Microsoft Word doc

Transcription

1 CD8+ T cell responses and antigen presentation during Mycobacterium tuberculosis infection in humans Jang-Eun Cho Department of Medical Science The Graduate School, Yonsei University

2 CD8+ T cell responses and antigen presentation during Mycobacterium tuberculosis infection in humans Directed by Professor Sang-Nae Cho The Doctoral Dissertation submitted to the Department of Medical Science, the Graduate School of Yonsei University in partial fulfillment of the requirements for the degree of Doctor of Philosophy Jang-Eun Cho June 2005

3 This certifies that the Doctoral Dissertation of Jang-Eun Cho is approved Thesis Supervisor Thesis Committee Member #1 Thesis Committee Member #2 Thesis Committee Member #3 Thesis Committee Member #4 The Graduate School Yonsei University June 2005

4 Acknowledgements 우리가알거니와하나님을사랑하는자곧그뜻대로부르심을입은자들에게는모든것이합렵하여선을이루느리라 " ( 로마서 8:28) 먼저여기까지인도하시고눈동자처럼지켜주신하나님께감사와찬양을드립니다. 박사과정동안맘껏실험하고공부할수있게지도해주시고한결같은모습으로본을보여주신조상래교수님께말로는다할수없는깊은감사를드립니다. 면역학에흥미를갖게자상하게가르쳐주신조성애교수님께진심으로감사드립니다. 언제나따뜻한관심으로조언해주신최인홍교수님, 바쁘신가운데논문지도를해주신김성규교수님과이경원교수님께감사드립니다. 조교생활을하는동안많은가르침을주신박전한교수님께감사드립니다. 자상하게대해주신김주덕교수님, 인자한모습으로때론날카로운질문으로깨닫게해주신김세종교수님과이원영교수님, 묵묵히지켜봐주신이봉기교수님, 김종선교수님, 이재면교수님과항상관심을가져주신신전수교수님께감사드립니다. 원주에서격려와응원을해주신박용석교수님과결핵연구원에서부터여기까지이끌어주신이혜영교수님, 멀리마산에서열심히연구하시는엄석용교수님, 늘관심을가지고지켜봐주신송택선교수님과한미영박사님께도감사의말을전하고싶습니다. 긴시간동안좋은실험실사람들을만났기에잘견뎌나갈수있었던것같습니다. 항상웃으시면서많은도움을주신김일휘선생님, 최유정선생님과고시환선생님, T 세포를기꺼이제공해주신장윤수선생님과

5 곽이섭선생님께도감사의말을전하고싶습니다. 또한함께한조교선생님들에게도감사의말을전합니다. 옆에서항상신경써주고힘이되어준 10년넘게함께한친구영미와은정이에게고맙다는말을하고싶습니다. 어리지만늘언니의말을잘들어주고따라준은희와대학원생활을같이하면서삶의활력을더해준성희와지숭, 착하고말잘듣는주영이, 지금은실험실을나갔지만잘챙겨주는정많은선화, 원주에서적응하느라바쁜의리파방혜은선생님과김연선생님, 이대에서열심히연구중인승은이, 마산에서즐겁게일하고있는은진이, 앞으로더열심히하고실험실을이끌어가야할은계, 정연, 민경, 아름, 재식, 지원, 지영, 원경, 희숙모두에게고맙다는말을전하고싶습니다. 워싱턴 FDA에서열심히연구하시는전보영선생님과함께해서든든했던박인호선생님, 여러가지로많이도와주고격려해준김지수선생님께도감사드립니다. 조교생활하면서기쁨과고생을함께나눈깜찍한은정이, 룸메이트를하면서알게된연변친구연희, 미생물학교실에와서더끈끈하게정이든형란이와수정이, 10년이상을이언니를믿고따라준사랑스런지영이와윤숙에게고마움을전합니다. 마지막으로여기까지어려움없이공부할수있게해주시고저를믿고아끼고사랑해준가족들에게이논문을바칩니다. 조장은씀

6 TABLE OF CONTENTS ABSTRACT...1 I. INTRODUCTION Current problems of tuberculosis Immune responses of CD8+ T cells to M. tuberculosis infection...6 A. T cell-mediated protective immunity to M. tuberculosis...6 B. Epitope-based approaches to vaccination...8 C. Immune responses of CD8+ T cells in TB patients MHC class I antigen processing pathway of M. tuberculosis somatic antigens Gene expression profiles of components of the MHC class I antigen processing machinery following infection with M. tuberculosis...12 II. MATERIALS AND METHODS Immune responses of CD8+ T cells to M. tuberculosis infection...16 A. Study subjects...16 B. HLA-A2 typing...16 C. Peptide sequences and synthesis...17 D. Cell lines and culture media...17 E. Cytokines and monoclonal antibodies...18 F. Bacteria culture...19 G. PBMCs separation...19 H. Intracellular cytokine staining of short-term cell lines (STCL)...20 I. Ex vivo IFN-γ elispot analysis of CD8+ T cells-specific for epitope peptides...20 J. In vitro induction of recall CTL responses from healthy subjects...21 i

7 K. CTL assay...22 L. Dimer staining...22 M. Immunocytochemistry of CD8+ CTL lines...23 N. Statistical analysis MHC class I antigen processing pathway of M. tuberculosis somatic antigens...24 A. Generation of CTL lines from healthy HLA-A*0201 and A*0206 subjects by in vitro immunization...24 B. Cytotoxicity of CTL lines for M. tuberculosis-infected macrophages...25 C. Metabolic inhibition of antigen presentation of M. tuberculosis-infected macrophages Gene expression of components of the MHC class I antigen processing machinery following infection with M. tuberculosis...26 A. Generation and infection of DCs with M. tuberculosis...26 B. Reverse transcriptase PCR (RT-PCR) analysis...27 III. RESULTS Characterization of HLA-A*0201-restricted CD8+ T cells specific for M. tuberculosis epitope peptides...29 A. HLA-A*02 allele types of the study subjects...29 B. CD8+ STCL generation from HLA-A*0201 PPD+ healthy subjects...31 C. Detection of IFN-γ producing CD8+ T cells from PPD+ subjects expressing HLA-A2 supertype...33 D. Quantification of the CD8+ T cell frequencies specific for M. tuberculosis- derived peptides from subjects expressing HLA-A2 supertype...35 E. Quantification of the frequencies of the CD8+ T cells specific for the M. tuberculosis-derived peptides in a pleural effusion from TB pleurisy patients...38 F. In vitro induction of recall CTL responses from healthy subjects expressing HLA-A2 supertype...40 G. Enumeration of frequencies of peptide-specific CD8+ T cell populations using HLA-A*0201 dimer complexes...43 H. Immunocytochemistry for perforin and granulysin expression in M. tuberculosis ii

8 peptide-specific CD8+ CTL lines MHC class I antigen processing pathway of M. tuberculosis somatic antigens...48 A. Generation of CD8+ T cell lines for M. tuberculosis somatic antigens...48 B. Kinetics of cytotoxic activities of RpoB specific CD8+ T cells for M. tuberculosis-infected macrophages...50 C. RpoB peptide presentation requires phagocytosis and bypass Golgi-ER transport in M. tuberculosis-infected macrophages Gene expression of components of the MHC class I antigen processing machinery following infection with M. tuberculosis...54 A. Effect of IFN-γ on the expression of MHC molecules of macrophages...54 B. The kinetics of IFN-γ induced gene expression of MHC class I Ag processing pathway affected by M. tuberculosis infection...56 C. Isolation and characterization of human immature DCs from peripheral blood derived adherent cell cultures D. The effect of IFN-γ treatment and M. tuberculosis infection on the expression of genes involved in MHC class I antigen processing...62 IV. DISCUSSION...70 V. CONCLUSION...78 REFERENCES...81 ABSTRACT (in Korean)...89 iii

9 LIST OF FIGURES Fig. 1. Intracellular cytokine staining for HLA-class I-restricted CD8+ T cell populations specific for M. tuberculosis peptides in HLA-A*0201 subjects...32 Fig. 2 Intracellular IFN-γ staining for HLA-class I-restricted M. tuberculosis peptide reactive CD8+ T cell populations in PPD+ subjects expressing HLA-A2 supertype...34 Fig. 3. Photomicrograph displaying IFN-γ specific elispot formation from CD8+ T cells Fig. 4. CD8+ T cell-mediated responses to M. tuberculosis-specific epitope peptides in TB patients and healthy subjects using ex-vivo IFN-γ elispot assay Fig. 5. CD8+ T cell-mediated responses to M. tuberculosis-specific epitope peptides in TB pleurisy patients using ex-vivo IFN-γ elispot assay...39 Fig. 6. Recall CTL response from healthy subjects expressing HLA-A2 supertype...41 Fig. 7. HLA-A*0201 dimer staining for M. tuberculosis peptide-specific STCL generated from healthy subjects...45 Fig. 8. Detection of perforin and granulysin in CTLs by confocal laser microsopy...47 Fig. 9. Generation of CTL lines from healthy HLA-A2 subjects by in vitro immunization...49 Fig. 10. AFB staining...50 Fig. 11. Kinetics of M. tuberculosis RpoB peptide processing inside macrophages...51 Fig. 12. Effect of metabolic inhibitors on presentation of M. tuberculosis-derived RpoB protein Fig. 13. Effect of IFN-γ concentrations in the expression of human MHC molecules...55 Fig. 14A. The kinetics of MHC class I gene expression in response to IFN-γ treatment and M. tuberculosis infection Fig. 14B. The kinetics of MHC class I gene expression in response to IFN-γ treatment and M. tuberculosis infection Fig. 14C. The kinetics of MHC class I gene expression in response to IFN-γ treatment and M. tuberculosis infection iv

