- Email: [email protected]
In vitro reconstitution of heat shock protein–peptide complexes for generating peptide-specific vaccines against cancers and infectious diseases
Methods 32 (2004) 25–28 www.elsevier.com/locate/ymeth
In vitro reconstitution of heat shock protein–peptide complexes for generating peptide-specific vaccines against cancers and infectious diseases Zihai Li* Center for Immunotherapy of Cancer and Infectious Diseases, University of Connecticut School of Medicine, 263 Farmington Avenue, Farmington, CT 06030-1601, USA Accepted 28 May 2003
Abstract Known commonly as molecular chaperones for proteins, heat shock proteins (HSPs) have also been found to chaperone small molecular weight cellular peptides. HSP–peptide complexes can prime T cell immunity specific against the peptides bound to HSPs, but not against HSPs per se. This immunomodulatory functions of HSPs are based on two intrinsic properties. One, HSPs are excellent adjuvants due to their ability to activate dendritic cells (DCs). Two, HSPs can bind directly to their receptors on DCs to then channel HSP-associated peptides to associate with MHC molecules. When a specific antigenic peptide is defined, this peptide can also be complexed with either tissue derived or recombinant HSPs in vitro to generate HSP–peptide complexes as peptidespecific vaccines. This article focuses on the methods commonly used to reconstitute HSP–peptide complexes, and discusses assays to verify the efficiency of complexing for immunotherapy against cancers and infectious diseases. 2003 Elsevier Science (USA). All rights reserved.
1. Introduction HSPs are a family of evolutionally conserved ‘‘housekeeping’’ molecules that play fundamental roles in assisting protein folding and unfolding [1,2]. They are sub-classified customarily according to their molecular weight (kDa), such as the family of HSP90, HSP70, and HSP60, low and high molecular weight HSPs [3]. There are multiple members in each family of HSPs. For example, HSP90 family encompasses cytosolic HSP90a, HSP90b, and the endoplasmic reticular counterpart of HSP90, gp96, or grp94 [4]. Over the last two decades, the surprising immunological properties of HSPs in high vertebrates have been uncovered. It is found that HSPs not only bind physically to antigenic peptides inside of cells, they can also bind to their surface receptors on antigen presenting cells (APCs) [5]. Such an interaction between HSPs and
* Fax: +860-679-1265. E-mail address: [email protected].
their receptors triggers the release of inflammatory cytokines by APCs and the transfer of peptides chaperoned by HSPs to MHC molecules for priming of adaptive immunity. HSP–peptide complexes can thus be used as vaccines against tumors or intracellular organisms [6–8]. For this vaccine to be effective, HSPs and peptides have to be complexed together either covalently or non-covalently [9–11]. Mixing of the two alone without complexing is clearly ineffective. This article focuses on approaches used to complex HSPs with peptides non-covalently. An alternative approach is to generate recombinant HSP–peptide fusion proteins, which will be discussed by H. Udono in this issue of Methods.
2. List of materials In addition to obtain pure HSPs and peptides, the following materials and buffers are required. It is important to have peptides further purified by a reverse phase HPLC using a C18 column. When highly
1046-2023/$ - see front matter 2003 Elsevier Science (USA). All rights reserved. doi:10.1016/S1046-2023(03)00183-X
26
Z. Li / Methods 32 (2004) 25–28
hydrophobic peptides are used, it is advisable to dissolve the peptides first with dimethylsulfoxide (DMSO) (Sigma, D8418) and then dilute it in aqueous buffer. Be sure to use the lowest concentration of DMSO possible. 2.1. Iodination • Iodo-Gen precoated iodination tubes (Pierce, 28601), • Na125 I (Amersham IMS-30), • Sep-Pak C18 cartridges (Waters). 2.2. HSP70–peptide complex Binding buffer. Phosphate-buffered saline (Sigma, P3813), 1 mM ADP (Sigma, A-2810), and 1 mM MgCl2 (Sigma, M9272). • Microcon or Centricon 50 (Amicon, VS043). 2.3. Peptide complexing with HSP90, gp96, and calreticulin Binding buffer. About 20 mM Hepes (Sigma, H4034), pH 7.2, 20 mM NaCl (Sigma, S3171), and 2 mM MgCl2. • Microcon or Centricon 50 (Amicon, VS043). 2.4. ELISPOT Red blood cell lysis buffer. 0.15 M NH4 Cl (Sigma, A5666), 10 mM KHCO3 (Sigma, P4913), and 0.1 M Na2 EDTA (Sigma, E6511), pH 7.2–7.4. • Anti-CD8 antibody conjugated with magnetic beads (Myltenyi Biotec, 130-049-401). • 96-well IP-Multiscreen plates (Millipore, MAIPS4510). • Mouse anti-mouse IFN-c Ab (clone R4-6A2, PharMingen, 551216). Gammacell irradiator. • Biotinylated anti-IFN-c (XMG1.2, PharMingen, 554410) mAb. • HRP-based streptavidin (Vector Lab, PK-6100). • 3-Amino-9-ethylcarbazole (Sigma, A-6926) and H2 O2 (Sigma, H-1009). • Zeiss ELISPOT reader.
