Patent Application
20060019323
The invention relates to recombinant adenylate cyclase of Bordetella sp. for diagnostic and immunomonitoring.
This invention relates to diagnostic testing and immunomonitoring of diseases, as well as immunomonitoring of any T cell response following stimulation of T cells by an antigen.
The incidence of tuberculosis (TB) in cattle, caused by Mycobacterium bovis, has dramatically increased over the last decades in the British national herd. This increase constitutes a significant animal welfare, economic, and potential public health problem (Krebs et al., 1997). To control this zoonotic disease, better and more specific diagnostic reagents, as well as effective vaccines, are urgently needed. The U.K. government has initiated a research program to develop such reagents and vaccines.
Diagnosis of bovine tuberculosis in cattle is done almost exclusively in skin tests with tuberculin Purified Protein Derivative (PPD). The specificity of this test is limited because of the undefined and cross-reactive nature of PPD. A blood-based test measuring tuberculin-induced production of IFN-γ is also currently in limited field use (Wood et al., 1994). The specificity of tuberculin-based reagents is compromised, though, following vaccination with the human TB vaccine M. bovis BCG (BCG) (reviewed in Buddle et al, 2003). Therefore, diagnostic reagents allowing the differential diagnosis of M. bovis infected and vaccinated animals are needed before effective TB vaccines can be developed for cattle.
M. tuberculosis is also a major threat to human health, being responsible for more deaths globally than any other bacterium. The vaccine against, and immunological diagnosis of, TB are not fully satisfactory. For instance, the skin test reagent, PPD, used to aid diagnosis of both active and latent tuberculosis lacks specificity and sensitivity. Bacille Calmette Guerin (BCG) vaccine is very widely used to prevent TB, but its protective efficacy in adults is also limited.
Besides vaccination, an alternative control strategy to prevent the progression of latent infection by M. tuberculosis (LTBI) to clinical TB is through the use of preventative antituberculous drug therapy (PT). One aspect of this control strategy is diagnostic testing, but the tuberculin skin test (TST), used to identify healthy individuals with latent infection, has several operational drawbacks. First, the TST reagent, PPD, is cross-reactive because it contains epitopes found in many mycobacteria. TST reactivity can arise through sensitization by environmental mycobacteria or from the BCG vaccine. Second, the sensitivity of the TST is reduced by HIV infection (Johnson, J. L., et al. 1998). Third, the TST requires two clinic visits, one for administration and one for reading. The test is also operator-dependent. These limitations impair identification of LTBI and, therefore, wider application of PT. While there is a need in the art for TB vaccine candidates of greater efficacy than BCG, there is also a need for development of immunodiagnostic methods of greater sensitivity, specificity, and practicality than TST skin testing.
Previously, it has been shown that the specificity of diagnostic reagents can be improved by using antigens that are highly expressed by M. bovis or M. tuberculosis but are deleted from the genome of BCG. Such antigens allow not only the differential diagnosis of infected and BCG vaccinated animals or humans, but also improve the specificity of tuberculin per se in the absence of vaccination.
A major advance in tuberculosis research has been the identification of a genomic segment (designated Region of deletion 1—RD1) that is present in pathogenic members of the M. tuberculosis complex, but absent from all attenuated BCG strains (Gordon, S. V., et al. 1999; Behr, M. A., et al. 1999; and Mahairas, G. G., 1999). Molecules encoded on this segment can contribute to virulence (Pym, A. S., et al. 2003), or stimulate species-specific T cell responses of protective potential (Weinrich Olsen, A., et al.: 2001; Pym et al., 2003). In addition, a great deal of interest has focused on the potential of RD1 encoded antigens to improve the immunodiagnosis of TB (Arend, S. M., et al., 2000; Ewer, K., et al. 2003). However, protein subunits tend to inefficiently stimulate T cell responses and even the most promising experimental vaccine preparations require powerful adjuvants that are not licensed for use in humans. Similarly, the best immunodiagnostic methods previously known rely on peptide mixtures and ELISPOT analysis that are likely too complex for use in medically-underserved environments (Arend, S. M., et al. 2002). The antigens ESAT-6 and CFP-10, which are encoded in the RD1 region of M. bovisi/M. tuberculosis, a region that is deleted in all strains of BCG, have shown particular promise as diagnostic reagents when used as recombinant proteins or synthetic peptides in the IFN-γ test (Buddle et al. 1999, Vordermeier et al., 1999 and 2001), but there is still a need in the art for simple methods by which T cell responses to M. tuberculosis antigens can be enhanced (Wilkinson, K. A., et al. 2000; Wilkinson, K. A., et al., 1999).
Under the classical pathway of antigen (Ag) presentation, exogenous and endogenous Ags of pathogens are generally processed in Ag presenting cells (APCs) by two distinct pathways to generate peptides for major histocompatibility complex (MHC)-restricted presentation (Germain, R. N. 1994). Exogenous Ags are taken up and degraded by proteases along the endocytic pathway. These processed peptides then bind nascent MHC class II molecules and are presented to CD4+ T cells at the APC cell membrane (Villadangos, J. 2001). After this specific recognition and interaction with co-stimulatory molecules, the activated CD4+ T cells can provide help to either B cells or CD8+ T cells by secreting cytokines. Endogenous proteins are degraded by the proteasome into the APC cytoplasm to generate MHC class I-restricted peptides that are transported to the endoplasmic reticulum where they bind to nascent MHC class I molecules. MHC I-peptide complexes are then exported and presented to CD8+ T cells at the APC cell membrane (Rock, K. L., and A. L. Goldberg. 1999).
In addition to these classical pathways of Ag presentation, it is now well documented that some exogenous cell-associated or particulate Ag can be cross-presented on MHC class I molecules through alternative pathways of processing (Jondal, M., et al., 1996; Heath, W. R., and F. R. Carbone. 2001; Reimann, J., and R. Schirmbeck. 1999; Moron, G., et al. 2002). One particular approach to inducing CTL responses against exogenous Ag takes advantage of the capacity of certain proteins, mainly bacterial toxins, to enter the cytosol of APC, to be processed along MHC class I presentation pathway, and to then be presented to CD8+ T cells. Thus, several vaccinal strategies using recombinant bacterial toxins have been designed in different laboratories to generate CTL responses against exogenous Ag (Ballard, J. D., et al. 1996; Bona, C. A. et al., 1998; Goletz, T. J., et al. 1997; Haicheur, N., et al. 2000).
An attractive approach to vaccine design is the delivery of proteins by non-replicating protein vectors such as bacterial toxins or toxoids. Bordetella pertussis secretes a calmodulin-activated adenylate cyclase toxin, CyaA, that primarily targets myeloid phagocytic cells that express the αMβ2 integrin receptor (CD11b/CD18), and include professional antigen presenting cells, such as neutrophils, macrophages, NK cells, and dendritic cells (Guermonprez, P., et al., 2000). CyaA is able to deliver its N-terminal catalytic adenylate cyclase domain (400 amino acid residues) into the cytosol of eukaryotic target cells directly through the cytoplasmic membrane (Guermonprez, P., et al., 2000; Sebo, P., et al., 1995).
The CyaA is such a vector system that has shown promise in mice models. Peptide and small proteins can be inserted and expressed as fusion proteins with CyaA or chemically bound to CyaA. CyaA facilitates direct translocation across the plasma membrane of target cells. Importantly, it has been shown that vaccination with CyaA can induce MHC class I restricted CD8+ T cell responses (e.g. Gueromonprez et al., 1999).
Genetically detoxified CyaA can be used as a vehicle to deliver both CD4+ and CD8+ T-cell epitopes to antigen presenting cells when the epitopes are inserted within the adenylate cyclase activity domain (AC) of the CyaA toxoid in the first 600 amino acids. The antigen-presenting cells then trigger specific T cell responses (Dadaglio, G., et al., 2000; Saron, M. F., et al., 1997; Osicka, R., et al., 2000; Loucka, J., et al., 2002; Fayolle, C., et al., 1996). CyaA delivers its N-terminal catalytic domain (AC domain) into the cytosol of eukaryotic cells bearing the αmβ2 integrin (CD11b/CD18). CD8+ T cell epitopes inserted into a genetically detoxified CyaA AC domain are delivered into CD11c+CD11bhigh DC cytoplasm both in vitro and in vivo (Guermonprez, P., et al. 2002). This mechanism of targeted delivery of CD8+ T cell epitopes into MHC class I pathway results in efficient presentation followed by robust and protective CTL responses (Fayolle, C., et al. 1996, 1999, and 2001). Moreover, the T cell responses generated in vivo by this delivery system are strongly polarized toward Th1 (Dadaglio, G., et al., 2000), and CD8+ T cells activation does not require CD4+ T cell help or CD40 signaling (Guermonprez, P., et al., 2002). Therefore, CyaA appears to be a safe and potent vehicle for in vivo targeted Ag delivery to CD11 bhigh DCs (El Azami El ldrissi, M., et al., 2002) leading to CD8+ T cell priming. Several studies have demonstrated that the generation of optimal CD8+ T cell responses in anti-tumoral prophylactic and therapeutic immunity, as well as against some infectious pathogens, may depend on the simultaneous activation of CD4+ T cells responses (Kern, D. E., et al., 1986; Toes, R. E., et al., 1999; Schnell, S., et al., 2000; Pardoll, D. M., et al., 1998; Wong, P., et al., 2003; Zajac, A. J., et al., 1998). Optimal vaccinal strategies may require the simultaneous delivery of both CD4+ and CD8+ T cell epitopes for T cell priming.
Previously, it has been shown that a MalE CD4+ T cell epitope inserted within the 600 first amino acids of the CyaA is efficiently targeted into MHC class II presentation pathway of APCs and presented to specific T cell hybridoma (Loucka, J., et al., 2002). Co-delivery of CD4+ and CD8+ T cell epitopes into MHC class I and class II-restricted presentation pathways, respectively, can be demonstrated with a recombinant detoxified CyaA carrying the MalE CD4+ T cell epitope and the OVA CD8+ T cell epitope in its AC domain. The capacity of this protein to deliver both epitopes for MHC-peptide complexes formation is also important.