10 Fig. 15. Cell surface molecule expression of human immature dendritic cells...61 Fig. 16. RT-PCR analysis for MHC class I antigen processing genes in human monocyte-derived dendritic cells (DCs)...64 Fig. 17. Densitometric quantitation of the relative induction of genes involved in MHC class I antigen processing...65 Fig. 18. RT-PCR analysis to exam the mrna expression pattern of genes involved in MHC class I antigen processing in macrophages...66 Fig. 19. Densitometric quantitation of the relative induction of genes involved in MHC class I antigen processing...67 Fig. 20. RT-PCR analysis for genes involved MHC class I antigen processing in human monocyte-derived macrophages...68 Fig. 21. Densitometry quantitation of the relative induction of genes involved in MHC class I antigen processing...69 v

11 LIST OF TABLES Table 1. Sequences of primers used for RT-PCR...27 Table 2. HLA-A genotype frequencies...30 Table 3. Demography of the study subjects participated in this study...31 Table 4. Frequency of peptide-specific CD8+ T cells detected by IFN-γ intracellular staining and HLA-A*0201 dimer staining...44 vi

12 ABBREVIATION TB, tuberculosis MTB, Mycobacterium tuberculosis AIDS, acquired immune deficiency syndrome MDR-TB, multidrug-resistant tuberculosis MHC, major histocompatibility complex HLA, human leukocyte antigen TAP, transport associated with antigen processing β 2 m, β 2 -microglobulin BCG, Bacillus Calmette-Guerin CTL, cytotoxic T lymphocyte PPD, purified protein derivative PBMCs, peripheral blood mononuclear cells PFMNCs, pleural fluid mononuclear cells EBV, Epstein-Barr virus DCs, dendritic cells GM-CSF, granulocyte-macrophage-colony-simulating factor MOI, multiplicity of infection ER, endoplasmic reticulum LMP, low-molecular-weight polypeptide JAK, Janus kinase GAS, gamma-activated sequence IRF, IFN regulatory factor RT-PCR, reverse transcriptase polymerase chain reaction

13 ABSTRACT CD8+ T cell responses and antigen presentation during Mycobacterium tuberculosis infection in humans Jang-Eun Cho Department of Medical Science The Graduate School of Yonsei University (Directed by Professor Sang-Nae Cho) Cell-mediated immune responses are a major protective mechanism against Mycobacterium tuberculosis infection. These cells mediated immune responses are composed of T cells and macrophages. In order to understand the immune mechanism for tuberculosis (TB), three aspects of the immune responses regarding the major histocompatibility complex (MHC) class I-restricted CD8+ T cells and the processing of M. tuberculosis antigens were researched. The three aspects included : CD8+ T cell responses to M. tuberculosis-derived peptides, the antigen processing mechanism of M. tuberculosis somatic antigen, and the effect of M. tuberculosis on interferon

14 gamma (IFN-γ)-induced genes. First, ThyA 30-38, RpoB , PstA , and 85B were previously identified as M. tuberculosis-derived peptides specific for HLA-A*0201-restricted CD8+ T cells. This study characterized these peptide-specific CD8+ T cells in individuals that were either actively (TB patients) or latently infected with M. tuberculosis (PPD+). CD8+ T cell responses to these peptides were induced in all study groups (PPD+ healthy subjects, pulmonary TB patients, and TB pleurisy patients), except the PPD- subjects. However, the M. tuberculosis antigen-specific CD8+ T cell immunity appeared to be depressed in patients with advanced stages of TB. These M. tuberculosis-derived peptides also induced CD8+ T cell responses in subjects expressing the subtypes HLA-A*0203, A*0206, and A*0207, suggesting that these epitopes are A2 supertype peptides. Among these four peptides, the immunodominant peptide that induced the highest number of IFN-γ secreting CD8+ T cells differed depending on the subjects. Short-term cell lines specific for these peptides proliferated in vitro and secreted IFNγ upon antigenic stimulation in PPD+ subjects. HLA-A*0201 dimer assays indicated that the PstA specific CD8+ T cell population in PPD+ healthy subjects was functionally heterogeneous, since only one-half or one-fourth of the cells produced IFN-γ upon peptide stimulation. In addition, by assaying cytotoxic T lymphocyte (CTL) activities, we observed that CTL responses specific for these M. tuberculosisderived peptides could be induced in Bacille Calmette-Guerin (BCG)-vaccinated subjects. This result suggests that CD8+ T cells may be involved in controlling TB in BCG-vaccinated or PPD+ healthy people

15 Secondly, M. tuberculosis resides and replicates inside macrophages. In our previous publication, CD8+ T cell-mediated immune responses specific for the peptide RpoB , which was derived from the RNA polymerase beta-subunit of the M. tuberculosis protein, could be induced in TB patients. In order to demonstrate that CD8+ T cells can recognize RpoB that was processed by M. tuberculosisinfected macrophages, CD8+ T cell lines specific for the RpoB peptide were generated from the peripheral blood mononuclear cells (PBMCs) of healthy HLA- A*0201 and A*0206 subjects, using in vitro immunization techniques. These CD8+ T cell lines specifically recognized and destroyed M. tuberculosis infected-macrophages. In addition, the presentation of the M. tuberculosis-derived epitope peptide, RpoB , to CD8+ T cells did not seem to be inhibited by brefeldin-a treatment, which blocks the classical MHC class I-restricted antigen presentation pathway in macrophages. Therefore, the RpoB peptide may be processed by accessing the alternative MHC class I processing pathway, which was previously suggested as the processing pathway for the cytoplasmic proteins of M. tuberculosis. Since the RpoB gene of M. tuberculosis was reported to be actively expressed inside macrophages, the RpoB protein or derived peptides may be useful for the development of TB vaccines. This study also suggests that not only secreted but also somatic proteins of M. tuberculosis need to be screened for TB vaccines and therapeutic agents. Lastly, the effect of M. tuberculosis infection on the expression of IFN-γ induced genes involved in MHC class I-restricted antigen processing pathway of the host cells was investigated. IFN-γ is a principal mediator of the bactericidal activation of - 3 -

16 macrophages. M. tuberculosis is either resistant to the IFN-γ responsive microbicidal mechanisms of macrophages or alternatively may block the macrophage response to IFN-γ. This study demonstrates that the M. tuberculosis infection selectively affected the transcription of IFN-γ responsive genes. While the transcription of CD64 decreased, IFN-γ responsive genes involved in the MHC class I Ag processing pathway were either unaffected or induced by M. tuberculosis. Further studies are needed to elucidate the underlying mechanisms whereby M. tuberculosis inhibits cellular responses to IFN-γ. Key words : Mycobacterium tuberculosis, CD8+ T cells, Peptide epitopes, HLA-A2, IFN-γ, Macrophage, RpoB peptide, IFN-γ responsive gene - 4 -

17 CD8+ T cell responses and antigen presentation during Mycobacterium tuberculosis infection in humans Jang-Eun Cho Department of Medical Science The Graduate School of Yonsei University (Directed by Professor Sang-Nae Cho) I. INTRODUCTION 1. Current problems of tuberculosis Tuberculosis (TB) caused by Mycobacterium tuberculosis was discovered by Robert Koch in M. tuberculosis is a facultative intracellular pathogen and acid fast bacillus that replicates within human macrophages. The World Health Organization (WHO) estimates that one-third of the world s population has been infected with M. tuberculosis, only 10% of infected people break down with the disease, 90% of infected people remains clinically latent. Therefore, M. tuberculosis is often in a dormant state in infected hosts, usually produces a chronic disease 2 and - 5 -

18 has a tendency to reactivate many years after the initial infection. TB is a major global health problem, especially in developing countries. In addition, 8 million new TB cases and two to 2.5 million cases of TB-related deaths occurs annually 3. The introduction of rifampicin, pyrazinamide and ethambutol in recent years offered in an ear of short-course treatment. But, the advent of HIV infection, the acquired immune deficiency syndrome (AIDS) pandemic in the 1980s, increases the risk of developing TB. Strains of M. tuberculosis resistant to both isoniazid and rifampicin with or without resistance to other drugs have been termed multidrugresistant strains 4. Multidrug-resistant tuberculosis (MDR-TB) is among the most worrisome elements of the pandemic of antibiotic resistance because TB patients that fail treatment have a high risk of death. 2. Immune responses of CD8+ T cells to M. tuberculosis infection A. T cell-mediated protective immunity to M. tuberculosis Clinical and experimental evidence suggests that immunity against M. tuberculosis is mainly controlled by the cell-mediated immune responses, involving T cells and macrophages. Studies on humans and animal models have demonstrated that CD4+ T cells are essential to the immune response to M. tuberculosis infection. In mouse models, CD8+ T cells have been reported to play significant roles in the containment of TB. In the mouse model, for example, β 2 -microglobulin (β 2 m)

19 deficient, transporter associated with antigen processing (TAP)-deficient and CD8+ T cell-deficient mice were more susceptible to M. tuberculosis infection than normal control mice 5-7. Recent study using a mouse model and reactivation suggests that CD8+ T cells may be even more important than CD4+ T cells in controlling the latent phase of TB infection 8. Three different roles of the CD8+ T cells have been defined : the release of IFNγ, the lysis of the infected targets and the direct antimicrobial activity. IFN-γ secretion from T cells induces both antigen presentation and the bactericidal activity of macrophages where bacilli mainly reside and replicate. A CD8+ T cell adoptive transfer experiment demonstrated that increased immunity to M. tuberculosis infection was the result of increased levels of IFN-γ secreting CD8+ T cells 9. In a mouse model of a low dose aerosol infection, the IFN-γ production by CD8+ T cells was significantly involved in the protection against chronic M. tuberculosis infection 8, 10. While CD8+ T cells are considered critical for the control of M. tuberculosis infection in the mouse model, its role in human infections remains less well understood. In humans, CD8+ T cells exhibit cytotoxic activity or growth inhibition against intracellular M. tuberculosis by releasing granules such as granulysin or direct contact via Fas (CD95), respectively 11, 12. Perforin molecule does not seem to play a role in early protective response against M. tuberculosis, but only in the late phases on infection. There are two pathways of cytotoxicity. One is granule exocytosis and the other is Fas-Fas ligand (FasL) pathways. Both induce apoptosis in the target cells. The - 7 -