3. Description of methods 3.1. Complexing of peptides with HSP70 Peptide binding to HSP70 is tightly regulated by adenosine nucleotide and ATP hydrolysis [12]. ATP binding and ATP hydrolysis by HSP70 can lead to dramatic conformational changes of the peptide binding domain and the release of peptides. HSP70-ADP interaction however can stabilize and promote peptide binding to HSP70. Based on this feature, peptide can be
easily loaded onto HSP70. The concentration of HSP70 should be pre-determined accurately with a Bradford assay (Bio-Rad). HSP70 is then incubated with peptides of interests at 1:10 molar ratio in the binding buffer (phosphate-buffered saline supplemented with 1 mM ADP and 1 mM MgCl2) at 37 C for 1 h. Depending on the subsequent assay, free peptides may have to be removed by size exclusion using Microcon 50 (molecular weight cutoff 50 kDa). Be sure to prewash the column with the binding buffer at least five times before use. If endogenous peptides need to be removed first before complexing, HSP70 can be treated with phosphate buffered saline supplemented with 1 mM ATP and 1 mM MgCl2 for 1 h followed by washing away peptides with Microcon 50. An endoplasmic reticulum HSP70, BiP (immunoglobulin heavy chain binding protein) seems to preferentially bind to peptides containing a subset of aromatic and hydrophobic amino acids in alternating positions, suggesting that peptides bind in an extended conformation, with the side chains of alternating residues pointing into a cleft on the BiP molecule [13,14]. This principle has been exploited by one group for more efficient in vitro reconstitution of HSP70–peptide complexes [15]. Hybrid peptides were constructed that contained a T cell epitope, a linker (GSG in single-letter amino acid code), and an octapeptide sequence (HWDFAWPW), which is known to bind with high affinity to BiP. Such a hybrid peptide can then be loaded onto HSP70 using ADP-containing binding buffer as described above. 3.2. Peptide complexing with HSP90, gp96, and calreticulin In an attempt to define the parameters related to gp96peptide interaction, it was found, serendipitously, that the gp96-peptide complex is thermal stable [4]. Incubation of gp96 with a radioactive peptide at 4 C results in no association of radioactive peptides with gp96. Raising the temperature up to 55 C led to complex formation, as measured by acquisition of radioactive material by gp96 [10]. Surprisingly, the complex is stable and not sensitive to SDS, since the radioactive complex can be resolved on SDS–PAGE and visualized directly by autoradiography. This simple heat-dependent folding assay can be reproducibly used to complex exogenous peptides not only with gp96, but also with HSP90, calreticulin [16], and more recently with a2 macroglobulin [17]. HSPs are incubated with peptides of interests at 1:10 molar ratio in the binding buffer (20 mM Hepes, pH 7.2, 20 mM NaCl, and 2 mM MgCl2 ) at 55 C for 10 min. The samples were then incubated at room temperature for additional 30 min. Depending on the subsequent assay, free peptides can then be removed by size exclusion as described above for HSP70.
Z. Li / Methods 32 (2004) 25–28
Gp96 can also be complexed with peptides in the presence of 2–3 M NaCl without heat treatment[10]. For this purpose, gp96 and free peptides are co-incubated in sodium phosphate buffer containing 2–3 M NaCl at room temperature for one hour.
