WO1992008806A1

WO1992008806A1 - Method for stimulating production and determining presence of an hla/mhc antigen

Info

Publication number: WO1992008806A1
Application number: PCT/US1991/008218
Authority: WO
Inventors: Sune Kvist; Frederic Levy
Original assignee: Ludwig Institute for Cancer Research Ltd; Ludwig Cancer Research
Current assignee: Ludwig Institute for Cancer Research Ltd; Ludwig Cancer Research
Priority date: 1990-11-09
Filing date: 1991-11-05
Publication date: 1992-05-29
Anticipated expiration: 1993-05-09
Also published as: AU9073091A

Abstract

This invention relates to a method for provoking a restricted T cell response, and for identifying agents which are capable of causing such a response. Essentially, peptide fragments are contacted to a cell free expression medium containing nucleic acids sequences which code for or express the components of an MHC or HLA antigen. By adding test peptides to the cell free medium and then determining the extent of assembly of the components to form a complete antigen molecule, it is possible to identify substances which provoke restricted T cell responses. Also described is a method by which MHC or HLA antigens and the cells which present them can be assayed, using a labelled peptide specific for the antigen, and a binding partner to the label.

Description

METHOD FOR STIMULATING PRODUCTION '

AND METHOD FOR DETERMINING PRESENCE OF AN HLA/MHC ANTIGEN

FIELD OF THE INVENTION

This invention relates to cellular immunology. More particularly, it relates to the process by which cell surface antigens of antigen processing cells ("APCs") are manufactured, leading to generation, stimulation, or provocation of restricted T cell responses. Additionally, it relates to methods for identifying HLA antigens, proteins in general cells or fragments thereof presenting them, and reagents useful for this purpose. BACKGROUND AND PRIOR ART

The last two decades have seen an explosion in the amount of information available about the mammalian immune system. Various mechanisms of action and molecular pathways have been elucidated, and various theories have been proposed by the manner in which an organism responds to a foreign substance.

It is not possible to do complete justice to the complex interactions which are involved in an immunologic response, but the following synopsis is intended to give some background, and to present what is known and what is not known about this area of science.

It has now been established that T cells are an essential part of the immune response, being involved in the identification of foreign molecules and in the production and stimulation of molecules involved in an immune response, such as antibodies, interleukins, and so forth. It is now known that in order to ensure that unprovoked responses do not take place, T cells must recognize the foreign substances and "ignore" those materials which are not problematic, for example "SELF" antigens.

The art now accepts that this process of discernment takes place on several levels, starting with the action of so-called antigen presenting cells, or "APCs". These cells, which include macrophages, literally engulf foreign materials, such as proteins, which are then broken down into smaller fragments, i.e., peptides. These peptides then interact with particular molecules known either as "major histocompatibility complexes" ("MHC") or human lymphocyte antigens ("HLA") . The former generally refers to molecules which originate from mice, the latter from humans, although the manner in which the molecules function is such that the terms are used interchangeably, as will be the case herein.

These MHC molecules are divided into class I and class II type of molecules. Much is known about their structure and interaction with cell membranes of APCs. The MHC-I molecules are made up of a glycosylated polypeptide molecule, known as the "heavy chain", which weighs about 45 kd, and a B2 microglobulin molecule, weighing about 12 kd. The MHC II molecules are made of two glycoprotein chains, a "heavy" chain (the "α" chain) , and a "light" chain (the "β" chain) , weighing 30-34 and 20-29 kds, respectively. Further information about their structure and interaction with cell membranes may be gathered from any of the standard textbooks on the subject, such as Male et al. , Advanced Immunology (1987, Gower

Medical Publishing) , especially Chapter 5, the disclosure of which is incorporated by reference. These molecules are in turn divided into different polymorphic classes, known by various terms such as "H-2K^b", "H-2K ", "HLA-A3", "HLA- B27", and so forth.

The two different classes of MHC/HLA molecules are involved in mechanisms that are somewhat different. The MHC-I molecule in concert with a peptide fragment, reacts with a cytolytic or cytotoxic T cell ("CTL" hereafter) . One of the results of the interaction is the production of IL- 1, leading to maturation and proliferation of the CTLs. The CTLs in turn react with APCs thereby deleting them from the organism. In contrast, MHC-II molecules, in association with the peptide antigen, react with so-called "helper" T cells ("T_H" cells) . The interaction also provokes maturation of the T_H cells, which are then involved in the stimulation of antibody production by B cells, among other steps. This simplification does not address the full gamut of issues involved in an immunological response, but it should be noted that each type of response requires the interaction of peptide fragment with the MHC/HLA molecule.