20 granule exocytosis pathway involves directed and regulated secretion of the lytic granule constituents, including perforin (a Ca2+ dependent, pore forming protein related to the membrane attack complex of complement) and granzymes (serine esterases that activate the caspase cascade). B. Epitope-based approaches to vaccination Several laboratories have identified the M. tuberculosis peptides that are presented by human major histocompatibility complex (MHC) class I molecules to T cells as well as the role these T cells play in containing infection. For instance, peptides derived from a 19-kDa protein (membrane-bound lipoprotein) and an esat-6 protein (early secretory protein) have been identified as being immunogenic for MHC class I-restricted CD8+ T cells 13, 14. In our previous study, four M. tuberculosisderived epitopes for HLA-A*0201-restricted CD8+ T cells were also defined 15. Three of these were derived from the somatic M. tuberculosis proteins, ThyA, RpoB, PstA1 and one was derived from antigen 85B, one of the major secreted proteins. Somatic antigens seem to be mainly derived from protein turn-over of bacilli during the late phase of infection. In contrast, secreted antigens from mycobacteria are produced by metabolically active organisms and related to the early or active phase of infection. These M. tuberculosis peptide-specific CD8+ T cells could release IFN-γ upon the recognition of M. tuberculosis-infected cells and reduce the number of intracellular bacilli, suggesting that M. tuberculosis somatic antigens can be processed - 8 -

21 and recognized by MHC class I-restricted CD8+ T cells for the protective immunity in humans. The definition of epitope peptides contributes not only to the study of the functional implication of CD8+ T cell on TB but also to the development of a vaccine for TB. Indeed, the development of a new TB vaccine is increasingly needed since the efficacy of Bacillus Calmette-Guérin (BCG), which is the only currently available vaccine for TB, is estimated to be 0-80 % effective. To develop universal subunit vaccines, it is essential to identify whether these newly defined epitopes are A2 supertype peptides. Human leukocyte antigen (HLA)-A2 is one of the most frequent HLA-A specificity in the human population with an allele frequency of 10-40% in different ethnic groups HLA-A2 supertype is a family of HLA subtypes that share specificities for a degenerate ligand with HLA-A*0201. Molecules of the A2 supertype are characterized by preferences for peptides of 9 or 10 residues in length bearing small or alipathic hydrophobic residues (A, I, V, L, M, or T) in position 2 and at the peptide C-terminus. On the basis of these studies, the A2 supertype minimally includes A*0201, A*0202, A*0203, A*0204, A*0205, A*0206, A*0207, A*6802 and A*6901. The A2 supertype includes the A*0201, A*0202, A*0203, A*0204, A*0205, A*0206 and A*0207, A*6802 and A*6901 types, and A2 supertype epitope peptide can bind to the HLA-A molecules of supertype members. Therefore, the part one of this study was performed to determine if these epitope peptides are the A2 supertype peptides

22 C. Immune responses of CD8+ T cells in TB patients It has been known that IFN-γ responses in PBMCs is decreased in many TB patients, a demonstrated by reports showing that the IFN-γ therapy was effective for treating TB patients. On the contrary, latently infected people with M. tuberculosis usually react positive to M. tuberculosis purified protein derivatives (PPD) skin test, and the disease can be contained because their protective immunity is effective. Recent studies have demonstrated that the M. tuberculosis protein specific CD8+ T cells are in fact highly induced in healthy household TB contacts or PPD+ healthy individuals 19. Therefore, in order to investigate the role for CD8+ T cells in protective immunity and to design an effective vaccine for TB, it is necessary to identify the correlations of the CD8+ T cell immune responses in latently infected subjects and in chronically infected TB patients. The part two of the study was subsequently designed to compare the distribution of these epitope-specific IFN-γ secreting CD8+ T cells in latently infected subjects and active TB patients expressing HLA-A2 supertype. In addition, these CD8+ T cells specific for M. tuberculosis-derived peptides were further examined to determine if they are also induced in the pleural effusion from TB pleurisy patients. Tuberculous pleuritis usually manifests as lymphocytic exudate pleural effusions 20. This local immune response is orchestrated by T cells and represents the protective immune response responsible for controlling the spreading of M. tuberculosis infection. CD4+ T cells play an important part in the development of tuberculous pleural effusion by cell-mediated immunity, inducing strong protective

23 immunity by concentrating a high level of IFN-γ in the localized disease sites 21. Since the importance of CD8+ T cells in tuberculous pleurisy has not been defined yet, these M. tuberculosis peptide-specific CD8+ T cells were examined to determine if are present in a pleural effusion from TB pleurisy patients. 3. MHC class I antigen processing pathway of M. tuberculosis somatic antigens In addition to cytotoxic activity, it is known that CD8+ T cells have significant roles in the protective mechanism against M. tuberculosis infection by releasing IFNγ. However, little is known about the mechanism by which antigens from M. tuberculosis gain access to the MHC class I-restricted presentation pathway. Previous studies using BCG-activated CD8+ CTL have shown that proteosome and Golgi inhibitors reduced CTL activity to zero 22. These data imply that the majority of CD8+ T cells against mycobacteria are restricted by MHC-class I. Recently, it was reported that BCG could activate CD8+ CTL by the MHC class I-restricted presentation pathway which is both proteasome-dependent and Golgi-endoplasmic reticulum (ER) transport pathway-dependent 22. On the other hand, it was shown that M. tuberculosisspecific CD8+ T cells could recognize an antigen which was processed by the proteasome-dependent pathway, but which was not transported through the Golgi-ER pathway. The immune responses of these M. tuberculosis-reactive CD8+ T cells were not inhibited by brefeldin A, while inhibited by MHC class I blocking antibody. M. tuberculosis-derived antigen presentation was found to require proteasomal

24 processing, but to be presented in a manner that was brefeldin A and hence TAP independent 23. In this report, we present the result that one of cytosolic M. tuberculosis proteins, RNA polymerase subunit B (RpoB) is processed by the alternative MHC class I presentation pathway and recognized by MHC class I- restricted CD8+ T cells. Alternative MHC class I processing pathway may permit the processing of phagosomal antigens. This pathway seems to be preferentially utilized by particulate antigens, including bacteria. M. tuberculosis induces apoptosis of host cells and the formation of apoptotic blebs. These blebs could be engulfed by macrophages or dendritic cells and presented in a TAP-dependent fashion to CD8+ T cells 24. The mechanism of this Golgi-ER independent presentation pathway needs to be investigated furthermore. 4. Gene expression profiles of components of the MHC class I antigen processing machinery following infection with M. tuberculosis Infection with M. tuberculosis induces a cellular immune response including CD4+ and CD8+ T cells that secrete IFN-γ. IFN-γ, the predominant activator of microbicidal functions of macrophages, is detectable at sites of M. tuberculosis infection, but is unable to stimulate macrophages to kill M. tuberculosis. IFN-γ acts by causing changes in gene expression through both transcriptional and posttranscriptional regulation with most genes being transcriptionally regulated by IFN-γ 25. Moreover, the effect of M. tuberculosis on the response to IFN-γ at the

25 molecular level and the aspects of regulation that are affected remain poorly understood. M. tuberculosis interferes with cellular signal transduction pathways that are activated by IFN-γ and thereby avoids being killed within macrophages. M. tuberculosis decreases IFN-γ stimulated mrna amount of the IFN-γ-regulated genes, Fcγ receptor I (CD64) and class II transactivator 26, 27. Binding of IFN-γ to cell surface receptors results in activation of the tyrosine kinase Janus kinases 1 (JAK1) and JAK2, leading to phosphorylation of cytoplasmic signal transducers and activators of transcription 1 (STAT1) 26. Tyrosine-phosphorylated STAT1 homodimerizes through interaction of the Src homology-2 (SH2) domain on one molecule with phosphotyrosine on another and translocates to the nucleus. In the nucleus, STAT1 homodimers activate transcription of specific genes that possess γ-activation sequences (GAS; consensus sequence is TTNCNNNAA) in the promoters of IFN-γ stimulated genes. Human genes that contain GAS include Fcγ receptor I (CD64), guanylate binding protein 2, class II transactivator, indoleamine-2,3-dioxygenase, TAP In addition, IFN-γ alters proteasome activity qualitatively. Vertebrate have three IFN-γ inducible β subunits (LMP2, LMP7, and LMP10), the former two being encoded in the MHC. Each inducible subunit is homologous with a constitutive catalytic subunit (LMP2/Y, LMP7/X, and LMP10/Z) and can replace its homologue during proteasome assembly. IFN-γ inducible proteasome subunits LMP2, LMP7 and LMP10 are called immunosubunits. The TAP2, LMP7, LMP10 and PA28 promoters contain different elements, but in all cases the IFN-consensus sequences 31, suggesting

26 a distinct regulation of their constitutive expression and a coordinated regulation upon IFN-γ treatment. In the ER, different chaperones, like tapasin, calnexin, calreticulin stabilize the MHC class I molecules during their folding and assembly or facilitate their loading with peptides The promoter of most chaperons has not been identified so far. Presumably, these seem to be additional factors regulating the expression of MHC class I, TAP, LMP and chaperone which are important for efficient MHC class I antigen processing and presentation. It is not known whether regulation of interferon regulatory factor 1 (IRF-1) expression in response to IFN-γ is solely transcriptional, because IRF-1 expression can be regulated posttranscriptionally. GAS and IRF-1 regions are important for the induction of TAP1 expression in response to IFN-γ. IRF-1 is known as a key factor in the induction of type I IFN gene expression during M. tuberculosis infection 35. IFN-γ increases TAP dependent peptide transport more rapidly than HLA class I molecule expression. IFN-γ-activated STAT1α/STAT1α binds to a GAS in the promoter of TAP1 and the promoter of the transcription factor IRF-1, which mediates the delayed response of HLA class I promoter. Infection with M. tuberculosis does not inhibit STAT1 tyrosine or serine phosphorylation, dimerization, nuclear translocation, or recognition of specific DNA sequences 27. Infection with M. tuberculosis inhibits IFNγ responses by directly or indirectly disrupting the essential interaction of STAT1α with the transcriptional coactivators CBP and p To further understand the mechanisms whereby M. tuberculosis inhibits cellular responses to IFN-γ, the effect of M. tuberculosis infection on the expression of IFN-γ