4. Confirmation of HSP–peptide complex formation Before subjecting further immunological analysis, the reconstituted HSP–peptide complexes have to be confirmed biochemically. Several methods can be used such as dissociation of the complex by acid extraction coupled by analysis of the low molecular weight peptides by a reverse phase HPLC or mass spectrometry. It has been found that HSP–peptide complexes are stable in electrophoretic conditions even in the presence of 0.1% SDS. Based on this principle, a convenient and a reliable method was developed to determine if peptide loading is efficient. First, the peptide of interests is iodinated (for peptide containing Try residue) using Iodo-Gen precoated iodination tubes (Pierce) following detailed instructions supplied by the manufacturer. Free iodine is removed by loading the sample to a Sep-Pak C18 cartridges (Waters), washed with H2 O and eluted with 80% acetonitrile in H2 O. HSPs are then complexed with radioactive peptides by the methods described above. An aliquot of the product (corresponding to 1–2 lg of HSPs) is mixed with SDS-PAGE loading buffer (0.1% SDS, 20% glycerol, and 5% bromophenol blue) and subjected to two identical 10% SDS–PAGEs without boiling. Proteins on one gel should be stained by silver staining or Brilliant Blue R Staining Solution (Sigma, B6529). The other gel should be dried and analyzed by a quantitative autoradiography. If reconstitution is successful, a clear band, corresponding to the position of HSPs, should be visible after autoradiograph. For peptides that cannot be easily labeled with radioactivity (such as peptides without Try residue), cold peptide competition should be done to gauge peptide binding efficiency. Briefly, a 15-mer model peptide, KRQIYTDLEMNRLGK (derived from vesicular stomatitis virus G protein, single amino acid designation) can be labeled with I125 . In vitro reconstitution of HSPs with this peptide should be done in the absence and presence of increasing concentration (from 1, 10, and 100) of nonradioactive peptides of interests. If the peptide of interests efficiently competes against radiopeptides, the reconstitution is deemed effective. On the other hand, if there is no competition or little binding even to the radioactive peptide, complexing is thus inefficient. One has to optimize the conditions of complexing by changing the HSP–peptide ratio, the duration of incubations, the ADP concentrations in the case of HSP70 (1–50 mM), and temperature (50–65 C) of complexing in the case of other HSPs.
27
5. Testing immunogenicity of HSP–peptide complexes HSP–peptide complexes reconstituted by the above method should be expected to be immunologically active in eliciting peptide-specific T cell responses and tumor immunity [10]. Interested readers should go to other chapters in this volume for detailed instructions. For example, S. Basu has detailed techniques for study of HSP–APC interactions in vitro, particularly with regards to the re-presentation of peptides from HSPs to MHC molecules. The most reliable test for assaying the functionality of HSP–peptide complexes is to determine if they can elicit functionally T cell responses in vivo. For this purpose, the appropriate method should already be established after consideration of the following variables. First, the host has to have the appropriate restriction alleles of MHC to be able to bind and present peptides to stimulate T cells. Second, assays for the presence of antigen specific T cells have to be available. The examples of T cell assays are direct killing of antigen-expressing target cells, ELISPOT assay to measure the frequency of cytokine producing T cells in response to antigen-specific stimulation on a single cell level, intracellular cytokine assays, and enumeration of antigenspecific T cells with MHC class I-peptide tetramers using flow cytometry. Third, the optimal dosage and schedule of immunization with HSP–peptide complexes to generate immune responses might be dependent on the precussor frequency of na€ıve antigen-specific T cells, the efficiency of cross-presentation of the particular peptide to MHC molecules, and the affinity of the cognate T cell receptor recognizing such a peptide in the complex with MHCs. A brief discussion of T cell assay using an ELISPOT ensues to illustrate some of the above variables that one needs to consider. A MHC I peptide or its precursor peptide is complexed onto HSPs as above. Since free peptides do not immunize very well, it is not necessary to remove free peptides before immunization. A total of 5, 25 or 100 lg of HSP–peptide complexes (based on the weight of HSPs) is injected in a volume of 100 ll intradermally to naive mice or other appropriate host. As a control, the host should be immunized with peptide alone, HSP alone or buffer only. One week later, a second injection is given to boost the immune responses. After an additional 1 week, spleen or draining lymph nodes are then collected and crashed in a Petri dish to make a single cell suspension. Red blood cells are lysed by lysis buffer (see above). If the peptide epitope is MHC I-restricted, CD8þ T cells should be further purified by positive selections using anti-CD8 antibody conjugated with magnetic beads (Myltenyi Biotec). A 96-well IPMultiscreen plates (Millipore, Badford, MA) should be precoated with 100 ll of mouse anti-mouse IFN-c Ab (10 lg/ml, clone R4-6A2, PharMingen, San Diego, CA).
