UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Functional relevance and structural requirements of peptide transport in a murine carcinoma cell line Reid, Gregor S.D. 1996

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1997-196437.pdf [ 8.18MB ]
Metadata
JSON: 831-1.0088072.json
JSON-LD: 831-1.0088072-ld.json
RDF/XML (Pretty): 831-1.0088072-rdf.xml
RDF/JSON: 831-1.0088072-rdf.json
Turtle: 831-1.0088072-turtle.txt
N-Triples: 831-1.0088072-rdf-ntriples.txt
Original Record: 831-1.0088072-source.json
Full Text
831-1.0088072-fulltext.txt
Citation
831-1.0088072.ris

Full Text

A B S T R A C T Major His tocompat ib i l i ty Complex Class I molecules mediate the presentation of endogenously synthesised antigens to the cytotoxic cells of the immune system. The expression of Class I molecules on the surface of most somatic cells, therefore, provides a mechanism by which the immune system can identify and destroy infected or cancerous cells. However , down-regulat ion of surface Class I expression is frequently observed in vira l ly infected and cancerous cells, a situation that al lows these cells to evade detection by the immune system. The study of cells exhibiting deficiencies in Class I expression has yielded many insights into the cellular processes invo lved in Class I restricted antigen presentation. In this study the antigen processing deficiency of the murine small cel l lung carcinoma ce l l line C M T . 6 4 is characterised. C M T . 6 4 cells express greatly reduced levels of Class I on their ce l l surface and are unable to present viral peptides to cytolytic T lymphocytes , unless treated wi th interferon-y. A l though these cells synthesise reduced amounts of both the heavy chain and beta-2-microglobul in components of the Class I complex, these deficiences do not account for the failure to present viral peptides. In addition, C M T . 6 4 cells fail to express the genes encoding the T A P 1 and T A P 2 components of the T A P - d e p e n d e n t peptide transporter. The res tora t ion of expression of these genes is sufficient to restore presentation of endogenously synthesised antigens. W h i l e being unable to present antigens to cytotoxic T cells , C M T . 6 4 cells are eff iciently lysed by natural k i l l e r ce l l s . T h i s recogni t ion is unaffected by T A P gene expression, stabilisation of Class I by peptide puls ing or interferon-y i i t rea tment . The TAP-dependent peptide transporter plays a cr i t ical role in the assembly of Class I complexes in the endoplasmic ret iculum. Current models propose that this transporter is composed of both the T A P 1 and T A P 2 proteins. However, some results suggest that alternative forms of the T A P transporter may be functional. Studies with the T A P 2 deficient R M A - S ce l l line revealed vi ra l presentation in the absence of a T A P heterodimer. In this study, the role of the T A P 1 protein in R M A - S presentation is addressed using an antisense approach. This strategy indicates that the single subunit contributes to Class I expression. To further investigate this activity, the T A P deficient C M T . 6 4 cel l line was uti l ized. When expressed individual ly , both T A P 1 or T A P 2 proteins are capable of increasing surface Class I expression and presentation of viral and allogeneic peptides by C M T . 6 4 cells. The detailed characterisation of the C M T . 6 4 ce l l l ine provides novel insights into Class I assembly and antigen presentation. This work represents the first demonstration that the lack of expression of T A P genes by carcinoma cells is sufficient to inhibit antigen presentation by these cel ls . This information may be relevant in the development of strategies aimed at reducing tumour growth. The demonstration that alternative forms of the T A P transporter are functional revises the model of peptide transport into the endoplasmic ret iculum. This result also provides an explanation for the basal level of peptide transport observed in other antigen processing deficient cells. i i i TABLE OF CONTENTS A B S T R A C T i i T A B L E OF CONTENTS i v LIST OF TABLES .7. i x LIST O F FIGURES x A C K N O W L E D G E M E N T S x i i C H A P T E R 1 G E N E R A L I N T R O D U C T I O N 1 1.1 Specif ici ty of the vertebrate immune response 2 1.1.2 Specific antigen receptors 3 1.2 The Major Histocompatibil i ty Complex 4 1.2.1 The Class I and Class II loci 5 1.2.2 M H C restriction 5 1.2.3 Recognition of M H C antigens 6 1.3 Structure and assembly of M H C antigens 7 1.3.1 Class II biosynthesis 7 1.3.2 Class I biosynthesis 8 1.3.3 Peptide binding 1 0 1.4 Peptide generation for Class I assembly 13 1.5 Peptide Transport into the E R 16 1.5.1 Evidence for T A P involvement in peptide t r anspo r t 1 7 1.5.2 The nature of the T A P complex 1 8 1.5.3 In vitro analysis of T A P structure and function 19 1.5.4 Peptide selectivity of T A P transporters 2 1 1.5.5 Alternative pathways of peptide transport 2 2 i v 1.5.6 Presentation of exogenously derived peptides 2 4 1.6 M H C Class I and N K cells 2 5 1.7 Objectives and Approach 2 7 1.7.1 The C M T . 6 4 cell line 2 8 1.7.2 Vi rus presentation assays 2 9 C H A P T E R 2 M A T E R I A L S A N D M E T H O D S 3 0 2.1 Cel lular Methods 3 0 2.1.1 Tissue Culture 3 0 2.1.2 Transfection 3 1 2.1.3 Fluorescence activated cel l sorter ( F A C S ) analys is 3 1 2.1.4 Generation of effector cel l populations 3 2 2.1.5 Cytotoxici ty assays 3 3 2.1.6 Antibodies 3 4 2.1.7 Peptides 3 4 2.1.8 Viruses 3 5 2.1.9 Animals 3 5 2.2 Nucleic A c i d Techniques 3 5 2.2.1 Plasmids 3 5 2.2.2 Oligonucleotides 3 6 2.2.3 R N A isolation 3 6 2.2.4 Northern analysis 3 7 2.2.5 Reverse transcriptase polymerase chain reaction (PCR) analysis 3 7 2.2.6 3 2 P labell ing of hybridisation probes 3 8 2.3 Protein Techniques 3 9 v 2.3.1 C e l l labeling and Pulse-chase analysis 3 9 2.3.2 Immunoprecipi ta t ion 3 9 2.3.3 SDS-polyacrylamide gel electrophoresis ( P A G E ) . . . . 4 0 2.3.4 Western blotting 4 0 CHAPTER 3 3.1 I N T R O D U C T I O N 4 1 3.1.1 Evasion of the immune response by tumour cells 4 1 3.1.2 Mechanisms causing reduced Class I expression on tumour cell 4 3 3.1.3 Rationale for this study 4 4 3.2 R E S U L T S ". 4 5 3.2.1 C e l l surface M H C Class I expression in C M T . 6 4 cells 4 5 3.2.2 Virus presentation by C M T . 6 4 cells 4 7 3.2.3 M H C Class I processing in C M T . 6 4 cells 4 7 3.2.4 V i r a l peptide presentation by C M T . 6 4 cells 5 3 3.2.5 (32m expression and virus presentation 5 4 3.2.6 Peptide generation and delivery in C M T . 6 4 cells 5 7 3.2.7 N K cell lysis of C M T . 6 4 cells 6 7 3.3 DISCUSSION 7 3 3.3.1 The nature of the C M T . 6 4 deficiency 7 4 3.3.1.1 Relevance of decreased Class I heavy chain and P2m expression 7 5 3.3.1.2 Implications of the block in peptide transport ....7 7 3.3.1.3 Proteasome subunit requirements 7 9 3.3.2 Comparison of C M T . 6 4 with other described antigen processing deficient carcinomas 8 0 3.3.3 Possible underlying causes of the C M T . 6 4 phenotype 8 1 v i 3.3.4 C M T . 6 4 cells and N K recognition 8 2 C H A P T E R 4 4.1 I N T R O D U C T I O N 8 6 4.1.1 Evidence for non-heterodimeric forms of the T A P t r anspo r t e r '. 8 6 4.1.1.1 Human cell studies 8 6 4.1.1.2 Mouse cel l studies 8 8 4.1.2 Rationale for this study 8 9 4.2 R E S U L T S 9 1 4.2.1 Confirmation of the phenotype of the R M A - S ce l l line 9 1 4.2.2 Investigation of the functional role of T A P 1 in R M A - S ...9 7 4.2.3 Analysis of the function of T A P sub-units in C M T . 6 4 cells 1 0 0 4.3 DISCUSSION 1 14 4.3.1 Effect of T A P 2 on Peptide Transport in R M A - S Cel ls 115 4.3.2 The Role of T A P 1 in Peptide Transport in R M A - S Ce l l s . . . l 17 4.3.3 C M T . 6 4 Cells as a Mode l for T A P Transporter Ac t iv i t y . . . . l 2 0 4.3.3.1 C M T . 6 4 provide a T A P 1 and 2 deficient b a c k g r o u n d 1 2 0 4.3.3.2 Individually Expressed T A P Sub-units can Transport Peptides .121 4.3.3.3 Comparison of different T A P transporters 123 4.3.4 Possible explanations for T A P sub-unit transport efficiency 12 5 4.3.5 Contradictions with other peptide transport systems 1 2 6 C H A P T E R 5 5. G E N E R A L C O N C L U S I O N S 1 2 9 v i i 5.1 C M T . 6 4 cells and evasion of the immune response 129 5.2 Role of alternative T A P transporters in M H C Class I a s s e m b l y 1 3 2 6. NOMENCLATURE 13 5 PREFERENCES 136 v i i i LIST O F T A B L E S Table 1: Stabilization of cell surface Class I by exogenous peptide and p 2 m 4 8 Table 2 : Surface expression of M H C Class I complexes by C M T r l 2 clones 6 3 Table 3: Effect of increased K b and [32m expression in C M T . 6 4 on cel l surface Class I levels 6 5 Table 4: Comparison of K b expression by R M A and R M A - S 9 2 Table 5: Partial restoration of R M A - S surface K b expression by T A P 2 9 3 Table 6: Class I expression levels on R M A - S . p t m l clones 9 9 Table 7: Surface Class I expression by the C M T T A P t ransfectants 1 0 5 ix LIST O F FIGURES Figure 1: Schematic representation of the M H C Class I and Class II antigens 9 Figure 2: M H C antigen biosynthesis 12 Figure 3: M H C Class I expression by C M T . 6 4 cells 4 6 Figure 4: Virus presentation by C M T . 6 4 cells 4 9 Figure 5: Intracellular transport of M H C Class I molecules in C M T . 6 4 cells 5 1 Figure 6: Presentation of viral peptides by C M T . 6 4 cells 5 5 Figure 7: Presentation of V S V virus by C M T . 6 4 cells expressing elevated levels of (32m 5 6 Figure 8: Expression of T A P 1 and T A P 2 genes by C M T . 6 4 cells 5 8 Figure 9: T A P 1 and T A P 2 expression by C M T . 6 4 transfectants 6 2 Figure 10: Effect o f T A P 1 and T A P 2 expression on v i r a l presentation by C M T . 6 4 6 6 Figure 11: N K lysis of R M A , R M A - S and R M A - S . m t p 2 6 8 Figure 12: N K lysis of C M T . 6 4 cells and r l 2 clones 7 0 Figure 13: Effect of peptide and IFN-y treatment on N K ki l l ing of C M T . 6 4 cells 7 1 Figure 14: V S V presentation by R M A , R M A - S and R M A - S . m t p 2 9 5 Figure 15: Comparison of virus presentation by R M A , R M A - S and C M T . 6 4 9 6 Figure 16: T A P 1 levels are reduced in R M A - S clones expressing an antisense T A P 1 construct 9 8 Figure 17: R T - P C R analysis of T A P expression by C M T . 6 4 cells 101 x Figure 18: T A P m R N A expression by C M T r l and r2 transfectants 1 0 3 Figure 19: Pulse-chase analysis of K b and D b from C M T . 6 4 and T A P 1 transfected C M T (clone r l -4) cells 107 Figure 20: Virus presentation by C M T r l and r2 clones 1 0 9 Figure 21: Comparison of Inf l .A presentation efficiency by C M T . T A P clones 1 10 Figure 22: Al logeneic peptide presentation by C M T . 