Cattle are an ideal model to test CyaA-based constructs in an actual target species of tuberculosis. CyaA fusion proteins with mycobacterial antigens are candidates not only for subunit vaccines in cattle, but also for diagnostic antigens, particularly when they are recognized in cattle more effectively than conventional recombinant proteins.
The increased efficiency of these fusion proteins results from enhanced sensitivity because they are recognized at lower protein concentrations. The latter consideration can have major cost benefits because it can significantly reduce the amount of antigen that must be produced to implement testing, potentially by several million tests per year.
In general, there is a need in the art for diagnostic reagents for the detection of TB in animals and humans. This need exists so that differential diagnosis of M. bovis infected and vaccinated animals, such as cattle, can be made and effective TB vaccines can be developed. In humans, there is a need for the development of immunodiagnostic methods of greater sensitivity, specificity, and practicality than TST skin testing. Such immunodiagnostic methods will result from methods that allow for enhanced T cell responses to M. tuberculosis.
This invention aids in fulfilling the needs in the art by providing immunodiagnostic methods, especially immunodiagnostic methods carried out in vitro, that allow for enhanced T cell responses to M. tuberculosis, more particularly, this invention provides a novel system for diagnostic testing and immunomonitoring that uses genetically detoxified Bordetella sp. CyaA as a delivery system.
The invention provides methods of diagnostic testing and immunomonitoring with peptides genetically fused or chemically bound to CyaA. The results of tests with recombinant CyaA are quantitative and, therefore, can provide immunomonitoring, as well as simple diagnostic testing.
In one embodiment, the invention is a method of diagnosing or immunomonitoring a disease or immunomonitoring any T cell response following a T cell stimulation by an antigen in an animal comprising: (A) exposing a recombinant protein wherein the recombinant protein comprises a Bordetella CyaA, or a fragment thereof, and a peptide that corresponds to an antigen with which T cells of said mammal are suspected to have been previously stimulated, to a T cell of said animal; and (B) detecting a change in activation of the T cell.
In another embodiment, the invention is a kit for diagnosis or an immunomonitoring test for a disease or immunomonitoring of a T cell response following stimulation of T cells by an antigen in an animal comprising: (A) a recombinant protein wherein the recombinant protein comprises a Bordetella CyaA, or a fragment thereof, and a peptide that corresponds to an antigen with which T cells of said animal are suspected to have been previously stimulated, and (B) reagents for detecting a change in the activation of the T cell.
In embodiments of the invention, the recombinant protein comprises one or more peptides that correspond to one or more antigens.
In embodiments of the invention, the Bordetella CyaA is from Bordetella pertussis, Bordetella parapertussis, or Bordetella bronchiseptica.
In embodiments of the invention the diagnostic tests and immunomonitoring strategies can be for human or animal diseases, for example, but not limited to, cattle diseases.
In particular embodiments of the invention the disease is an infectious disease, such as tuberculosis, or is a cancer, such as melanoma.
In embodiments of the invention, the recombinant protein is CyaA-ESAT-6 or CyaA-CFP10.
In embodiments of the invention, the antigen for which the test is employed can include, but is not limited to, an infectious agent, an allergen, or an antigen from a cancer cell, such as a melanoma.
This invention will be described in greater detail with reference to the Figures in which:
Cattle with TB Infection
Human Individuals with TB Infection
As shown in the Examples, T cells previously stimulated by a given antigen can be restimulated in vitro by the same antigen comprised in CyaA. Based on this discovery, the invention includes diagnostic tests and immunomonitoring for TB by providing a delivery system, which can deliver the M. tuberculosis immunodominant proteins ESAT-6 and CFP-10, as well as other proteins, as proteins comprising CyaA.
The invention also provides a simplified whole blood model to detect tuberculosis infection, wherein the frequency of positive responses to CFP-10 is increased by CyaA delivery (p=0.021). This increased frequency of positive response is an important attribute that can help identify latent infection in at risk populations, and thus facilitate better prevention of active tuberculosis.
The invention has been shown to be effective for improved diagnostic testing and immunomonitoring in animals such as cattle, as well as humans. Specifically, bovine T cells recognize CyaA fusion proteins with ESAT-6 or CFP-10 in vitro.
In addition, the invention provides for diagnostic tests and immunomonitoring for diseases other than TB. The AC domain of CyaA can co-deliver a CD8+ T cell epitope, OVA, and a CD4+ T cell epitope, MalE, into BMDCs MHC class I and class II presentation pathways, respectively. As these epitopes are not from TB, they demonstrate the utility of the non-TB embodiments of the invention. Upon CyaA delivery, there is a strong potentiation of the CD4+ T cell peptide presentation as compared to the MalE protein, which is abrogated by blocking CyaA interaction with its receptor by anti-CD11b mAbs. After its internalization, the AC domain is processed along either conventional endocytic routes or a cytoplasmic route to generate MalE and OVA peptides, respectively. In vivo, CyaA induces specific Th1 CD4+ and CD8+ T cell responses against MalE and OVA epitopes.
Therefore, the CyaA delivery system is useful for novel diagnostic tests because it targets DCs, delivers MHC I and II-restricted T cell epitopes for efficient presentation, and induces Th-1 polarized CD4+ T cell and robust CTL response in vivo.
As used herein, the term “immunomonitoring” refers to tracking the progression of or recovery from a disease with immunological assays. It refers to testing the immune responses, especially T cell responses of mammals, after stimulation by an antigen. For instance, immunomonitoring refers to testing the T cell response of vaccinated individuals, for example in clinical trials. Testing the immune response according to the invention, is especially carried out in vitro, on a biological sample. The invention is especially directed to diagnostic and immunomonitoring of tumor evolution including a tumor clearance in a human patient or in an animal, as a result of immunomonitoring of the T cell response. It is especially indicated that T cell response monitoring is in some instances of tumor immunomonitoring more appropriate than monitoring of B cell response.
As used herein, the term “antigen” refers to a heterologous peptide that can elicit an immune response. In specific embodiments, an antigen or molecule of interest is a heterologous antigen. As used herein, the term “heterologous” refers to an antigen derived from the antigen of a species other than the CyaA that is used in the vector or from an antigen of a species identical to the CyaA that is used in the vector but said antigen located in CyaA in a location where it does not naturally occur.
As used herein, the term “epitope” refers to the minimal peptide sequence of an antigen that can elicit an immune response.
As used herein, the terms “a peptide that correspond to an antigen” or “a peptide of an antigen” encompass an antigen, an epitope, or an antigen or an epitope flanked by naturally or non-naturally present flanking regions which, for example, specifically enhance antigen/epitope processing by the antigen presenting cells.
The term “restimulated” refers to the T cells of the claimed method, which were originally stimulated by the antigen upon infection, vaccination, or other exposure to antigen, especially in vivo, and are stimulated again, in vitro, in the method of the invention. The “restimulation” test according to the present invention relies on the fact that in the tested biological sample, T cells which are contacted with a determined antigen, can “respond” to this antigen (e.g., by significantly producing a cytokine, e.g., interferon) only if the patient providing the sample has previously been in contact with the agent (including infectious, tumoral or other pathogenic agent) carrying said antigen.
It has been shown in the present invention, that the recombinant protein used that comprises adenylate cyclase (CyaA) or a fragment thereof, elicits a significant increase in sensibility, to the “restimulation” test of the invention.
As used herein, the term “immunogenic” refers to a characteristic of a protein as being able to elicit an immune response.
The term “Bordetella sp. CyaA” or “Bordetella CyaA” refers to the adenylate cyclase toxoid of a pathogen of Bordetella species. Such a Bordetella CyaA can be from Bordetella pertussis, Bordetella parapertussis, or Bordetella parapertussis.
The terms “Bordetella CyaA” or “Bordetella adenylate cyclase” encompass Bordetella CyaA protein, or a fragment thereof, either modified or not, but in which the specific binding to CD11b/CD18 receptor and the process of translocation of the catalytic domain are not affected. For example, Bordetella CyaA can be modified in order to be detoxified.
The term “peptide” refers to a series of amino acids linked by amide bonds, comprising at least three amino acids and preferably more than six amino acids.
The term “tumor antigen” refers to a substance from a tumor that elicits an immune response and reacts specifically with antibodies or T cells.
The antigen portion of the recombinant protein used in the tests of the invention can be localized to any permissive site of the CyaA adenylate cyclase toxoid (see WO 93/21324). In addition, the invention encompasses tests that utilize only fragments of the CyaA adenylate cyclase in the recombinant protein (see EPO 03/291,486.3, which corresponds to U.S. Pat. Nos. 5,503,829, 5,679,784 and 5,935,580; see also El-Azami-El-Idrissi, et al., 2003, Interaction of Bordetella pertussis Adenylate Cyclase with CD11b/CD18, J. Biol. Chem., vol. 278, pp. 38514-21).
The antigen of the invention can be fused or chemically bound to CyaA (PCT/EP01/11315).
As used herein the term “fragment of the CyaA adenylate cyclase” relates to a fragment of said protein, including the CyaA protein wherein one or several amino acids which are not in the terminal parts have been deleted and the desired functional properties of the adenylate cyclase toxin are not substantially affected, i.e. the domains necessary for the specific binding to CD11b/CD18 receptor and the process of translocation of the catalytic domain are not affected. For example, a CyaA wherein the amino acids 224 to 240 have been deleted.
As used herein, the term “permissive site” relates to a site where the heterologous peptide can be inserted without substantially affecting the desired functional properties of the adenylate cyclase toxin, i.e. without affecting the domains necessary for the specific binding to CD11b/CD18 receptor and advantageously without affecting the process of translocation of the catalytic domain.
