjuvant, however, cannot be used in connection with human vaccination.

The foregoing discussion points to the desirability, and need for effective and useful methodologies for producing antigen specific CTLs. It would be useful to have a methodology available which can be employed in both an in vivo and in vitro context to generate desired CTLs. (It must be understood that the process by which CTLs are generated in vitro is the same as that in vivo, the sole difference being the contact of the APCs to a CTL containing sample rather than via in vivo administration to a host subject) . Previous work by Luescher et al. , J. Immunol. 148: 1003

(1992), and Nature 351: 72 (1991) both of which are incorporated by reference in their entirety, involved photoaffinity labelling of MHC molecules. Specifically, a

Plasmodium berαhei circumsporozoite peptide i.e. "PbCS 253- 260", which has amino acid sequence

Tyr lie Pro Ser Ala Glu Lys lie (SEQ ID NO: 1) when labelled with the photoreactive molecule iodo, 4- azidosalicylic acid ("IASA") , selectively and efficiently labelled H-2K^d molecules, to which the peptide is restricted. In contrast, when the peptide was labelled with either of IASA, or 4-azidobenzoic acid ("ABA"), coupled at

it did not photoaffinity label the MHC molecule, but did label TCRs on cloned CTLs specific for the PbCS/H-2K^d complex. See Romero et al., J. Immunol. 150: 3825 (1993); Romero et al. , J. Exp. Med. 177: 1247 (1993), the disclosures of both being incorporated by reference herein. These papers demonstrates that particularly efficient labelling of T cell receptors found on CTLs which recognize the peptide of interest was accomplished when the TCR was labelled with both. When the IASA group was photoactivated by ultraviolet radiation at 350- 410 nm, covalent attachment to the H-2K^d molecules resulted.

It has now been found that exogenous peptides, labelled with a photoactivatable group, bind much more efficiently and strongly to their restricted MHC binding partner following photoactivation. The resulting complexes are efficient provokers of CTL responses. In addition, it has been found that complexes of exogenous peptide and soluble MHC derivatives which are the functional equivalent of cell bound MHC molecules also provoke the CTL response. Thus, the invention involves methodologies for provoking a CTL response to a specific complex of peptide and CTL, be it in vivo or in vitro, using the labelled peptides described herein. The invention also involves a methodology by which one can determine the presence of CTLs specific for particular peptide/MHC complexes in a sample, such as a biological fluid, using the labelled APCs, as described herein.

These, and other aspects of the invention, are explained in greater detail in the Detailed Description of Preferred Embodiments which follows. BRIEF DESCRIPTION OF THE FIGURES Figure 1 shows the influence of temperature and S2 microglobulin on the binding of IASA to H-2K^d molecules presented on concanavalin A activated spleen cells. The temperatures were - either 26°C (the circles) , or 37°C (triangles) . Filled in symbols represent work where S2 microglobulin was used, and empty symbols where it was not. The 100% standard is set with reference to binding at 26°C, in the presence of S2 microglobulin, after six hours of incubation.

Figures 2A and 2B show the specificity of photoaffinity labelling. Specifically, in figure 2A, immunoprecipitation experiments are depicted where monoclonal antibodies specific for H-2L^d, H-2D^d, or H-2K^d were added to precipitate any bound, labelled material . The first lane shows results obtained with a cell lysate. In lanes 5-8 of figure 2B, the same experiments were carried out, but this time using different peptides, as shown, at a 300 fold excess. Figure 3 shows how dissociation of the photoaffinity labelled peptide is prevented.

Figure 4 presents data showing how, in vivo, the photolabelled APCs induced strong CTL responses. Figure 4A shows results of cross linking, while figure 4B shows results following pulsing with no cross linking.

Figure 5 demonstrates the specificity of the induced CTLs. The data involve the ability of the CTLs to lyse cells which have been treated to present targeted peptide.

Figure 6A presents data showing the binding of labelled photoactivatable peptide to cytolytic T cell clone N9. The peptide is presented to the T cell in the form of cell bound ligand, and is photoactivated. The cells are lysed and immunoprecipitated, with the immunoprecipitates being subjected to densitometric analysis thereafter. Figure 6B presents results from a study of the dissociation of the peptide from cells. Presenting cells of cell line P815 were preincubated with the peptide, and then cultured in fresh medium, to determine the rate of dissociation. Figures 7A and 7B depicts results from antigen recognition experiments, where a chromium release assay was carried out. Figure 8A sets forth SDS-PAGE analysis of experiments involving soluble complexes of the K^dQ10 molecule and the labelled peptide discussed herein.

Figure 8B shows the results obtained in competition experiments where the soluble complex was tested in the presence of differing amounts of K^dQ10, or of a monoclonal antibody.

Figure 9A shows the results of a study designed to determined the kinetics of binding between the peptide and the T cell receptor on cytolytic T cell clone N9.

Figure 9B presents results of experiments where the kinetics of dissociation were studied.

Figure 10 analyzes the binding of cell associated peptide to N9 T cell receptors. Figures 11A and 11B, respectively, present data obtained from competition experiments involving either cell bound ligand (11A) , or soluble complexes (11B) . The experiments were designed to study the mechanism of binding between receptor and ligand. DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS Example 1

As a preliminary matter, the requisite reagents were made. All non-peptide starting materials were purchased from Bachem Finechemical and from Sigma Chemie. Any peptide or N- terminal conjugate described herein was made on a manual RaMPS peptide synthesizer, using Fmoc for transient N-terminal protection, following manufacturer's instructions. Deprotected peptides and N-terminal conjugates were purified by C-18 reverse phase HPLC, using a Waters 600E system on line with a 1000S diode array UV detector, on a semipreparative C-

18 column (1x25 cm, 5μm) . The radioactive derivatives described herein were made on an analytical C-18 Vydac column

(4.6x250 mm, 5μm) . HPLC columns were eluted via linear gradient of acetonitrile on 0.1% trifluoroacetic acid in water, rising from 0% to 75% in one hour. Amino acid composition of peptides and conjugates was assessed according to the well known DABSYL method, and were 100% in agreement with expected ratios.