27 responsive genes involved in MHC class I-restricted antigen processing pathway was investigated in this study

28 II. MATERIALS AND METHODS 1. Immune responses of CD8+ T cells to M. tuberculosis infection A. Study subjects The HLA-A types of 60 healthy blood subjects, 84 chronically infected pulmonary TB patients and 38 TB pleurisy patients were determined. These subjects were previously vaccinated with BCG. Before blood sampling, informed consent was obtained from individuals after explaining the goals and methods of this project. Healthy PPD+ and PPD- subjects were selected randomly. The patients with mild, moderate and far advanced pulmonary TB from the Ewha woman s university hospital (Seoul, Korea) or the National Masan tuberculosis hospital (Masan, Korea) were classified based on a chest X-ray radiograph (CXR) (Table 2). The subject was defined as being PPD positive (+) if the hypersensitivity skin reaction with 2 T.U. tuberculin/0.1ml (Statens Serum Institut, Copenhagen S, Denmark) was > 6 mm at three days after the tuberculin injection. B. HLA-A2 typing To define HLA-A2 type, flow cytometry analysis was performed by using anti-hla- A2 monoclonal antibody (mab); BB7.2 (BD-Pharmingen, San-Jose, CA, USA)

29 HLA-A2 subtypes from each individual were subsequently identified by direct DNA sequence analysis of the polymorphic exons 2 and 3 of the HLA-A gene at the DNA Sequencing Facility at Hallym University (Ahnyang, Korea). C. Peptide sequences and synthesis Five HLA-A*0201-restricted CD8+ T cell-specific peptides were synthesized at the Korea Basic Science Institute (Seoul, Korea). Sequences were confirmed by mass spectrometry analysis. Sequences of the peptides used for the assays are as follows: ThyA is RLPLVLPAV (IC50 = 5.1 nm), derived from thymidylate synthase, RpoB is MTYAAPLFV (IC50 = 13.8 nm), derived from RNA polymerasesubunit, 85B is LMIGTAAAV (IC50 = 79.0 nm), derived from antigen 85B, and PstA is SLYFGGICV (IC50 = 10.6 nm), derived from putative phosphate transport system permease protein A-1 of M. tuberculosis. A*0201-binding epitope peptide derived from the influenza A virus matrix protein (Flu : GILGFVFTL) was used as the positive control. D. Cell lines and culture media.221a2 (provided by Robert DeMars, University of Wisconsin-Madison) is an Epstein-Barr virus (EBV)-transfected B-cell line mutagenized and selected for loss of HLA antigens, then transfected with HLA*0201, were used as target for the recall

30 CTL response. To generate B lymphoblastoid cell line (B-LCL), 5ⅹ10 5 peripheral blood mononuclear cells (PBMCs) from healthy subjects were transformed by EBVcontaining supernatants from B-95.8 in RPMI 1640 (GIBCO, Grand Island, NY, USA) that had been supplemented with 2 mm L-glutamine, 100 U/ ml penicillin and 100 μg / ml streptomycin (GIBCO, Grand Island, NY, USA) and 10% heat-inactivated fetal bovine serum (JBI, Taegu, Korea). After 14 days, the transformed cells were expanded in RPMI containing 10% FBS. Iscove s modified Dulbecco s medium (IMDM, GIBCO, Grand Island, NY, USA) was used in vitro immunization. Hank s balanced salt solution (HBSS : Sigma, St Louis, MO, USA) was used to wash PBMCs. E. Cytokines and monoclonal antibodies Recombinant human (rh) IL-2, rhil-7, rhgm-csf and rhil-4 were purchased from R&D systems Inc. (Minneapolis, MN, USA). And rhifn-γ was purchased from Pierce Endogen (Rockford, USA). Monoclonal antibodies (mab) specific for CD1a, CD83 and CD28 were supplied by BD Pharmingen (San-Jose, CA, USA). For the analysis of cell surface antigens, the following mab specific for PerCP-conjugate CD3, FITC-conjugate CD8, FITC conjugate CD14, FITC conjugate HLA-DR and intracellular mab specific PE conjugate IFN-γ and IL-4 (BD, Mountain View, CA, USA). W6/32 hybridoma producing Ab specific for human class I MHC was obtained from American Type Culture Collection (ATCC). Human IFN-γ elispot pair (purified anti-human IFN-γ and biotinylated anti-human IFN-γ), streptavidin-horseradish

31 peroxidase conjugate and anti-perforin Ab were purchased from BD Pharmingen (San-Jose, CA, USA). Alexa Fluor 488 goat anti-mouse IgG and Alexa Fluor 568 goat anti-mouse IgG1 were purchased from Molecular probes (Eugene, OR, USA). And anti-granulysin DH2 Ab was provided from Alan Krensky (Stanford Univ., USA). F. Bacteria culture Mycobacteria tuberculosis H37Rv (ATCC 27294) and M. bovis BCG (Pasteur strain 1173P2) used in this study was grown for about 10 days at 37 as a surface pellicle on Sauton medium enriched with 0.4% sodium glutamate and 3.0% glycerol. The surface pellicles were collected and disrupted with 6 mm glass beads by gentle vortexing. After clumps had settled out, the upper suspension was collected and aliquots were stored at -70 until used. After thawing, viable organisms were counted by plating serial dilutions on Middlebrook 7H11 agar (Difco, Detroit, MI, USA). G. PBMCs separation PBMCs derived lymphocytes from heparinized venous blood or pleural fluid mononuclear cells (PFMNCs) from pleural effusion were isolated by density gradient centrifugation over Ficoll-Paque TM PLUS (Amersham Biosciences, Uppsala, Sweden). The cells were washed twice with culture medium and once with 1X HBSS

32 H. Intracellular cytokine staining of short-term cell lines (STCLs) STCLs were generated by stimulating the PBMCs of the healthy PPD+ subjects with each peptide for 10 days in the presence of IL-2 (10 units/ml) and IL-7 (10 ng/ ml ) at 37 C in a CO 2 incubator. For induction of recall CD8+ T cell responses, the cultures were re-stimulated for an additional week with irradiated (3000 rad) autologous monocytes pulsed with each peptide. Subsequently, the EBV lines were stimulated with each peptide overnight. Next day, cells were incubated with 1 μg / ml brefeldin A (Sigma, St Louis, MO, USA) for 10 h. The cells were stained with anti-cd3 mab PerCP-conjugate and anti-cd8 mab FITC-conjugate. Intracellular cytokine staining for IFN-γ or IL-4 was performed using the Cytofix/Cytoperm kit (BD-Pharmingen, San-Jose, CA, USA) according to the manufacturer s instructions. The cells were incubated in fixation solution for 15 min at RT. The cells were washed and incubated with Perm/Wash solution and anti-ifn-γ mab PE-conjugate, anti-il-4 mab PEconjugate for 15 min at RT. Subsequently cells were fixed with 2% paraformaldehyde (PFA, Sigma, St Louis, MO, USA) and analyzed by a FACScalibur (Becton Dickinson, San Jose, CA, USA); >10 5 events were acquired for each sample. I. Ex vivo IFN-γ elispot analysis of CD8+ T cells-specific for epitope peptides Ninety-six-well polyvinylidene difluoride (PVDF)-backed plates (Millipore, Molsheim, France) were precoated with anti-ifn-γ mab. The PBMCs or PFMNCs

33 derived lymphocytes were depleted of the CD4+ T cell population using magnet beads (Dynal Biotech ASA, Oslo, Norway), and plated with each synthetic peptide (10 μg / ml ) in 96 well plates in the presence of rhil-2 (10 units/ ml ) and anti-cd28 mab for approximately 2 days in a CO 2 incubator at 37 C. Subsequently the plates were washed ten times with phosphate buffered saline (PBS)/0.05% Tween 20 to remove cells and were incubated for 24 h with biotinylated anti-ifn-γ mab. Streptavidin-horseradish peroxidase conjugate was added for 2 h in room temperature (RT) and subsequently chromogenic 3-amino-9-ethylcarbazole (AEC) peroxide substrate (Sigma, St Louis, MO, USA) was added. After 15 min, the colorimetric reaction was terminated by washing the plate with tap water. The immuno-spotsspecific for single cells secreting IFN-γ were developed according to the manufacture s instruction and counted using an automatic elispot reader system (Zeiss, Germany). J. In vitro induction of recall CTL responses from healthy subjects PBMCs were resuspended in RPMI 1640 medium plus 10% pooled human serum and plated in a 24 well plate at 3ⅹ10 6 cells/well. Synthetic peptides were added to the PBMC cultures at a final concentration of 10 μg / ml. On days 3 and 6, rhil-2 was added to each well at a concentration of 10 units/ ml. On day 8, the cultures were restimulated with irradiated (3000 rad) autologous monocytes that were pulsed with peptide in the presence of 3 μg / ml of β 2 m (Sigma, St Louis, MO, USA) for 2h. On

34 days 10 and 13, 10 units/ ml of rhil-2 was added into each well. The cytolytic activity of cultured PBMCs was tested on day 14 or 15. K. CTL assay Cytolytic activity was measured by a standard 4 h 51 Cr release assay. Approximately 2ⅹ10 6 target cells (.221A2 cells or macrophages) were pulsed with peptide (10 μg / ml ) and β 2 m (3 μg / ml ) overnight. These cells were then labeled with 100 uci Na 51 CrO 4 (NEN, Boston, MA, USA) for 1h at 37. After washing, target cells were incubated with different ratios of effector cells in a 200 μl total reaction volume of RPMI supplemented with 10% FBS in a 96 well plate. After 4 h incubation, 100 μl of supernatant was removed from each well, and its radioactivity was counted in a γ- counter. Specific lysis (%) was calculated by using the formula 100 x (experimental release-spontaneous release)/(maximal release-spontaneous release). Maximal release was determined from supernatants of target cells lyses by the addition of 1% Triton X-100. Spontaneous release was determined from supernatants of target cells incubated with media only. Spontaneous release was < 20% of the maximum release in all assays. L. Dimer staining Dimeric complexs (BD-Pharmingen, San-Jose, CA, USA) were prepared by mixing