While it has long been acknowledged that specific MHC/HLA molecules interact only with certain types of peptide fragments (see, e.g., Marx, Science 235: 843:844 (1987); Hackett et al., in Immune Recognition of Protein Antigens (Cold Spring Harbor Laboratory, 1985), pg. 45-55; Buus et al., Science 235: 1353-1358 (1987); Unanue et al., Science 236: 551-557 (1987); Marrack, Science 2 5: 1311-1313 (1987); Allen et al., Immunol. Rev. £8.: 171-187 (1987); Babbit et al., Proc. natl. Acad. Sci. 83.: 4509-4513 (1986), the disclosures of which are incorporated by reference herein, the particulars of the interaction have, until recently remained unclear. For example, all of the evidence regarding the formation of complexes between peptide fragments and class I antigens has remained indirect, until the recent work of Chen et al.. Nature 337: 743-745 (1989); Bouillot et al.. Nature 339: 473-475 (1989). Others have suggested that complex formation between peptide and class I antigen occurs in the endoplasmic reticulum compartment. Nuchtem et al.. Nature 339: 223-226 (1989); Yewdel et al. Science 244: 1072-1075 (1989). The work of Townsend et al.. Nature 340: 443-448 (1989), has shown that peptides can induce assembly of the class I complex. Townsend used the urine cell line RMA-S, and studied expression of murine H- 2 antigens in a whole cell system.

It has now been found that the assembly of MHC/HLA antigens can be stimulated by peptides which normally binds to the antigen and by labeled derivatives of the peptides, such as biotinylated derivatives in a cell free medium, such as a cell lysate. It is this discovery that is the basis of one aspect of the invention described herein.

Further, since the labeled peptide fragment which stimulates the assembly can also be identified by, e.g., reacting it with a solid phase bound binding partner, another aspect of this invention is to identify MHC/HLA antigens, both on the surface of cells expressing them, as well as in cell free medium. Thus, the invention also provides a way for cell typing various components of the cellular repertoire.

Since proliferation of restricted T cells relies on the production of MHC/HLA antigens with which they complex, yet another aspect of this invention is the stimulation of restricted T cell responses by contacting a cell sample capable of expressing MHC/HLA antigens with a peptide which stimulates assembly of the surface antigen with which the restricted T cell type interacts.

How these and other aspects of the invention are achieved will be seen from the disclosure which follows. BRIEF DESCRIPTION OF THE FIGURES

Figure 1 shows cell free translation of HLA-B27 heavy chain and B2-microglobulin in a cell free expression system containing radioactive amino acids.

Figure 2 presents analysis of the content of HLA antigen in Raji microso al vesicles.

Figure 3 depicts the effect of NP 384-394 peptide on the assembly of HLA-B27 heavy chain and B2-microglobulin.

Figure 4 shows the in vitro assembly of HLA-B27 and B2- microglobulin at various concentrations of NP 384-394.

Figure 5 shows the results of experiments showing stimulation of assembly of HLA-B27 heavy chain and B2 microglobulin by biotinylated NP384-394.

Figure 6 shows how biotinylated NP384-394 interacts directly and specifically with HLA-B27 heavy chain.

Figure 7 presents data showing competition between biotinylated and non-biotinylated NP384-394 for binding to HLA-B27 heavy chain.

Figure 8 shows that the biotinylated NP384-394 peptide must cross the microsomal membrane to associate with HLA- B27 heavy chain. Figure 9 shows the in vitro translation of HLA-B27 and human B2 microglobulin in microsomes deficient of B2 microglobulin.

Figure 10 shows results obtained when the experiment generating the data depicted in Figure 9 was repeated at different temperatures.

DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS Example l

This example describes the translation of HLA-B27 and &2 microglobulin in a cell free system medium, i.e., a lysate of reticulocytes. To carry this out, cDNA for HLA-B27 and B2 microglobulin, both of which were provided by other investigators, were inserted into the EcoRI site of the plas id pGEM-3Zf(-), following standard techniques. In addition, a modified, translation enhancing fragment of the E3/19K gene of adenovirus type 2 was inserted upstream of the AUG initiation condon for the HLA-B27 cDNA. The mRNA for these cDNA clones was transcribed, using SP6 RNA polymerase, following the directions provided by the supplier of the enzyme.