The photoreactive derivatives were made under dimmed light in accordance with Luescher et al. , J. Immunol. 148: 1003 (1992) : romero et al. , J. Immunol. 150: 3825 (1993) , and Romero et al . , J. Exp. Med. 177: 1247 (1993), all of which are incorporated by reference in their entirety. For completeness, however, the synthesis is discussed herein.

As a start, the peptide of SEQ ID NO: 1, protected at the N-terminus by Fmoc, was reacted with N-hydroxysuccimidyl- 4-azido benzoic acid ("ABA-ONSu") . The reaction took place in the presence of 1 -hydroxybenzotriazole and diisopropylethylamine. The resulting product was N- deprotected and precipitated with ethyl acetate. It eluted from HPLC at 37.5% acetonitrile, and demonstrated UV absorption maxima at 270 nm. This material, referred to as "YIPSAEK(ABA) I" , was prepared by reacting it in anhydrous dimethylsulfoxide with N-hydroxysuccinimidyl iodo,4- azidosalicylate (IASA-ONsu) , in the presence of 1- hydroxybenzotriazole and diisopropylethylamine. Product eluted at 56.2% acetonitrile, and displayed UV adsorption maxima at 325 nm and 272 nm.

Purified material was reconstituted in PBS at 1 mM concentration, filtered and stored in the dark at -20°C.

In order to make radiolabelled peptides, radioiodinated ¹²⁵IASA-ONSu YIPSAEK(ABA) I was prepared, following Luescher et al., 1992, supra. HPLC purified and lyophilized material was reconstituted in PBS at 2-4x10° cpm/ml. Specific radioactivity was measured as approximately 2000 Ci/mMol. Any radioactive peptide was stored in the dark at 4°C, and was used within four days of synthesis. Example 2

The peptides prepared in Example 1 were then used to label cells. Normal BALB/c spleen cells (6xl0⁶ cells/1) , were cultured for two days in DMEM supplemented with 5% FCS, lOμM HEPES and 2.5 ug/ml of concanavalin A ("ConA") , at 37°C and 5% C0₂. The cells were washed, twice, in DMEM supplemented with 0.5% FCS and 10 mM HEPES, and resuspended at 3xl0⁶ viable cells/ml in the medium.

Samples of cells (3xl0⁶) were incubated with the radiolabelled peptide of Example 1 (2xl0⁷ cpm in 50 ml PBS) , either with or without E2 microglobulin (2.5 ug/ml) . The incubation was in 6 well plates, and at either 26°C or 37°C. Incubation was over a period of time ranging from 0 to 6 hours. Measurements were taken at one hour intervals.

Following incubation, cells were UV irradiated at room temperature for 30 seconds, using a 3000W "SUPUVASUN" irradiator, emitting between 350-410 nm, at a distance of 1 meter. The IASA group is "photoactivated" at this wavelength.

Labelled cells were washed, once, and solubilized with

NP-40 (0.7%) detergent, in the presence of leupeptin (10 ug/ml) , PMSF (0.5 mM) , and iodoacetamide (10 mM) . The H-2K^d molecules were immunoprecipitated with either of monoclonal antibody SF1-1111 or 34-1-2S, both of which are specific for this MHC molecule. To test for specificity of binding of the peptide, H-2L^d and H-2D^d were also immunoprecipitated, using specific mAbs 28-14-8S and 34-4-20S, both of which were obtained from the American Type Culture Collection.

Immunoprecipitates were analyzed on 10% linear SDS-PAGE under reducing conditions, in accordance with the Luescher papers, cited supra and incorporated by reference. Dried gels were exposed at -80°C to XAR 5 films in cassettes containing amplifying screens. Quantitation was determined via densitometry and peak integration was carried out, in both cases using standard methodologies.

The results of these experiments are shown in figures 1, 2A and 2B. Figure 1 shows that more efficient binding took place at 26°C, although binding at 37°C was still good. Binding increased considerably, at both temperatures, when β2 microglobulin was present. Best binding took place after 4-6 hours of incubation. As such, the preferred protocol for preparing the materials was one where the incubation was at 26°C in the presence of human S2 microglobulin, for 4-6 hours. Unless otherwise indicated, this is the standard labelling protocol used hereafter. The analysis of the immunoprecipitates, as presented in figures 2A and 2B, show one major labelled band of 45 KDa ⁽see, e.g., figure 2A, lane 1) . The results with the MHC specific mAbs confirm that it must be H-2K^d. In additional experiments only summarized here (figure 2B) , the inhibition of photoaffinity labeling in the presence of excess unlabelled peptide (lane 5) , and H-2K^d restricted peptide P198^_14-22 (lane 7) , but not in the presence of other peptides restricted to different MHC molecules, demonstrate the selective photoaffinity labelling of H-2K^d. These data in fact confirm the work shown by Luescher et al. , supra. Example 3

Labelled cells were studied to determine relative dissociation, and the results are shown in figure 3. The dissociation of conjugate IASA-YIPSAEK(ABA) I from H-

2K^d molecules was rapid at 37°C, 50% dissociation being reached after about 45 minutes. Slightly lower dissociation rates were found for IASA-YIPSAEK(biotin) I, as per Luescher et al. , J. Immunol. 148: 1003 (1992) , and the difference is probably attributable to lower binding affinity for the ABA derivative.