Permissive sites of the Bordetella pertussis adenylate cyclase include, but are not limited to, residues 137-138 (Val-Ala), residues 224-225 (Arg-Ala), residues 228-229 (Glu-Ala), residues 235-236 (Arg-Glu), and residues 317-318 (Ser-Ala) (see Sebo et al., 1985). The following additional permissive sites are also included in embodiments of the invention: residues 107-108 (Gly-His), residues 132-133 (Met-Ala), residues 232-233 (Gly-Leu), and 335-336 (Gly-Gin). (See generally, Glaser et al., 1988 Bordetella pertussis adenylate cyclase: the gene and the protein, Tokai J. Exp. Clin. Med., 13 Suppl.: 239-52.)
The invention encompasses diagnostic tests and immunomonitoring systems that detect any change caused by the activation of T lympocytes. These changes include, but are not limited to changes in IL-2, IL-4, IL-5 or IFN-γ production.
The invention also encompasses diagnostic tests and immunomonitoring systems wherein the test sample can be peripheral blood mononuclear cells (PBMC), whole blood, or fractions of whole blood, for example.
The diagnostic tests and immunomonitoring systems of the invention include, but are not limited to, detection methods such as the ELISPOT assay and ELISA, or other assays using antibodies, assays using tetramers and any other assay to detect T cell activation.
Yet other embodiments of the invention include the nucleotide sequences of the inserts of the plasmids pT7CACT336/ESAT-6 and pT7CACT336/CFP-10. These plasmids were prepared as follows: The open reading frames of Mycobacterium tuberculosis H374v genes esat-6 and cfp-10 were amplified by PCR with the primers shown Table 1 and using as template the pYUB412 cosmid clone of RD1 region (Gordon, et al. 1999). The PCR product was digested by BsrG I at the sites incorporated into the PCR primers and the purified fragments encoding the antigens were inserted in-frame between codons 335 and 336 of cyaA of the pT7CACT-336-BsrG I expression vector (Osicka, et al. 2000). The exact sequence of the cloned inserts were verified by DNA sequencing. Escherichia coli XL1-Blue (Stratagene) was used throughout this work for recombinant DNA construction and for expression of antigens inserted into CyaA. Bacteria transformed with appropriate plasmids derived from pT7CACT1 (Gordon et al. 1999) were grown at 37° C. in Luria-Bertani medium supplemented with 150 μg of ampicillin per ml.
Plasmid pT7CACT336/CFP-10 was deposited on Nov. 18, 2003, at C.N.C.M. under the accession number 1-3135. Plasmid pT7CACT336/ESAT-6 was also deposited on Nov. 18, 2003, at C.N.C.M., Paris, France, under the accession number 1-3136.
In addition, plasmid XL1/pTRACES5-Tyros369, expressing CyaA-Tyr, was deposited on May 31, 2003, at C.N.C.M. under accession number 1-2679. Plasmid pTRACE-5-Tyros369 is a derivative of the expression vector PTRACG that expresses the cyaC and cyaA genes from Bordetella pertussis under the control of the λ phage Pr promoter (PTRCAG also harbors an ampicillin resistance selectable marker and the thermosensitive λ repressor CI857). In pTRACE5-Tyros369, the cyaA gene is modified by insertion of a dipeptide Leu-Gln between codons 188 and 189 of wild-type CyaA (resulting in the inactivation of the adenylate cyclase activity) and by insertion of a DNA sequence encoding the following peptide sequence PASYMDGTMSQVGTRARLK (SEQ ID NO: 5) inserted between codons 224 and 240 of CyaA. The underlined peptide (YMDGTMSQV) (SEQ ID NO: 6) corresponds to the amino acids sequence 369-377 of tyrosinase. Plasmid XL1/pTRACES-GnTV, expressing CyaA-GnTV, was deposited on Oct. 16, 2003, at C.N.C.M., Paris, France, under accession number 1-3111. Plasmid pTRACE5-GnTV is a derivative of the expression vector pTRACG that expresses the cyaC and cyaA genes from Bordetella pertussis under the control of the λ phage Pr promoter (PTRCAG also harbors an ampicillin resistance selectable marker and the thermosensitive A repressor CI857). In pTRACE5-GnTV, the cyaA gene is modified by insertion of a dipeptide Leu-Gin between codons 188 and 189 of wild-type CyaA (resulting in the inactivation of the adenylate cyclase activity) and by insertion of a DNA sequence encoding the following peptide sequence PASVLPDVFIRCGT (SEQ ID NO: 7) inserted between codons 224 and 240 of CyaA. The underlined peptide (VLPDVFIRC) (SEQ ID NO: 8) corresponds to the HLA-A2 restricted melanoma epitope NA17-A derived from the N-acetylglucosaminyl-transferase V gene. (G. Dadaglio, et al. (2003) Recombinant adenylate cyclase of Bordetella pertussis induces CTL responses against HLA-A2-restricted melanoma epitope. Int. Immuno.)
Results regarding induction of a T cell response against tumoral antigens are illustrated in a publication Dadaglio G. et al. (International Immunology, 2003, vol. 15, No. 12, pp. 1423-1430). A separate U.S. patent application is filed by Dadaglio et al. on the same day as the present application concerning “Recombinant Adenylate Cyclase Toxin of Bordetella sp. Induces T Cell Responses Against Tumoral Antigens” (attorney docket no. 03495-0327) and claiming U.S. Provisional Application 60/523,632, filed Nov. 21, 2003, for priority. This reference and application are hereby incorporated by reference in regard to induction of a T cell response against tumoral antigens.
The data presented herein shows that the CyaA-CFP-10 fusion protein is recognized in vitro by bovine T cells more efficiently than CFP-10 alone, both with respect to higher maximum values and the reduced antigen concentrations needed to achieve equivalent stimulation. This recognition is CD11b-mediated. CyaA-based fusion proteins can be applied to whole blood IFN-γ tests. Both ESAT-6 and CFP-10 based CyaA fusion proteins are more strongly recognized than their non-fusion protein counterparts. CyaA-CFP10 created increased sensitivity over that created by CFP-10 alone, particularly at the lower test concentration. The Examples provided demonstrate that: CyaA fusion proteins fused to the mycobacterial antigens CFP-10 and ESAT-6 are recognized by bovine T cells and that this recognition is CD11b-mediated. These CyaA-based recombinant fusion proteins are recognized by bovine T cells more efficiently than the corresponding non-fusion proteins, allowing a reduced test concentration. The CyaA-based fusion proteins of the diagnostic tests of the invention can be applied to whole blood IFN-γ tests, and these test formats can be used in the field. The Examples show that these CyaA-based reagents are useful diagnostic reagents in cattle and subunit vaccine candidates in cattle.
Examples 3-5 show that CyaA fusion proteins are recognized in cattle via a CD11b mediated mechanism, as has been described before in the murine system. These CyaA fusion proteins that target bovine DC can also be used as subunit vaccines to induce immune responses in vivo in a similar manner, as has been described in mice. However, the unique sensitivity and specificity of ESAT-6 and CFP-10 as immuno-diagnostic reagents must be considered if they are to be used for subunit vaccination.
As demonstrated by the data shown in Examples 3-5, CyaA based fusion proteins are in vitro diagnostic reagents detecting bovine tuberculosis in cattle. The practicality of their use can be determined in large numbers of cattle with bovine tuberculosis collected from farms (field reactors), as well as cattle from herds free of bovine tuberculosis, in order to determine their sensitivity and specificity, respectively, in the field. Another determination of the practicality of those reagents in large-scale field applications is the ease with which they can be produced in large quantities and their production costs as compared to conventional recombinant proteins or synthetic peptides.
CyaA toxoids carrying ESAT-6 or CFP-10 were able to restimulate T cells from over 91.1% of TB patients and healthy sensitized donors. Delivery of antigen by CyaA decreased by 10 fold the amount of ESAT-6 and CFP-10 required to restimulate T cells and, in low responders, the overall frequency of IFN-γ producing cells detected was increased. Delivery of these antigens by CyaA enhanced the response of both CD4+ and CD8+ T cells and this response could be blocked by inhibition MHC Class II or classical MHC Class I antigen processing respectively. Antigen processing of toxoids was required as a simple mixture of CyaA carrier and ESAT-6 did not enhance the response. In addition, CD4 recognition of toxoids was sensitive to inhibition by chloroquine. In a simplified whole blood model to detect LTBI the frequency of positive responses to CFP-10 was increased by CyaA delivery, a potentially important attribute that could help identify LTBI in at risk populations, thus facilitating the better prevention of active infectious TB.
The data provided in the Examples are consistent with the interpretation that antigen delivery by CyaA increases the availability of processed M. tuberculosis derived peptide to nascent MHC molecules. It is known that CyaA toxoids become accessible to proteosomic cleavage in the cytoplasm processing as CyaA is specifically taken up via CD11b/CD18 (Guermonprez, P., et al. 2001). In vivo, CyaA has been demonstrated to be delivered efficiently to the cytosol of dendritic cells (Guermonprez, P., et al., 2001). Thus, CD8 responses are more readily detected when comparing the response to soluble recombinant antigen because it is typically processed in the endosome and, thus, less accessible to MHC Class I. CD8+ T cells potentially contribute to the human protective response against tuberculosis (Pathan, A. A., et al. 2000; Lalvani, A., et al., 1998), but the detection of antigen specific responses has so far been limited by the necessity to use peptide pools or recombinant Vaccinia viruses that express the antigen of interest (Pathan, A. A., et al. 2000; Lalvani, A., et al., 1998; Wilkinson, R. J., et al., 1998). Delivery of antigens by CyaA represents a novel method by which the response of CD8+ T cells to whole M. tuberculosis proteins can be assayed.