35 HLA-A*0201 : Ig protein with specific peptide at 640 molar excess at 37 overnight incubation. Peptide-specific STCLs were incubated with 10 μl human serum at RT for 10 min. Cells were incubated with 1-2 μg of peptide loaded HLA-A*0201 : Ig protein for each sample for 60 min at 4. Cells were washed with FACS buffer (PBS with 0.1% BSA and 0.02% sodium azide). Cells were given 10 μl of human serum, then incubated for 10 min at RT. Subsequently, cells were incubated at 4 for 30 min in FACS buffer containing anti-cd8 mab, anti-cd3 mab and anti-mouse IgG1 (BD- Pharmingen, San-Jose, CA, USA). Then cells were fixed with 2% PFA and analyzed by FACScalibur. M. Immunocytochemistry of CD8+ CTL lines Slides were coated for 30 minutes with poly-l-lysin solution (Sigma, St Louis, MO, USA). Peptide-specific STLCs were plated for 30 min to allow cells to stick to slide, then washed three times in PBS. Slides were fixed for 10 min at RT with 4% PFA and then washed in PBS. Slides were added with Perm/Block solution (10% human serum, 10% goat serum, 0.01% saponin, 0.1% Triton X, 1% dry milk in PBS) and incubated for 1 h at RT. Then slides were washed in PBS then added antibody to perforin in phosphate-buffered saline (PBS) with dilution buffer (2% goat serum, 0.01% saponin and 0.5% milk) or nonbinding isotype-matched IgG2b control. Subsequently cells were incubated for 1 h, then washed with dilution buffer and incubated with Alexa Fluor 488 goat anti-mouse IgG in dilution buffer for 1 h at RT. Subsequently, samples

36 were stained with either anti-granulysin DH2 or IgG1 control, followed by additional wash with dilution buffer and incubated with Alexa Fluor 588 goat anti-mouse IgG in dilution buffer for 1 h at RT. Samples were fixed in 4% PFA for 5 min. For confocal microscopy, samples were mounted on glass slides in DAKO fluorescent mounting medium (DAKO, Carpinteria, CA, USA). N. Statistical analysis A two-sample Student t-test for the means was performed to determine if there is a statistically significant difference in the number of IFN-γ secreting CD8+ T cells in response to M. tuberculosis-derived peptides between groups. 2. MHC class I antigen processing pathway of M. tuberculosis somatic antigens A. Generation of CTL lines from healthy HLA-A*0201 and A*0206 subjects by in vitro immunization PBMCs from HLA-A*0201 and HLA-A*0206 healthy subjects were pulsed with 50 μg / ml of peptide at cells/well in IMDM at 37 for 90 min. These cells were washed and plated at cells/well in 10% pooled human serum with rhil-7 (10 ng/ ml ) and keyhole limpet hemocyanin (5 μg / ml, Sigma, St Louis, MO, USA). Cultures were re-stimulated weekly with peptide-pulsed and -irradiated autologous

37 monocytes, and supplemented with rhil-2 at 10 units/ ml every 3-4 days 11. After four to five cycles of re-stimulation, the cytotoxic activity of the CD8+ T cells was determined by using the chromium release CTL assay (target cell :.221A2 cell lines). B. Cytotoxicity of CTL lines for M. tuberculosis-infected macrophages Macrophages were generated by culturing adherent monocytes in antibiotics-free RPMI 1640 containing 10% FBS for 3-4 days. These macrophages were then infected with M. tuberculosis (H37Rv) at a multiplicity of infection for 4 h. Extracellular nonphagocytosed M. tuberculosis was removed by three times washing. The infected cells were cultured for 1-4 days before use as targets in CTL assays. The cell viability of macrophage populations was > 90% and about 70% of the cells were infected according to Ziehl-Neelson method for acid-fast bacteria. For cytotoxicity experiments, 1ⅹ10 6 macrophages were labeled with 100 µci of 51 CrNa 2 O 4 for 1 h at 37, and then added to the wells of 96 well U-bottom plates at 5-7ⅹ10 3 cells/well. CD8+ T cell lines were added at various effector to target ratios. Specific lysis (%) was determined as described above. C. Metabolic inhibition of antigen presentation of M. tuberculosis-infected macrophages One hour before the infection of macrophage with M. tuberculosis (3MOI), brefeldin

38 A (3 μg / ml, Sigma, St Louis, MO, USA), cytochalasin D (10 μg / ml, Sigma, St Louis, MO, USA) or lactacystin (40 um, Sigma, St Louis, MO, USA) were added to the culture medium. After 18 h of coincubation with M. tuberculosis, macrophages were used as target cells for the assays. 3. Gene expression of components of the MHC class I antigen processing machinery following infection with M. tuberculosis A. Generation and infection of DCs with M. tuberculosis DCs were generated from human monocytes by culturing with rhgm-csf (800 U/ml) and rhil-4 (500 U/ml) for 3 days in 6 well plates. Cells were then infected with M. tuberculosis for 4 h at a multiplicity of infection of 10. Extracellular bacteria were removed by 3 times washing with RPMI medium. The percentage of infection was estimated by staining aliquots of cells by the Ziehl-Neelson method. Routinely, the infectivity of DCs was approximately 70%. After one day, FACS analysis of cell surface markers for DCs were performed using Abs against MHC class II (anti-dr- FITC), MHC class I (anti-w6/32), B7.1 (anti-cd80), B7.2 (anti-cd86), anti-cd83, and anti-cd1a. All staining procedures were performed in FACS buffer for 30 min at 4. Cell were fixed with 2% PFA and analyzed by FACScaliber

39 B. Reverse transcriptase PCR (RT-PCR) analysis Macrophages and DCs were cultured under various experimental conditions as shown in results. Then total RNA was isolated from cultured cells by lysis with Trizol (Gibco Invitrogen, Carlsbad, CA, USA). The cdna was synthesized by reverse transcription with 2 μg total RNA, 0.25 μg of random hexamer (Gibco Invitrogen, Carlsbad, CA, USA) and 200 unit of Murine Molony Leukemia Virus Reverse Transcriptase (MMLV-RT, Gibco Invitrogen, Carlsbad, CA, USA). Subsequent PCR amplification using 0.2 units of Taq polymerase (Takara, Seoul, Korea) was performed in a thermocycler (PerkinElmer, USA) for cycles (1 min at 94, 1 min at 54-58, 1 min at 72 ). An aliquot of the PCR products was electrophoresed in 1% agarose gel in Tris-borate buffer. Bands were visualized by ethidum bromide staining and photographed. The appropriate forward and reverse primer sequences for genes involved in MHC class I antigen processing pathway are shown in table Table 1. Sequences of primers used for RT-PCR Gene Primer sequence Temp.( ) TAP1 5'-CTC TGA GTG AGA ATC TGA GC-3' 57 gi: '-GAG ACA TCT TGG AAC TGG AC-3' TAP2 5'-CGC CTT CTT CTT CCT TGT CC-3' 54 gi: '-CTG AGC ATG AAG CCA TAC AG-3'

40 LMP2 5'-CAT CTA CTG TGC ACT CTC TG-3' 57 gi: '-CAG CTG TAA TAG TGA CCA GG-3' LMP7 5'-CGA ACA CGA ACA TGA CAA CC-3' 58 gi: '-GCC ACA TGA GTG TCT TAC TG-3' LMP10 5'-CAA GAG CTG CGA GAA GAT CC-3' 54 gi: '-GTA GCG GCC AGA CCT CTT CA-3' X 5'-GAC GGT GAA GAA GGT GAT AG-3' 58 gi: '-TTG ACT GCA CCT CCT GAG TA-3' Y 5'-CGC CAA TCG AGT GAC TGA CA-3' 58 gi: '-AAG CGA GAG CAT TGG CAG TG-3' Z 5'-CGG CTG TGT CGG TGT ATG CT-3' 58 gi: '-CTC ACA CCT GTA CCG GCC AA-3' calnexin 5'-GGA AGT GGT TGC TGT GTA TG-3' 54 gi: '-TTC ACA TAG GCA CCA CCA CA-3' calreticulin 5'-AAG TTC TAC GGT GAC GAG GA-3' 54 gi: '-CTC TCC GTC CAT CTC TTC AT-3' PA28α 5'-GGA GCC AGC TCT CAA TGA AG-3' 54 gi: '-GCA TCA CCA CGC TCA GAG AA-3' PA28β 5'-GGA GGT CTT CAG GCA GAA TC-3' 54 gi: '-ATA GGC TGC CTC ATC TCG CT-3' HLA-A2 5'-GAG AAG GCC CAC TCA CAG A-3' 57 gi: '-TAT CTG CGG AGC CAC TCC AC-3' HLA-DR 5'-GCT CCA ACT CTC CGA TC-3' 54 gi: '-CCA CGT TCT CTG TAG TCT CTG G-3' CD64 5'-ATG GCA CCT ACC ATT GCT CAG G-3' 54 gi: '-CCA AGC ACT TGA AGC TCC AAC TC-3' GAPDH 5'-CGG GAA GCT TGT GAT CAA TGG-3' 55 gi: '-GGC AGT GAT GGC ATG GAC TG-3'