When the IASA derivatives were photocrosslinked to the H-

2K^d molecule, using the ultraviolet irradiation discussed supra, rapid dissociation was prevented. In fact, as figure

3 shows, when SF1.111 was used to immunoprecipitate H-2K^d, nearly 95% of the photolabelled molecules were precipitated after 4 hours of incubation at 37°C. Although not shown herein, the same results were obtained with mAb 34-1-2S, which is conformation dependent, and only precipitates H-2K^d when it is associated with β2 microglobulin (Godeau et al. , Int. Immunol. 4: 265 (1992) . These results indicate that photocrosslinked, H-2K^d/peptide derivative complexes retain native conformation. Example 4

The photocrosslinked material (i.e., spleen blasts, photocrosslinked with the IASA labelled peptide discussed supra) , were then used in an .in vivo experiment.

Con A activated BALB/c spleen cells (5-6x10° cells/ml) , were incubated with 2 μM of the conjugate, in DMEM, supplemented with 0.5% FCS, 10 mM HEPES, and 2.5 ug/ml of human S2 microglobulin, at 26°C for 4-6 hours. Cells were washed, briefly, at 4°C, and then resuspended in 12 ml of ice cold DMEM containing FCS. UV irradiation was carried out in accordance with Example 2. The cells were washed, once, adjusted to 50xl0⁶ cells/ml, in DMEM, and 0.5 ml samples of suspension were injected, interperitoneally, into 8-10 week old (BALB/c x C57BL/6)F_!, mice. Control mice were injected with similar spleen cells which were peptide pulsed twice, but not irradiated.

Anywhere from 9-14 days after injection, peritoneal exudate lymphocytes (PELs) were harvested by washing the peritoneal cavities of the mice with two 5 ml volumes of DMEM, supplemented with 5% FCS. PELs (3-4xl0⁶ cells) were cultured with 7-irradiated, peptide derivative pulsed P815 cells (lxlO⁵) in 2 ml of DMEM, supplemented with 5% FCS, 10 mM HEPES, 5xl0'⁵m 2-mercaptoethanol, and EL-4 supernatant containing 30 U/ml of murine IL-2 in 12 well plates, following Romero et al. , J. Immunol. 148: 1871 (1992), incorporated by reference herein. After 7-8 days, viable cells were restimulated by coculture with irradiated, peptide pulsed P815 (H-2K^d) cells (2xl0⁵) , in 2 ml of the same medium, in 12 well plates. Cells (2-3xl0⁵) were restimulated in the same fashion, every week. Seven days after harvest, cytolytic activity of the cells was determined in a ⁵¹Cr release assay, as described by Romero et al . , J. Immunol. 148: 1871 (1992), the disclosure of which is incorporated by reference, but is described herein. Effector cells were serially diluted in V-bottom microtiter plates to obtain Effector/T cell ratios ranging from 100/1 to 0.001/1. Target cells (lxlO⁶ P815 cells, clone 444/A1.1) were labeled with lOOμ Ci of Na₂ ⁵¹Cr0₄ (NEN) and 100 ml of Dulbecco's Tris buffer, supplemented with 2 μg/ml of BSA, and 2 μm of peptide. After one hour of incubation at 37°C, cells were washed, three times, and adjusted to lxlO⁴ cells/ml. Labeled target cells (1000 cells/well) were added to the effector cells in 10 ul/well aliquots. After incubation for 4 hours at 37°C, in a 5% C0₂ atmosphere, ⁵¹Cr content of 100 ul aliquots of supernatant was determined. The formula for the calculation is:

[ER-SR] 100 x

[MR-SR] where ER is experiment release, SR is spontaneous release, and MR is maximum release. "SR" is the release obtained from the cells when CTLs are not present, and "MR" refers to the release obtained from control cells treated with 0.5M HC1. Experimental values were calculated in duplicate, and controls in quadruplicate. The relative number of cytolytic cells generated was quantitated in terms of lytic units (LU) with 1LU being defined as the number of CTLs required to obtain 50% lysis of a given target.

The results of these experiments are presented in figures 4A and 4B. Figure 4A show that bulk cultures of cells, derived from mice immunized with the photocrosslinked cells, did efficiently lyse cells (P815) which had been sensitized with the peptide, i.e. "IASA-YIPSAEK (ABA) I". Cultures taken from mice which had been immunized with pulsed cells, however, showed much lower levels of lytic activity, as will be seen in figure 4B.

As explained, supra, the number of CTLs present in the PEL bulk cultures was expressed as LU' s per 10⁶ cells. Table 1, which follows summarizes results. The mice immunized with the photocrosslinked blasts had much higher LU cells as compared to pulsed cells - levels that are 14-720 LUs higher.