6 4 and the C M T . T A P clones 1 1 2 x i A C K N O W L E D G M E N T S The single name on the cover of this thesis belies the incalculable contribution of many other people to its completion. To my supervisor, Dr. Wi l f Jefferies, goes my thanks for providing, not only many of the ideas and facilities that are central to this work, but also a lab filled wi th great people. Too numerous to mention by name, I would like to thank all the members of the lab wi th whom it has been my pleasure to work over the past years. Special mention must go to Daphne Blew and Renee LeNobel for keeping the lab running smoothly despite the rest of us, and Nancy Carpenter and Doug Chidgey for their efficiency in keeping Biotech ticking along. The direction of this research has been influenced by several people. Drs. Reinhard Gabathuler and Katherine Barbey were ins t rumenta l i n the development of the ideas explored in this thesis and provided the requisite expertise to follow them through (in addition, they generously provided data included in this thesis); Gerry Koliatis and Caro l -Ann Sari educated me in the ways of C T L assays; several investigators, acknowledged in the text, generously provided reagents needed for many experiments; and my supervisory committee members, Drs. H u g h Brock, Kei th Humphries, Fumio Takei and Terry Snutch, provided critical input in the development of the final manuscript, as d id Cachou. Financial ly I owe my thanks to the Department of Zoology and the Biotechnology Laboratory for providing the wherewithal to keep me in perogies for the duration of my studies. M y teaching assistant duties wi th Zoology have added significantly to my learning experience at U B C . The generosity of the Scottish International Education Trust in providing support for the init ial years of my research was greatly appreciated. Dur ing my time at U B C , the highs would not have been so high and the lows a whole lot lower if not for the great friends around me. The "Old School" of Roger, Ian, Cyp , Joe, Kathy and Mike , as well as Forest, Jacquie and Alex made even the bad days in the lab enjoyable and provided the often much needed encouragement dur ing the sessions at the Barn. The Weenies (Andy , Ar t , Gareth, Deb and Barry), Katherine, Jeff, Steve, C y n d i , Cachou, Penelope and Duncan maintained my sense of humour throughout and introduced me to the local distractions needed to stop me from graduating too quickly. Although many miles away, the contribution of my family has been great. Their visits, letters, support, and even slagging, have constantly reaffirmed the solid foundation upon which this work is built. To my parents, especially M u m , the gratitude I feel for the support and freedom to make my own decisions cannot be fully expressed. Finally, my heartfelt thanks go to Kathy. The spirit and opt imism wi th which she fills the day more than makes up for the fact she keeps winning the hockey pool. Her patience, support and company have meant more to me than I can express and her influence on my life w i l l continue long after I've forgotten all about T A P transporter specificity. x i i 1. G E N E R A L INTRODUCTION The ability of organisms to survive and reproduce is dependent on their capacity to withstand the many threats posed by the environment. The fundamental importance of this ability is reflected by the presence of some form of defence mechanism in nearly al l l iv ing organisms (1). The spectrum of defence mechanisms ranges from external behaviour, such as "fight or flight" responses, to reactions occurring within the organism, examples of which are the presence of restriction enzymes in many bacteria and the production of highly specific antibodies by vertebrates. The science of immunology is classically concerned with the study of the internal processes that mediate protection of the organism. A s the most sophis t ica ted and m e d i c a l l y important defence mechanisms have evolved in vertebrates, these have become the primary concerns o f modern immunology . The highly specialized vertebrate immune system is composed of many diverse funct ional components. The ab i l i ty to mount a successful immune response i n v o l v e s complex in terac t ions be tween these different components. These interactions can be defined generally as belonging to either the innate or adaptive immune systems. The innate immune system represents a constant barrier of protection and shares many features with the mechanisms present in lower organisms. A t the cel lular level , the primary effectors of the innate immune system are the phagocytic cells, such as macrophages, that can attack almost any foreign material. However, their ability to respond is not influenced by previous interactions with the invading entity and is not specif ica l ly 1 directed to that entity. The adaptive immune response, as the name suggests, is inf luenced by prior exposure to the invading material . Mediated pr imar i ly by lymphocytes, the adaptive immune response is targeted specif ica l ly to ind iv idua l antigens and is characterized by a more rapid response upon subsequent contact with that antigen. Both the antibody production by B cells and the cytotoxic or helper responses of T cells display this antigen specificity. It is this arm of the vertebrate immune system that sets it apart from the systems of other organisms. 1.1 S p e c i f i c i t y of the ver tebra te i m m u n e r e s p o n s e The exquisi te specif ic i ty displayed by lymphocytes of the vertebrate immune system was a source of wonder to immunologists for many years. The demonstration that the system was capable of recogniz ing and mount ing a h igh ly specif ic response against sma l l man-made molecules , to wh ich there was no poss ib i l i ty of previous exposure, prompted many explanations (2). The correct solution was provided by Burnett 's "c lonal selection hypothesis", in which it was predicted that i n d i v i d u a l s con ta ined many lymphocy te c lones , each d i s p l a y i n g receptors with a single specificity (3). These clones are stimulated to divide by interact ion wi th the specific l igand recognized by their surface receptor. W h i l e the original model was proposed to explain the production of specific antibodies by B cells, the general premise was subsequently shown to apply also to T cel l responses. 1.1.2 Specific antigen receptors Despite their similarity in mediating clonal selection, the antigen receptors on B and T cells are structurally and functionally distinct. B cell specificity is mediated by surface immunoglobulin (slg) molecules. These are comprised of two heavy and two light chains, forming a complex with two identical antigen binding sites (reviewed in 1). In the case of T cells, the T-cell receptor (TcR) is formed by two chains, more commonly a and (3 but sometimes y and 8, which combine to form a single recognition site (4). The great range of specificity demonstrated by both slg and T c R molecules is the result of somatic gene rearrangement. By this mechanism, the coding region for a single chain of the receptor is generated by the splicing together of smaller regions, each encoding part of the chain. By having many different genes for each region, random rearrangement of these genes can produce chains with many different binding sites. The combining of these chains to form the functional antigen binding complex further increases the amount of receptor diversity generated. The difference in antigen recognition structures on B and T cells likely reflects the different functions of these molecules. In the case of B cells, the primary function of immunoglobulin is mediated by the soluble forms of the molecules, antibodies. Antibodies, secreted by activated B cells, bind directly to their specific target either directly altering it or targeting it for destruction by other cells. To achieve this, antibodies must be capable of binding in solution to antigenic epitopes expressed by large complex structures. By contrast, the T c R and the specific ligand 3 it recognises only exist in a membrane bound form. This al lows other molecules expressed on the surface of the cells involved to contribute to the b ind ing . U n l i k e antibody, which recognizes epitopes on native antigens, the T c R recognizes primari ly short fragments of the antigen, produced by intracellular processing (5). These fragments are expressed on the ce l l surface in association with major histocompatibili ty complex ( M H C ) antigens. 1.2 T h e Major H i s t o c o m p a t i b i l i t y C o m p l e x The abi l i ty of host defence mechanisms to protect against infect ion requires an efficient system for d iscr iminat ing self from non-self, a l l o w i n g specif ic attack of the foreign entities. In the vertebrate immune system this function is performed pr imar i ly by the M H C antigens. There are two main types of M H C antigen, Class I and Class II, which are distinct in their molecular structure, expression pattern and function. Together, however, these two molecules provide the means for identifying and eradicating a broad range of pathogens. The evidence indicating the role for these proteins in specifying self was obtained primari ly from work on tissue transplantation in mice (6,7). This work es tab l i shed that the h i s t o c o m p a t i b i l i t y genes c o n t r o l l e d t issue compat ib i l i ty between indiv iduals of the same species; a transplant between indiv iduals differ ing in the expression of these genes, an allograft, wou ld be rejected due to the immune response mounted against the histocompatibil i ty antigens. In a l l species studied, the genes promoting the strongest rejection reactions were found to be c losely linked, which gave rise to the name Major Histocompatibili ty Complex. 4 1.2.1 The Class I and Class II loci The two best studied M H C regions are those of mouse, the H-2 region on chromosome 17, and human, the H L A region on chromosome 6. The M H C regions of a l l the vertebrates in which they have been studied show a high degree of conservat ion. A l t h o u g h the order on the chromosome may change, the same classes of loc i are present. There are three distinct classes of loci within the M H C , although only Classes I and II encode histocompatibility antigens. The classical Class I antigens, also known as Class la , are encoded by the K , D and L loc i in the H-2 region and the A , B and C loci in the H L A region. A second group of Class I antigens, called Class lb , are encoded by several other genes in the Class I loc i . These latter antigens have a more l imited expression and are less polymorphic than the Class l a molecules (reviewed in 8), although their precise function is , as yet, unclear. In this thesis, unless otherwise stated, the term "Class I" refers to the Class l a complex type. Class II antigens are encoded in the H L A region by the D O , D P , D Q and D R genes and in the H-2 region by the I -A and I-E genes. 1.2.2 MHC restriction Although the role of lymphocytes in mediating allograft rejection had been demonstrated much earlier, the relat ionship between antigen recognition by T cells and the M H C antigens was not demonstrated until the virus presentation studies of Zinkernagel and Doherty (9). These experiments demonstrated that T cells generated by virus infection of mice were specific only for virus-infected cells sharing the M H C 5 antigens of the o r ig ina l mice. This phenomenon was termed M H C restriction and it has subsequently transpired that it is relevant for most antigen presentation to T cells. This dual specificity of the T cell for virus and M H C antigen is mediated by a single T c R complex. The development of a T c R repertoire restricted to self M H C is the result of selection of developing T cells within the thymus. 