The response of M. tuberculosis specific CD4+ T cells was also enhanced, consistent with previous findings (Loucka, J., et al., 2002). Enhancement of the response to antigens fused to CyaA was especially pronounced in donors who have a low response to free antigen. This can be because soluble recombinant antigen is less efficiently taken up by pinocytosis (and thus less available for endosomal processing) than the macromolecular CyaA antigen conjugate that binds specifically to the CD11b/CD18 integrin receptor of antigen presenting cells (Guermonprez, P., et al., 2001) whereupon it is rapidly endocytosed (Loucka, J., et al. 2001). This would explain why on a molar basis 10-20 fold less toxoid antigen could restimulate the same response (
Several years ago replacement of the TST by a test that assays the in vitro production of IFN-γ produced by T cells in response to defined M. tuberculosis antigens was discussed (Jurcevic, S., et al., 1996). This approach has been refined and improved by incorporation of the highly immunogenic RD1 encoded antigens ESAT-6 and CFP-10 (Sorensen, A. L., et al., 1995; Berthet, F. X., et al., 1998). Several studies have shown that in vitro responses to RD1 encoded antigens differentiate immune sensitization by BCG from infection by pathogenic mycobacteria (Arend, S. M., et al., 2000; Cockle, P. J., et al., 2002, Lalvani, A., et al., 2001). The IFN-γ ELISPOT response to multiple peptides of ESAT-6 can be utilized to detect latent or overt tuberculosis infection with a sensitivity of 96% and a specificity of 92%. (Lalvani, A., et al., 2001). The very high frequency of recognition of the ESAT-6 and CFP-10 antigen toxoids that are observed in M. tuberculosis sensitized subjects closely accords with these estimates. The more practical approach of using antigen stimulated whole blood cultures is unfortunately associated with a fall in sensitivity to 72% (Brock, I., et al., 2001). However, the data suggest that the use of CyaA toxoid as a delivery system may overcome this deficiency. Furthermore, delivery by CyaA best enhanced the detection of IFN-γ in low responding subjects, an attribute that could be an obvious advantage in the setting of HIV.
The results of the study demonstrate that the AC domain of the CyaA is delivered in vitro into both MHC class I and class II-restricted presentation pathways of BMDCs. A high potentiation of class II presentation by CyaA-MalE was observed as compared to the presentation of MalE protein or MalE peptide. This potentiation is dependent on CD11b-CyaA interaction as it is blocked by anti-CD11b mAbs. Using drugs and TAP1 deficient BMDCs in presentation assays, it is clear that after receptor-mediated endocytosis, the AC domain of CyaA is either translocated into the cytosol of BMDCs to be processed along conventional MHC class I processing routes or is degraded along the endocytic route of processing. In vivo, CyaA simultaneously delivers MalE and OVA peptide for CD4+ and CD8+ T cell priming and induces CTL against OVA peptide and Th-1 cytokine production specific for both MalE and OVA epitopes.
MalE CD4+ T cell epitope inserted into the AC domain of the genetically detoxified adenylate cyclase of Bordetella pertussis is very efficiently presented by BMDCs to CRMC3, a MalE-specific CD4+ T cell hybridoma. The MHC class II-restricted presentation obtained is 100 times more efficient than the presentation observed with an equivalent concentration of the purified MalE protein. It is well demonstrated that even when APCs are incubated with high concentrations of exogenous Ag, only a few MHC class II molecules present the peptides derived from that Ag (Lich, J. D., et al., 2000). Here, the potentiation of MHC class II epitope delivery by CyaA into endocytic pathway is abrogated when the interaction of CyaA with its cellular receptor CD11b is blocked by anti-CD11b mAbs. These results show that the interaction of CyaA with CD11b promotes the generation of MHC class II-restricted peptides for presentation to T cell hybridoma. The potentiation of MHC class II presentation is still observed with CyaA bearing both MalE and OVA CD8+ T cell epitopes. In this case, CyaA simultaneously delivers the OVA and MalE epitopes into their respective presentation pathway as efficiently as CyaAs carrying only one of these epitopes.
It has been repeatedly shown that CyaA delivers its N-terminal AC domain into target cell cytosol by a translocation that is thought to be direct and followed by AC domain processing along conventional cytosolic pathway (Ladant, D., and A. Ullmann. 1999). As CyaA AC domain is also very efficiently delivered into MHC class II presentation pathway, the processing mechanism implicated in such dual delivery was analyzed. Several studies have reported that MHC class II-restricted presentation of peptides derived from cytosolic Ags can be generated by alternative processing pathway (Rudensky, A., et al., 1991; Mukherjee, P., et al., 2001). However, the MHC class II processing of CyaA AC domain does not require proteasome activity nor TAP transporters, but is performed by endocytic proteases that are activated after vesicle acidification. These results also confirm that MalE peptide presentation requires MHC class II molecules neosynthesis. Therefore, AC domain processing for MHC class II-restricted presentation occurs along the conventional endocytic route.
This result and the CD11b requirement for CyaA presentation suggests that this toxin may enter the cell also by receptor-mediated endocytosis, followed either by the rapid translocation of AC domain from vesicles to target cell cytosol or by the degradation of this domain along endocytic vesicles. Alternatively, the AC domain is either directly translocated from cell membrane into the cytosol or taken up to enter the vesicles of the endocytic pathway. CyaA uptake does not require phagocytosis, macropinocytosis, nor caveolae-mediated endocytosis. However, the results show that MHC class II-restricted presentation of CyaA depend on CyaA internalization through a receptor-mediated endocytosis. This suggests that CD11b-mediated endocytosis is one of the mechanisms of CyaA entry into target cells. CyaA AC domain can then translocate into the cytosol to be further processed along MHC class I presentation pathway.
CyaA is a very efficient vector that targets CD11b positive cells (Guermonprez, P., et al., 2001 and 2002); and delivers peptides into MHC class I presentation pathway. This targeted delivery was shown to induce protective CTL in vivo (Fayolle, C., et al., 2001). In this study, CyaA in vivo co-delivers OVA and MalE T cell epitopes into their respective presentation pathway and induces CD8+ and CD4+ T cell responses. One injection of 50 μg of the CyaA by i.v. route, without adjuvant, induced CD4+ and CD8+ T cell responses that are polarized toward Th-1. Among CD11b+ cells, the CD11b CD8− DC subset is responsible for the in vivo presentation of CyaA (Guermonprez, P., et al., 2002). This murine DC subpopulation has been reported to be the most efficient in CTL induction (Schlecht, G., et al., 2001; Ruedl, C., et al., 1999) but also to bias the CD4+ T cell responses mostly towards Th-2. However, after activation by certain microbial compounds, this DC subset acquires the capacity to induce Th-1 T cell responses (Manickasingham, S. P., et al., 2003; Boonstra, A., et al., 2003). As shown in the Examples, the T cell responses induced are strongly polarized towards Th-1, suggesting that CyaA may promote DC maturation in addition to delivering the inserted epitopes. Additional studies can explain the nature of the signal that allows CyaA to generate Th-1 responses.
The simultaneous induction of robust Th-1 CD4+ and CTL CD8+ T cell responses is one goal of vaccination. Indeed, most of the subunits vaccines used at this time generate Th-2 polarized CD4+ T cell responses. In infectious diseases induced by viruses or intracellular pathogens as well as in anti-tumoral immunity, CD8+ T cell and Th-1 CD4+ T cell type of responses are required. CyaA is able to generate both CD4+ and CD8+ T cell responses very efficiently. Therefore the potentiation of MHC class II presentation observed combined with the great efficiency of CyaA in class I epitope delivery into MHC class I presentation pathway render this vector very useful for novel diagnostic tests and immunomonitoring, as well as very promising for vaccine design.
This invention will be described in greater detail in the following Examples.
Bovine (PPD-B) and avian (PPD-A) tuberculin were obtained from the Tuberculin Production Unit at the Veterinary Laboratories Agency—Weybridge and used in culture at 10 μg/ml. Recombinant ESAT-6 was supplied by Dr. A. Whelan (VLA Weybridge), recombinant CFP-10 was obtained from Lionex Ltd., Braunschweig, Germany. CyaA, CyaA-CFP-10, and CyaA-ESAT-6 was provided by Dr. C. Leclerc, Institut Pasteur, Paris. Identical batches of proteins were used throughout.
M. bovis infected cattle (Vordermeier et al., 1999) Calves were infected with a M. bovis field strain from GB (AF 2122/97) by intratracheal instillation of between 5×103 and 5×104 CFU. Infection was confirmed by the presence of tuberculous lesions in the lungs and lymph nodes of these animals as well as by the culture of M. bovis from tissue collected at the post-mortems performed approximately 20 weeks after the infection. Heparinized blood samples were obtained at least six weeks after infection when strong and sustained in vitro tuberculin responses were observed.
Interferon-gamma ELISPOT assay (Vordermeier et al., 2002). Peripheral blood mononuclear cells (PBMC) were isolated from heparinized blood by Histopaque-1077 (Sigma) gradient centrifugation and cultured in tissue culture medium (RPMI1640 (Life Technologies, Paisley, Scotland, U.K.) supplemented with 5% CPSR-1 (Controlled process serum replacement type-1, Sigma Aldrich, Poole, UK), non-essential amino acids (Sigma Aldrich), 5×10−5 M 2-mercaptoethanol, 100 U/ml penicillin, and 100 μg/ml streptomycin sulphate)). Direct ELISPOTs were enumerated, as described earlier. Briefly, ELISPOT plates (Immunobilon-P polyvinyidenefluoride membranes, Millipore, Molsheim, France) were coated overnight at 4° C. with the bovine IFN-γ specific monoclonal antibody 2.2.1. Unbound antibody was removed by washing and the wells were blocked with 10% FCS in RPM 11640 medium. PBMC (2-5×105/well suspended in tissue culture medium (RPMI1640 supplemented with 5% CPSR-1)) were then added and cultured at 37° C. and 5% CO2 in a humidified incubator for 24 h. Spots were developed with rabbit serum specific for IFN-1 followed by incubation with an alkaline phosphatase-conjugated monoclonal antibody specific for rabbit IgG (Sigma Aldrich). The monoclonal antibody 2.2.1 was kindly supplied by Dr. D. Godson, (Veterinary Infectious Disease Organization, Saskatoon, SK, Canada). The spots were visualized with BCIP-NBT substrate (Sigma Aldrich).