Table 1

ACTIVITY INDUCED AFTER IMMUNIZATION WITH PEPTIDE-PULSED

OR PEPTIDE-CROSSLINKED Con A BLASTS

Exp. N Conditions LTJ/10⁶Cells

Peptide-Specific

Mouse Control Peptide-pulsed LU/Culture N Target Target

Pulsed 1 <10 <10 <10

Pulsed 2 <10 <10 <10

Crosslinked 3 <10 200 233

I

Crosslinked 4 <10 370 952

Crosslinked 5 <10 2500 7250

Crosslinked 6 <10 500 1350

Pulsed 7 <10 100 190

Pulsed 8 10 133 295

Pulsed 9 33 91 249

II Crosslinked 10 37 500 3334

Crosslinked 11 25 2083 15640

Crosslinked 12 10 1111 4404

The cytolytic assays, discussed supra, are presented in figure 5. The results demonstrate that when peptides are covalently complexed to their target MHC molecules, via a photoactivatable group, a strong CTL response is provoked. Example 5

The labelled peptide derivatives, described supra, were used in further experiments to label either cell associated HLA-K molecules, or the soluble HLA-K^d derivative K^dQ10. This derivative consists of the first, second and third domain of

K^d, and C terminal sequence of Q10^b, in place of transmembrane and cytoplasmic domains. This molecule is functionally equivalent to the HLA-K^d molecule. In particular, the soluble K^dQ10 molecule is indistinguishable from HLA-K^d with respect to antigenic binding properties. Background for this type of molecule may be found in, e.g., Luescher et al . , J. Immunol. 148: 1003-1011 (1991) , which is incorporated by reference in its entirety. In the cell associated ligand experiments, the cell line used was P815, a cell line known to express HLA-K^d molecules on its surface. To photoaffinity label the K^dQ10 molecules, 50 ug of the solubilized molecule, in 100 ul PBS, was added to lyophilized ¹²⁵IASA-YIPSAEK(ABA) I, and incubated at 37°C for one hour. Following ultraviolet irradiation (UV irradiator emitting at 350-410 nm, cutoff out 350 nm; lamp-sample distance l , irradiation time 45 seconds) , the incubation mixture was diluted with PBS (300 ul) , and subjected to gel permeation chromatography in PBS on a Spadex protein WS-803 column (0.8x50 cm, separation range 100 to 500 kDa) . The main radioactive material eluted with an apparent mass of about 55 kDa, and was ≥ 95% pure, as assessed by SDS-PAGE. The photoaffinity labelled KQ10 was diluted with PBS to a final concentration of 1.2xl0⁸ cpm/ml.

To label the cells, 5xl0⁶ cells/ml were incubated in Petri dishes (7 ml/dish) with ¹²⁵IASA-YIPSAEK(ABA) I, at l-2xl0⁸ cpm/ml, in the presence of human S2 microglobulin (2.5 ug/ml) , at 26°C, for four hours. The cells were subjected to UV radiation as above, were washed 2x with DMEM (2% FCS) , and then resuspended in the same medium, at 2.7xl0⁷ cells/ml. The amount of cell associated peptide derivative was approximately 3.10⁶ cpm/lxlO⁶ cells. Example 6

Binding and dissociation experiments were carried out using cell line P815, PbCS 253-260 specific cytolytic T cell clone N9, and peptide ¹²⁵IASA-YIPSAEK(ABA) I ("the peptide") . In these experiments, either P815 or N9 cells were incubated with the peptide. 3xl0⁶ cells/ml in DMEM, supplemented with 20 mM HEPES and 0.5% fetal calf serum (DMEM 0.5% FCS) were incubated in 6 well plates (1 ml/well) in a humidified incubator with 5% C0₂, in the presence of human S2 microglobulin. Incubation was at either 37°C or 26°C, as elaborated upon infra. The mixtures of cells and photoactivatable peptides were treated by ultraviolet radiation which emitted at 350-410 nm, with a cutoff at 350 nm. The distance between the lamp and sample was 1 meter, and irradiation time was 45 seconds. The conditions completely photoactivate the IASA group after 20 seconds, with no impact on the ABA group.

Following labelling, the labelled cells were incubated with protease inhibitors leupeptin (10 ug/ml) , PMSF (0.5 mM) , and iodoacetamide (20 nM) , followed by lysis with NP40 detergent (0.7%) . Immunoprecipitation was then carried out using the K^d specific mAb SF 1.111, obtained from the American Type Culture Collection, following Luescher, Electrophoresis 8: 508-514 (1987) . (Immunoprecipitations were performed on nitrocellulose membrane, coupled with affinity purified mAbs) . Any immunoadsorbed proteins were analyzed by SDS-PAGE on 10% acrylamide gels, under reducing or non-reducing conditions, following Romero et al. , J. Immunol. 150: 3825-3831 (1993) , and Romero et al . , J. Exp. Med. 177: 1247-1256 (1993) , both of which are incorporated by reference in their entirety. Dried gels were assessed by densitometry, in accordance with Luescher et al. , J. Immunol. 148: 1003-1011 (1991), incorporated by reference in its entirety.

Representative results are set forth in figures 6A and 6B. In these figures, 6A shows binding, and 6B dissociation. P815 cells are represented by circles, and N9 cells by squares. Filled in symbols are at 37°C, and empty symbols at 26°C.

Over a period of time (0-6 hours) it will be seen that at 37°C, binding to N9 cells increased over the first two hours, and remained constant thereafter. At 26°C, binding increased continuously over the six hours, with binding levels being nearly 10 fold higher. Binding by P815 was more efficient. Levels of binding at 37°C for P815 were similar to those found for N9 at 26°C; however, at 26°C, the level of binding for P815 was 10 fold that of binding at 37°C, and nearly 100 times that on N9 cells at 37°C.

In the dissociation experiments (figure 6B) P815 cells were preincubated with the peptide, washed, and incubated in fresh medium at 37°C or 26°C (• or O) . The dissociation was rapid; for 37°C, half life was approximately 40 minutes, and at 26°C, approximately one hour.

Additional dissociation experiments were also carried out. When the peptides were photocrosslinked, as described in the previous examples, dissociation at 37°C was prevented. After four hours of incubation, over 90% of photoaffinity labelled K^d was identified in i munoprecipitates. The same results were obtained when mAb 34-1-2S was used. This mAb only precipitates K^d molecules associated with S2 microglobulin, indicating that the photocross-linked K^d molecules largely preserved native configuration.