In addition a lysate of rabbit reticulocytes was prepared, following Pelham et al., Eur. J. Biochem 67: 247- 256 (1976); Meth. Enzymol. 96: 50-74 (1983). A total of 50 ng of HLA-B27 mRNA and 150 ng of B₂-microglobulin mRNA were added to the lysate, which was supplemented with 3 ul of purified microsomes (60 A_2g0/ml) , from the Raji lymphoid cell lines, as described by Garoff et al., J. Mol, Biol. 124: 587-600 (1978) . A reaction mixture totalling 65 ul in total resulted, of which 46 ul was lysate, 2-4 ul was mRNA, 6 ul of ³⁵S methionine (90 μCi, > 1000 Ci/mmol) , microsomal membranes, 3 ul, RNasin 10 units, phenylmethylsulfonyl fluoride (PMSF) , 2 ug, and aprotinin, 10 units. Translation started by adding the mRNA, followed by incubation at 37°C, for 90 minutes. After 90 minutes, translation was terminated by changing temperature to 0°C and adding 135 ul of TNE buffer (20 mM Tris-HCl, 150 mM NaCl, 5 mM EDTA) . Membranes were pelleted in an Eppendorf centrifuge for 5 minutes, and solubilized in a physiological buffer containing 5 mM EDTA and 1% Triton X-100. Non-solubilized material was pelleted following the centrifugation protocol outlined and supernatant was subjected to immunoprecipitation, as described by Burgett et al.. Cell 41: 987-997 (1985) , the disclosure of which is incorporated by reference. The i muno-precipitate was subjected to SDS- PAGE, following Kvist et al.. Cell 19: 61-69 (1982), the disclosure of which is incorporated by reference, using rabbit antiserum to HLA heavy chains. Antiserum was absorbed on β-microglobulin coupled to Sepharose beads, and is non-reactive with B2-microglobulin.

As is seen in figure 1, HLA-B27 heavy chain appears as a doublet having a molecular weight of approximately 39 kd, while β₂-microglobulin migrates as one distinct band, having a molecular weight of about 12 kd. These bands will be seen in Figure 1, lines 1 and 2. Co-translation of both mRNAs however, followed by immuno-precipitation with monoclonal antibody mAb W6/32 which recognizes only assembled forms of HLA-B27/B₂-microglobulin (see Barnstable et al.. Cell 14.: 9- 20 (1978) shows that only a small fraction of translated material was present in assembled form. This is shown in Figure l, lane 3. Translation efficiency, however, was not changed. Example 2

The HLA-B27 antigen in Raji microsomal vesicles was analyzed. Following the protocol described, supra, mRNA for HLA-B27 was translated in vitro in the presence of Raji microsomes (3 ul, A_2g0=60 units/ml) . The major difference in this experiment was that increasing amounts of mRNA were used. The amount of mRNA was either Ong, 8ng, 32ng, 128ng or 512ng. After completion of translation samples were solubilized and each separate translation was divided into two equal parts. One part used for immuno- precipitation with either anti HLA heavy chain rabbit antiserum, or antihuman &₂ microglobulin antiserum. The second portion was used for Western blotting using the same antisera. In Figure 2, "A" shows immunoprecipitation, while "B" shows western blotting. Lanes 1-5 show increasing amounts of mRNA (0, 8, 32, 128, 512 ng) , while lanes 6-10 show the same amount of microsomes used in lanes 1-5, analyzed via Western Blotting using rabbit antiserum against &2 microglobulin.

The data show that the HLA-B27 band becomes visible at approximately 32 ng, and increases intensity to become a thick band at 512 ng. While translation was efficient and a large amount of radiolabelled HLA-B27 heavy chain was seen in lane 5, an increase in the total amount of HLA heavy chains in the microsomes was not seen (Figure 2 , lanes 6- 10) . This leads to the conclusion that the HLA-B27 heavy chain translated in vitro in the reticulocyte lysates did not contribute to any significant extent to the large amount of HLA antigen heavy chains endogenously present within the microsomes.

Analysis of the data depicted in Figure 2B shows that the same conclusions are reached with β₂ microglobulin, i.e., that a large excess of endogenous &2 microglobulin is present in microsomes and that the amount translated in vitro does not contribute to the extent that it can be visualized by an increase in western blotting. Thus, the endogenous HLA, both heavy and light chains are present in excessively large amounts over newly translated radiolabelled HLA-B27 heavy chain and B2-microglobulin. Example 3

The effect of a peptide, i.e., NP384-394 on the assembly of HLA-B27 heavy chain and B2 microglobulin was studied. NP 384-394 is a peptide which is identical to residues 384-394 of the nuclear protein of influenza A virus. This peptide is known to constitute a CTL target in HLA-B27 restricted manner.

To test this, the translation protocol described in example 1 was carried out, with the exception that peptides were added prior to incubation at 37°C, at concentrations of 1.0, 4.0, and 16.0 μm. The peptides were NP384-394, described by Huet et al.. Int. Immunol. 2.: 311-316 (1990), and NP 147-158, described by Taylor et al., Immunogenetics 26: 267-272 (1987) . Both peptide amino acid sequences are known.