1.2.3 Recognition of MHC antigens The Class I and Class II M H C antigens are recognized by distinct sub-sets of T cel ls , dis t inguishable by their expression of co-receptor molecules (10). T cells expressing the C D 4 molecule recognize Class II antigens. The C D 4 + T cells are mainly of the helper (Th) phenotype. Recogn i t ion of Class II antigen complexes by these ce l l s results primarily in the release of cytokines that influence the function of other cells of the immune system, a cr i t ica l event in the generation of a successful immune response. Perhaps due to its central role i n stimulating an immune response, Class II antigen expression is l imi ted primari ly to specialized cells of the immune system, including dendritic cells and activated B cells and macrophages, although expression can be induced on some other ce l l types (11). Class I antigen recognition is achieved by T cells expressing the C D 8 molecule, which are mostly cytotoxic T cells ( C T L ) (10). Class I antigens are expressed on most nucleated cells, although the level of expression varies, being highest on cells of the immune system (12). There are, however, some notable exceptions, including neuronal cel l types (13) and trophoblast cells (14). The almost ubiquitous expression of Class I molecules allows the C D 8 + 6 C T L to identify and eliminate infected or altered cells throughout the body, a process known as immunosurveillance. 1.3 Structure and assembly of MHC antigens The different functions of the two M H C antigens is reflected in their pathways of biosynthesis and structural properties. The focus of the research described in this thesis involves Class I mediated antigen processing events. For this reason, the process of Class II antigen assembly w i l l be described only briefly, to al low comparison wi th the events required for Class I biosynthesis. 1.3.1 Class II biosynthesis A n assembled Class II complex consists of two non-cova len t ly associated chains, a and (3, both encoded by the M H C , and a peptide fragment (Figure 1). The a chain is the larger of the two, wi th a molecular mass of approximately 32 ki lodal ton (kDa) compared to 28 k D a for the (3 chain. Both chains are of s imilar structure, wi th two disulphide l inked extracellular domains, a transmembrane domain and a cy toplasmic ta i l . The peptide b inding region is formed by the membrane distal domains of each chain, a l and (31 respectively. These latter domains are also the site of the majority of the polymorphisms (reviewed in 15). Both the a and [3 Class II chains are inserted into the endoplasmic reticulum (ER) membrane by virtue of their signal sequences. A t this 7 point the chains assemble with a third n o n - M H C protein, the invariant (Ii) chain, into large complexes containing three copies of each chain (16). The presence of targeting sequences within the amino terminus of the Ii chain directs this complex to the plasma membrane v ia the endocytic pathway (17,18). In addition, the Ii chain appears to block the peptide binding cleft of the Class II complex unti l the complex is delivered to the appropriate site for interaction with peptides (19-21). Upon reaching a currently poorly defined compartment of the endocytic pathway (15), the Ii chain is cleaved, leaving only a small fragment, the C L I P fragment, bound to the Class II molecule (18,22). This fragment is subsequently removed by the action of the D M protein (23), a l lowing the Class II molecule to bind peptides in the compartment prior to transport to the ce l l surface (Figure 2). 1.3.2 Class I biosynthesis Correctly assembled Class I antigens are a trimeric complex comprised of heavy chain , be ta-2-microglobul in ((32m) and peptide (F ig 1). O f these, only heavy chain is encoded by the M H C complex; the (32m gene is located on a different chromosome (25) and the peptide is derived from other endogenously synthesized proteins (5). The 45 k D a heavy chain contains three extracellular domains, a transmembrane domain and a short cytoplasmic tai l . It possesses from 1 to 3 N- l inked glycosylat ion sites. The immature carbohydrate is added in the E R and subsequently modified as the molecule travels through the secretory pathway to the ce l l surface (26,27). The majority of the po lymorph i sm between different Class I genes is located in the a l and a2 extracellular domains. 8 Figure 1: Schematic representation of the M H C Class I and Class II antigens The general structural characterises of the two M H C ant igens are depicted. The Class I antigen consists of the the a c h a i n , non-cova len t ly ssociated with P2 m and a short peptide fragment. T h e pept ide b i n d i n g site is formed by the a l and a 2 extracellular domains of the Class I heavy chain. A short cytoplasmic tail extends into the cytoplasm. The Class II molecule is composed of an a and (3 chain w h i c h combine to form the peptide binding site. As with Class I, the chains are non-covalent ly associated. In both figures, peptide is l a b e l e d w i t h " P " . (Adapted from 24) 9 The oc2 and a 3 domains are stabilized by disulphide bonds. The P2m protein is 12 kilodaltons and forms a single domain with a disulphide b r idge . The formation of stable Class I molecules requires the presence of the three components in the E R (reviewed in 11). Class I heavy chain and P2m entry into the E R is mediated by their respective signal peptides. The heavy chain remains attached to the E R membrane whi le p 2 m , wh ich lacks a transmembrane region, is secreted into the lumen. Peptides required for Class I complex formation are generated in the cytoplasm and then act ively transported into the lumen of the E R . Successful assembly of the complex causes an alteration in the overal l conformation (28,29) which results in release of the complex from the E R (30) and subsequent transport to the cel l surface (Figure 2). Several proteins exhib i t ing chaperone activities have been impl ica ted in the assembly of Class I complexes. The best studied of these is calnexin, also referred to as p88, wh ich appears to retain incompletely assembled complexes in the E R (30-33). Upon successful interaction of the three components conformational changes signal the release of the heavy chain from calnexin a l lowing transport of the complex to the ce l l surface. This retention of a l l but correctly assembled complexes may be required to prevent complexes deficient of peptide from reaching the ce l l surface where they could bind and present inappropriate peptides. 1.3.3 Peptide binding It is clear, even from the brief overview given above, that the pathways 1 0 of biosyntheisis of Class I and Class II complexes are distinct. The choice of maturation pathway taken by the M H C antigen dictates the location at which peptide binding to the M H C complex can occur. This provides the basis for the presentation of separate peptide pools by the different M H C molecules. In general, Class I complexes contain peptides derived from endogenously synthesized proteins, whi le the peptides bound in Class II complexes are primarily from extracellular proteins. The difference in the nature of the peptides bound by the M H C complexes is reflected in the fine structure of the respective binding clefts. In the case of Class I, the peptide nestles in a binding cleft formed by the a l and a 2 domains. The sides of this cleft are delineated by two a-helices while the floor is composed of a (3-pleated sheet (34). Th is b ind ing cleft contains several "pockets", sites that mediate interaction wi th specific amino acid side chains of the peptide (35). Peptides f i l l the binding cleft like ink in a gravure block, with side chain groups nestling in suitable pockets. The pockets at either end of the peptide groove, the A and F pockets, are formed by highly conserved residues and provide influential sites for peptide binding (35-38). The A and F pockets interact with the NH2 and C O O H ends of the peptide respectively. The requirement that the peptide interact wi th both these sites dictates the min imal length of peptide that can be bound in the groove. However , the contribution of the two sites to the binding of part icular peptides may vary (39). In addi t ion to determining the min imum peptide length, the A and F pockets form a closed structure preventing binding of longer peptides. This is not, however, absolute as binding of longer peptides has been reported (40). 1 1 F i g u r e 2: M H C ant igen b iosynthes is The general features of the biosynthetic pathways of both Class I and Class II antigens are shown schematical ly. The distinct location of peptide binding to Class I and Class II molecules is clearly shown. As discussed in the text, there are exceptions to these pathways. This length res t r ic t ion represents the most s igni f icant difference between the peptide binding sites of Class I and II molecules. Class II molecules have an open ended peptide groove which permits peptides of greatly varying lengths to bind (41). In this case, pockets within the groove, rather that at the ends, provide the anchoring interactions. W h i l e the conserved residues of the A and F pockets provide the anchoring for Class I bound peptides, the other pockets dictate the specificity of peptide binding. Wi th the rapid increase in the number of peptides ident i f ied , the general b inding motifs of many Class I molecules have been described (42,43). W h i l e not representing the absolute binding specif ici ty, knowledge of these motifs is useful in prediction of potential peptide epitopes. It is clear that peptides are an integral part of M H C complexes and that the peptide process ing pathways are in t r ica te ly l i n k e d w i t h the functioning of the immune system. The primary focus of the work described in this thesis concerns Class I restricted antigen presentation by a carcinoma cel l line. For this reason, several aspects the processing pathway of Class I molecules w i l l now be introduced in greater detail. 1.4 P e p t i d e g e n e r a t i o n for C l a s s I a s s e m b l y The primary source of peptides presented by Class I complexes is the cy top la sm. The majori ty of peptides are der ived f rom proteins synthesized in the cy toplasm and subsequently degraded. Peptide generat ion from cy top lasmic proteins invo lves the act ion of the proteasome, a mu l t i ca t a ly t i c complex present i n the cy top l a sm 1 3 (reviewed in 44). The first evidence l i n k i n g the proteasome with antigen presentation came from the immunoprecipi ta t ion of a low molecular weight protein ( L M P ) complex with a n t i - M H C antisera (45). Th is complex had many structural s imilar i t ies wi th the proteasome. Since then, a great deal of evidence in support of the role of the proteasome has been obtained, although the precise nature o f the complex involved in the generation of antigenic peptides has yet to be determined. The proteasome is a multisubunit complex, which exists in several distinct forms, distinguished by their subunit components and sedimentation coefficients. The two primary complexes are the 20S and 26S proteasomes. The 20S form is an essential component of the larger 26S form. The 26S proteasome is required for the ubiquitin-dependent degradation pathway (46) This ubiquitin-dependent pathway has been implicated in Class I antigen processing (47-51), suggesting a role for the 26S proteasome. The contribution of the 20S proteasome to antigen presentation has yet to been assessed. The reason for the or iginal immunoprecipitation of the proteasome with a n t i - M H C antisera is the presence of two subunits encoded in the Class II region of the M H C (52-56). These two components, L M P 2 and L M P 7 , are not essential to the functioning of the proteasome; cells containing a deletion of the genes for these subunits are viable. Whi le the position of these genes wi th in the M H C provided an intriguing l ink with antigen processing, it was clearly