28
Z. Li / Methods 32 (2004) 25–28
Purified CD8þ T cells are to be plated at a concentration of 2 105 cells/well. The target cells should be cells expressing the antigen that contains the peptide, as well as the appropriate MHC I. Generally, 7000 rad irradiated target cells are used (5 104 cells/well). After incubation for 20 h at 37 C, cells can be removed by extensive washing with PBS/0.05% Tween 20. This is followed sequentially by washing, incubation with biotinylated anti-IFN-c (XMG1.2, PharMingen, San Diego, CA) mAb, and a HRP conjugated streptavidin (Vector Lab, Burlingame, CA). The spots can be developed with 3-amino-9-ethylcarbazole and H2 O2 (Sigma, St. Louis, MO), counted with a Zeiss ELISPOT reader, and reported as the number of IFN-c spots per 1 106 cells. If the HSP–peptide complex is immunologically active, one expects to observe significantly more antigen-specific IFNc spots per 1 106 CD8þ T cells, comparing with mice immunized with HSPs or peptide alone.
6. Concluding remarks The ability of complexing HSPs with given peptides in vitro provides a way of generating more efficient peptide-based vaccines. The increased efficiency is due to the fact that HSPs can, simultaneously, activate professional APCs and mediate cross-presentation of peptides from HSPs to MHC molecules. This approach is clearly effective in preclinical models [10] and is now being tested clinically for infectious diseases. For example, recombinant HSP70 has been used to complex with peptides derived from human papilloma virus E7 protein and human herpes simplex virus glycoprotein, to
prevent/treat cervical cancer and herpes virus infections, respectively. Further studies are needed to best optimize the technique of in vitro complexing of HSPs with peptides and to translate this principle to clinics for immunotherapy of cancer and infectious diseases. Acknowledgment This work is supported by a NIH Grant CA90337. References [1] S. Lindquist, Annu. Rev. Biochem. 55 (1986) 1151–1191. [2] R.I. Morimoto, Genes Dev. 12 (1998) 3788–3796. [3] M.E. Feder, G.E. Hofmann, Annu. Rev. Physiol. 61 (1999) 243– 282. [4] Z. Li, J. Dai, H. Zheng, B. Liu, M. Caudill, Front. Biosci. 7 (2002) 731–751. [5] P.K. Srivastava, Nat. Rev. Immunol. 2 (2002) 185–194. [6] Z. Li, Semin. Immunol. 9 (1997) 315–322. [7] P.K. Srivastava, A. Menoret, S. Basu, R.J. Binder, K.L. McQuade, Immunity 8 (1998) 657–665. [8] P.K. Srivastava, Nat. Immunol. 1 (2000) 363–366. [9] R. Suto, P.K. Srivastava, Science 269 (1995) 1585–1588. [10] N.E. Blachere et al., J. Exp. Med. 186 (1997) 1315–1322. [11] H. Udono, T. Yamano, Y. Kawabata, M. Ueda, K. Yui, Int. Immunol. 13 (2001) 1233–1242. [12] D. Brehmer et al., Nat. Struct. Biol. 8 (2001) 427–432. [13] G.C. Flynn, T.G. Chappell, J.E. Rothman, Science 245 (1989) 385–390. [14] G.C. Flynn, P. J., M.T. Flocco, J.E. Rothman, Nature 353 (1991) 726–730. [15] Y. Moroi et al., Proc. Natl. Acad. Sci. USA 97 (2000) 3485–3490. [16] S. Basu, P.K. Srivastava, J. Exp. Med. 189 (1999) 797–802. [17] R.J. Binder, D. Karimeddini, P.K. Srivastava, J. Immunol. 166 (2001) 4968–4972.