41 III. RESULTS The significant roles of MHC class I-restricted CD8 + T cells in protective immunity for M. tuberculosis have been recognized in recent years. In order to understand the protective immune mechanism mediated by MHC class I-restricted CD8+ T cells, we have performed the following three different approaches ; 1) Characterization of HLA-A*0201-restricted CD8+ T cells specific for M. tuberculosis epitope peptide 2) MHC class I antigen processing pathway of M. tuberculosis somatic antigens 3) Gene expression of components of the MHC class I antigen processing machinery following infection with M. tuberculosis. 1. Characterization of HLA-A*0201-restricted CD8+ T cells specific for M. tuberculosis epitope peptides A. HLA-A*02 allele types of the study subjects Among 182 HLA-A typed individuals, 91 appeared to carry one of the HLA-A*02 alleles. As showed in Table 2, 63.6% of healthy subjects carries HLA-A*02 alleles. Among these, 30.3% expresses a HLA-A*0201 and the remaining 33.3% expresses other HLA-A*02 alleles. In pulmonary TB patients, 51.2 % of them were HLA-A*02 positive. Among these, 26.2% carries HLA-A*0201 and remaining 25% carries other HLA-A*02 alleles. Only HLA-A*0201, 0203, 0206 and 0207 of HLA-A*02 alleles

42 were identified in the Korean population in our study. In contrast to pulmonary TB patients, only 26.3% of TB pleurisy patients were HLA-A*02 positive. This is about 50% less than pulmonary TB patients. Even though only 38 TB pleurisy patients were typed in this study, this result implies that HLA-A*02 positive individuals may be more efficiently protected from TB pleurisy than HLA-A*02 negative individuals. Among the subjects expressing HLA-A2, 12 healthy PPD+ subjects, 7 PPD- subjects, 15 patients with mild or moderate (mild-moderate) TB, 5 patients with far-advanced TB and 3 TB pleurisy patients were selected in this study (Table 3). Table 2. HLA-A genotype frequencies HLA-A type Healthy subjects TB patients A* % (18) 26.2% (22) A* % (2) 1.2% (1) A* % (10) 15.5% (13) A* % (8) 8.3% (7) Total 63.6% (38/60) 51.2% (43/84)

43 Table 3. Demography of the study subjects participated in this study Group N* Age range M / F PPD+ healthy subjects (33) 10 / 2 PPD- healthy subjects (28) 6 / 1 Mild-moderate TB (35.4) 10 / 2 Far-advanced TB (45.7) 5 / 0 TB pleurisy (52) 1 / 2 *N designates number of participants in each group. The number in parenthesis indicates the mean ages of subjects. M and F designate male and female, respectively. B. CD8+ STCL generation from HLA-A*0201 PPD+ healthy subjects STCLs (short term cell lines) were generated by stimulating PBMCs for days with each peptide in the presence of IL-2. Cells were subsequently re-stimulated with each peptide overnight and stained for intracellular cytokine production for either IFN-γ or IL-4 (Fig 1). We observed each peptide-specific CD8+ STCL was all generated from PBMCs of PPD+ healthy subjects and produced IFN-γ upon peptide stimulation; however, these cells were not able to produce IL-4 indicating that these cells are Tc1 type cells. Flu peptide-specific for HLA-A*0201-restricted CD8+ T

44 cells was previously found to be able to bind at least four alleles of the HLA-A2 supertype including A*0201, A*0203, A*0206 and A* Fig. 1. Intracellular cytokine staining for HLA-class I-restricted CD8+ T cell populations specific for M. tuberculosis peptides in HLA-A*0201 subjects. STCL from PPD+ subjects expressing HLA-A*0201 were stimulated with each peptide overnight (PstA , ThyA 30-38, RpoB , 85B 15-23, and Flu ). On next day, STCLs were treated with brefeldin A and stained with IFN-γ, CD3+ and CD8+ specific mab labeled with PE, PerCP and FITC, respectively. For the intracellular staining of IL-4, anti-il-4 mab labeled with PE was used in the assay. The frequencies of the peptide-specific IFN-γ or IL-4 secreting CD8+ T cell population were measured from the CD3-gated lymphocyte population

45 C. Detection of IFN-γ producing CD8+ T cells from PPD+ subjects expressing HLA-A2 supertype To identify whether these newly defined epitopes are A2 supertype peptides, the four peptides-specific for CD8+ T cell-mediated immune responses were examined using STCL generated from PBMCs of A*0203, A*0206 and A*0207 subjects (Fig 2). Using an intracellular IFN-γ staining method, we observed that CD8+ T cells specific for each peptide produced IFN-γ upon stimulation with each peptide in PPD+ healthy subjects expressing HLA-A2 supertype (A*0203, A*0206 and A*0207). However, PstA and RpoB peptide-specific CD8+ T cell response was not induced in HLA-A*0203 subjects although only two healthy subjects were screened. It may be necessary to screen more subjects to explain whether PstA and RpoB peptide-specific CD8+ T cell responses can be induced in HLA-A*0203 subjects. It is known that > 50% of the East-Asian population expresses the HLA-A*0201, A*0203, A*0206 or A*0207 subtype. Therefore, these peptides may be useful as effective vaccine components for the prevention of TB in East-Asia, one of the TB endemic areas in the world

46 Fig. 2. Intracellular IFN-γ staining for HLA-class I-restricted M. tuberculosis peptide reactive CD8+ T cell populations in PPD+ subjects expressing HLA-A2 supertype. STCLs from PPD+ subjects expressing the HLA-A2 supertype (A*0203, A*0206 and A*0207) were stimulated with each peptide overnight (PstA , ThyA 30-38, 85B 15-23, RpoB and Flu ). On next day, each STCL was treated with brefeldin-a and stained with IFN-γ, CD3+ and CD8+ specific antibodies labeled with PE, PerCP and FITC, respectively. The frequencies of peptide-specific IFN-γ secreting CD8+ T cell population were measured from the CD3-gated lymphocyte population

47 D. Quantification of the CD8+ T cell frequencies specific for M. tuberculosisderived peptides from subjects expressing HLA-A2 supertype To determine the correlations of the CD8+ T cell immune responses in latently infected subjects and in chronically infected TB patients, the frequency of CD8+ T cells specific for each peptide in PBMCs from PPD+ and PPD- healthy subjects and TB patients was quantified using IFN-γ elispot assay. To induce the IFN-γ production from peptide-specific CD8+ T cells after depletion of CD4+ T cells, rhil-2 and anti- CD28 mab were included in the assay. These peptide-specific CD8+ T cell responses were detected in subjects expressing the A*0201, A*0203, A*0206 and A*0207 subtypes. Again, PstA specific CD8+ T cell responses were low in A*0203 subjects. In the IFN-γ elispot assay, there was no hierarchy of the frequencies or spot sizes based on the subtypes (Fig. 3), which suggests that these peptides may bind to any of the subtypes with similar affinities in this assay system except PstA peptide. The presence of the peptide reactive CD8+ T cell responses suggests that these peptides were processed and presented to the T cells during a natural infection in latently or actively infected individuals. Fig. 3. Photomicrograph displaying IFN-γ specific elispot formation from CD8+ T cells

48 The frequencies of the CD8+ T cell populations specific for these peptides ranged from 1 to 40 in the 5.5 x 10 5 circulating PBMCs in the study subjects. Among the four peptides tested, the frequencies of the CD8+ T cells specific for ThyA 30-38, RpoB or 85B peptides were higher in the PPD+ healthy subjects than either of the PPDhealthy subjects or in the far advanced TB patients. There were no statistically significant differences among the four groups regarding the CD8+ T cell response to the PstA peptide. The frequency of the RpoB specific CD8+ T cell population in the PPD+ subjects was significantly higher than that in the PPDsubjects (p=0.0001), mild-moderate TB patients (p=0.001) or the far-advanced TB patients (p=0.001). The statistical differences were also observed in the frequencies of the RpoB (p=0.049) and 85B (p=0.05)-specific CD8+ T cell population by comparing the epitope-specific frequency of CD8+ T cells from the patients with mild-moderate TB as well as that from the patients with far-advanced TB. In the mildmoderate TB patients, no significant difference in the CD8+ T cell frequency was observed compared with the PPD+ healthy subjects except in the RpoB specific CD8+ T cells. Therefore, the immune response of the CD8+ T cells for these M. tuberculosis antigens appears to be induced in both the PPD+ healthy subjects and patients with mild-moderate TB for the protective immunity to M. tuberculosis. On the other hand, the CD8+ T cell-mediated immune responses for some of these M. tuberculosis antigens appeared to be decreased in the far-advanced TB patients. However, this depression of the CD8+ T cell responses may depend on the M. tuberculosis antigens, as shown in Fig. 4. The frequencies of the control peptide,

49 Flu 58-66, specific CD8+ T cell populations were similar among the four groups. Fig. 4. CD8+ T cell-mediated responses to M. tuberculosis-specific epitope peptides in TB patients and healthy subjects using ex-vivo IFN-γ elispot assay. An ex-vivo IFN-γ elispot assay was used to quantify the frequency of the circulating epitope-specific IFN-γ secreting CD8+ T cells in healthy subjects (PPD+ and PPD-), patients with mild or moderate TB (mild TB) and patients with far-advanced TB

50 (severe TB). Each dot designates the number of IFN-γ secreting CD8+ cells per 5.5x10 5 PBMCs for one person (A*0201: red, A*0206: blue, A*0207: green and A*0203: pink dots). The frequencies of the SFCs (spot forming cells) were calculated as a mean of two duplicate wells and the values are expressed as number of spots per 5.5x10 5 PBMCs after subtracting the number of spots in the un-stimulated PBMCs. The significant difference compared with PPD+ (*P<0.05 and **P<0.001), a significant difference compared with mild-moderate TB (# P<0.05) and a significant difference compared with PPD- (+P<0.05) was statistically calculated using twosample Student t-test for means. a) PstA , b) ThyA 30-38, c) RpoB , d) 85B and e) Flu peptides were used for IFN-γ elispot assay. E. Quantification of the frequencies of the CD8+ T cells specific for the M. tuberculosis-derived peptides in a pleural effusion from TB pleurisy patients Pleural tuberculosis is a localized disease where the protective immune responses are active. Therefore, an IFN-γ elispot assay was performed to examine if these epitopespecific CD8+ T cell immune responses are concentrated in the pleural effusion from TB pleurisy patients (Fig. 5). Interestingly, the epitope-specific release of IFN-γ by CD8+ T cells was observed in all three patients (two patients : A*0201 and one patient : A*0206). This suggests that these peptide-specific immune responses are involved in the protective immunity at the site of disease. The frequencies of CD8+ T cells specific to the M. tuberculosis peptides in TB pleural effusion were equivalent to those in the PBMCs