In Figure 2, lane 1, data are presented for when no peptide is used. In lanes 2-4, NP 384-394 at concentrations of 1.0, 4.0 and 16 mM, while lanes 5-7 show the same concentrations of NP 147-158.

Figure 3 shows that in the absence of the restricted peptide, only small amounts of the assembled protein are produced. Indeed, since the mAb W6/32 was used, this conclusion can be reached. Little or no increase in production was observed at 1 μ , but enhanced assembly was evident, and became much greater at concentrations of 4 μM and 16 μM, respectively. Increase is due solely to enhanced production of HLA-B27 as B2 microglobulin bands show equal intensities regardless of NP 384-394 concentrations. The NP 147-158 peptide did not induce increased assembly when used at the same concentrations as NP 384-394, even though the β2 microglobulin band showed the same intensity. These data lead to the conclusion that assembly of an HLA antigen, a prerequisite to presentation at the cell surface with concomitant stimulation of a restricted T cell response, can be provoked by introducing an appropriate concentration of a peptide which associates with an MHC receptor. Example 4

The data generated in Example 3 suggested that B2 microglobulin complexes which were newly translated and radioactive complex with large excesses of unlabelled HLA heavy chains presented endogenously within microsomal vesicles. Similarly, radiolabelled HLA-B27 heavy chain is most likely to assemble with unlabelled B2-microglobulin. Only a small amount of newly translated B2 microglobulin (radiolabelled) associates with HLA-B27 heavy chain, and does not visibly contribute to labelled B2 microglobulin bands. This explains the constant band of B2-microglobulin in Figure 3, and the conclusion is supported by the finding in Figure 2, and the fact that the amount of radiolabelled β2 microglobulin shown in Figure 3 constitutes only a very small portion of what is totally labelled. This in turn suggests that it should be possible to observe promotion of heavy chain assembly to B2 microglobulin in the absence of in vitro translation of mRNA for B2 microglobulin. The following experiment does this.

Again, in vitro translation was carried out as described in Example 1. Messenger RNA for HLA-B27 heavy chain (50 ng) was translated in the presence of Raji microsomes (3 ul) at different concentrations of NP 384-394 peptide. As depicted in Figure 4, lanes 1 and 5 show translation with no peptide, lanes 2 and 6, 1 μM; lanes 3 and 7, 4 μM, lanes 4 and 8, 16 μM. Immunoprecipita-tion was carried out using mAb W6/32 in lanes 1-4, and BBM.l, in lanes 5-8. The latter mAb is specific for B2-microglobulin.

As Figure 4 shows, W6/32 mAb precipitates increasing amounts of HLA-B27 heavy chains as NP 384-394 concentration increases. This is the same effect of assembly stimulation shown supra. even when mRNA for B2 microglobulin is not translated. The mAb BBM.l confirms this.

The data of Examples 1-4 show that a peptide which is known to induce CTL response against target cells showing a particular HLA antigen stimulates assembly of that antigen in a cell free medium favoring expression of the nucleic acids which code for or express the antigen components. The assembly occurs in the lumen of microsomal vesicles which mimic endoplasmic reticulum compartments. Increase in assembly is of the order of 3-8 fold, as compared to the absence of the peptide. These data suggest that not only can assembly be studied, but that one can test for a peptide which provokes a restricted T cell response. This is done by adding a peptide of interest to a cell free medium such as the medium described supra. and determining assembly of MHC/HLA antigen. Since the "mapping" of these antigens to particular types of T cell response in vivo is a process that has been studied, one can determine which peptides would be responsible for an in vivo response. One can then take the diagnostic aspect of the invention further. After determining a peptide which provokes assembly, the very same peptide may be added to an in vivo system to provoke the T cell response.

In a similar fashion, one can determine controlling agents or inhibitors of excessive or inappropriate T cell response by determining inhibitors of MHC/HLA antigen assembly. One carries out the same process as described supra. only now one looks for a decrease in assembly. Such a system is useful in determining pharmacological agents for treatment of inter alia autoimmune diseases. EXAMPLE 5

Following the results obtained and shown in Examples l- 4, studies were carried out to determine if assembled MHC/HLA antigens could be detected by contacting samples containing them with biotinylated peptides specific for the heavy chain portion thereof.

To do this, complete cDNAS for HLA-B27, human B2 microglobulin and H-2K^k were used. Both HLA-B27 and B2 microglobulin cDNA were inserted into the EcoRI site of pGem32f(-), as described in Example 1. The H-2 ^k cDNA was also inserted in this vector, but at a BamHI site. In each case, a translation enhancing fragment for E3/19K gene of Ad2 was inserted upstream.