In vitro reconstitution of heat shock protein–peptide complexes for generating peptide-specific vaccines against cancers and infectious diseases Zihai Li* Center for Immunotherapy of Cancer and Infectious Diseases, University of Connecticut School of Medicine, 263 Farmington Avenue, Farmington, CT 06030-1601, USA Accepted 28 May 2003
Abstract Known commonly as molecular chaperones for proteins, heat shock proteins (HSPs) have also been found to chaperone small molecular weight cellular peptides. HSP–peptide complexes can prime T cell immunity specific against the peptides bound to HSPs, but not against HSPs per se. This immunomodulatory functions of HSPs are based on two intrinsic properties. One, HSPs are excellent adjuvants due to their ability to activate dendritic cells (DCs). Two, HSPs can bind directly to their receptors on DCs to then channel HSP-associated peptides to associate with MHC molecules. When a specific antigenic peptide is defined, this peptide can also be complexed with either tissue derived or recombinant HSPs in vitro to generate HSP–peptide complexes as peptidespecific vaccines. This article focuses on the methods commonly used to reconstitute HSP–peptide complexes, and discusses assays to verify the efficiency of complexing for immunotherapy against cancers and infectious diseases. 2003 Elsevier Science (USA). All rights reserved.
1. Introduction HSPs are a family of evolutionally conserved ‘‘housekeeping’’ molecules that play fundamental roles in assisting protein folding and unfolding [1,2]. They are sub-classified customarily according to their molecular weight (kDa), such as the family of HSP90, HSP70, and HSP60, low and high molecular weight HSPs [3]. There are multiple members in each family of HSPs. For example, HSP90 family encompasses cytosolic HSP90a, HSP90b, and the endoplasmic reticular counterpart of HSP90, gp96, or grp94 [4]. Over the last two decades, the surprising immunological properties of HSPs in high vertebrates have been uncovered. It is found that HSPs not only bind physically to antigenic peptides inside of cells, they can also bind to their surface receptors on antigen presenting cells (APCs) [5]. Such an interaction between HSPs and
* Fax: +860-679-1265. E-mail address: [email protected].
their receptors triggers the release of inflammatory cytokines by APCs and the transfer of peptides chaperoned by HSPs to MHC molecules for priming of adaptive immunity. HSP–peptide complexes can thus be used as vaccines against tumors or intracellular organisms [6–8]. For this vaccine to be effective, HSPs and peptides have to be complexed together either covalently or non-covalently [9–11]. Mixing of the two alone without complexing is clearly ineffective. This article focuses on approaches used to complex HSPs with peptides non-covalently. An alternative approach is to generate recombinant HSP–peptide fusion proteins, which will be discussed by H. Udono in this issue of Methods.
2. List of materials In addition to obtain pure HSPs and peptides, the following materials and buffers are required. It is important to have peptides further purified by a reverse phase HPLC using a C18 column. When highly
1046-2023/$ - see front matter 2003 Elsevier Science (USA). All rights reserved. doi:10.1016/S1046-2023(03)00183-X
26
Z. Li / Methods 32 (2004) 25–28
hydrophobic peptides are used, it is advisable to dissolve the peptides first with dimethylsulfoxide (DMSO) (Sigma, D8418) and then dilute it in aqueous buffer. Be sure to use the lowest concentration of DMSO possible. 2.1. Iodination • Iodo-Gen precoated iodination tubes (Pierce, 28601), • Na125 I (Amersham IMS-30), • Sep-Pak C18 cartridges (Waters). 2.2. HSP70–peptide complex Binding buffer. Phosphate-buffered saline (Sigma, P3813), 1 mM ADP (Sigma, A-2810), and 1 mM MgCl2 (Sigma, M9272). • Microcon or Centricon 50 (Amicon, VS043). 2.3. Peptide complexing with HSP90, gp96, and calreticulin Binding buffer. About 20 mM Hepes (Sigma, H4034), pH 7.2, 20 mM NaCl (Sigma, S3171), and 2 mM MgCl2. • Microcon or Centricon 50 (Amicon, VS043). 2.4. ELISPOT Red blood cell lysis buffer. 0.15 M NH4 Cl (Sigma, A5666), 10 mM KHCO3 (Sigma, P4913), and 0.1 M Na2 EDTA (Sigma, E6511), pH 7.2–7.4. • Anti-CD8 antibody conjugated with magnetic beads (Myltenyi Biotec, 130-049-401). • 96-well IP-Multiscreen plates (Millipore, MAIPS4510). • Mouse anti-mouse IFN-c Ab (clone R4-6A2, PharMingen, 551216). Gammacell irradiator. • Biotinylated anti-IFN-c (XMG1.2, PharMingen, 554410) mAb. • HRP-based streptavidin (Vector Lab, PK-6100). • 3-Amino-9-ethylcarbazole (Sigma, A-6926) and H2 O2 (Sigma, H-1009). • Zeiss ELISPOT reader.