demonstrated in a number of systems that these subunits were not required for v i r a l presentation (57-59) . However , the generation of mice deficient in either L M P 2 or L M P 7 (60,61), as we l l as further studies in deficient ce l l lines (62), have 1 4 provided evidence supporting a role for these components in Class I processing. Interestingly, the expression of several proteasome subunits is influenced by I F N - y (63-65), a cytokine that exerts a profound influence on Class I expression. Three subunits, inc luding both L M P 2 and L M P 7 , are up-regulated by IFN-y , while three others are down-regulated. Consistent wi th such changes in proteasome component expression, different proteasome complexes possess subtly different protealytic activit ies (61,66). The general proteolytic act ivi ty of the proteasome complex yields mainly peptides in the range of 8 to 10 amino acids in length. It is perhaps not coincidental that this is the preferred length of peptides for binding to Class I complexes. It seems clear from the accumulated data that the digestion of proteins in the cytoplasm represents the primary source of peptides for binding to Class I molecules. However, a role for other sources of peptides has not been discounted. The main alternative w o u l d appear to be degradation of proteins within the E R . It is wel l known that proteins can be degraded in the E R , for example as a result of improper folding. Evidence in favour of such degradation contributing to Class I assembly has been obtained by directing peptides or proteins to the E R via signal sequences or targeting specific lumenal proteases. In some cases this has been sufficient to achieve presentation of peptides. (67,68). M a n y other examples of this approach, however, have fai led to detect this activity, suggesting that it is, at best, a minor contributor of Class I binding peptides (see section 1.5.5). 1 5 1.5 P e p t i d e T r a n s p o r t into the E R Class I complex assembly and transport to the cel l surface requires the availability of peptides within the lumen of the E R , where they can bind with heavy chain and p 2 m - A role f ° r two proteins, called transporter associated with antigenic peptide ( T A P ) 1 and T A P 2 , in the peptide transport necessary for Class I expression and antigen presentation was described, almost simultaneously, in human, mouse and rat ce l l systems. W o r k with human cells centred around the use of a series of mutants which contained deletions within the Class II region of the M H C (69,70). The mapping of peptide del ivery to this region was consistent wi th findings from the study of Class I expression in rats (71). It was known that the class I modifier (cim) locus, responsible for altered recognition by a l loreac t ive C T L , mapped to this same reg ion (72). These observations quickly led to the identification of two genes wi th in this region in human (73-76), rat (72), and mouse (77) cells. These genes, T A P 1 and T A P 2 , share considerable homology across species and are members of the A T P b ind ing cassette ( A B C ) fami ly of proteins. Interestingly, other members of this gene family have been reported to be involved in transport of several substrates including peptides (78). Al though the A B C protein family has members in both the prokaryotic and eukaryot ic k ingdoms , they a l l share several character is t ics (reviewed in 78). Each member is thought to function in complexes cons is t ing of two mul t ip le membrane spanning regions, and two cytoplasmic regions each containing an A T P binding site. This complex may be encoded by a single gene or may be divided into as many as 1 6 four genes. The previously described eukaryotic members of this family have been of the single gene type. The T A P genes however appear to contain only two domains; one multiple membrane spanning region and one cytoplasmic A T P binding region. This f inding, ^coupled with those from studies on the prokaryotic A B C proteins that share the T A P two domain structure, suggests that the T A P proteins function as dimers. 1.5.1 Evidence for TAP involvement in peptide transport Strong support for the involvement of the T A P proteins in peptide transport was obtained from their abil i ty to complement the Class I processing deficiencies of several human mutant ce l l l ines. Use of the T A P 1 and 2 deficient T2 and .174 cells (79,80), the T A P 1 deficient .134 (80,81) and the T A P 2 deficient B M 3 6 . 1 (82) suggested that both the proteins were necessary for efficient peptide delivery, as measured by Class I expression and allogeneic and v i ra l antigen presentation. The requirement for both proteins was also observed wi th the mouse mutant ce l l l ine R M A - S . Init ially selected on the basis of low Class I expression (83), this ce l l line was speculated to have a deficiency in peptide loading of Class I in the E R (28). This defect is the result of a premature stop codon in the one copy of the T A P 2 gene present in the cel l line (84). Transfection of either mouse (85,86), rat (87) or human (88) T A P 2 significantly increased both ce l l surface Class I expression and allogeneic and vira l antigen presentation. In rat, the cim phenotype was shown to be the result of allelic differences in the T A P 2 gene (89). Expression of the different alleles leads to transport of a different set of peptides, altering the peptide repertoire presented at the ce l l surface. 1 7 Consistent with a role in peptide delivery for Class I assembly, T A P proteins are primarily localized to the E R (90,91), although they are also reported to be found in the c / s -Go lg i (91). These locations fit wel l with the proposed function of conveying peptides from the cytoplasm to the site of Class I assembly. The in vivo relevance of the T A P transporter has been demonstrated by the generation of a T A P 1 deficient ( T A P l ~ / ~ ) mouse (92). As predicted by the heterodimer model of the T A P transporter, cells from the T A P 1 - / - mouse have a profound defect in peptide transport and Class I mediated events (92-97). 1.5.2 The nature of the TAP complex The apparent requirement for both T A P proteins, together wi th the mechanism of action of other A B C proteins, suggests that peptide transport is mediated by a heterodimer of T A P 1 and 2. The abil i ty of the proteins to form a heterodimer has been shown by the co-immunoprec ip i ta t ion of the proteins by antisera to either sub-unit (80,82,90,98,99) . S i z e - e x c l u s i o n chromatography analys is of T A P structures in insect cells indicates that the complex is composed of two sub-units , presumably T A P 1 and T A P 2 (99). In teres t ingly , co-immunoprecipitation of T A P proteins from T2 cells infected with a T A P expressing vaccinia virus, frequently yielded higher amounts of T A P 1 than T A P 2 (90) This suggests that not al l the T A P subunits present are in the form of a heterodimer. 1 8 1.5.3 In vitro analysis of TAP structure and function Detailed analysis of the mechanism of T A P dependent peptide transport has been facil i tated by the development of several in vitro assays (90,94.98-102). The role of A T P in peptide transport, inferred by the sequence s imi la r i ty of the T A P proteins to members o f the A B C transporter family (72,76,77,103), was confirmed in an assay u t i l i z ing microsomes isolated from the T A P 1 ~l~ mouse (94). Us ing radiolabeled peptide derived from the Sendai virus nucleoprotein, microsomes from w i l d - t y p e mice accumulated peptide at a greatly increased rate compared to the T A P 1 deficient microsomes. This accumulat ion was specific and was dependent on availabili ty of A T P . In longer assays at 3 7 ° C , however, it was noted that both T A P 1 + and T A P 1 ~ microsomes accumulated peptide in the absence of A T P , although this was always to a l ower l eve l than the A T P dependent peptide accumula t ion . The amount of peptide accumulation correlated we l l wi th the amount of class I complex formed wi th in the microsomes, as determined by immunoprecipi ta t ion with conformation dependent antibodies. A similar ratio of ATP-dependent and ATP-independent transport was obtained using permeabi l ized cells (98-100,104). In the presence of the T A P complex and A T P , much lower concentrations of peptide were required to al low transport of heavy chain out of the E R . (100). Both T A P 1 and T A P 2 proteins bind A T P directly, v ia their hydrophil ic C-terminal domain (90,101,102). The binding of A T P can be achieved by either sub-unit expressed ind iv idua l ly , and therefore is not dependent on heterodimer formation (90). Competi t ion analysis of binding using 1 9 other nucleotides demonstrates A T P is the primary substrate involved but suggests that G T P may also play a role (90,101,102). A s yet, no ATPase activity has been demonstrated directly by the T A P sub-units, although the use of non-hydrolysable analogues of A T P yie ld transport levels equal to those in the absence of A T P . The ini t ia l binding of the peptide to the T A P transporter complex is not dependent on the presence of A T P (98,105,106). Consistent with the results described earlier ind ica t ing length requirements for peptides " that bind Class I molecules, the T A P complex was capable of binding a s imilar size range of peptides (98,105). There is a strong correlation between peptide binding to the T A P complex and peptide transport, suggesting that the A T P independent binding of peptide is a necessary event in T A P mediated transport (98,105-107). In an insect ce l l system, the release of peptide from the T A P complex upon binding of A T P has been observed. The hydrolysis of A T P did not appear to be required for this release. Interest ingly, substantial release, al though less than achieved in the presence of A T P , was observed at 3 7 ° C in the absence of A T P (98). Cross- l inking experiments indicate that peptide binds to both T A P 1 and T A P 2 (106,107). The consensus is that there is no specific peptide b inding when T A P sub-units are expressed i n d i v i d u a l l y (106,107). However, in one of these studies, peptide binding to T A P 1 was observed to di f fer ing degrees wi th different peptides, al though compet i t ion studies suggested that, for one peptide at least, this was non-specific (106). Tryp t ic digest ion of the TAP-pep t ide complex has defined a 20 region of T A P 1 involved in peptide binding. This region includes two putative transmembrane domains closest to the A T P binding site (106). Studies u t i l i z ing the specificity of peptide transport observed for rat T A P 2 a and T A P 2 U alleles indicated two polymorphic regions as being important for peptide specificity. These regions mapped to hydrophi l ic loops protruding into the cytoplasm (108). 