Prior to the translation, cDNAS were linearized and purified by phenol extraction/ethanol precipitation. Transcription was carried out as per Example 1, and, after transcription was completed, the DNA was digested with DNase I. The transcript mRNA was then purified via ammonium activate/ethanol precipitation. The remaining mRNA was quantified spectrophotometrically. During the transcription, 10 units of RNase inhibitor were present. The mRNA was dissolved in autoclaved water containing 0.1% of diethylpoly-carbonate (DEPC) and was stored at -85°C.

In vitro translation was then carried out. Rabbit reticulocyte lysate was prepared following Pelham et al. , supra. and Jackson et al., supra. The translation mixture contained 50 nanogra s of one of the heavy chain mRNAs, 200 nanogra s of mRNA for β2 microglobulin (only some experiments; see infra) ; 3 microliters of microsomal membranes (60 A₂₈₀ units/ml) , 90 microCi of ³⁵S methionine (>1000 Ci/mole) , RNasin 10 units; 2 micrograms of PMSF and 10 unites of aprotinin. One of NP 384-394 and NP147-158, described supra (HLA-B27; NP 384-394; H-2K ; NP 147-158), was added at a final concentration of 8-32 micromolar. The total incubation mixture was 65 microliters. Translation was started by adding mRNA, and incubation of the mixture for 60-90 minutes, or other lengths of time as indicated infra. at 37°C. Translation was terminated by shifting temperature to 0°C and by adding 135 micro-liters of TNE buffer (20 mM NaCl, 5 mM EDTA). Membranes were pelleted by Eppendorf centrifuge for five minutes, and the resulting pellet was washed in TNE buffer once, followed by solubilizations in physiological buffer containing 5 mM EDTA and 1% Triton X-100. Nonsolubilized material was pelleted, also by centrifugation, and supernatant was used for one of immunoprecipitation or streptavidin-sepharose precipitation. The peptide NP 384-394 was added in either normal form or its biotinylated derivative. The latter was prepared by mixing NHS-LC-biotin, which was dissolved at a concentration of 10 mM in 100 mM sodium phosphate buffer, pH 7.5. A 5 molar excess was added to peptide solution, and the reaction was allowed to continue for two hours at room temperature under continuous agitation. After the reaction, Tris-HCl, pH 7.4 was added to a final concentration of 10 mM. This consumes excess NHS-LC-biotin, preventing further reactions with other proteins in the translation mixture. The efficacy is between 70-90%, as measured by iodination and precipitation with streptavidin-agarose. Biotinylated peptides were stored at -85°C when not being used.

In carrying out the translations, different concentrations of peptide were used. Figure 5 shows these results. The type and amount of protein are as follows: LANE PEPTIDE

1 no peptide

2 NP 384-394, 8 micromolar

3 NP 384-394, 32 micromolar

4 no peptide

5 biotinylated NP 384-394, 8 micromolar

6 biotinylated NP 384-394, 32 micromolar Following translation, immunoprecipitation was carried out using mAb W6/32, described supra. and precipitates were analyzed via SDS-PAGE (Figure 5) .

The data depicted in Figure 5 shows a doublet of HLA-B27 heavy chain in lanes 1-4, and also show that NP 84-395 and biotin modified NP 384-394 stimulate assembly indistinguishably. This was confirmed by densitometric analysis, showing approximately 5 fold stimulation by both forms of the peptide at 32 micromolar concentration. Example 6

The direct action of biotinylated NP 384-394 with HLA- B27 was determined. Translation of HLA-B27, as in Example 5, as well as H-2K^k was carried out. Following this, precipitation experiments were carried out. In these, streptavidin-agarose was used, together with biotinylated NP 384-394, and rabbit anti-H-2 antiserum. In Figure 6, lanes 1-6 show use of streptavidin agarose with different amounts and types of peptide. In lane 7, the antiserum was also used. LANE REAGENTS

1 no peptide, agarose-streptavidin

2 8 micromolar biotinylated NP 384-394 streptavidin

3 32 micromolar biotinylated NP 384-394 streptavidin

4 32 micromolar unmodified NP 384-394 streptavidin

5 biotin solution, no peptide streptavidin

6 (H-2K^k) 32 micromolar biotinylated NP 384-394 streptavidin 13

7 (H-2K^k) 32 micromolar biotinylated NP 384-394, anti H-2 antiserum Figure 6 shows that at 8 micromolar concentration of biotinylated NP 384-394, two faint bands are seen (lane 2), and this band is strong at 32 micromolar concentration (lane 3) . No bands are seen in the absence of peptide, thus indicating that newly synthesized HLA-B27 heavy chains can be visualized by interaction with biotinylated peptide. Interaction actually occurs via biotin, as was shown by using unmodified peptide at 32 micromolar. No bands were seen, as per lane 4 of Figure 6.