3. Description of methods 3.1. Complexing of peptides with HSP70 Peptide binding to HSP70 is tightly regulated by adenosine nucleotide and ATP hydrolysis [12]. ATP binding and ATP hydrolysis by HSP70 can lead to dramatic conformational changes of the peptide binding domain and the release of peptides. HSP70-ADP interaction however can stabilize and promote peptide binding to HSP70. Based on this feature, peptide can be
easily loaded onto HSP70. The concentration of HSP70 should be pre-determined accurately with a Bradford assay (Bio-Rad). HSP70 is then incubated with peptides of interests at 1:10 molar ratio in the binding buffer (phosphate-buffered saline supplemented with 1 mM ADP and 1 mM MgCl2) at 37 C for 1 h. Depending on the subsequent assay, free peptides may have to be removed by size exclusion using Microcon 50 (molecular weight cutoff 50 kDa). Be sure to prewash the column with the binding buffer at least five times before use. If endogenous peptides need to be removed first before complexing, HSP70 can be treated with phosphate buffered saline supplemented with 1 mM ATP and 1 mM MgCl2 for 1 h followed by washing away peptides with Microcon 50. An endoplasmic reticulum HSP70, BiP (immunoglobulin heavy chain binding protein) seems to preferentially bind to peptides containing a subset of aromatic and hydrophobic amino acids in alternating positions, suggesting that peptides bind in an extended conformation, with the side chains of alternating residues pointing into a cleft on the BiP molecule [13,14]. This principle has been exploited by one group for more efficient in vitro reconstitution of HSP70–peptide complexes [15]. Hybrid peptides were constructed that contained a T cell epitope, a linker (GSG in single-letter amino acid code), and an octapeptide sequence (HWDFAWPW), which is known to bind with high affinity to BiP. Such a hybrid peptide can then be loaded onto HSP70 using ADP-containing binding buffer as described above. 3.2. Peptide complexing with HSP90, gp96, and calreticulin In an attempt to define the parameters related to gp96peptide interaction, it was found, serendipitously, that the gp96-peptide complex is thermal stable [4]. Incubation of gp96 with a radioactive peptide at 4 C results in no association of radioactive peptides with gp96. Raising the temperature up to 55 C led to complex formation, as measured by acquisition of radioactive material by gp96 [10]. Surprisingly, the complex is stable and not sensitive to SDS, since the radioactive complex can be resolved on SDS–PAGE and visualized directly by autoradiography. This simple heat-dependent folding assay can be reproducibly used to complex exogenous peptides not only with gp96, but also with HSP90, calreticulin [16], and more recently with a2 macroglobulin [17]. HSPs are incubated with peptides of interests at 1:10 molar ratio in the binding buffer (20 mM Hepes, pH 7.2, 20 mM NaCl, and 2 mM MgCl2 ) at 55 C for 10 min. The samples were then incubated at room temperature for additional 30 min. Depending on the subsequent assay, free peptides can then be removed by size exclusion as described above for HSP70.
Z. Li / Methods 32 (2004) 25–28
Gp96 can also be complexed with peptides in the presence of 2–3 M NaCl without heat treatment[10]. For this purpose, gp96 and free peptides are co-incubated in sodium phosphate buffer containing 2–3 M NaCl at room temperature for one hour.