1.5.4 Peptide selectivity of TAP transporters The demonstration that the cim phenotype observed in rat cells was the result of the different transport activities of alternative T A P transporter complexes led to much speculation on the role of T A P selectivity in peptide presentation. Potential involvement of selective T A P transport in disease was also speculated. The subsequent studies have provided much confl ict ing data but it appears that the T A P transporter sub-units exhibi t only l imi t ed se lect iv i ty . T A P dependent transport is most efficient for peptides in the size range of 8-13 amino acids (105,109), although there have been several reports indicat ing the b inding and transport of considerably longer peptides. The cut off of transport of peptides below the cr i t ica l length is more defined, wi th no peptide shorter than 8 amino acids demonstrating binding or translocation. The reported size range fits wel l with the length of peptides that bind Class I molecules leading to speculation that the T A P transporter specifically selects peptides with the correct characteristics for binding. The composit ion of the peptides does not appear to affect transport by different T A P alleles. The cim phenotype described in rat cells is the 2 1 only clear exception to this rule (89,110). T A P transporters from different species, which have considerably less similarity than do alleles wi th in a species, exhibit similar transport activities. In general terms, the rat T A P a and human T A P transporters are relatively non-specific in terms of peptides transported, whi le the rat T A P U and mouse T A P transporters more eff ic ient ly transport peptides wi th part icular C -terminal amino acids (111-114). These, findings argue against a role for T A P selectivity in disease, although the direct analysis of this question has yielded contradictory results. A recent report, however, described the selective restriction of peptide transport by the T A P transporter (115). In this system, the presentation of an immunogenic self-peptide was blocked at the level of the T A P complex. It remains to be seen to what extent this phenomenon is occurring. 1.5.5 Alternative pathways of peptide transport The profound reduction in Class I expression by cel ls wi th T A P de f i c i enc i e s argues s t rongly that this represents the p r i m a r y mechanism for delivery of peptides into the E R . There is evidence to suggest that it may not be the only pathway, although studies into possible alternative peptide delivery strategies have not, as yet, clearly defined the relevance of any such pathways. Conceptual ly , the most attractive T A P independent process for delivery of peptides into the E R is v ia signal sequence mediated translocation. As proteins containing N H 2 - t e r m i n a l signal sequences gain entry into the E R , these potentially provide a source of peptides for Class I binding which does not require T A P transporter ac t iv i ty . Several systems have been studied that 2 2 provide support for this model. T2 cells, which lack both T A P genes, express s ignif icant levels of H L A - A 2 at their ce l l surface. The subsequent isolation and sequencing of peptides bound to these Class I molecules revealed that they were derived from s ignal sequences (116,117). On ly seven dominant peptides were isolated from H L A - A 2 molecules on T2 cel ls , compared to over two hundred from T A P expressing cells (117), indicating that the contribution of this pathway is much less than the T A P mediated peptide delivery. It is noteworthy that the peptides isolated from H L A - A 2 of T2 cells were derived from signal peptides and not from other regions of proteins present in the E R . The relevance of the signal sequence mediated pathway for peptides contained within the main protein sequence is inferred by the study of ant i-HIV-1 C T L (118,119). In this system, the co-translational insertion into the E R was sufficient for processing and presentation of the H I V - 1 envelope protein by T2 cells. In addition, attempts to deliver specific peptides to the E R by attaching them to a signal sequence have, in many cases, been successful (120-123) Taken together, the above results support a role for signal sequences in the delivery of peptides for presentation by Class I complexes. However, there is also a substantial body of evidence to suggest that in many cases this pathway is not capable of mediating the necessary peptide transport. T w o of the clearest examples have involved presentation of signal sequence encoded epitopes (124,125). A C T L response directed against the gp33 epitope from lymphocyt ic choriomeningit is virus is T A P dependent despite the presence of this epitope wi th in the signal sequence of gp33 (124). S i m i l a r l y , presentation of peptides derived 2 3 from the signal sequences of H - 2 L ^ or by the Class lb molecule Qa-1 requires the presence of T A P 1 (125). The apparently contradictory findings from the studies described above are perhaps a reflection of the location of the proteases required to generate the relevant peptide, rather than of the peptide delivery system uti l ized. In short, it seems possible that signal sequence mediated translocation into the E R is sufficient for the delivery of peptides for Class I binding but that this w i l l only be relevant when generation of the peptide by E R proteases is poss ib le . In addition to the widely studied signal sequence pathway, evidence for other peptide transport mechanisms has been observed, but, as yet, not expla ined . T 2 infected wi th Sendai virus eff ic ient ly presents v i r a l epitopes to specific C T L , despite the absence of the T A P proteins (126-128). Surpris ingly, unlike both the T A P and signal sequence mediated peptide transport, this presentation of Sendai epitopes is resistant to Brefe ld in A treatment. This pathway of presentation appears restricted to only some cel l types and to Sendai virus. A l so in the T2 cel l system, the express ion of several v i r a l epitopes encoded by mini -gene constructs resulted in efficient presentation of the peptides (122). The mechanism underlying this presentation is unknown but appears to be unrelated to the hydrophobicity of the peptides involved. 1.5.6 Presentation of exogenously derived peptides W h i l e the description of Class I molecules presenting endogenously synthes ized antigens and Class II presenting exogenous ones is 2 4 generally true, the requirements of T cel l activation dictate that there must be exceptions (129). The almost ubiquitous expression of Class I molecules provides the means for identification of infected or altered cells where ever they occur. K i l l i n g of target cells, however, requires prior activation of the specific C T L clones. This process of activation requires the action of specialized antigen presentation cells. These cells provide not on ly the peptide antigen, but also accessory signals required for in i t ia l activation of T cells. Normal somatic cells are not capable o f T c e l l ac t iva t ion . C l e a r l y this process requires the presentation of Class I peptides by cells which do not endogenously express the relevant protein. Cel ls with the ability to present exogenous antigens in the context of Class I complexes appear to belong primari ly to the macrophage l ineage (130,131). The requirement for T A P transporters i n this process is unclear as two independent studies u t i l i z i n g the T A P 1 - / - mouse model obtained contradictory f indings (132,133). It remains to be determined i f Class I presentation of exogenous peptides involves a novel processing pathway or s imply follows the defined route after the peptides are somehow released into the cytoplasm. 1.6 MHC Class I and NK cells The detailed description of the events leading to the assembly and transport of Class I complexes reveals how viral ly infected or cancerous cells present specific peptides for recognition by C T L . The ce l l surface expression of Class I complexes, however, also influences the activity of a second cy to ly t ic lymphocyte population, namely the natural k i l l e r 25 ( N K ) cells. N K cells are classically defined on the basis of their ability to lyse target cells in a n o n - M H C restricted fashion and without the need for previous exposure to the targets. Despite their distinct activities, C T L and N K cells share several characteristics. One important s imi la r i ty between the ce l l types is their pr imary cy to ly t i c mechanism. The development of a per for in-def ic ien t mouse c l ea r ly showed the requirement for perforin in the cytolytic activity of both C T L and N K cells (134). The most significant difference between C T L and N K cells involves their receptors for recognition of target cells. As was described earlier, the specificity of C T L recognition is mediated by a single T c R molecule, produced as the result of gene rearrangement. Successful interaction of this receptor wi th the correct p e p t i d e / M H C complex triggers the cytolytic action of the C T L . In contrast, the recognition and k i l l i n g of target cells by N K cells appears to involve a plethora of receptor molecules, the resulting action being determined by the interaction of the competing signals (reviewed i n 135). To date, none of the putative N K receptors are the product of gene rearrangement and are thus distinct from the other primary lymphocyte receptors. The relat ionship between N K c e l l recogni t ion and M H C Class I expression has been we l l documented in many systems, both human and murine. Indeed, many of the mutant c e l l l ines used in the investigation of Class I assembly and transport have also been ut i l ized in analysis of N K c e l l ac t iv i ty (83,136-140). These studies have supported the "missing self" model of N K cel l k i l l i n g . In this proposed 2 6 model, the N K cells are activated to k i l l by the absence of normal Class I expression on the target cell surface. In the case of the Class I deficient cells, restoration of Class I expression switches the cells from being N K sensitive to N K resistant (136,137,141,142). In these systems, therefore, it appears that recognition of Class I molecules imparts a negative signal to the N K cel l , inhibiting the lytic activity. The sensitivity of N K ki l l ing to M H C Class I expression is not universal. M a n y ce l l types, most often derived from solid tumours, do not reveal this correlation (143). It thus appears that the origin of the potential target cel l is an important factor in determining the outcome of the interaction with N K cells. The number of candidate receptor molecules reflects the complex nature of N K ce l l recognit ion. In the murine system the two most studied molecules are the L y - 4 9 and N K R - P 1 famil ies of receptors (144,145). Both these molecules are C-type lectins. L y - 4 9 demonstrates the properties expected of a negative signaling M H C Class I receptor, and may bind directly to M H C Class I by virtue of a conserved glycosylation site in the a l domain of heavy chains. N K R - P 1 molecules, in contrast, exert a stimulatory influence on N K lytic activity, although the precise l igand for these receptors has not yet been defined. The description of N K recognition and signaling is not complete. However , the role of M H C Class I in the interaction between N K and a number of target cells is clearly important in determining the final outcome of such in terac t ions . 2 7 1.7 Objectives and Approach Insights into both Class I antigen biosynthesis and immunosurvei l lance mechanisms have been obtained from the study o f mutant and cancerous cells . This thesis presents the results obtained in both these areas from the analysis of the mouse small cel l lung carcinoma ce l l line C M T . 6 4 . In chapter 3, the nature of the antigen processing deficiency in C M T . 6 4 cel ls is investigated and the implicat ions of this defect to tumour growth is discussed. Chapter 4 details studies of T A P transporter function using the C M T . 