By incubating translation solution with biotinylation solution without peptide, the question of whether biotin reagent reacted with the HLA-B27 heavy chain. The absence of any band in lane 5 shows that it does not. In the final part of this experiment, the specificity of biotinylated NP 384-394 for HLA-B27 was tested. Specifically, mouse MHC H- 2K^k mRNA was translated, and tested with the biotinylated peptide. Lane 6 shows no interaction, but H-2K heavy chain was precipitated by using the rabbit antiserum (shown in lane 7) .

The conclusion reached herein is that the assembled HLA/MHC antigen can be assayed by using a biotinylated peptide specific for that antigen, followed by contact with a receptor for the biotin, such as streptavidin or avidin- agarose. Example 7

In this experiment, the question of whether NP 384-394 and its biotinylated derivative competed for binding to HLA- B27 heavy chain.

To test this, 50 ng of mRNA for HLA-B27 heavy chain were translated as described supra in the presence of 8 micromolar concentration of biotin-NP 384-394, in connection with varying amounts of NP 384-394, and NP 147-158. As indicated, the concentration of biotinylated derivative was kept constant. Unmodified NP 384-394 or NP 147-158 at increasing concentrations was added before start of translation. Referring to Figure 7, the reagents used were as follows:

LANE REAGENT (+ 8 micromolar biotinylated

NP 384-394

1 O peptide

2 2 micromolar unmodified NP 384-394

3 4 micromolar unmodified NP 384-394

4 8 micromolar unmodified NP 384-394

5 16 micromolar unmodified NP 384-394

6 32 micromolar unmodified NP 384-394

7 2 micromolar NP 147-158

8 4 micromolar NP 147-158

9 8 micromolar NP 147-158

10 16 micromolar NP 147-158

11 32 micromolar NP 147-158

As shown in Figure 7, when completing NP 384-394 was absent, 2 heavy chain bands were observed. As the NP 384-394 is added in increasing concentrations the bands weaken and disappear (lane 6) . Control peptide NP 147-158 did not compete with the biotinylated NP 384-394. This shows that the biotinylated and non-biotinylated molecules compete for the same site. Example 8

The assay system provided herein is simple enough to permit its use in, e.g., dissection of the entire process by which MHC presentations takes place. One essential question in this process and studied herein is where heavy chain and peptide interact.

To study this, HLA-B27 heavy chain mRNA was translated in vitro and as described supra, in the absence of, and in the presence of increasing concentrations of biotinylated NP 384-394. (8 micromolar, 32 micromolar). In parallel experiments, translation was either terminated in the manner described supra or by adding cycloheximide to block further translation, followed by addition of proteinase K to digest proteins in the lysate outside microsomal membranes. After digestion for 30 minutes on ice, phenyl methylsulphonid fluoride was added to stop digestion, and membranes were washed and solubilized, as for the parallel experiment. All experiments were then analyzed via incubation with streptavidin-agarose and SDS PAGE. These results, shown in Figure 8, are divided as follows: LANE TREATMENT

1 no biotin peptide, normal treatment

2 8 micromolar biotin NP 384-394 treatment

3 32 micromolar biotin NP 384-394 treatment

4 no biotin peptide, cycloheximide treatment

5 8 micromolar biotin NP 384-394 treatment

6 32 micromolar biotin NP 384-394 treatment Data from lanes 1-3 show that normal stimulation of assembly took place, with molecular weights of the heavy chain band being identical to those in other experiments. These experiments in which proteinase K digestion took place, however, showed a decrease in molecular weight corresponding to a lose of 30-40 amino acids at the carboxy terminus, normally protruding at the cytoplasmic side of the ER membrane, indicating that proteinase K digestion was efficient and. degraded membrane integrated protein close to the membrane. Although proteinase K digestion was efficient, identical stimulation of assembly of HLA-B27 heavy chains and B2 microglobulin was observed for both series. Addition of proteinase K prior to peptide addition resulted in no stimulation of assembly.

From this, it can be concluded that the interaction between peptide and heavy chain must take place at the luminal side of the ER membrane, since it was protected from proteinase K digestion, and that the peptide must be present within the lumen of the microsomal vesicles, eventually crossing the ER membrane. Example 9

All of Examples 1-8 rely on a medium for in vitro translation which contains microsomal membranes prepared from Raji cells. These contain high amounts of endogenous B2 microglobulin which, as has been shown, supra. assembles with newly translated, radioactive HLA-B27 heavy chain. In connection with this, experiments were carried out to determine the role of B2 microglobulin in assembly, by using Daudi cell microsomes. This cell line has the initiation codon for the B2 microglobulin gene mutated, and as such, the cells cannot express B2 microglobulin.