The same type of experiments were carried out, using soluble K^dQ10 molecule. Maximal binding was achieved in one hour, in contrast to the longer periods of time described for the cell associated K^d molecules. Dissociation was similar to that shown in figure 6B. Example 7

Antigen recognition experiments were carried out, using the ⁵¹Cr release assay developed in the examples, supra, and not repeated here. Two cytolytic T cell clones, i.e., N9 and H2, were used, to measure the specific lysis of ⁵¹Cr labeled P815 cells, which had been sensitized in the presence of one of (i) PbCS 253-260, (ii) IASA-YIPSAEK(ABA) I, (iii) biotin- YIPSAEK(IASA) I, (iv) YIPSAEK(IASA) I, and (v) biotin- YIPSAEK(ABA) I. The concentrations used ranged from 10^"14 to 10^"5 M, as figures 7A and 7B indicate.

Using the ⁵¹Cr release assay, it had been found that when IASA is replaced by ABA, K^d competitor activity did not change (See Romero et al. , J. Immunol. 150: 3825-3831 (1993)) ; however, the efficiency of recognition improved approximately 11-fold. Deletion of the N-terminal biotin group reduced efficiency of recognition only 7 fold. This indicates that the group is not essential for recognition of the peptide derivative by this clone. When the experiments were repeated, increase efficiency of recognition of biotin-YIPSAEK(ABA) I varied in the range of 10-50 fold. Reduction in efficiency of recognition of YIPSAEK(IASA) I ranged from 1-7 fold. Substitution of N-terminal biotin of biotin-YIPSAEK(ASA) I with IASA considerably reduced its antigenic activity. When compared with biotin-YIPSAEK(IASA) I, the IASA-YIPSAEK(ABA) I derivative was recognized only 10 times less efficiently. One may explain this, at least in part, by the lower relative K^d competitor activity of IASA-YIPSAEK(ABA) I as compared to biotin-YIPSAEK(IASA) I (2.5 and 0.1, respectively) . In contrast to the N9 clone generated in response to biotin- YIPSAEK(IASA) I, the PbCS reactive, H2 clone recognized PbCS 253-260 well, but did not recognize pbCS derivatives containing a modified Lys residue at position 259, as shown in figure 7B. These results show that residue 259 plays an important role in peptide recognition by T cells, but not in the binding of peptides to the K^d molecule.

In figures 7A and 7B, please note the following key:

O PbCS 253-260

# IASA-YIPSAEK(ABA) I

Δ biotin-YIPSAEK(IASA) I

H YIPSAEK(IASA)I biotin-YIPSAEK(ABA) I .

Example 8

A series of photoaffinity labeling experiments were carried out using the soluble ligand, K^dQ10. Either N9 or another PbCS (253-260) specific CTL, i.e., H2 was used. In one set of experiments, K^dQ10 was crosslinked with the peptide, and then N9 was incubated with the crosslinked complexes. Increasing concentrations of complex (l.lxl0^"10M, 3.3xlO^"10M, and 9.9xlO^"10M) were used. The N9 cells were resuspended in DMEM (2% FCS) (lOxlO⁶ cells/ml) , and placed in 450 ul aliquots in 12 well plates. The indicated amounts of photocrosslinked complexes indicated were added, and incubated at 37°C for three minutes. Ultraviolet radiation was applied (45W mercury fluorescence lamp, emission maximum at 312 nm, bandwidth of approximately 80 nm, distance of 3 cm for 30 seconds at 4°C) . Complete photoactivation of the ABA group occurs after 25 seconds, using these conditions.

T cell receptors were then immunoprecipitated using the protocol elaborated in example 6, supra, except mAb H57-597 available from the American Type Culture Collection, which is anti-TCR, was used.

Figure 8A shows these results, in lanes 1, 2 and 3 SDS- PAGE analysis, under reducing conditions, of the immunoprecipitated TCR showed two labeled bands at approximately 86 and 88 kDa, the amount of which increased with the concentration of soluble ligand.

Figure 8B shows competition experiments. Either H2 or N9 cells were incubated with a constant amount of cross-linked K^dQ10/¹²⁵IASA-YIPSAEK(ABA)I (3.3x10' ) , in the absence or presence of increasing amounts of K^dQ10. Lanes 1-5, which involve work in N9, used 0, 9.9xlO"¹⁰M, 3.3xlO^"10M, 9.9xlO'¹⁰M of K^dQ10, or mAb 20-88-45 (5 ug/ml), which is anti-K^d and was obtained from the American Type Culture Collection. Lane 6 used cytolytic T cell clone H2. The same protocols of binding, ultraviolet treatment, and immunoprecipitation used in Example 6 were employed here.

No inhibition of TCR labelling was observed, indicating that the affinity of the complex for TCR was considerably higher than that of the HLA-K^d derivative. Further, TCR photo¬ affinity labelling was inhibited by the anti-K^d mAb 20-8-45, and was not at all detectable on the PbCS specific H2CTL. Example 9

The kinetics of interaction of K^dQ10, crosslinked with ^I25IASA-YIPSAEK(ABA) I with TCR on N9 cells were studied. In a first set of experiments, incubation of N9 with 3.3xl0^"10M of cross-linked complex was carried out over 180 minutes, at 37°C, 18°C, and 4°C, and then subjected to immunoprecipitation and SDS-PAGE analysis, as above. Figure 9A shows that ligand binding at 37°C was rapid, with optima being reached at 2-4 minutes, with time dependent decrease. At 18°C, the transient maximum binding was reached after about 30 minutes. At 4°C, in contrast, binding was slower, and increased continuously over three hours.