The involvement of CD11b was determined by addition (50 μl/well of ELISPOT plate) of the mouse mAb CC94 and ILA15 (both IgG1, kindly provided by Dr C. Howard, IAH, Compton, UK) to 2×105 PBMC dispensed in 100 μl . After 30 minutes pre-incubation at 37° C., serial dilutions of CyaA-CFP10 was added and the cultures incubated for 24 h as described above, followed by ELISPOT analysis.
CD4+ and CD8+ T cell subpopulations were depleted by magnetic negative selection using the ant-bovine CD4 or CD8 specific mAb CC30 and CC58 (C. Howard, IAH) in conjunction with the MACS system (goat anti-mouse IgG coated beads, LS separation columns, Miltenyi Biotec Ltd, Bergisch-Gladbach, Germany) as described earlier (Vordermeier et al., 2001).
Interferon-gamma (IFN-γ) assay (Wood et al., 1994; Vordermeier et al., 1999). Whole blood cultures were performed in 96-well plates in 0.2 ml/well aliquots by mixing 0.1 ml of heparinized blood with an equal volume of antigen containing-solution. Supernatants were harvested after 24 h of culture and interferon-gamma (IFN-γ) determined using the Bovigam EIA kit (CSL, Melbourne, Australia) (Vordermeier et al., 2002). The data are expressed as OD450 units (OD450×1000). CyaA background levels were subtracted from CyaA-ESAT6 and CyaA-CFP10 values.
Statistical analysis. Statistical analysis was performed using Instat v3.0a (GraphPad, San Diego, Calif., USA) on an iMac personal computer. Data were analysed using the one- or two-tailed Wilcoxon signed rank matched pairs test. See figure legends for further details.
Escherichia coli XL1-Blue (Stratagene) was used for recombinant DNA construction and for expression of antigens inserted into CyaA. Bacteria transformed with appropriate plasmids derived from pT7CACT1 (Osicka et al., 2000) were grown at 37° C. in Luria-Bertani medium supplemented with 150 μg of ampicillin per ml. The open reading frames of Mycobacterium tuberculosis H37Rv genes esat-6 and cfp-10 were amplified by PCR from the pYUB412 cosmid clone of the RD1 region (Gordon et al., 1999) using the following primers:
The PCR product was digested by BsrG I at the sites incorporated into the PCR primers and the purified fragments encoding the antigens were inserted in-frame between codons 335 and 336 of the cyaA gene open reading frame born on the pT7CACT-336-BsrG I expression vector (Osicka et al., 2000). The exact sequence of the cloned inserts was verified by DNA sequencing.
The control detoxified mock CyaA and the recombinant CyaA proteins carrying the ESAT-6 and CFP-10 antigens, respectively, were produced in E. coli, purified from inclusion bodies by a combination of ion-exchange chromatography on DEAE-sepharose and hydrophobic chromatography on Phenyl-sepharose, as described previously (Karimova et al., 1998). In the final step, the proteins were eluted with 8 M urea, 50 mM Tris-Cl pH 8, 2 mM EDTA and characterized as previously described (Karimova et al., 1998). The resulting proteins were free of any detectable adenylate cyclase enzymatic activity.
PBMC were prepared from experimentally infected cattle and incubated with serial dilutions of antigens (recombinant ESAT-6, CFP-10, CyaA-ESAT6, CyaA-CFP10, and CyaA control). The antigen-induced IFN-γ responses were determined after 24 h culture using a sensitive ELISPOT assay. The number of spot-forming cells (SFC) found without antigen added (medium controls) were subtracted, the number of SFC obtained after CyaA stimulation were subtracted from the number of SFC induced after CyaA-ESAT6 and CyaA-CFP10 stimulation. To illustrate how the data were subsequently expressed and compared, a representative result for CFP-10 tested in one calf is given in
Subsequently a further batch of six experimentally M. bovis infected calves were tested and the results interpreted identically. As demonstrated by the comparison of peak responses induced by CFP-10 or CyaA-CFP10, and the reduced concentration needed for 50% maximal responses, the CyaA-CFP10 fusion protein was superior to its non-fusion counterpart (
The IFN-γ responses induced by ESAT-6 or the CyaA-Esat-6 fusion proteins were not significantly different from each other (
To determine whether the recognition of CyaA-CFP10 is mediated via a CD11b-dependent mechanism (as has been recently shown for mice), PBMC from an infected calf were stimulated with CyaA-CFP10 in the presence of two mAb of the same isotype (IgG1) specific for bovine CD11b (kindly provided by Dr C. Howard, IAH, Compton, UK). One of these mAb, (ILA15) interfered with the interaction of CyaA-CFP10 with CD11b as the number of SFC was reduced significantly, whereas the non-blocking isotype control mAb (CC94), did not (
CyaA-CFP10 were also shown to be recognized by both CD4+ and CD8+ T cells. This was analysed by depleting either sub-population with magnetic beads. Cattle at early stages of bovine tuberculosis display only weak or undetectable CD8+ T cell responses (Pollock et al., 1996, Vordermeier, unpublished observation). Consequently, all of the experimentally infected animals available for this study were tested relatively early following infection (i.e. approximately 4-6 months post-infection), and significant PPD-B and CFP-10-specific CD8+ T cell responses were observed in only in one of four cows tested. Nevertheless, the results obtained from the adult cow that was infected several years previously, indicated that CyaA-CFP10 was recognized by CD4+ and CD8+ T cells, as was the recombinant protein. However, CyaA-CFP10 induced higher in vitro CD8+ T cell responses compared to the recombinant protein (CD8/CD4 ratio of responding cells: CFP-10: 0.8, CyaA-CFP10: 1.05, data not shown), though this difference was not statistically significant.
The IFN-γ test was applied as diagnostic assay in the field in the format of a whole blood assay (BOVIGAM test). In this format, blood collected on farms was heparinized and incubated with either tuberculins or specific antigens. After a 24 h incubation period, the amount of antigen-induced IFN-γ in plasma supernatants was determined by ELISA. To determine the performance of the CyaA fusion proteins with ESAT-6 and CFP-10, blood was obtained from a second batch of eight experimentally M. bovis infected calves. These blood samples were stimulated with ESAT-6, CFP-10, CyaA-ESAT6, and CyaA-CFP10 at 4 and 20 nM concentrations. The results of the ELISA assay conducted 24 h later are shown in
When the diagnostic outcome was evaluated using the commonly applied cut-off of 100 OD450 units, six of eight tested animals were deemed positive for bovine TB using ESAT-6 and CyaA-ESAT6 applied at both test concentrations (
One of the eight test negative animals presented without tuberculous lesions at a post-mortem carried out several months after this experiment was performed, though M. bovis could be cultured from any tissue samples taken. This animal was also tuberculin skin test negative. Taken together, this suggests that the experimental infection in this animal was contained and did not result in disease. As expected, no IFN-γ was induced in the blood of this calf after stimulation with either PPD-B, CyaA-ESAT-6, CyaA-CFP10, ESAT-6, or CFP-10, thus highlighting the specificity of these reagents (
Human studies were conducted with ethical approval from the Harrow Local Research Ethics Committee (Harrow LREC 1646 and 2414). Patients with tuberculosis and their healthy contacts were recruited from Northwick Park Hospital, Harrow (North West London Hospitals NHS Trust). Three groups of people with distinct clinical phenotypes were selected. The first group was adults with overt (i.e. culture or biopsy positive) tuberculosis (n=21, 14 M, 7 F, average age 35.1 years). The second group consisted of asymptomatic adults with normal chest radiographs who nevertheless exhibited strongly positive TST reactions (Heaf Grade 3 and above) and were thus thought likely to have LTBI (n=44, 26M, 18F, average age 34.7 years). A third control group consisted of healthy adults with no documented exposure to TB and whose skin test reactions were negative (n=7, 3M, 4F, average age 37.6 years). The first two groups were chosen to maximize the chances of T cell reactivity to M. tuberculosis specific antigens and thus allow the comparison of response to recombinant and CyaA toxoids. All subjects were subsequently advised and, if indicated, treated according to British Thoracic Society guidelines (see Thorax 55:887-901, 2000).
Cells. PBMC were separated from 20 mls of blood by centrifugation over Ficoll-Paque Plus (Pharmacia, Uppsala, Sweden), and suspended in RPMI supplemented with 2 mM L-glutamine, penicillin 100 U/ml, gentamicin 5 μg/ml and 10% heat-inactivated fetal calf serum (Sigma, St. Louis, Mo.) (R10). CD4+ and CD8+ T cells were depleted using anti-CD4 or anti-CD8 mAb conjugated to ferrous beads (Dynabeads M-450, Dynal, Oslo, Norway) according to the manufacturer's instructions. These depletions consistently yielded cells populations with 97-99% purity. Anti MHC Class II blocking antibody (L243, Leinco Technologies), anti MHC Class I blocking antibody (W6/32, Leinco) and isotype control antibody (Mouse IgG2a, Leinco) were used at 5 μg/ml 30 minutes after addition of antigens. Chloroquine (Sigma) at 10 μg/ml was added to the cultures just before the antigens.
Ex vivo enzyme-linked immunospot (ELISPOT) assay for single cell IFN-γ release. 96-well PVDF-backed plates (MAIPS45, Millipore, Bedford, Mass.), pre-coated with 15 μg/ml of anti-IFN-γ mAb 1-D1 K (Mabtech, Nacka, Sweden), were blocked with R10 for 2 hrs. 3×105 PBMC were added in 100 μl R10/well. Duplicate wells of CyaA toxoids and recombinant ESAT-6 and CFP10 were used at the optimum concentrations derived from
Recombinant antigen and CyaA toxoid construction. Recombinant native ESAT-6 was prepared as previously described and was a gift from the Veterinary Laboratories Agency, Weybridge, Surrey KT15, UK. Recombinant CFP-10 was obtained commercially from Lionex (Braunschweig, Germany). N-terminal sequencing confirmed the identity of the cloned antigen. Escherichia coli XL1-Blue (Stratagene) was used throughout this work for recombinant DNA construction and for expression of antigens inserted into CyaA. Bacteria transformed with appropriate plasmids derived from pT7CACT1 (Osicka, R., 2000) were grown at 37° C. in Luria-Bertani medium supplemented with 150 μg of ampicillin per ml. The open reading frames of Mycobacterium tuberculosis H37Rv genes esat-6 and cfp-10 were amplified by PCR from the pYUB412 cosmid clone of the RD1 region (Gordon, S. V., et al., 1999) using the following primers:
The PCR product was digested by BsrG I at the sites incorporated into the PCR primers and the purified fragments encoding the antigens were inserted in-frame between codons 335 and 336 of CyaA on the pT7CACT-336-BsrG I expression vector (Osicka, et al., 2000). The exact sequence of the cloned inserts was verified by DNA sequencing.