6 4 cells as a peptide transport deficient background. W h i l e these chapters have distinct foc i , several components are common to both and are introduced below. 1.7.1 The CMT.64 cell line The C M T . 6 4 ce l l l ine was established from a spontaneous lung tumour arising in a female C57B1/6 mouse (146). The ce l l l ine is capable of metastasis when injected into syngeneic mice. The C M T . 6 4 c e l l l ine expresses greatly reduced levels of ce l l surface Class I (147,148). This expression can be restored to high levels by treatment wi th IFN-y . The cells express reduced amounts of both Class I heavy chain and (32m proteins, although studies of Class I transport suggested that neither was the l imi t ing factor in the low surface levels (147). Simi lar to many tumour cells, the reduced Class I expression correlates wi th an inabi l i ty of C M T . 6 4 cells to present endogenously synthesized antigens to C T L (59,148). The antigen presentation ab i l i ty is restored by I F N - y treatment. Incubation of the cells with exogenous peptide s imi l a r ly 28 restored their ab i l i ty to be recognized by peptide specif ic C T L (59 ,148 ,149) . 1.7.2 Virus presentation assays In this study two virus systems w i l l be used to determine the ability of C M T . 6 4 cells to present endogenous antigens to specific C T L . Vesicular Stomatitis virus ( V S V ) is a member of the rhabdovirus family . The immune response against this virus is directed pr imar i ly against the nucleocapsid (N) and glycoprotein (G) proteins (150). In mice of the H -2 D haplotype the a n t i - V S V C T L response is directed against a single immunodominant epitope spanning amino acids 52 to 59 of the cytoplasmic N protein (150,151). The dominance of this epitope is such that greater than 90% of a n t i - V S V C T L from H - 2 D mice are directed against this single peptide. The V S V N52-59 peptide is presented by the K D complex. In contrast to the V S V response, infection of H - 2 D mice wi th inf luenza A virus strain A / P R / 8 / 3 4 leads to a p r imar i ly D D restricted response (152). The primary D D b inding peptide is derived from the N P protein and encompasses amino acids 366 to 380. The use of these two viruses, therefore, al lowed the abili ty of C M T . 6 4 cells to present antigen restricted to both Class I complexes to be addressed. This abil i ty may be important given the selective down regulation of Class I alleles by some tumour cells (153-156). 2 9 2. MATERIALS AND METHODS 2.1 Cellular Methods 2.1.1 Tissue Culture. The three cel l lines used in these studies originated from the C 5 7 B L / 6 mouse strain which have the H - 2 b haplotype. C M T . 6 4 (kindly provided by Dr . L . M . Franks, Imperial Cancer Research Fund, London , U K ) is derived from a spontaneous small cell lung carcinoma. C M T . 6 4 cells and their derivatives were cultured in D M E M supplemented wi th 10% foetal calf serum (FCS) , 20 m M H E P E S , and 2 m M L-glutamine (henceforth referred to as complete medium). C M T . 6 4 cells were harvested or passaged using 0.05% trypsin (wt/vol) , in phosphate buffered saline (PBS) containing 1 m M ethylenediaminetetraacetic acid ( E D T A ) , for 5 minutes at 3 7 ° C . For passaging, 85-95% confluent plates were t rypsinised and cel ls replated in fresh media. R M A and R M A - S (obtained from Dr . P. Cresswell , Ya le Un iv . , Conneticut and D r . H - G Ljunggren, L u d w i g Institute, Stockholm) are Rauscher v i rus- induced leukemia cel l lines. R M A - S cells were selected for reduced M H C Class I expression fol lowing chemical mutagenesis (139,140). R M A , R M A - S and their derivatives were cultured in complete R P M I medium at between l x l O 5 and l x l O 6 cel ls /ml. Passaging was by diluting the cells (1:100) in new media. Ant ib io t ics were not used in any culture media. A l l cells were incubated at 3 7 ° C in a humidified, 5% CC»2/95% air environment. Where indicated cells were treated with IFN-y (Genzyme) . I F N - y was added at 2 0 0 U / m l to complete tissue culture medium and the cells 3 0 incubated under normal conditions for the time indicated., 2.1.2 Transfection. C M T . 6 4 cells were transfected using Lipofect in reagent ( G i b c o B R L ) . A total of 5-10 u.g of uncut plasmid D N A was used and the transfection carr ied out as per the manufacturer's instructions. 24 hours after transfection the cells were split into two plates and incubated for a further 1-2 weeks in the presence of l m g / m l neomycin ( G i b c o B R L ) . N e o m y c i n - r e s i s t a n t c lones were i so l a t ed us ing c l o n i n g r i n g s . Transfection of R M A - S cells was achieved v ia eletroporation using a B i o r a d Gene Pulser . 1 0 8 cel ls , in 1ml of serum-free R P M I culture medium, were pulsed at 1500V, 25 I U F D in a 0.4mm gap cuvette. Fo l lowing estimation of cel l survival using 0.4% Trypan Blue ( G i b c o B R L ) the cells were resuspended at 1 0 6 / m l i n R P M I conta in ing l m g / m l neomycin. Selected cells were cloned by l imit ing dilution. 2.1.3 Fluorescence activated cell sorter (FACS) analysis. For each ce l l type, 10 6 cells were washed twice wi th F A C S buffer ( D M E M with 0.5% wt/vol bovine serum albumin ( B S A ) , 20 m M H E P E S and 20 m M NaN3) and then incubated wi th the indicated primary antibody, in the form of lOOul of hybridoma ce l l supernatant, for 45 minutes at 4 ° C . After two washes with F A C S buffer, the cells were incubated wi th 100|Ltl of 10 | ig/ml stock of the appropriate f luoroscein isothiocyanate (FITC) conjugated secondary antibody (Jackson) in F A C S buffer at 4 ° C for 45 minutes. The cells were then washed once with 3 1 F A C S buffer, once with P B S containing 20 m M sodium azide, and fixed in 1.5% ( v o l / v o l ) para-formaldehyde in P B S . A n t i b o d y labe l led ce l l populations were analysed on a Becton-Dickinson flow cytometer using the F A C S c a n or Lys i s II programme. 5000 gated events were collected for each sample. 2.1.4 Generation of effector cell populations. Virus specific C T L populations were generated by infecting C57B1/6 mice with 10 6 tissue culture infection dose (TCID) units of V S V in the foot pads and ears or 700 H A units of Influenza i.p.. V S V C T L were derived from draining lymph nodes and spleen, collected on day 5 post immunization. Prior to use in a C T L assay, lymph node ce l l suspensions were cultured at 4 x l 0 6 ce l ls /ml for 3 days and spleen cells at 5 x l 0 6 cel ls /ml for 7 days in the absence of any restimulation. Influenza C T L were derived from splenocytes, isolated 1 week post-immunization, and cultured i n the presence of Influenza infected stimulators for 6 days. Stimulators were freshly isolated C57B1/6 splenocytes infected for 4 hours with Influenza A virus and then irradiated (2200 rads). A l l the above C T L were maintained i n complete R P M I culture medium con ta in ing 5 x l 0 " 5 M (3-mercaptoethanol. B u l k populat ions o f virus specific C T L were maintained by weekly rest imulat ion wi th virus infected stimulator splenocytes. Freshly isolated C57B1/6 splenocytes were infected wi th either V S V ( M O I of 10) or Influenza ( 2 0 0 H A u n i t s / 1 0 6 cells) for 4 hours then irradiated. Irradiated stimulator cells and C T L were incubated together at a ratio of 4:1 in complete R P M I culture medium containing 20 units/ml I L - 2 . 3 2 B u l k H - 2 d ant i -C57Bl/6 alloreactive C T L s were generated from B A L B / c mice previously immunized twice in t raper i toneal^ wi th 1 0 7 C57B1/6 spleen cells. Responder splenocytes (25 x 10 6 ) were stimulated in vitro with irradiated C57B1/6 splenocytes (25 x 10 6 ) , in 40 ml complete R P M I containing 1 m M non-essential amino acids ( G i b c o B R L ) , 5 m M sodium pyruvate, 5 x 10" 5 M (3-mercaptoethanol and 2 0 U / m l I L - 2 . Seven days later, this bulk population was used in a C T L assay, or restimulated in vitro using 10 6 responder splenocytes and 3 x l 0 6 irradiated C57B1/6 stimulator spleen cells. N K cells were isolated from C57B1/6 mice injected intraperitoneally 24 hours previously with 200|i.g poly inosinic-poly cyt id i l ic acid (poly I :C, S igma) . Spleens were removed and homogenised i n a Dounce homogeniser . Splenocytes were washed twice i n cul ture media . Enrichment for N K cells and lymphocytes was achieved by density centrifugation. Splenocytes, resuspended at 1 0 7 / m l , were layered onto lOmls of F i c o l l (Pharmacia) and centrifuged at 1800rpm (Beckman G P R centrifuge) at 2 0 ° C for 30 minutes. Cel ls located at the interface were removed, washed and resuspended in complete R P M I . Isolated bulk N K cel l populations were used immediately in cytotoxicity assays. 2.1.5 Cytotoxicity assays . Target cells for both C T L and N K assays were loaded with 5 1 Cr by i n c u b a t i n g 1 0 6 cel ls wi th l O O u X i of 5 1 C r (as sod ium chromate, Amersham) in 200uT complete R P M I media for 1 hour. F o l l o w i n g 3 3 washing, the target cells were incubated with effector cel ls at the indicated ratios in release assays of 4 hour duration in the case of C T L and 4-6 hours for N K cells . The 5 1 C r release was measured b y a compugamma computer ( L K B Instruments) and the specif ic release calculated using the fol lowing equation:-[(experimental-media control)/( total-media c o n t r o l ) ] x l 0 0 % Spontaneous release did not exceed 25% of maximal release obtained by lysis of the cells with a 5% Tri ton-X 100 ( B D H ) solution. 2.1.6 Antibodies. The fol lowing monoclonal antibodies (mAbs) were used in these studies: Y - 3 and 142-23.3 (both anti-H-2 K b , a l and a 2 domains) , 28-1 l -5s (anti-H-2 D b , a l + a 2 domains), 28-14-8s (anti H-2 D b , a3 domain) and B B M . l (anti-human p^m). The rabbit anti-serums uti l ised were against exon-8 of H-2 K b (kindly provided by Dr . B . Barber, U n i v . of Toronto, Ontario, Canada) and mouse T A P 1 and T A P 2 (87) (obtained from Dr . G . Butcher, A F R C , Cambridge, U K ) . 2.1.7 Peptides. Synthetic peptides representing the immunodominant epitopes of V S V and Influenza A viruses were purchased from the Universi ty of V ic to r i a Peptide Synthesis Fac i l i ty (Victor ia , B . C . , Canada). These peptides were V S V N (52-59) R G Y V Y Q G L , and Inf luenza A N P (365-380) I A S N E N M D A M E S S I L E . 3 4 2.1.8 Viruses. Vesicu la r Stomatitis Vi rus (Indiana strain) was providied by Dr . F. Tufaro (Univ. of B . C . , Vancouver, Canada). Stocks were grown on Vero cells ( A T C C ) in complete D M E M medium and kept at - 8 0 ° C as culture supernatant at a titre of 10 8 T C I D units/ml. Influenza A virus (strain A /PR/8 /34 ) was obtained from Caro l Murano ( L C D C , Ottawa, Canada). The virus stock was at a concentration of 2000HA units/ml in allantoic f lu id . V a c c i n i a virus containing the human P2tn gene was generously provided by Dr. J. Yewde l l (National Institutes of Health, Bethesda, M D ) . 2.1.9 Animals. M i c e were housed in the breeding facil i ty at the Univers i ty of Br i t i sh C o l u m b i a , under the auspices of W i l l e m Schoor l and maintained according to the guidelines of the Canadian C o u n c i l on A n i m a l Care. C57B1/6 ( H - 2 b ) and B A L B / c ( H - 2 d ) inbred strains were used in these studies. M i c e used in the experiments were between 5 and 15 weeks of age, and were sacrificed, where necessary, by CO2 asphyxiation. 