Messenger RNA (50 ng) for HLA-B27 heavy chain was translated, either alone or with mRNA for B2 microglobulin (200 ng) . As indicated, the same medium as used previously was used here, except that microsomal membranes from Daudi cells were used, in the same concentration as the Raji isolate. In precipitation experiments, protocols as used supra were employed.

In Figure 9, which follows, lanes l and 2 represent data where HLA-B27 was translated without β2 microglobulin mRNA present, while lanes 3-5 used the two templates together. In lane 1, rabbit anti-HLA antiserum was used. Lanes 2 and 4 show precipitation using streptavidin agarose, lane 3, precipitation using mAb W6/32, and lane 5, mAb BBM.l, all of which have been described herein. In all cases, the translation took place in the presence of 32 micromolar biotinylated NP 384-394.

Figure 9 shows that HLA-B27 was efficiently translated, as is verified by lane 1. There was no precipitation of HLA-B27 heavy chain by streptavidin agarose when B2 microglobulin was not translated as well; however, when cotranslation took place, the streptavidin-agarose precipitated both the heavy chains and B2-microglobulin (lane 4) , although this was considerably less than mAb W6/32 did (as shown in lane 3) . From lane 5, it is clear that considerably more B2 microglobulin is available.

This experiment shows that the β2 microglobulin is a requirement for stabilizing HLA-B27 heavy chain interaction with the peptide. In addition the fact that mAb W6/32 precipitated about 5 times more of both components than can streptavidin indicates that assembly in Daudi microsomes may occur either with or without endogenous peptides being involved.

Example 10

The experiments of example 9 were repeated, at a temperature of 26°C. Prior to precipitation, the temperature was increased to 37^βC for 15 minutes.

The intensity of the HLA-B27 heavy chain in the absence of NP 384-394 biotin decreased to background levels (lane 7, Figure 10) . However, when this peptide was present, an increase in T° did not lower band intensity (lane 8) , and peptide still bound heavy chain (lane 10) . This shows that the peptide stabilizes the ternary complex, and that the HLA-B2/B2 microglobulin complex is not stable at 37°C. Translation of mRNA for HLA-B27 heavy chain in the presence of the biotinylated derivative but not B2 migroglobulin, followed by precipitation, did not show interaction with peptide at T° (lanes 11, 12) . This appears to indicate that assembly of the heavy chain and β2 microglobulin requires presence and binding of peptide.

Examples 5-10 show that it is possible to identify particular MHC-HLA antigens by using a biotinylated peptide which is specific for the heavy chain component of the antigen. To the skilled artisan, the data using NP 384-394 will be seen to be extrapolatable to other HLA/MHC surface antigens, may of which are known to react with particular peptides. It will be evident that by labelling the specific peptide, it is possible both to precipitate the antigen of interest, and also to identify expression of the molecule on the cell surface. Thus, the invention involves, e.g., a method for identifying an MHC'HLA antigen in a cell sample by contacting a sample or lysate of cells to a labelled peptide under conditions favoring binding of the specific peptide to the assembled antigen, followed by binding of the resulting peptide-antigen complex to a binding partner for the label of the peptide, and then determining the binding.

"Label" as used herein is not meant in the sense used in immunology for e.g., enzyme labels. Rather, "label" refers to a member of a binding pair which is not an antibody. As indicated in the data, the label may be biotin, either modified or unmodified by chemical means known in the art. When the label is biotin, its binding partner may be, e.g., streptavidin or avidin, bound to an inert solid phase material, such as agarose.

It will be understood that the label and binding partner may in fact be reversed, with streptavidin or avidin being the label, and biotin coupled to solid phase acting as the binding partner. Any inert solid may serve as the solid phase, such as silica, glass, various polymers, sepharose, agarose, and so forth.

The nature of biotin-avidin interaction is such that every molecule of avidin\streptavidin binds to 4 biotin molecules. As a result, agglutination and complexing occurs rather easily, and the resulting complexes can be measured by, e.g., assaying precipitate, via turbidometric analysis in solution, etc. It is also possible to link the binding partner to a true, non-colloidal solid phase, such as the side of a tube, or other laboratory apparatus. In such a case, the antigen or cell type of interest is removable from the sample, and assaying can then take place.

While MHC/HLA class II type antigens do not contain &2 microglobulin, it is well understood that this type of antigen also interacts with particular peptides in the same general manner as do MHC-I molecules. Thus this invention, in all of its aspects, is equally applicable to the stimulation of MHC/HLA class II antigens and the resulting restricted T cell type stimulation, as well as assaying for the molecules and cell types themselves.

It will be understood that the specification and examples are illustrative but not limitative of the present invention and that other embodiments within the spirit and scope of the invention will suggest themselves to those skilled in the art.