The control detoxified mock CyaA and the recombinant CyaA proteins carrying the ESAT-6 and CFP-10 antigens, respectively, were produced in E. coli, purified from inclusion bodies in 8 M urea, 50 mM Tris-Cl pH 8, 2 mM EDTA and characterized as previously described (Sebo, et al., 1999). The resulting proteins were free of any detectable adenylate cyclase enzymatic activity.
Whole blood assay and Interferon-γ ELISA. Venous blood was collected (BD Na Heparin vacutainer, Cat 368480) and processed within 4 hours of sampling. Whole blood was diluted 1:10 in RPMI (supplemented with glutamine and penicillin/streptomycin). 180 μl of the diluted blood was plated in 96-welled round-bottomed plates with stimulating antigens in duplicate wells. The final concentrations of the antigens were: 250 nM (rESAT-6), 50 nM (CyaA-ESAT-6), 500 nM (rCFP-10), 50 nM (CyaA-CFP-10), 50 nM (mock CyaA toxoid), 5 μg/ml (PHA: positive control) and 20 μl/ml (RPMI: negative control). Stimulated whole blood was cultured at 37° C. in a CO2 incubator. Supernatants from duplicate wells were harvested after 60-72 hours of culture, pooled and immediately frozen for later IFN-γ measurements by ELISA. The optimal concentrations of stimulants and the timing of harvesting had been previously determined by dose-response and time-course experiments. ELISA reactions were performed in accordance with antibody manufacturers' instructions. Briefly, 96-welled flat-bottomed plates were coated overnight at 4° C. with purified mouse anti-human IFN-γ (BD Pharmigen 554548). After blocking and washing, wells were plated with supernatants (1:2 dilutions in duplicate) and standards (standard curve dilutions from 15 pg/ml-10 000 pg/ml, duplicate measurements) after which the plates were again incubated at 4° C. overnight. After washing, the wells were incubated with biotinylated mouse anti-human IFN-γ (BD Pharmingen, 554550) for 1.5 hours at room temperature, washed again and incubated with Streptavidin (Sigma Cat no A3151) for 30 mins. OPD was used as a substrate for detection and 2N H2SO4 to stop color development. Optical densities were read at 490 nm on a plate reader and IFN-γ concentrations were calculated from standard curves. The Spearman rank correlation coefficients between independent variables were calculated using SPSS-10.
The optimum stimulatory dose of ESAT-6 and CFP-10 and the respective CyaA toxoids in vitro was determined, using the equivalent dose of antigen inserted in the recombinant molecule (i.e. the same molar amount of protein) as the CyaA toxoid. The numbers of IFN-γ SFC were then enumerated in an overnight ELISPOT assay. These experiments with CyaA toxoids were controlled by subtracting the number of IFN-γ SFC in wells containing the same amount of CyaA toxoid into which no antigenic stimulus had been inserted (mock toxoid). In nine healthy TST+ve donors who responded to ESAT-6, optimal recognition of this molecule was observed at a dose of 500 nM. Ten fold less ESAT-6 (50 nM) was required when the antigen was presented as CyaA-ESAT-6 (
Interestingly, increasing the dose of CyaA-ESAT-6 to 500 nM indicated that delivery by CyaA vector could lead to an overall increase in IFN-γ SFC detected. However, further increase in the dose of CyaA toxoid was associated with a decrease in SFC that was due the high urea content of solutions necessary to solubilize the CyaA toxoid (data not shown). In ten similar donors who responded to rCFP-10, CyaA fusion similarly shifted the dose response curve to the left. Approximately 10-20 times less CFP-10 expressed as a CyaA toxoid elicited the same response as native antigen (
Based on the results shown in
When the subjects' responses were analyzed according to their response to the recombinant antigens into low (<50 IFN-γ SFC/106 PBMC), intermediate (>50<100 SFC/106 PBMC) or high responders (>100 SFC/106 PBMC), clear enhancement in the group of low responders was seen. Thus, the average number of detected IFN-γ SFC/million increased from 26.1±2.7 to 48.2±7.3 (n=27, p=0.009) for ESAT-6 and from 17.5±2.6 to 36.8±4.3 (n=34, p=0.0002) in the case of CFP-10 (
In order to define the T cell subset that recognized the CyaA toxoids, populations were enriched by performing prior immunomagnetic depletion of either CD4+ or CD8+ T cells from PBMC. The remaining cells were set up in the ELISPOT assays and stimulated overnight with the ESAT-6 or CFP-10 and detoxified CyaA incorporating the same antigens. Eight donors were tested for the CyaA-ESAT-6 and five donors for the CyaA-CFP-10 and the corresponding recombinant antigens. Both CD4+ and CD8+ responses were seen to the recombinant antigens, the CD4+ response being dominant (
To determine whether the M. tuberculosis protein has to be covalently linked to CyaA, the effect of mixing ESAT-6 with mock CyaA toxoid was tested. PBMC from four subjects were set up with ESAT-6 (500 nM), CyaA-ESAT-6 (50 nM) or the mixture of rESAT-6 (500 nM) and CyaA (50 nM). The median IFN-γ SFC/million for these stimulants was 113, 147 and 58 respectively, showing that covalent linkage between the antigen and carrier is required for enhancement to occur (data not shown). In fact, it appeared that simple mixture of rESAT-6 with CyaA might have actually decreased the response to rESAT-6.
Next, whether the CD4+ and CD8+ T cells that recognized CyaA toxoid were classically MHC Class I or Class II restricted was determined. CD4 or CD8 depleted cells were set up on the ELISPOT plates and stimulated with CyaA-ESAT-6 (5 donors) or CyaA-CFP-10 (4 donors). Anti MHC Class I blocking antibody, anti MHC Class II blocking antibody, or isotype control (all at 5 μg/ml) was added to selected wells and the plates were incubated overnight. The IFN-γ response of CD4 depleted (interpreted as CD8) T cells in response to CyaA-ESAT-6 and CyaA-CFP-10 was 48% and 83% inhibited by anti MHC Class I antibody respectively (
Detection of IFN-γ secreted into the supernatant of whole blood cultures requires less blood and is potentially more applicable to field conditions than the ex-vivo IFN-γ ELISPOT assay. However, it appears that such whole blood assays, while retaining specificity, are less sensitive than ELISPOT detection. Therefore, it was examined whether the enhanced response to CyaA toxoids carrying ESAT-6 or CFP-10 could compensate this deficiency. Thirty-three patients and healthy sensitized subjects were tested in parallel using the two read-out assays for IFN-γ production. All 33 donors responded to rESAT-6, and 31 donors responded to rCFP-10. The ELISPOT and whole blood IFN-γ responses to CyaA-ESAT-6 and CyaA-CFP-10 were positively correlated (r=0.58 and 0.64 respectively, p<0.001 in both cases,
Splenocytes of C57BL/6 mice infected (s.c. or i.v.) with 1×106 or 1×107 CFU of a BCG strain, stably complemented with the RD1 chromosomal region of M. tuberculosis (referred to as BCG::RD1) (Pym, 2002; Pym, 2003) produced substantial levels of IFN-γ upon in vitro stimulation with ESAT-6:1-20 peptide or r-CyaA-ESAT-6 construct (
Mice, Infection, Immunization. Female, specific pathogen-free BALB/c (H-2d) or C57BL/6 (H-2b) mice (Iffa Crédo, L'Arbresle, France) were used at 6-12 weeks of age. Mice were infected (s.c. or i.v.) with 1×106 or 1×107 CFU/mouse of BCG::RD1 and were maintained in isolators in ABL-3 biohazard conditions in Pasteur Institute's animal facilities. T-cell responses were studied 3-4 weeks post-infection. Mouse immunization with r-CyaA was performed by one or two i.v. injections with 10 or 50 pg of appropriate r-CyaA in PBS. T-cell responses were studied 10-12 weeks post-immunization.
T-cell proliferation and cytokine production assays. Single-cell suspensions of spleen or lymph node cells were plated (1×106 cell/well) onto 96-well flat-bottom plates in synthetic HL-1 medium (BioWhittaker, Walkersville, Md.) complemented with 2 mM L-glutamine, 100 IU penicillin/ml and 100 μg streptomycin/ml in the presence of various concentrations of synthetic peptides (Neosystems, Strasbourg, France) or 1-10 μg/ml of r-CyaA. For lymphoproliferation assays, cultures were pulsed with 1 μCi [methyl-3H]-thymidine (ICN, Orsay, France) for 16 h and cells were harvested for cpm counting.
For cytokine assays, culture supernatants were collected at 48 h for IL-2 detection and at 72 h for the other cytokines. IL-2 was quantified using a standard CTLL-2 bioassay. IL-4, IL-5 and IFN-γ were quantified by a sandwich ELISA using, respectively, BVD4-1 D11, TRFK5 and R4-6A2 as capture monoclonal antibodies and biotin-conjugated BVD6-24G2, TRFK4 and XMG1.2 monoclonal antibodies (BD PharMingen, San Diego, Calif.). Standard curves were obtained with recombinant murine cytokines (BD PharMingen).
Mice. Female C57BL/6 (H-2b) mice from Iffa Credo (L'Arbresle, France) were used between 6 and 10 weeks of age. Female TAP1 knockout mice (Van Kaer, L., et al., 1992) onto a C57BL/6 background were a gift from A. Bandeira (Institut Pasteur, Paris, France) and were bred in our animal facilities.