4. Confirmation of HSP–peptide complex formation Before subjecting further immunological analysis, the reconstituted HSP–peptide complexes have to be confirmed biochemically. Several methods can be used such as dissociation of the complex by acid extraction coupled by analysis of the low molecular weight peptides by a reverse phase HPLC or mass spectrometry. It has been found that HSP–peptide complexes are stable in electrophoretic conditions even in the presence of 0.1% SDS. Based on this principle, a convenient and a reliable method was developed to determine if peptide loading is efficient. First, the peptide of interests is iodinated (for peptide containing Try residue) using Iodo-Gen precoated iodination tubes (Pierce) following detailed instructions supplied by the manufacturer. Free iodine is removed by loading the sample to a Sep-Pak C18 cartridges (Waters), washed with H2 O and eluted with 80% acetonitrile in H2 O. HSPs are then complexed with radioactive peptides by the methods described above. An aliquot of the product (corresponding to 1–2 lg of HSPs) is mixed with SDS-PAGE loading buffer (0.1% SDS, 20% glycerol, and 5% bromophenol blue) and subjected to two identical 10% SDS–PAGEs without boiling. Proteins on one gel should be stained by silver staining or Brilliant Blue R Staining Solution (Sigma, B6529). The other gel should be dried and analyzed by a quantitative autoradiography. If reconstitution is successful, a clear band, corresponding to the position of HSPs, should be visible after autoradiograph. For peptides that cannot be easily labeled with radioactivity (such as peptides without Try residue), cold peptide competition should be done to gauge peptide binding efficiency. Briefly, a 15-mer model peptide, KRQIYTDLEMNRLGK (derived from vesicular stomatitis virus G protein, single amino acid designation) can be labeled with I125 . In vitro reconstitution of HSPs with this peptide should be done in the absence and presence of increasing concentration (from 1, 10, and 100) of nonradioactive peptides of interests. If the peptide of interests efficiently competes against radiopeptides, the reconstitution is deemed effective. On the other hand, if there is no competition or little binding even to the radioactive peptide, complexing is thus inefficient. One has to optimize the conditions of complexing by changing the HSP–peptide ratio, the duration of incubations, the ADP concentrations in the case of HSP70 (1–50 mM), and temperature (50–65 C) of complexing in the case of other HSPs.
27
5. Testing immunogenicity of HSP–peptide complexes HSP–peptide complexes reconstituted by the above method should be expected to be immunologically active in eliciting peptide-specific T cell responses and tumor immunity [10]. Interested readers should go to other chapters in this volume for detailed instructions. For example, S. Basu has detailed techniques for study of HSP–APC interactions in vitro, particularly with regards to the re-presentation of peptides from HSPs to MHC molecules. The most reliable test for assaying the functionality of HSP–peptide complexes is to determine if they can elicit functionally T cell responses in vivo. For this purpose, the appropriate method should already be established after consideration of the following variables. First, the host has to have the appropriate restriction alleles of MHC to be able to bind and present peptides to stimulate T cells. Second, assays for the presence of antigen specific T cells have to be available. The examples of T cell assays are direct killing of antigen-expressing target cells, ELISPOT assay to measure the frequency of cytokine producing T cells in response to antigen-specific stimulation on a single cell level, intracellular cytokine assays, and enumeration of antigenspecific T cells with MHC class I-peptide tetramers using flow cytometry. Third, the optimal dosage and schedule of immunization with HSP–peptide complexes to generate immune responses might be dependent on the precussor frequency of na€ıve antigen-specific T cells, the efficiency of cross-presentation of the particular peptide to MHC molecules, and the affinity of the cognate T cell receptor recognizing such a peptide in the complex with MHCs. A brief discussion of T cell assay using an ELISPOT ensues to illustrate some of the above variables that one needs to consider. A MHC I peptide or its precursor peptide is complexed onto HSPs as above. Since free peptides do not immunize very well, it is not necessary to remove free peptides before immunization. A total of 5, 25 or 100 lg of HSP–peptide complexes (based on the weight of HSPs) is injected in a volume of 100 ll intradermally to naive mice or other appropriate host. As a control, the host should be immunized with peptide alone, HSP alone or buffer only. One week later, a second injection is given to boost the immune responses. After an additional 1 week, spleen or draining lymph nodes are then collected and crashed in a Petri dish to make a single cell suspension. Red blood cells are lysed by lysis buffer (see above). If the peptide epitope is MHC I-restricted, CD8þ T cells should be further purified by positive selections using anti-CD8 antibody conjugated with magnetic beads (Myltenyi Biotec). A 96-well IPMultiscreen plates (Millipore, Badford, MA) should be precoated with 100 ll of mouse anti-mouse IFN-c Ab (10 lg/ml, clone R4-6A2, PharMingen, San Diego, CA).