SUBSTITUTE SHEET

Claims

We Claim:

1. Method for determining a peptide capable of provoking a restricted T cell response, comprising: contacting a test peptide to a first nucleic acid sequence which codes for a heavy chain of an MHC or HLA antigen, and a second nucleic acid sequence which codes for β2 microglobulin in a cell free expression medium containing amino acids sufficient for expression of said heavy chain and said B2 microglobulin and an identifiable label which is incorporated into an expression product of one of said first and second nucleic acid sequence, under conditions favoring expression of said first and second nucleic acid sequences, and determining assembly of said first and second expression products to form an MHC or HLA antigen as a determination of a peptide capable of provoking a restricted T cell response.

2. Method of claim 1, wherein said first nucleic acid sequence codes for an HLA heavy chain.

3. Method of claim 1, wherein said first and second nucleic acid sequences are mRNA.

4. Method of claim 1, wherein said cell free expression medium is a cell lysate.

5. Method of claim 1, wherein said cell free expression medium contains a label which is incorporated into one of said first and second expression products to give a detectable signal.

6. Method of claim 1, wherein at least one of said first and second nucleic acid sequences is a DNA sequence.

7. Method of claim 1, wherein said MHC or HLA antigen is an MHC-I antigen.

8. Method of claim 1, wherein said test peptide is a fragment of a larger peptide produced by an infectious agent.

9. Method of claim 8, wherein said infectious agent is a virus.

10. Method of claim 1, wherein said restricted T cell response is proliferation of cytotoxic T cells.

11. Method of claim 5, wherein said label is a radioactive amino acid.

12. Method of claim 1, further comprising determining assembly of said first and second expression product by contacting said expression medium with a binding agent which specifically binds to assembled first and second expression product but not to the unassembled components thereof.

13. Method of claim 12, wherein said binding agent is a monoclonal antibody.

14. Method for provoking an in vivo restricted T cell response, comprising: administering to a subject infected suffering from an immune disorder an amount of a peptide fragment which is non-toxic but which stimulates production and assembly of MHC or HLA antigen which present peptide fragments of said infectious agent thereby provoking a restricted T cell response, sufficient to stimulate production and assembly of said MHC or HLA antigen.

15. Method for determining presence of an HLA or MHC antigen in a biological sample comprising contacting said sample with a labelled peptide which specifically binds with said HLA or MHC antigen under conditions favoring formation of complexes therebetween, contacting said sample with a solid phase bound binding partner of said label under conditions favoring binding, and determining formation of binding between said complex and said solid phase bound binding partner as a determination of said HLA or MHC antigen.

16. Method of claim 15, wherein said antigen is a class

I antigen.

17. Method of claim 15, wherein said antigen is a class

II antigen.

18. Method of claim 15, wherein said sample is a cell lysate.

19. Method of claim 15, wherein said sample contains whole cells.

20. Method of claim 15, wherein said label is biotin or a biotin derivative and said binding partner is avidin or streptavidin.

21. Method of claim 15, wherein said label is avidin or streptavidin and said binding partner is biotin or a biotin derivative.

22. Method of claim 15, further comprising separating bound solid phase and complex from said biological sample prior to determining said MHC or HLA antigen.

23. Reagent useful in determining an HLA or MHC antigen, said reagent comprising a peptide which specifically binds to an HLA or MHC antigen labelled with biotin or a biotin derivative.

24. Reagent useful in determining an HLA or MHC antigen, comprising separate portions of each of:

(i) a peptide which specifically binds to an HLA or MHC antigen labelled with a member of the group consisting of biotin, avidin and streptavidin, and

(ii) a solid phase bound binding partner for said label, wherein said binding partner is selected from the group consisting of biotin avidin and streptavidin.

25. Method for inhibiting an in vivo restricted T cell response comprising: administering to a subject suffering from excessive restricted T cell response an amount of a non-toxic inhibiting agent which prevents assembly of MHC or HLA antigen associated with said restricted T cell response sufficient to inhibit said response.

26. Method for determining a peptide capable of inhibiting restricted T cell response, comprising: contacting a test peptide to a first nucleic acid sequence which codes for a heavy chain of an MHC or HLA antigen, and a second nucleic acid sequence which codes for B2 microglobulin in a cell free expression medium containing amino acids sufficient for expression of said heavy chain and said β2 microglobulin and an identifiable label which is incorporated into an expression product of one of said first and second nucleic acid sequence, under conditions favoring expression of said first and second nucleic acid sequences, and determining inhibition of assembly of said first and second expression products to form an MHC or HLA antigen as a determination of a peptide capable of inhibiting a restricted T cell response.