In dissociation experiments (figure 9B) , N9 cells were preincubated with the complexes, and then incubated in fresh medium, over 180 minutes. Analysis was carried out as in the previous examples. Dissociation was rapid at 37°C, with a halftime of about one minute. Dissociation was slower at 18°C and 4°C, with halftimes of about 6 minutes and 20 minutes.

Example 10 Binding of P815 cell associated ligand to the N9 cell

TCRs was assessed.

P815 cells, previously photoaffinity labelled, were added in a volume of 50 ul (DMEM, 2% FCS) , to a sample of N9 cells.

Equal numbers of cells were used. The incubations were at 37°C (O) , 18°C (Δ) , or 40°C (|) , over a period of time (140 minutes) . Binding was determined following ultraviolet radiation, as in Example 8. Immunoprecipitation with SF 1.111, and densitometric analysis was then carried out, also as set forth supra. Figure 10 shows that at 37°C, intercellular TCR binding reached maximum after 40-60 minutes, following a linear increase over the first 30 minutes. At 18°C, ligand binding was slower, with transient maximum being reached at about 110 minutes. Maximal binding was about 25% higher at 18°C then at 37°C. No binding took place at 4°C. Example 11

The ability of various agents to inhibit the labelling of the N9 TCR was tested. N9 was incubated with P815 cell associated ligand as previously described in Example 10, either in the absence or presence of one of EGTA (5 mM) , EDTA (5 mM) anti-LFA-1 antibody (FD 18.5, 5 ug/ml) , anti-CD8 antibody (H35; 5 ug/ml), or cytochalasin D (4 μM) , at 37°C for 50 minutes. The same type of treatment (i.e., ultraviolet radiation, immunoprecipitation, SDS-PAGE analysis or denistometric analysis), was carried out. Figure 11A shows this. Parallel experiments, using soluble complexes of K^dQ10 crosslinked with the peptide were also carried out. Figure 11B shows these experiments.

While binding of cell associated ligand increased in presence of EGTA (approximately 20%) , it was nearly inhibited in the presence of EDTA. In contrast, these agents had no significant effect on the binding of soluble ligand. Similarly, TCR photoaffinity labeling by cell associated ligand was inhibited by over 70% in the presence of anti-LFA- mAb, but the mAb did not effect binding by soluble ligand. In contrast, the anti-CD8 mAb inhibited the cell associated ligand by about 50%, but nearly completely inhibited the labeling by the soluble ligand. When anti-LFA-1 and CD8 mAbs were used in combination (same concentration) inhibition of either ligand occurred. Cytochalasin D is known to interfere with icrofilament function. It inhibited the binding of cell associated ligand, but not soluble ligand.

The foregoing examples demonstrate the ability to generate CTLs of a desired specificity. Specifically, the method involves the use of cells, e.g., which present, on their surface, cross-linked complexes of covalently complexed peptides and MHC molecules, wherein the peptides have been labelled with a photoactivatable group. This group, when activated by appropriate light, covalently crosslinks the peptide to the MHC molecule. Also, soluble complexes of MHC molecule or derivatives thereof and covalently complexed peptides can also be used.

It is especially preferred that when APC's are used they be treated with a proliferation enhancing agent. Exemplary of proliferation enhancing agents are the mitogens, such as concanvalin A ("Con A") phytohemagluttin (PHA) , pokeweed mitogen (PWM) , and so forth.

The use of a photoactivatable group as a component of the peptide presented to the MHC molecule and thus covalently complexed thereto is an essential part of the invention. "Photoactivatable" as used herein refers to the ability of the moiety to become chemically reactive, i.e., capable of covalently complexing to an MHC molecule, when exposed to light in a particular frequency range. It is especially preferred that the photoactivatable substance be activatable in the presence of ultraviolet light, but this is not essential . Examples of photoactivatable groups include the "ABA" and "IASA" molecules exemplified herein, as well as other materials with which the art will be familiar.

It has been found that β>2 microglobulin optimizes the cross-linking of the photoactivatable group labeled peptides to the MHC molecule, and thus in one aspect of the invention, where APC's are used, β2 microglobulin is present. Also part of the optimization of the invention is the incubation of labelled peptide within a particular temperature range for a preferred period of time. It is preferred to incubate the materials at a temperature of from about 20°C to about 40°C, more preferably at some temperature between 20°C and 30°C. The time of incubation should be less than about 10 hours, an incubation time about at least 3 hours being preferred. Most preferably, incubation times of anywhere from about 4 hours to about 6 hours are preferred.

Any cell which presents an MHC molecule on its surface is usable in the aspect of the invention involving APC's, "MHC" being understood to refer to a generic family of molecules which includes, e.g., human leukocyte antigens ("HLAs") and others. Such cells include not only those which normally present these molecules, but also those which have been transformed to express these molecules. Cells of eukaryotic origin are preferred, especially mammalian cells. In one embodiment, the cells derive from a mammalian spleen.

When soluble complexes are used, the labeled peptide is covalently bound to either an MHC molecule or an MHC derivative. "MHC derivative", as used herein, refers to a molecule such as, but not limited to K^dQ10, which possess antigenic binding properties that are the same as those of native, membrane bound MHC molecules.

As was pointed out, supra, the photoactivatable group is attached to the a ino acid which covalently complexes the peptide to the HLA or MHC molecule. The amino acid will vary from system to system. For example, with the PbCS peptide exemplified herein, the N-terminal amino acid anchors the peptide to the MHC molecule. It is well known in the art, however, that the "anchoring" amino acid can be anywhere along the sequence of the peptide. The ability to identify the salient amino acid is well within the skill of the art. Given that the peptides presented by MHC molecules of both class I and class II are quite small, it is fairly easy to identify the amino acid of interest when it is not already known. Identification can be made via, e.g., truncation experiments. The invention, as exemplified, may be used jin vivo as well as in vitro. In the case of the first embodiment, the presenting cells may be administered to the organism in any of the standard forms used for such materials. These include, e.g., intraperitoneal and subcutaneous administration, which are the preferred modes of administration.