Peptides and proteins. The synthetic peptides SIINFEKL and NGKLIAYPIAVEALS, corresponding respectively to the CD8+ T cell epitope encompassing the ovalbumin residues 257-264 (Bevan, M. J., 1976) and to the CD4+ T cell epitope corresponding to E. coli MalE protein residues 100-114 (REF) were purchased from Neosystem (Strasbourg, France). MalE protein was kindly given by J. M. Clement (Institut Pasteur) and ovalbumin was purchased from Sigma (Saint-Quentin Fallavier, France). Both were dissolved in PBS at 1 mg/ml.
Construction, production and purification of recombinant CyaA toxins with inserted CD4+ MalE and CD8+ OVA epitopes. To construct the hybrid cyaA alleles encoding the CyaA proteins carrying simultaneously the MalE and the OVA epitopes, appropriate unique restriction sites along the cyaA alleles were used for recombination of cyaA alleles encoded on a set of pT7CACT1-derived plasmids and carrying oligonucleotide inserts encoding for either the CD4+ MalE epitope (Loucka, J., et al., 2002) or the CD8+ OVA epitope (Osicka, R., et al., 2000), respectively. The insertion and the orientation of both oligonucleotides in cyaA gene were verified by restriction analysis of plasmids, the length of the corresponding expressed CyaA proteins was verified by 7.5% SDS-PAGE. The recombinant CyaA used in this study bear the NGKLIAYPIAVEALS sequence between amino acids 108 and 109 (CyaA-MalE), the SIINFEKL sequence between amino acids 336 and 337 (CyaA-OVA), or both sequences in their respective insertion site (CyaA-MalE-OVA). All constructs were genetically detoxified by insertion of a dipeptide sequence between residues 188 and 189.
The E. coli XL-1 Blue strain (Stratagene) was transformed with the constructed plasmids derived from pT7CACT1 and containing the accessory gene cyaC required for post-translational acylation of ACT (Osicka, R., et al., 2000). The cells were grown as described previously (Osicka, R., et al., 2000) and the expression of recombinant proteins was induced by adding of 1 mM IPTG. The CyaA proteins were extracted with 8M urea (Sebo, P., et al., 1991) and purified by DEAE-Sepharose and Phenyl-Sepharose chromatographies (Karimova, G., et al., 1998). The homogeneity of purified toxins was verified by 7.5% SDS-PAGE. Purified recombinant CyaA proteins concentrations were determined by the Bradford method.
CyaA E5, a genetically detoxified CyaA without insert, was kindly provided by D. Ladant (Institut Pasteur) and was used as a negative control.
Culture medium. Complete medium (CM) consisted of RPMI 1640 containing L-Alanyl-L-Glutamine dipeptide supplemented with 10% fetal calf serum (Valbiotech, Paris, France), 5×10−5 M of 2-ME and antibiotics (penicillin 100 U/ml, streptomycin 100 μg/ml).
Cell lines. The H-2b restricted hybridoma CRMC3, specific for the 100-114 sequence of the MalE protein from E. coli was generated in our laboratory as previously described (Lo-Man, R., et al., 2000) and was maintained in CM. B3Z (Karttunen, J., et al., 1992), the CD8+ T cell hybridoma specific for the Kb restricted OVA 257-264 peptide was a generous gift from N. Shastri (University of California, Berkeley, Calif.), and was maintained by adding 1 mg/ml of G418 and 400 μg/ml of hygromycin B to the CM. The EL-4 thymoma was obtained from American Type Culture Collection (Manassas, Va.) and maintained in CM.
BMDC generation. BMDCs were generated from bone marrow precursors as previously described (Inaba, K., et al., 1992). Briefly, bone marrow cells from C57BL/6 or TAP1 knockout mice were harvested, washed, and plated at 2.105 cells/ml in CM with 1% of a GMCSF-containing supernatant. After 3 days of culture at 37° C., 7% CO2, medium was added in the plates. The non-adherent and semi-adherent cells were recovered at day 7 or 8 by flushing the plates with PBS EDTA (5 mM) and washed before use. The recovered cells usually contained 60 to 70% of CD11c positive cells that all expressed CD11b. These BMDCs were CD40lo and CD86lo.
Antigen presentation assays. The stimulation of CRMC3 or B3Z T-cell hybridoma (105 cells/well) was monitored by IL-2 release in the supernatants of 18-h cell cultures in the presence of BMDCs (105 cells/well) in 96-well culture plates. In most experiments, BMDCs were pulsed for 4 to 5 hours with proteins or peptides at various concentrations (see legends of the figures) and washed three times before adding 105 T cell hybridoma in 0.2 ml of CM. In the drug inhibition assay, the BMDCs were fixed with 0.05% glutaraldehyde (Sigma) after being pulsed and washed, and then the hybridoma were added. After 18 hours, culture supernatants were frozen for at least 2 hours at −80° C. Then, 104 cells/well of the IL-2 dependent CTL-L cell line were cultured with 100 μl of these supernatants. After 48 hours, [3-H]-thymidine (50 μCi/ml, ICN, Orsay, France) was added to the wells and the cells were harvested 6 hours later with an automated cell harvester (Skatron, Lier, Norway). Incorporated thymidine was detected by scintillation counting. In all experiments, each point was done in duplicate.
Inhibitors and antibodies. Cycloheximide (CHX, used at 5 μg/ml), brefeldin A (BFA, 5 μg/ml), cytochalasin B (CCB, 5 μg/ml), leupeptin (50 μg/ml), pepstatin (50 μg/ml), chloroquine (50 and 150 μM), N-acetyl-L-leucinal-L-norleucinal (LLnL, 12 μg/ml) and N-acetyl-L-leucinal-L-methioninal (LLmL, 12 μg/ml), were all from Sigma-Aldrich (Saint-Louis, Mo.) and were dissolved in appropriate solvent according to manufacturer's advises. Lactacystin (Biomol, research Labs., Inc., Plymouth Meeting Pa.) was dissolved in water at 1 mg/ml and used at 10 μM final. The purified mAbs specific for murine CD11b (M1/70, rat IgG2b,K) and the corresponding isotype control were purchased from Pharmingen (Le Pont de Claix, France) and were used at 10 μg/ml.
Inhibition studies. For inhibition studies, BMDCs were first incubated with the drugs or antibodies for one hour in 0.1 ml of CM at 37° C., 7% CO2. Then, Ags were added in 0.1 ml of CM at the final concentrations indicated in the legends of the figures, in the continuous presence of the inhibitors. In the assays using anti-CD11b or isotype control antibodies, the cells were washed three times after 5 hours of incubation with both Ags and antibodies, and 105 T cell hybridomas were added. In the assays using drugs, the cells were washed after the 5-hours incubation and fixed using glutaraldehyde 0.05% for 2 min at 37° C. (Sigma) and lysine 0.2 M (Sigma). After washing three times, the T cell hybridoma were added to the wells in 0.2 ml CM.
For inhibition of clathrin-mediated endocytosis by K+ depletion following hypotonic shock, DC (105/well) were incubated for 30 min in serum-free synthetic OptiMEM medium (Life Technologies) supplemented with 5.10−5 M 2-ME, 100 U/ml penicillin and 100 μg/ml streptomycin. DCs were then incubated for 5 min in hypotonic medium (OptiMEM medium and ultrapure H2O, 50/50) and finally for 30 ml in K+-free (140 mM NaCl, 20 mM HEPES-NaOH, 1 mM CaCl2, 1 mM MgCl2, 1 mg/ml glucose and 0.5% BSA) or K+-containing (10 mM KCl, 130 mM NaCl, 20 mM HEPES-NaOH, 1 mM CaCl2, 1 mM MgCl2, and 0.5% BSA). Ags were added to the wells at the concentrations indicated in the figure legends and 1 hour later, DCs were washed in PBS and CM was added for 4 hours to allow Ag processing. DCs were washed and fixed as described previously and T cell hybridomas were added to the wells for 18 hours.
Mouse immunization. C57BL/6 mice were i.v. injected with 50 μg of CyaA-OVA, CyaA-MalE, CyaA-MalE-OVA or CyaA E5 diluted in 0.1 ml of PBS.
In vitro cytotoxicity assays. Splenocytes from immunized mice were isolated 7 days after CyaA injection and in vitro restimulated for 5 days with OVA257-264 peptide (1 μg/ml) in the presence of syngeneic irradiated naive spleen cells. The cytotoxic activity was determined in a 5-hour in vitro [51Cr]-release assay as previously described (Fayolle, C., et al., 1996). Briefly, EL4 (H-2b) tumor cells loaded with 50 μM of the OVA257-264 peptide were used as target cells for H-2b effector cells. Various effector to target ratios were used and all assays were done in duplicate. In each assay, EL-4 cells incubated in the absence of the peptide were used as control for nonspecific lysis. [51Cr]-release in each well was counted using a MicroBeta Trilux liquid scintillation Counter (Wallac, Turku, Finland). Percentage of specific lysis was calculated as 100× (experimental release−spontaneous release)/(maximal release−spontaneous release). Maximum release was obtained by adding 10% Triton X-405 to target cells and spontaneous release was determined with target cells incubated in CM.
Cytokine ELISA assay. Splenocytes from immunized mice were restimulated in vitro in the presence or absence of 1 μg/ml of MalE100-114 or OVA257-264 peptides and the culture supernatants were harvested after 72 hours. IL-4, IL-5, and IFN-α concentrations were then measured in these supernatants by a standard sandwich ELISA. Maxisorp plates (Nunc, Roskilde, Denmark) were coated with unconjugated anti-IL-4, anti-IL-5, or anti-IFN-δ capture antibodies (BVD4-1 D 1, TRFK5, R4-6A2 clones respectively, Pharmingen) and detection was done using corresponding biotinylated mAb (BVD6-24G2, TRFK4, XMG1.2 clones, Pharmingen). The plates were developed using streptavidin-HRP (Pharmingen) and o-Phenylenediamine (Sigma-Aldrich) as substrate. All dosages were performed in duplicate. The assays were standardized with recombinant murine cytokines (Pharmingen) and results are expressed in pg/ml.