2.2 Nuc le i c A c i d Techn iques 2.2.1 Plasmids. Mammal ian expression vectors containing the rat T A P 1 and T A P 2 genes were kindly provided by Dr. G . Butcher ( A F R C , Cambridge, U . K . ) . These contained ful l length T A P c D N A cloned into the pHp A P r - l n e o vector. 3 5 Expression of the c D N A was driven by the human (3-actin promoter. The T A P 1 antisense expression construct was created by exc is ion of the T A P 1 c D N A using E c o R I (Boehringer) and subsequent religation into the same vector but in the reverse orientation. 2.2.2 Oligonucleotides. A l l ol igonucleot ides used in these studies were synthesised by the Nucle ic A c i d and Peptide Synthesis Uni t (University of B . C . , Vancouver, Canada). The oligonucleotide sequences used are shown below (written 5' to 3'):-A T A C G A G A G C A G C T T T T C T C A T - T A P 2 (sense, primer) T C G C T C C A G G G C C T C C T T G T A G - T A P 2 (antisense, primer) T G C T G C G G A G C C T G G T G A A G G - T A P 2 (sense, probe) G T G G C C G C A G T G G G A C A A G A - T A P 1 (sense, primer) T C T T G T C C C A C T G C G G C C A C - T A P 1 (antisense, primer) G A G T G T C T C G G G A A T G C T G C - T A P 1 (sense, probe) C A G T C A G G T C C C G G C C A G C C A G G T - Act in (antisense, probe) C C A G A C G C A G G A T G G C A T G G G G G A G G 2.2.3 RNA isolation. Total cel lular R N A was prepared from cel l lines using the guanidium isothiocyanate (GITC) procedure (157). Br ief ly , pelleted cells were lysed in 6mls of 4 M G I T C , then centrifuged at 130,000g (Beckman SW41.1 rotor) for 16 hours at 2 3 ° C through a cushion of 4mls of 5 . 7 M cesium chloride. The purif ied R N A pelleted was precipitated i n ethanol and 3 6 resuspended in d ie thy l pyrocarbonate (DEPC) - t r ea t ed H 2 O . R N A concentrations were estimated by O D 2 6 O readings and solutions stored at - 8 0 ° C . 2.2.4 Northern analysis. 10 ng of each R N A sample was loaded and separated on a 1% agarose gel containing 2 . 2 M formaldehyde using the 3-[N-morpholino] propane sulfonic acid ( M O P S ) buffering system. The gel was capi l lary blotted onto H y b o n d N membrane (Amersham) and U V f i x e d before hybr idizat ion. Hybr id iza t ion with 3 2 P - l a b e l e d probes was performed at 4 2 ° C in buffer containing 0 .4M N a 2 H P 0 4 , 50% formamide, and 7% S D S . Several washes were performed at 4 2 ° C under conditions of increasing stringency and the filter exposed to f i lm ( X - O M A T A R , Kodak) . 2.2.5 Reverse transcriptase polymerase chain reaction (PCR) analysis. l | i g of total cellular R N A was used in each reaction. The R N A was denatured and renatured in the presence of l O f i M an t i s ense ol igonucleotide primer. First strand synthesis was then carried out at 4 2 ° C for 1 hour with reverse transcriptase (Superscript, Boehringer) in P C R reaction buffer ( l O m M T r i s - H C l , 1.5mM M g C l 2 , 5 0 m M K C 1 , pH8.3) containing I m M d N T P s . A t the end of the reaction the mixture was heated to 9 9 ° C for 5 minutes. The P C R reaction was performed using Taq polymerase (Boehringer) . The reaction mixture contained 1(J.M sense oligonucleotide primer, 1 x P C R reaction buffer, 5 m M M g C h and 3 7 dimethylsulphoxide ( D M S O ) . The reaction proceeded for 38 cycles of a 30 second denaturation step (94°C) , a 30 second annealing step (init ial ly 5 4 ° C and reducing to 4 8 ° C by the fourth cycle) and a 30 second e longat ion ( 7 2 ° C ) . A l l reactions were carried out on a Perkin Elmer GeneAmp 9600. Fo l lowing completion of the P C R , lOu.1 of the reaction mixture was electrophoresed on a 1% TAE-agarose gel and blotted onto hybond N membrane. The specificity of the P C R amplification was then determined by hybr idisa t ion of 3 2 P - l a b e l e d o l i g o n u c l e o t i d e probes specific for an internal region of the predicted fragment. 2.2.6 32P labelling of hybridisation probes. c D N A probes were labeled by random pr iming (Boerhinger random pr iming label ing ki t ) . 50ng of D N A was denatured at 9 5 ° C for 10 minutes and then m i x e d wi th l abe l ing buffer conta in ing random hexanucleotides, 2 units Klenow enzyme, 2 5 | i M each of d A T P , d G T P and d T T P and 50u.Ci [ 3 2 P ] - a - d C T P . The reaction mixture was incubated at 2 3 ° C for 1 hour. The labeled probe was denatured immediately prior to addit ion to the hybr id isa t ion solut ion. Ol igonucleo t ide probes were labeled by terminal transferase mediated addition of 3 2 P to the 3' end (TdT labeling kit, Boehringer). 50ng of oligonucleotide was incubated for 1 hour at 2 5 ° C in react ion mixture containing 25 units te rminal transferase, 2 . 5 m M C o C l , l x react ion buffer ( 2 0 0 m M potass ium cacodylate, 2 5 m M T r i s - H C I , 25mg/ml B S A , p H 6.6) and 50( iCi 3 2 P - y -d C T P . In both cases unincorporated 3 2 P was removed using a nick spin column (Amersham). 3 8 2.3 P r o t e i n T e c h n i q u e s 2.3.1 Cell labeling and Pulse-chase analysis. Cel ls were washed 1ml M E M medium without methionine 1 hour before label ing. Labe l ing was achieved with 150f iCi /ml of [ 3 5 S ] m e t h i o n i n e (Amersham). For pulse-chase experiments, cells were labeled for 15 minutes and then chased with normal medium containing an excess of cold methionine. Labeled cells were solubilised with 1ml of 2 0 m M Tris-HC1, p H 7.6, containing 0 .12M NaCI , 4 m M M g C l 2 , and 1% NP-40 and 20iug/ml phenyl-methylsulfonylf lour ide ( P M S F ) . After 15 minutes on ice , par t iculate matter was removed by cent r i fugat ion and the supernatant used for subsequent immunoprecipi tat ions. 2.3.2 Immunoprecipitation. Supernatants containing labeled antigens were pre-cleared wi th 2 (il normal rabbit serum for 45 minutes at 4 ° C followed by 50 u.1 Protein A -Sepharose (1:1 in lysis buffer) for a further 45 minutes at 4 ° C . Protein A-sepharose was removed by centrifugation. The supernatant was subsequently reacted with the indicated antibody or immune serum for 1 hour at 4 ° C . 35( i l of protein A-sepharose was added and the incubation continued for a further 30 minutes. After centrifugation, the beads were washed twice in 0.2% N P - 4 0 in l O m M T r i s - H C l , pH7 .5 , 0 .15M NaCI and 2 m M E D T A , and once in 0.2% NP-40 in l O m M T r i s - H C l , pH7.5 , 0 . 5 M NaCI and 2 m M E D T A and finally with l O m M T r i s - H C l , pH7 .5 . 3 9 2.3.3 SDS-polyacrylamide gel electrophoresis (PAGE). Gels wi th a 10-15% polyacrylamide gradient were poured on either large (25 x 30 cm) or small (10 x 7 cm) plates (BioRad) . A l l samples were heated to 9 5 ° C for 2 minutes, cooled to room temperature and alkylated with 5 0 m M iodoacetimide for 20 minutes, prior to loading onto the S D S - P A G E gels. U p o n complet ion of electrophoresis, gels containing radioactive samples were fixed in 30% v/v methanol, 10% acetic acid for 45 minutes at room temperature, fol lowed by a further 45 minute incubation in A m p l i f y solution (Amersham). Gels were then dried and exposed to f i lm (Kodak X A R ) at - 8 0 ° C . 2.3.4 Western blotting. Upon completion of S D S - P A G E separation of protein samples, gels were b lo t ted u s ing the B i o R a d M i n i P r o t e a n sys tem. Transfe r onto Immobi l ion-P polyvinyl idene difuoride ( P V D F ) membrane was achieved us ing the T r i s - g l y c i n e - m e t h a n o l transfer buffer. Membranes were subsequently dried and incubated wi th the indicated antibody for 1 hour. Fo l lowing three washes of 5 minutes duration (in 0.1% B S A , 0.05% Tween-20, 2 5 0 m M thimerosal in P B S ) the membranes were incubated wi th secondary ant ibody conjugated wi th horseradish peroxidase ( H R P O ) for 30 minutes. The above washing proceedure was repeated with a final 5 minute wash in P B S , and the membranes were subject to enhanced chemiluminescence ( E C L ) detection as per the manufacturer's instructions (Amersham) and exposure to f i lm (Kodak X A R ) . 4 0 3.1 I N T R O D U C T I O N Many insights into M H C Class I assembly and antigen presentation have been gained from the study of tumour cells and tumour cel l lines. It is a re la t ive ly common occurrence that tumour ce l l s express reduced amounts of surface Class I molecules and the study of the underlying mechanisms respons ib le for this reduct ion has been ex t remely informative. In addit ion to provid ing details of intracel lular antigen processing pathways, an understanding of these tumour phenotypes has shed light on the nature of recognition and surveillance by the immune system. As is also the case for many infectious pathogens (158), it appears that tumour cells develop mechanisms by which they evade detection by the immune system. Given the importance of C D 8 + C T L in the detection and el imination of abnormal cells (159,160), it is perhaps not surprising that M H C Class I down regulation is often observed on tumour cells. The frequency of this occurrence has led to the proposal that reduced Class I expression bestows a selective advantage on tumour cells by a l lowing them to grow and metastasize undetected. 3.1.1 Evas ion of the immune response by tumour ce l l s Direct support for the role of decreased Class I expression in tumour outgrowth is p r o v i d e d by studies o f mouse and rat systems (156,161,162). In the case of three independent tumour ce l l models, increased H-2 expression v ia transfection greatly reduced the growth and/or metastasis of the cells upon injection into animals. W h i l e these studies demonstrate the importance of Class I expression for clearance 4 1 of introduced cells , they obviously represent a system different from that of the normal growth of a tumour in the presence of a functioning immune system. In this regard, the results from human studies are more cont rovers ia l . W h i l e it is not a universa l phenomenon, the reduction or absence of Class I expression has been observed on many human tumour c e l l types (153,154,163-167) . A direct cor re la t ion between reduced C la s s I express ion and tumour g rowth and differentiation has been shown in some cases (155,168-170), although the relevance of this to patient survival is unclear (171). Support for the functional relevance of reduced Class I expression is provided by the demonstration that this reduction correlates wi th an inab i l i ty of the t umour c e l l s to present e n d o g e n o u s l y s y n t h e s i z e d ant igens (59,148,163,172). In addition, both murine and human C D 8 + T cells reactive against tumour antigens have therapeutic activity. Co l l ec t ive ly , these findings argue that, while other factors may be invo lved , the immunogenicity of the tumour cells may be important to the interaction between the tumour and the immune system and in the development of treatment strategies. In contrast to the above findings, several reports have demonstrated that elevating Class I expression on several tumour cells leads to reduced rates of clearance of the cells fo l lowing injection into mice (83,137,141). A s reduced Class I expression is the only feature common to these distinct tumour cells, it is probable that the Class I level , and not some additional factor, is the relevant parameter for determining clearance rates. It may be important that in a l l these cases the cells used originated from lymphomas. The clearance of the Class I deficient 4 2 tumour cells is thought to be mediated by N K cells, which are known to k i l l cells with reduced Class I levels (173) It appears, therefore, that the detection and e l iminat ion of tumour cells by the immune system is mediated by several ce l l types, p rovid ing many layers of defense against tumour formation. 