When used in an ±n vivo context, it may or may not be desirable to separate the CTLs which are generated. In the case of a therapeutic regime, for example, it is desirable not to purify the CTLs, as the point of the administration is to "boost" the subject's immune system such that CTLs are produced against a target antigen. Such an approach is useful when the subject is suffering from a condition where it is known that a particular peptide is presented on the cell surface. The art is replete with examples of such conditions, and as such a listing is not necessary.

It is also desirable to have CTLs to a particular complex of MHC and peptide available, for the reasons set forth at the outset of the specification. In these situations, the CTL's may be purified from the subject, in any of the standard ways of doing so. One preferred embodiment of this is the use of the invention in autologous transfer therapy. This methodology involves taking a T cell containing sample from a patient. That sample is then contacted with cells which have the covalent, cross-linked complexes described supra on their surface or soluble, non-cell bound complexes. Upon exposure to these complexes, any T cells capable of reacting with the complex will do so, with proliferation of particular CTL clones. As these derive from a particular patient, it would be expected that the CTLs could be reintroduced to the subject, without concern as to immune rejection.

Implicit in this discussion is another facet of this invention, which is a diagnostic methodology for identifying presence of CTLs of a particular specificity. Described herein is a well known chromium release assay, used for identifying CTLs. Other variations on the chromium release assay are known, as per Traversari et al . , Immunogenetics 35: 145 (1992) . Other types of assays are also available. In the diagnostic assay of the invention, cells which have been treated to cross link the peptide to its presenting molecule partner or soluble complexes of peptide and MHC or MHC derivative are contacted with a sample to determine presence of CTLs. Any of the foregoing methods may be adapted to use the materials described herein. As has been pointed out, the cross-linking of the peptide to the MHC leads to superior results, both in terms of the CTL response in vivo or in vitro, as well as in the ability to identify the presence of any CTLs in a sample, such a result was certainly not to be expected in this field.

In a preferred pertinent embodiment, one labels the peptide with two photoactivatable groups. The first of these groups is placed at the amino acid which joins the peptide to the HLA molecule. The second photoactivatable group is attached to the amino acid residue which contacts the TCR of specific CTLs.

The results presented herein all deal with the generation of a CTL response. It is well known that the "sine qua non" of such a response is the successful binding of a peptide to MHC molecule of class I. It should be borne in mind that the B cell response also depends upon the successful binding of other peptides to MHC molecules of class II. The binding of the peptides to MHC molecule takes place in a similar way, be they class I or class II molecules. Thus, the invention may be seen as a method for provoking or enhancing, i.e., generating, an immune response to a targeted peptide, via the use of an appropriate peptide, labelled with a photoa tivatable group at an appropriate site, which is then complexed to its partner MHC molecule or an MHC derivative either on an APC or in the form of a soluble complex. The resulting complex can then be used ±n vitro or in vivo, to stimulate, e.g., a T cell response, a B cell response or an antibody response, as well as other related responses.

Other aspects of the invention will be clear to the skilled artisan, and are not set forth here. The terms and expressions which have been employed are used as terms of description and not of limitation, and there is no intention in the use of such terms and expressions of excluding any equivalents of the features shown and described or portions thereof, it being recognized that various modifications are possible within the scope of the invention.

Claims

Claims :

1. Method for provoking an immune response in vivo, comprising:

(i) contacting an MHC molecule presenting cell with a peptide which specifically binds to said MHC molecule, wherein said peptide has been labelled with a photoactivatable group,

(ii) photoactivating said photoactivatable group to covalently crosslink molecules of said photoactivatable group to said MHC molecule in a covalent complex, and (iii) administering cells presenting cross linked peptide/MHC covalent complexes to a subject, in an amount sufficient to provoke an immune response.

2. The method of claim 1, further comprising contacting said cells with a proliferation enhancing agent prior to administering to said subject.

3. The method of claim 2, wherein said proliferation enhancing agent is a mitogen.

4. The method of claim 3, wherein said mitogen is concanavalin A.

5. The method- of claim 1, wherein said photoactivatable group is iodo 4-azidosalicylic acid.

6. The method of claim 1, further comprises contacting said peptide to said MHC molecule presenting cell in the presence of β2 microglobulin.

7. The method of claim 1, wherein said photoactivatable group is linked to said peptide at its contact site.

8. The method of claim 1, wherein said peptide is labelled with a second photoactivatable group at the amino acid of said peptide which contacts the T cell receptor of a cytolytic T cell.

9. The method of claim 8, wherein said second photoactivatable group is linked to said peptide at an amino acid which contacts a T cell receptor.

10. The method of claim 8, wherein said second photoactivatable group is 4-azidobenzoic acid.

11. The method of claim 1, comprising incubating said peptide and said MHC molecule presenting cell at a temperature of from 20°C to about 30°C.

12. The method of claim 11, comprising incubating said peptide and said MHC molecule presenting cell for a period of at least three hours.

13. The method of claim 12, wherein said period is less than 10 hours.

14. The method of claim 12, wherein said period ranges from about 4 to about 6 hours.

15. The method of claim 1, wherein said immune response comprises proliferation of cytolytic-T lymphocytes specific for said complex of peptide and MHC molecule.

16. The method of claim 1, wherein said immune response comprises proliferation of B cells which produce antibodies specific to an epitope formed by said peptide.