28
Z. Li / Methods 32 (2004) 25–28
Purified CD8þ T cells are to be plated at a concentration of 2 105 cells/well. The target cells should be cells expressing the antigen that contains the peptide, as well as the appropriate MHC I. Generally, 7000 rad irradiated target cells are used (5 104 cells/well). After incubation for 20 h at 37 C, cells can be removed by extensive washing with PBS/0.05% Tween 20. This is followed sequentially by washing, incubation with biotinylated anti-IFN-c (XMG1.2, PharMingen, San Diego, CA) mAb, and a HRP conjugated streptavidin (Vector Lab, Burlingame, CA). The spots can be developed with 3-amino-9-ethylcarbazole and H2 O2 (Sigma, St. Louis, MO), counted with a Zeiss ELISPOT reader, and reported as the number of IFN-c spots per 1 106 cells. If the HSP–peptide complex is immunologically active, one expects to observe significantly more antigen-specific IFNc spots per 1 106 CD8þ T cells, comparing with mice immunized with HSPs or peptide alone.
6. Concluding remarks The ability of complexing HSPs with given peptides in vitro provides a way of generating more efficient peptide-based vaccines. The increased efficiency is due to the fact that HSPs can, simultaneously, activate professional APCs and mediate cross-presentation of peptides from HSPs to MHC molecules. This approach is clearly effective in preclinical models [10] and is now being tested clinically for infectious diseases. For example, recombinant HSP70 has been used to complex with peptides derived from human papilloma virus E7 protein and human herpes simplex virus glycoprotein, to
prevent/treat cervical cancer and herpes virus infections, respectively. Further studies are needed to best optimize the technique of in vitro complexing of HSPs with peptides and to translate this principle to clinics for immunotherapy of cancer and infectious diseases. Acknowledgment This work is supported by a NIH Grant CA90337. References [1] S. Lindquist, Annu. Rev. Biochem. 55 (1986) 1151–1191. [2] R.I. Morimoto, Genes Dev. 12 (1998) 3788–3796. [3] M.E. Feder, G.E. Hofmann, Annu. Rev. Physiol. 61 (1999) 243– 282. [4] Z. Li, J. Dai, H. Zheng, B. Liu, M. Caudill, Front. Biosci. 7 (2002) 731–751. [5] P.K. Srivastava, Nat. Rev. Immunol. 2 (2002) 185–194. [6] Z. Li, Semin. Immunol. 9 (1997) 315–322. [7] P.K. Srivastava, A. Menoret, S. Basu, R.J. Binder, K.L. McQuade, Immunity 8 (1998) 657–665. [8] P.K. Srivastava, Nat. Immunol. 1 (2000) 363–366. [9] R. Suto, P.K. Srivastava, Science 269 (1995) 1585–1588. [10] N.E. Blachere et al., J. Exp. Med. 186 (1997) 1315–1322. [11] H. Udono, T. Yamano, Y. Kawabata, M. Ueda, K. Yui, Int. Immunol. 13 (2001) 1233–1242. [12] D. Brehmer et al., Nat. Struct. Biol. 8 (2001) 427–432. [13] G.C. Flynn, T.G. Chappell, J.E. Rothman, Science 245 (1989) 385–390. [14] G.C. Flynn, P. J., M.T. Flocco, J.E. Rothman, Nature 353 (1991) 726–730. [15] Y. Moroi et al., Proc. Natl. Acad. Sci. USA 97 (2000) 3485–3490. [16] S. Basu, P.K. Srivastava, J. Exp. Med. 189 (1999) 797–802. [17] R.J. Binder, D. Karimeddini, P.K. Srivastava, J. Immunol. 166 (2001) 4968–4972.