51 Fig. 5. CD8+ T cell-mediated responses to M. tuberculosis-specific epitope peptides in TB pleurisy patients using ex-vivo IFN-γ elispot assay. The values are expressed as a number of spots per 5.5x10 5 pleural effusion derived cells after subtracting the number of spots in the un-stimulated cells. The data is expressed as a mean + standard deviations. Flu is used as a positive control

52 F. In vitro induction of recall CTL responses from healthy subjects expressing HLA-A2 supertype Not only IFN-γ production but also cytotoxic T cell responses by CD8+ T cells in human as an important function to control M. tuberculosis infection. Therefore, we observed the IFN-γ production specific these peptides in healthy BCG vaccinated subjects. Recall responses measure memory cells but not necessarily effector T cells. memory T cells generally have poor cytotoxicity without in vitro restimulation 37. Thus, we in vitro stimulated PBMCs from BCG vaccinated subjects and tested the CTL activities. Recall CTL activities specific for PstA , RpoB , ThyA and 85B peptides were all observed in healthy BCG vaccinated subjects expressing HLA-A*0201 type (Fig. 6). Moreover, PstA , ThyA and 85B peptides stimulated CD8+ T cell lines showed CTL activity in healthy BCG vaccinated subjects who express HLA-A*0206 subtype

53 A) HLA-A*

54 B) HLA-A*0206 Fig. 6. Recall CTL response from healthy subjects expressing HLA-A2 supertype. Synthetic peptides were added to the PBMCs cultures at a final concentration of 10 μg / ml. On days 3 and 6, rhil-2 was added to each well at a concentration of 10 units/ ml. On day 8, the cultures were re-stimulated with irradiated (3000 rad) autologous monocytes that were pulsed with peptide in the presence of 3 μg / ml of β 2 m for 2 h. On days 10 and 13, 10 units/ ml of ril-2 was added into each well. The cytolytic activity of cultured PBMCs was tested on day 14. Cytolytic activity was measured by a standard 4 h 51 Cr release assay

55 G. Enumeration of frequencies of peptide-specific CD8+ T cell populations using HLA-A*0201 dimer complexes The HLA-A*0201-restricted CD8+ T cell specific peptides were used to synthesize A*0201 dimer complexs to stain the peptide-specific CD8 + T cells from PBMCs. To determine the specificity of each dimer, fresh blood PBMCs were obtained from HLA-A*0201 subjects. Dimer staining revealed that the frequency of circulating PstA specific CD8+ T cells was higher when measured by dimer staining compared with IFN-γ intracellular staining (Fig. 7). It indicated that PstA specific CD8+ T cell population in PPD+ healthy subjects is functionally heterogeneous since one-half or one-fourth of PstA specific CD8+ T cell population in PPD+ healthy subjects produced IFN-γ upon peptide stimulation. The ratio of CD8+ T cell population specific for IFN-γ production and A*0201-dimer was not distinctly different in other peptide-specific T cell population (Table 4)

56 Table 4. Frequency of peptide-specific CD8+ T cells detected by IFN-γ intracellular staining and HLA-A*0201 dimer staining Donors PstA ThyA RpoB B Flu ICS / dimer (%) ICS / dimer (%) ICS / dimer (%) ICS /dimer (%) ICS /dimer (%) / / / / / / / / / / / / / / / / / / / / / / / / / 1.6 *ICS : intracellular staining, dimer : HLA-A*0201 dimer

57 Fig. 7. HLA-A*0201 dimer staining for M. tuberculosis peptide-specific STCL generated from healthy subjects. Dimeric complexs were prepared by mixing HLA- A*0201 : Ig protein with specific peptide at 640 molar excess at 37 overnight incubation. Peptide-specific STCLs were incubated with 10 μl pooled human serum at RT for 10 min. Cells were incubated with 1-2 μg of peptide loaded HLA-A*0201 : Ig protein to each sample for 60 min at 4. Subsequently, cells were incubated at 4 for 30 min in FACS buffer containing anti-cd8 mab, anti-cd3 mab and anti-mouse IgG

58 H. Immunocytochemistry for perforin and granulysin expression in M. tuberculosis peptide-specific CD8+ CTL lines Granulysin is a protein present in cytotoxic granules of CTL and natural killer (NK) cells, which are released upon antigen stimulation. CD8+ CTLs lyses M. tuberculosis infected macrophages by a granule-dependent mechanism that results in killing of the intracellular pathogen. Moreover, granulysin directly killed extracellular M. tuberculosis changing the membrane integrity of the bacillus and in conjunction with perforin, decreased the viability of intracellular M. tuberculosis 11. Confocal microscopy analysis showed that cytotoxic granules, granulysin and perforin, are colocalized in PstA , ThyA 30-38, RpoB , 85B 15-23, and Flu specific CD8+ CTLs. This data indicates that these peptide-specific cell lines released perforin and granulysin may be lead to kill M. tuberculosis (Fig. 8)

59 Fig. 8. Detection of perforin and granulysin in CTLs by confocal laser microsopy. Fluorescent confocal images were obtained for granulysin expression (red, first panel of each row) and perforin expression (green, second panel of each row). The two images were then superimposed (third panel of each row) to show vesicles expressing both perforin and granulysin (yellow)

60 2. MHC class I antigen processing pathway of M. tuberculosis somatic antigens A. Generation of CD8+ T cell lines for M. tuberculosis somatic antigens To demonstrate that CD8+ T cell response against epitopes-derived from somatic antigens can be induced in healthy subjects expressing HLA-A*0201 and A*0206 type, in vitro CD8+ CTL induction experiments were performed. PBMCs from healthy HLA-A*0201 and A*0206 subjects were stimulated weekly for 4-5 weeks with autologous PBMCs pulsed with RpoB peptide. CD8+ T cell lines specific for RpoB peptide were generated. These CTL lines showed cytotoxic activity for RpoB peptide pulsed target cells as shown in Fig

61 Fig. 9. Generation of CTL lines from healthy HLA-A2 subjects by in vitro immunization PBMCs from HLA-A*0201 and HLA-A*0206 subjects were pulsed with 50 μg / ml of peptide at cells/well in 10% pooled human serum with rhil- 7 (10 ng/ ml ) and keyhole limpet hemocyanin (5 μg / ml ). Cultures were re-stimulated weekly with peptide-pulsed and irradiated autologous monocytes and cell culture was supplemented with rhil-2 at 10 units/ ml every 3-4 days. After four to five cycles of re-stimulation, the cytotoxic activity of the CD8+ T cells was determined by using the chromium release CTL assay (A*0201 target cell :.221A2 cell lines, A*0206 target : EBV-transformed line from A*0206 PBMC)

62 B. Kinetics of cytotoxic activities of RpoB specific CD8+ T cells for M. tuberculosis-infected macrophages The kinetic studies of specific killing of M. tuberculosis-infected (3 MOI) macrophages by RpoB specific CD8+ T cells suggest that RpoB protein is processed more efficiently when the infection period is increased longer (Fig 11). Infectivity of macrophages was >70% by Ziehl-Neelsen method (Fig. 10). Fig. 10. AFB staining. Human monocyte-derived macrophages were infected with H37Rv at 3MOI for 4 h. After three times washing, cells were fixed and stained with Ziehl-Neelsen method (infectivity >70%)

63 Fig. 11. Kinetics of M. tuberculosis RpoB peptide processing inside macrophages. Macrophages were generated by culturing adherent monocytes in RPMI 1640 with 10% FBS for 3-4 days. These macrophages were then infected with M. tuberculosis (H37Rv) at a multiplicity of infection of 3 for 1-4 days. Extracellular non-phagocytosed M. tuberculosis was removed by washing. For cytotoxicity experiments, 1ⅹ10 6 macrophages were labeled with 100µCi of 51 Cr for 1 hour at 37, and then added to the wells of 96 well U-bottom plates at 5ⅹ10 3 cells/well. CD8+ T cell lines were added at various effector to target ratios. Specific lysis (%) was determined as described above

64 C. RpoB peptide presentation requires phagocytosis and bypass Golgi-ER transport in M. tuberculosis-infected macrophages RpoB peptide-specific CD8+ T cell lines generated from both HLA-A*0201 and A*0206 subjects showed the specific lysis for M. tuberculosis-infected macrophages. Processing of RpoB (RNA polymerase β-subunit) protein of M. tuberculosis for the presentation of RpoB peptide to MHC class I-restricted CD8+ T cells was insensitive to brefeldin-a (Golgi-ER transport inhibitor) and lactacystin (proteasome inhibitor) treatment, although the recognition of M. tuberculosis-infected cells by RpoB specific CD8+ T cells was inhibited by anti-mhc class I blocking antibody (W6/32) or cytochalasin D (phagocytosis inhibitor). This result indicates that M. tuberculosis RpoB peptide is processed and presented from intracellular MTB antigen to these CD8+ T cells. It requires phagocytosis of the bacteria, with antigen entry to the cytoplasm. However, the absence of inhibition by brefeldin A demonstrates that the antigen processing may bypass the Golgi-ER pathway. In addition, the processing seems to be not inhibited by lactacystin. This result demonstrated RpoB antigen was not processed by proteasomal degradation. This suggests that the RpoB peptide may be processed by the alternative MHC class I- restricted presentation pathway (Fig. 12)

65 A) Cytotoxic assay using CTL line generated from HLA-A*0201 PBMCs B) Cytotoxic assay using CTL line generated from HLA-A*0206 PBMCs Fig. 12. Effect of metabolic inhibitors on presentation of M. tuberculosis-derived RpoB protein. RpoB specific CD8+ CTL lines were stimulated with macrophages that had been preincubated with metabolic inhibitors, brefeldin A (Golgi-ER transport; 3 μg / ml ), lactacystin (proteasome, 40µM) or cytochalasin D (phagocytosis; 10 μg / ml ) for 1 h before the addition of M. tuberculosis. Anti-MHC class I blocking antibody (W6/32, 10 μg / ml ) was added to target cells 1 h before CTL assay