Using a MalE CD4+ T cell epitope as reporter, it has been previously shown that recombinant CyaA-MalE delivers the NGKLIAYPIAVEALS MalE100-114 peptide into MHC class II presentation pathway of splenocytes (Loucka, J., et al., 2002). The efficiency of this delivery as compared to MalE protein was evaluated next. Using BMDCs, which are CD11b positive (data not shown), the presentation of CyaA carrying the MalE NGKLIAYPIAVEALS CD4+ T cell epitope at position 108 was compared to MalE protein. APCs were incubated with serial dilutions of each protein and I-Ab-NGKLIAYPIAVEALS complexes apparition at their surface was monitored with CRMC3, a CD4+ T cell hybridoma specific for this MHC-peptide complex (Lo-Man, R., et al., 2000). As expected (
To exclude that this potentiation was due to a non-specific stimulatory effect of the CyaA or of some component in the CyaA preparation, BMDCs were incubated with a constant concentration of CyaA E5 or CyaA-OVA and various concentrations of MalE100-114 peptide or MalE protein. The efficiency of 1-Ab-NGKLIAYPIAVEALS complexes presentation to CRMC3 was then monitored. As shown in
In a previous report, it was shown that CyaA carrying three different CD8+ T cell epitopes simultaneously induces in vivo protective CTL responses against these epitopes (Fayolle, C., et al., 2001). Therefore, it was determined whether CyaA could deliver both CD4+ and CD8+ T cell epitopes to BMDCs for Ag presentation to specific T cell hybridoma. Therefore, CyaA-MalE-OVA, a recombinant CyaA bearing both MalE (class II-restricted) and OVA (class I-restricted) epitopes was compared to CyaA-MalE and CyaA-OVA in a presentation assay. To make this comparison possible, the MalE CD4+ T cell epitope was inserted between amino acids 108 and 109 of CyaA-MalE and CyaA-MalE-OVA. The OVA CD8+ T cell epitope was inserted between amino acids 336 and 337 in CyaA-OVA and CyaA-MalE-OVA. To detect the presence of Kb-SIINFEKL complexes on BMDCs B3Z, a CD8+ T cell hybridoma specific for the OVA257-267 peptide (Karttunen, J., et al., 2002) was used. As shown in
The potentiation of MHC class II-restricted presentation on CyaA delivery could be explained by the specific interaction of this protein with its CD11b receptor (Guermonprez, P., et al., 2001), which is expressed on BMDCs. To test this hypothesis, BMDCs were first incubated either with 10 μg/ml anti-CD11b mAbs or with the same concentration of isotype control mAbs. As shown in
Previous studies have demonstrated that CyaA interaction with CD11b results in direct AC domain translocation into target cell cytosol. The subsequent processing of this domain to generate peptides for MHC class I-restricted presentation requires proteasome and is dependent on TAP transporters (Guermonprez, P., et I., 1999). It has been reported that some endogenous Ags are processed in the cytosol for MHC class II presentation by an alternative pathway that requires the proteasome and calpain (Lich, J. D., et al., 2000). The peptides released in the cytosol are then transported into endocytic compartments along a poorly understood mechanism. Therefore, the proteasome requirement of CyaA-MalE-OVA for MHC class II-restricted MalE100-114 peptide presentation to CRMC3 was tested. BMDCs were incubated for one hour with lactacystin, a 20S proteasome inhibitor (Fenteany, G., et al., 1995; Craiu, A., et al., 1997), and the Ags were then added.
As shown in
The effect of LLnL (a cathepsin and proteasome inhibitor) and LLmL (a cathepsin inhibitor) (Rock, K. L., et al., 1997) on MalE100-114 peptide presentation to CRMC3 by BMDCs was then compared. As shown in
To further confirm that the processing of class I and class II epitopes from the AC domain follows distinct pathways after AC domain delivery to APCs, the TAP requirement for CyaA presentation by MHC class II molecules was tested. BMDCs were generated from TAP1 knockout mice and used in a presentation assay to CRMC3. As shown in
After internalization of exogenous soluble Ag, peptide ligands for MHC 11 presentation are generated in endosomes and lysosomes by proteolysis of the proteins by a set of proteases that are sequentially activated (Villadangos, J. 2001). As cathepsin activity is required to generate MalE100-114 peptide presentation after CyaA-MalE-OVA delivery, whether others endocytic proteases are required for I-Ab-MalE100-114 complexes formation was tested. Leupeptin (Umezawa, H. 1976), an inhibitor for serine and cysteine proteases totally blocked MalE100-114 peptide presentation to CRMC3 when CyaA-MalE-OVA was used as Ag, but did not affect the presentation of the free peptide (
Vacuolar acidification is an important factor, which controls the sequential activation of endocytic proteases. Therefore, chloroquine, an inhibitor of endocytic vesicle acidification was used to confirm that MalE100-114 peptide presentation on CyaA delivery occurs after endocytic processing. As shown in
Presentation of MHC II-peptide complexes at APC surface requires the degradation of exogenous Ag but also the association of the generated peptides with MHC class II molecules (Gordon, S. V., et al., 1999). In the classical endocytic pathway, newly synthesized MHC class II molecules are required. These molecules leave the ER through the Golgi and reach the trans-golgi network (TGN) where they are sent towards the endocytic pathway.
To determine whether I-Ab-MalE100-114 peptide complexes generation after CyaA-MalE-OVA delivery requires nascent MHC class II molecules, cycloheximide (CHX), an inhibitor of protein synthesis was used. As shown in
To determine whether MHC class II molecules that present MalE100-114 peptide reach early and late endosomes towards Golgi, Brefeldin A (BFA), an inhibitor of Golgi transport (Doms, R. W., et al., 1989; Pelham, H. R. 1991) was used. Here again, the presentation of MalE100-114 peptide after its delivery to BMDCs by CyaA-MalE-OVA or MalE protein was totally abrogated when the APCs had been treated with BFA (
The internalization of CyaA and the subsequent MHC class I-restricted presentation of the OVA peptide inserted in its AC domain have already been shown to be independent on phagocytosis (Guermonprez, P., et al., 2000a, b). However, it can not be excluded that some molecules of CyaA translocate their AC domain into APCs cytoplasm whether others are captured and processed as classical exogenous Ag to give rise to MHC class II-restricted peptides. It was first tested whether actin-dependent capture was implicated in MalE epitope delivery for efficient MHC class II-restricted presentation. In this experiment, cytochalasin B (CCB), a drug that prevents actin filament polymerization and impairs macropinocytosis, phagocytosis, and also caveolae-mediated endocytosis (Gottlieb, T. A., et al., 1993) was used. As shown in
As CyaA interacts with CD11b on APC cell surface it was tested, whether CyaA was endocytosed by a clathrin-dependent process. K+ depletion following hypotonic shock (Larkin, J. M., et al., 1983; Madshus, I. H., et al., 1987; Bayer, N., et al., 2001) was used to test if clathrin coated pits were required for MHC class I and class II presentation of CyaA-MalE-OVA. K+ depletion following hypotonic shock was performed by BMDCs exposure to hypotonic medium followed by incubation in the absence of extracellular potassium. This treatment results in dissociation of clathrin coats from the plasma membrane and nonproductive assembly of clathrin cages in the cytoplasm. Internalization of membrane proteins that interact with AP2 clathrin adapter complex through cytoplasmic amino acid sequences is therefore impaired. As shown in
These results demonstrate that CyaA-MalE-OVA clathrin-mediated endocytosis is required for both class I and class II restricted presentation. This was surprising, as it was believed that CyaA directly translocated its AC domain into cytosol from plasma cell membrane, without being endocytosed. Instead, these results suggest that CyaA AC domain is translocated from clathrin-coated vesicles after its endocytosis.
The great efficiency of CyaA to induce CTL responses against different CD8+ T cell epitopes (Fayolle, C., et al. 2001), and proliferative responses against MalE CD4+ T cell epitope (Loucka, J., et al., 2002) has previously been demonstrated. After in vitro studies demonstrating that CyaA is a potent vehicle to deliver both CD4+ and CD8+ T cell epitopes to BMDCs for Ag presentation, the efficiency of CyaA-MalE-OVA in the simultaneous in vivo delivery of these epitopes was tested. Mice were immunized with 50 μg of CyaA-MalE, CyaA-OVA, CyaA-MalE-OVA or CyaA E5 by i.v. route, without adjuvant. The T cell responses were monitored seven days after injection.
As a readout for CD8+ T cell responses, the cytotoxic activity of splenocytes from immunized mice against target cells loaded with the OVA257-264 peptide was tested. As shown in
The CD4+ T cell responses induced by CyaA-MalE-OVA as compared to CyaA-OVA were also analyzed. As readout, the cytokine secretion of splenocytes in vitro restimulated with the MalE100-114 peptide was monitored. As shown in
These results further confirm the capacity of CyaA simultaneously to deliver both class I and class II epitopes for in vivo T cell priming. Moreover, the efficiency of such simultaneous delivery is similar to the single epitope delivery, and both CD4+ and CD8+ T cell responses appear to be Th1 polarized.
Furthermore, the ESAT-6 (Rv3875, 95 amino acids) or CFP-10 (Rv3874, 100 amino acids) M. tuberculosis genomic sequences can be delivered by CyaA and CyaA affects the dose-response or detection frequency of M. tuberculosis specific IFN-γ producing cells, which enhances immunodiagnosis of TB.
The following references are cited herein. The entire disclosure of each reference is relied upon and incorporated by reference herein.
The application claims the benefit of priority of U.S. Provisional Application No. 60/523,704 (attorney docket number 03495-6094), filed Nov. 21, 2003, which is incorporated herein by reference.
| Number | Date | Country | |
|---|---|---|---|
| 60523704 | Nov 2003 | US |