17. Method for improved binding of an exogenous peptide to an MHC presenting cell, comprising:

(i) contacting an MHC molecule presenting cell with a peptide which specifically binds to said MHC molecule which has been labelled with a photoactivatable group so as to form covalent complexes of said peptide and said MHC molecules, and

(ii) photoactivating said covalent complexes to crosslink said photoactivatable group to said MHC molecule.

18. Method for provoking in vitro proliferation of cytolytic T cells to a complex of peptide/MHC molecule, comprising:

(i) contacting an MHC molecule presenting cell with a peptide which specifically binds to said MHC molecule, wherein said peptide has been labelled with a photoactivatable group, (ii) photoactivating said photoactivatable group under conditions favoring crosslinking of molecules of said photo¬ activatable group to said MHC molecule to form covalent complexes of MHC molecule and peptide, and

(iii) contacting said cells to a sample which contains cytolytic T cells, in an amount sufficient to provoke proliferation of any cytolytic T cells present in said sample which are specific for complexes of said peptide and said MHC molecule .

19. Method for determining presence of cytolytic T cells specific for a complex of a peptide and an MHC molecule in a sample, comprising contacting said sample to a complex of said peptide and said MHC molecule, wherein said peptide and said MHC molecule have been crosslinked via a photoactivatable group present on said peptide, and determining either (i) proliferation of cytolytic T cells in said sample or (ii) lysis of cells presenting said complex as a determination of said cytolytic T cells in said sample.

20. The method of claim 19, wherein said peptide is labelled with a second photoactivatable group at the amino acid of said peptide which contacts the T cell receptor of a cytolytic T cell.

21. Method for provoking an immune response in vivo comprising: administering to a subject in need thereof a complex of (i) a soluble MHC molecule or soluble. MHC derivative and (ii) a peptide which specifically binds to (i) , wherein said peptide comprises a photoactivatable group, and (i) and (ii) have been covalently bound to each other via photoactivation of said photoactivatable group, in an amount sufficient to provoke an immune response thereto.

22. The method of claim 21, wherein said complex is a cell free, soluble complex.

23. The method of claim 21, wherein said complex is on the surface of a cell .

24. The method of claim 23, further comprising contacting said cells with a proliferation enhancing agent prior to administering to said subject.

25. The method of claim 24, wherein said proliferation enhancing agent is a mitogen.

26. The method of claim 25, wherein said mitogen is concanavalin A.

27. The method of claim 21, wherein said photoactivatable group is iodo 4-azidosalicylic acid.

28. The method of claim 23, further comprises contacting said peptide to said MHC molecule presenting cell in the presence of β2 microglobulin.

29. The method of claim 21, wherein said photoactivatable group is linked to said peptide at its site of contact to said MHC molecule or binding derivative.

30. The method of claim 21, wherein said peptide is labelled with a second photoactivatable group at the amino acid of said peptide which contacts the T cell receptor of a cytolytic T cell.

31. The method of claim 30, wherein said second photoactivatable group is linked to said peptide at an amino acid which contacts a T cell receptor.

32. The method of claim 30, wherein said second photoactivatable group is 4-azidobenzoic acid.

33. The method of claim 23, comprising incubating said peptide and said MHC molecule presenting cell at a temperature of from 20°C to about 30°C.

34. The method of claim 33, comprising incubating said peptide and said MHC molecule presenting cell for a period of at least three hours.

35. The method of claim 34, wherein said period is less than 10 hours.

36. The method of claim 34, wherein said period ranges from about 4 to about 6 hours.

37. The method of claim 21, wherein said immune response comprises proliferation of cytolytic-T lymphocytes specific for said complex of peptide and MHC molecule.

38. The method of claim 37, wherein said immune response further comprises proliferation of B cells which produce antibodies specific to an epitope formed by said peptide.

39. Method for improved binding of an exogenous peptide to an MHC presenting cell, comprising:

(i) contacting an MHC molecule presenting cell with a peptide which specifically binds to said MHC molecule which has been labelled with a photoactivatable group so as to form covalent complexes of said peptide and said MHC molecules, and (ii) photoactivating said covalent complexes to crosslink said photoactivatable group to said MHC molecule.

40. Method for provoking in vitro proliferation of cytolytic T cells, comprising contacting a cytolytic T cell containing sample with a complex of (i) an MHC molecule or binding derivative thereof and (ii) a peptide which specifically binds to (i) and which contains a photoactivatable group, wherein (i) and (ii) have been covalently bound to each other via photoactivation of said photoactivatable group, in an amount sufficient to provoke proliferation of any cytolytic T cells specific for said complex of (i) and (ii) in said sample.

41. The method of claim 40, wherein said complex is a cell free, soluble complex.

42. The method of claim 40, wherein said complex is on the surface of a cell.

43. Method for determining presence of cytolytic T cells specific for a complex of a peptide and an MHC molecule or binding derivative in a sample, comprising contacting said sample to a complex of said peptide and said MHC molecule, wherein said peptide and said MHC molecule have been crosslinked via a photoactivatable group present on said peptide, and determining either (i) proliferation of cytolytic T cells in said sample or (ii) lysis of cells presenting said complex as a determination of said cytolytic T cells in said sample.

44. The method of claim 43, comprising contacting said sample to a cell free, soluble complex of said peptide and said MHC molecule or binding derivative and determining proliferation of cytolytic T cells.

45. The method of claim 43, comprising contacting said sample to a cell presenting said complex on its surface.

46. The method of claim 43, wherein said peptide is labelled with a second photoactivatable group at the amino acid of said peptide which contacts the T cell receptor of a cytolytic T cell.