66 3. Gene expression of components of the MHC class I antigen processing machinery following infection with M. tuberculosis It has been known that M. tuberculosis infection affected the expression of enormous genes in host cells. M. tuberculosis infects and replicates mainly in either macrophages or dendritic cells. In our next experiment, therefore, we analyzed the profile of the gene expression of MHC class I antigen processing components either in M. tuberculosis-infected macrophages or dendritic cells. A. Effect of IFN-γ on the expression of MHC molecules of macrophages Monocytes-derived macrophages were treated with various concentrations of rhifn-γ for 18 h. The highest expression of MHC class I (W6/32) and MHC class II (DR) in human macrophages was observed when the concentration of IFN-γ was 20 ng/ml (Fig. 13)

67 Fig. 13. Effect of IFN-γ concentrations in the expression of human MHC molecules. Human monocyte-derived macrophages were treated with various concentrations of rhifn-γ for 18 h. Cells were stained with isotype control Ab (solid histograms), anti-class I Ab (W6/32 : black histograms) or class II Ab (DR : grey histograms) and then analyzed by flow cytometry

68 B. The kinetics of IFN-γ induced gene expression of MHC class I Ag processing pathway affected by M. tuberculosis infection After treating macrophages with various concentrations of rhifn-γ (20 ng/ ml ), the expression of genes involved in MHC class I processing machinery were examined by RT-PCR analysis. The expression of TAP2 gene by IFN-γ induction was downregulated by M. tuberculosis infection at 24 h, while the expression of TAP1 gene by IFN-γ induction was increased maximally by M. tuberculosis infection at 12 h (Fig. 14A). The mrna expression of LMP2, TAP2 and LMP10 genes by IFN-γ induction was not increased at the first 4 h after IFN-γ treatment. In addition, the mrna expression of TAP2 and LMP2 genes by IFN-γ induction was increased in a time dependent manner and it is maximized at 24 h (Fig. 14A). M. tuberculosis infection inhibited the IFN-γ induction of LMP2 gene expression at 24 h (Fig. 14A). The expression of CD64 gene by IFN-γ induction was maximal at 8 h (Fig. 14B). Up to 8-12 h, the expression of PA28α gene by IFN-γ induction was steadily maintained and its expression decreased at 24 h. But the expression of PA28β by IFN-γ induction was maintained up to 24 h (Fig 14B). The expression of other genes was not nearly changed throughout the whole incubation period with IFN-γ and M. tuberculosis infection (Fig 14B, C). Therefore, this result indicates that MHC class I processing genes were regulated differently during the time periods after IFN-γ or M. tuberculosis infection

69 Fig. 14A. The kinetics of MHC class I gene expression in response to IFN-γ treatment and M. tuberculosis infection. Densitometric quantitation of the relative induction of MHC class I processing genes were normalized with the amount of GAPDH mrna in human macrophages. The fold induction was calculated as (ratio of target gene/gapdh intensity in study group) / (ratio of target gene/gapdh intensity in control). The detailed information of primers is described in Table

70 Fig. 14B. The kinetics of MHC class I gene expression in response to IFN-γ treatment and M. tuberculosis infection. Densitometric quantitation of the relative induction of MHC class I processing genes were normalized with the amount of GAPDH mrna in human macrophages. The fold induction was calculated as (ratio of target gene/gapdh intensity in study group) / (ratio of target gene/gapdh intensity in control)

71 Fig. 14C. The kinetics of MHC class I gene expression in response to IFN-γ treatment and M. tuberculosis infection. Densitometric quantitation of the relative induction of MHC class I processing genes were normalized with the amount of GAPDH mrna in human macrophages. The fold induction was calculated as (ratio of target gene/gapdh intensity in study group) / (ratio of target gene/gapdh intensity in control)

72 C. Isolation and characterization of human immature DCs from peripheral blood derived adherent cell cultures. We generated DCs from adherent PBMCs cells by culturing with rhgm-csf and rhil-4. After 3-4 days of culture, analysis of cell surface markers was performed and it was demonstrated that cells expressed DCs specific marker, CD1a and CD83. These cells expressed high levels of MHC class I and MHC class II molecules as well as the co-stimulatory molecules such as CD80 and CD86. These DCs appear to have the phenotype and functional characteristics of immature DCs which are very effective at antigen capturing and processing 38, 39 (Fig. 15)

73 Fig. 15. Cell surface molecule expression of human immature dendritic cells. Human monocyte-derived immature DCs were treated with rhgm-csf and rhil-4 for 3 days. Cells were stained with IgG isotype control, or FITC conjugated Abs against DR, CD1a, CD80, PE conjugated Abs against CD83 and CD86 and subsequently examined by flow cytometry. A) DR and W6/32 (solid histograms), isotype control (open histograms), B) CD1a and CD80 (open histograms), isotype control (solid histograms), C) CD83 and CD86(solid histograms), isotype control (open histograms)

74 D. The effect of IFN-γ treatment and M. tuberculosis infection on the expression of genes involved in MHC class I antigen processing The expression of the genes involved in MHC class I antigen processing, TAP, proteasome subunits, proteasome activators, different chaperones and MHC class I antigens were analyzed by RT-PCR. After treating human monocyte-derived macrophages with IFN-γ (20 ng/ ml ), mrna expression levels of TAP-1, TAP-2, LMP2, LMP10, PA28α, PA28β and HLA-DR were increased (Fig. 17, 19). IFN-γ regulated differentially gene expression of proteasome inducible components, LMP10, PA28α, PA28β and constitutive subunits X, Y, Z 40. The promoter regions of TAP2, LMP7, LMP10 and PA28 all contain the IFN-consensus sequences 31, 41, suggesting a distinct regulation of these genes upon IFN treatment. But mrna of LMP7 after IFNγ treatment was not nearly increased in their experiment (Fig. 17, 19). In addition, M. tuberculosis infection inhibited IFN-γ induction of proteasome component LMP2 and TAP2 gene of macrophage and DCs by 10-20% (Fig. 17, 18). In addition, the IFN-γ induction of LMP10 was slightly reduced in macrophage and DCs by MTB infection (Fig 17, 19). Interestingly, IFN-γ increased the expression of TAP1 gene on M. tuberculosis infected DCs and macrophages 1.2 fold (Fig 17, 19). On the contrary, M. tuberculosis infection inhibited IFN-γ induction of TAP2 gene of macrophage and DCs by 10-20% (Fig 17-20). Interestingly, BCG infection inhibited IFN-γ induction of TAP1 and TAP2 genes of macrophages. Moreover, the M. tuberculosis infection affected

75 differently mrna expression of HLA-DR depending on cell types. Namely, infection of macrophages either with M. tuberculosis or BCG inhibited IFN-γ induction of DR, but infection of DCs with M. tuberculosis accelerated IFN-γ induction of DR expression. The inhibitory effect by M. tuberculosis and BCG infection on the induction of gene expression by IFN-γ seems to be various depending on the responsive genes. The gene expression levels involved in MHC class I processing pathway were different dependent on antigen presenting cells. Infection with M. tuberculosis inhibited IFN-γ signaling on CD64 gene in macrophages and DCs, similar to previously reported data. Moreover, infection with BCG inhibited IFN-γ signaling on CD64 gene in macrophage. The gene expression of chaperons such as calnexin, calreticulin, x, y and z was not changed according to IFN-γ treatment, M. tuberculosis infection or BCG infection. These results suggest that cellular signal transduction pathways which are activated by IFN-γ might be differentially interfered depending on gene with M. tuberculosis and BCG Infection. Accordingly, M. tuberculosis might survive within macrophages and DCs by evading protective immune mechanism of host cells

76 Fig. 16. RT-PCR analysis for MHC class I antigen processing genes in human monocyte-derived dendritic cells (DCs). Total cellular RNA from human DCs infected with M. tuberculosis (10MOI) and treated with 20 ng/ ml IFN-γ for 8 h was extracted and used for RT-PCR analysis. GAPDH is used as an internal control of RNA amount

77 Fig. 17. Densitometric quantitation of the relative induction of genes involved in MHC class I antigen processing. Results were normalized for the amount of GAPDH mrna in human monocyte-derived dendritic cells after IFN-γ and M. tuberculosis infection. The fold induction was calculated as (ratio of target gene/ GAPDH intensity in study group)/(ratio of target gene / GAPDH intensity in control). Data shown are representative of three experiments and are expressed as mean ± S. D

78 Fig. 18. RT-PCR analysis to exam the mrna expression pattern of genes involved in MHC class I antigen processing in macrophages. Total cellular RNA from human macrophages infected with M. tuberculosis for 24 h and subsequently treated with 20 ng/ ml. IFN-γ for 8h was extracted and used for RT-PCR analysis. GAPDH is used as an internal control of RNA amount in each sample

79 Fig. 19. Densitometric quantitation of the relative induction of genes involved in MHC class I antigen processing. Results were normalized for the amount of GAPDH mrna in human macrophages after IFN-γ and M. tuberculosis infection. The fold induction was calculated as (ratio of target gene / GAPDH intensity in study group)/(ratio of target gene / GAPDH intensity in control). Data shown are representative of three experiments and are expressed as mean ± S. D

80 Fig. 20. RT-PCR analysis for genes involved MHC class I antigen processing in human monocyte-derived macrophages. Total cellular RNA from human macrophages infected with BCG (10MOI) and treated with 20 ng/ ml IFN-γ for 8h was extracted and used for RT-PCR analysis. GAPDH is used as an internal control of RNA amount

81 Fig. 21. Densitometry quantitation of the relative induction of genes involved in MHC class I antigen processing. Results were normalized for the amount of GAPDH mrna in human macrophages after IFN-γ and BCG infection. Fold induction calculated as (ratio of target gene / GAPDH intensity in study group)/(ratio of target gene / GAPDH intensity in control group). Data shown are representative of two experiments and are expressed as mean ± S. D