3.1.2 M e c h a n i s m s c a u s i n g r e d u c e d C l a s s I e x p r e s s i o n o n t u m o u r c e l l s M H C Class I expression is a multi-factorial process. It is perhaps not surprising, therefore, that the study of Class I deficient tumour cells has revea led many dif ferent mechanisms u n d e r l y i n g this c o m m o n phenotype. Examples invo lv ing a l l the recognized Class I components have been described. For example, in the Daudi Burki t t ' s lymphoma ce l l l ine (3 2m express ion is abrogated by a c h r o m o s o m a l de le t ion encompassing the p 2 m gene (174). This results in a total loss of Class I expression at the ce l l surface. Reduced Class I heavy chain expression has been shown to be the cause of down-regulation of surface Class I by several neuroblastoma c e l l l ines (167) and Adenovi rus- t ransformed cells (175). A combined deficiency in both P2m and heavy chain gene expression has been described in several small ce l l lung carcinomas ( S C L C ) (163,164) and the mouse embryonic carcinoma line F9 (176). Recently, carcinomas wi th reduced expression of the T A P transporter have been described, impl ica t ing peptide loading deficiency in down-regulation of Class I by tumour cells (154,175,177-179). In several tumour cel l lines studied, of both mouse and human origin, the Class I expression can be restored by treatment of the cel ls wi th I F N - y (59,147,148,163,164,172,180,181). This treatment results in an increase 4 3 in expression of the heavy chain, (32m and T A P genes, as wel l as several other genes not thought to be directly involved in Class I assembly. The pleiotropic effects of IFN-v treatment make it difficult to determine the primary deficiency responsible for the reduced Class I expression, but clearly these cells are distinct from those in which relevant genes have been deleted as a result of chromosomal abnormalities. 3.1.3 Rat ionale for th is s tudy The C M T . 6 4 tumour ce l l line exhibits many features characteristic of a variety of tumour ce l l types. The cel l line has a profound deficiency in both Class I expression and endogenous antigen presentation. Al though both these deficiencies can be repaired by IFN-y treatment, the primary mechanism mediating the deficiency is unknown. To determine in more detail the nature of the Class I deficient phenotype of C M T . 6 4 cells, the role of each of the recognized components of the Class I complex was inves t igated independent ly . B y taking this approach the precise mechanism by which IFN-y treatment exerts its influence on the antigen processing should be revealed. This information may be relevant to the understanding of other similar tumour ce l l lines and provide alternative strategies for increasing the immunogenicity of such cells . In addition, the effect of increased Class I expression to recognition of C M T . 6 4 cells by both C T L and N K cells w i l l indicate the contr ibut ion of these mechanisms of the immunosurveillance to tumours of this type. 4 4 3.2 RESULTS 3.2.1 Cell surface MHC Class I expression in CMT.64 cells. It has been previously reported that cel l surface M H C Class I expression by C M T . 6 4 cells is extremely low but is restored by treatment of the cel ls wi th I F N - y (147,149). To confirm this f inding, the level of expression of Class I complexes on C M T . 6 4 cells, treated with or without I F N - y , was measured by fluorescence-activated ce l l sorting ( F A C S ) analysis (Figure 3). The two antibodies used in this study were 28.14.8s ( a n t i - D b ) and Y - 3 ( an t i -K b ) . A l l incubations with antibody were carried out at 4 ° C . Uninduced C M T . 6 4 cells express barely detectable levels of both Class I complexes at their ce l l surface. When C M T . 6 4 cells were incubated in the presence of 200 units/ml of recombinant mouse IFN-y for 48 hours the expression of both K b and D b was restored to high levels (Figure 3). The failure to detect ce l l surface Class I on uninduced C M T . 6 4 cells (Figure 3) could be the result of either an inabil i ty to assemble and transport the complexes, or s imply the transport of a reduced number of unstable complexes. The dissociat ion of unstable complexes upon arrival at the cel l surface has been reported for other Class I processing mutant cel l lines (28,182). Such unstable complexes can be stabilized by incubating the cells in the presence of exogenous peptides and (32m . C M T . 6 4 cells were, therefore, incubated overnight with 50u.M V S V N (52-59) peptide in the presence of exogenous p^m. This peptide binds with high affinity 4 5 CMT.64 CMT.64 H- IFN Log ol Fluorcr.r.onco Inlonr.iiv Figure 3: M H C Class I expression by C M T . 6 4 cells The cell surface Class I expression by C M T . 6 4 cells and C M T . 6 4 cel ls treated with IFN-y for 48 hours was measured by F A C S analysis us ing mAbs against D b (28.14.8s) and K b (Y-3). Panels labeled N F A arc c o n t r o l s in which no first antibody was used. The m e a n l inear f luorescence for each sample is given in the top right corner o f each p a n e l . A 6 to K b , b u t not D b , heavy chain (151). Us ing mAbs 142.23.3 ( K b ) and 28.11.5s ( D b ) , F A C S analysis revealed no detectable increase in the amount of either Class I complex present at the cell surface (Table 1). 3.2.2 Virus presentation by CMT.64 cells. To determine i f the lack of Class I expression had funct ional implications for C M T . 6 4 cells, the ability of these cells to present virus to specific C T L was investigated. The cells were treated with or without I F N - y for 48 hours prior to virus infection. C M T . 6 4 cells were infected with either V S V virus ( M O I of 10 for 12 hours) or Influenza A virus ( 2 0 0 H A units/10^ cells for 3 hours) and then incubated in the presence of the relevant virus specific C T L in a 4 hour 5 1 C r release assay (Figure 4). The use of these two wel l defined vira l systems allows measurement of the ability of C M T . 6 4 cells to present virus in the context of both D b and K b . Figure 4 A shows that IFN-y treated C M T . 6 4 cells infected with V S V virus were eff ic ient ly recognized by the V S V specif ic C T L . Uninduced C M T . 6 4 cells, however, were not recognized, as is indicated by the equal level of lysis seen with uninfected cells. This pattern of presentation is also observed fo l lowing infect ion wi th I n f l . A virus (Figure 4 B ) . IFN-y treatment induces a high level of recognition by C T L , whereas for untreated or uninfected cells no specific lysis is observed. 3.2.3 MHC Class I processing in CMT.64 cells Intracellular transport of class I heavy chain to the ce l l surface is 4 7 Table 1: Stabilization of cell surface Class I by exogenous peptide and Cell line peptide + (32m Db K b CMT .64 5 3 CMT . 64 + 8 5 C M T . 6 4 were incubated overnight with 50p:M of V S V N 52-59 peptide in the presence of (32m. Control cells were left untreated. The level of D b and K b expression was then measured using conformat ion specif ic ant ibodies 28.11.5s ( a n t i - D b ) and 142.23.3 ( a n t i - K b ) . The numbers shown represent mean linear fluorescence of each sample. 4 8 40 C M T / V S V C M T + In f -y /VSV C M T + Inf-y/-10 100 E:T 10 100 E:T Figure 4: Virus presentation by CMT.64 cells C M T . 6 4 cells were infected with (A) Vesicular Stomatitis Virus ( V S V ) for 12 hours or (B) Influenza A virus ( Inf l .A) for 3 hours or left uninfected (-). I F N - y treatment of uninfected cells (+inf-y) was for 48 hours with 200 U / m l . Lys i s by specific C T L was determined in a 4 hour 5 1 C r release assay. 4 9 accompanied by processing to a higher molecular weight form by modif ica t ion of the N - l i n k e d glycans during successive exposure to Golgi-specif ic enzymes. S D S - P A G E analysis of immunoprecipitated Class I complexes can, therefore, be used to investigate the transport of the complexes out of the E R . Immunoprecipitation of assembled Class I complexes pulls down both heavy chain, of approximately 45kD, and the 14kD (3 2m. Pulse-chase experiments were carried out to determine whether any such processing events were taking place in C M T . 6 4 cells (Figure 5). In the absence of IFN-y treatment both Class I heavy chains, K b and D b , fa i led to obtain the higher molecu la r weight form characteristic of transport out of the E R , even after a 4 hour chase period. In the case of the D b chain, it is apparent that there is a progressive loss of signal over the pulse period (compare 0 and 4 hours), most l i k e l y due to degradation of incomple te ly assembled complexes in the E R (Figure 5A) (27,183). The antibody used to detect D b , 28.14.8s, recognizes both assembled and unassembled D b molecules. It is clear from the lack of co-immunoprecipitation of p 2 m in uninduced C M T . 6 4 cells that most of the D b heavy chain detected is not correctly assembled into the Class I complexes. The poor signal seen for K b in uninduced cells is probably due to the specif ici ty of the antibody, 142.23.3, used for detection. A s this antibody recognizes only correctly assembled complexes, this indicates that K b complexes are not being correctly assembled. T o al low for better detection of K b processing, the experiment was repeated using an antibody which is specific for the cy top lasmic 5 0 Figure 5: Intracellular transport of M H C Class I molecules in C M T . 6 4 cells A ) Cel l s were labeled with 3 5 S methionine and chased in excess co ld methionine for the times indicated in hours. So lub i l i zed antigens were immunoprecipi ta ted wi th 142.23.3 m A b for K b and 28.14.8s for D b . Treatments are indicated at the top of the panels and the migration of molecular weight markers are shown on the left side. B) The experiment was ca r r i ed out as i n A ) except that K b m o l e c u l e s we re immunoprec ip i t a t ed w i t h a rabbit an t i -exon 8 ant iserum w h i c h recognizes the cytoplasmic tail of free and assembled K b heavy chain. Radioactive proteins are detected after 4 day exposure to f i lm . 5 1 K b Db CMT 64 -iFN-g + I F N . g Time: 0 0.5 1 2 4 0 0.5 l 2 4 46 kd 14.3 kd — -•PN-g + IFN-g 0 0.5 1 2 4 0 0.5 1 2 4 B C M T 64 Time: 46 led -IFN-g 0 0.5 1 2 4 8 K b + IFN-g 0 0.5 1 2 4 8 29 kd — 5 2 region of the molecule and is thus unaffected by complex formation (Figure 5B) . Use of this antibody showed more clearly that the K b chain is not processed to a higher molecular weight form and, similar to D b , is rapidly degraded in the E R . Treatment of C M T . 6 4 cells with IFN-y for 48 hours prior to the pulse-chase analysis restored normal processing of both K b and D b complexes (Figures 5 A and 5 B ) . In both cases, maturation to the higher molecular weight forms is clearly seen after a 1 hour chase period. Successful formation of Class I complexes is also demonstrated by the increased co- immunoprec ip i ta t ion of (32 m. In addition to inducing Class I transport, the treatment wi th I F N - y also increased the levels of expression of both the heavy chains and ^2m proteins (Figure 5 A ) . 3.2.4 Viral peptide presentation by CMT.64 cells The immunoprecip i ta t ion of K b and D b shown in Figures 5 A and 5B demonstrated that both heavy chains were synthesized by C M T . 6 4 cells. Whi l e the F A C S analysis of C M T . 6 4 cells suggested that no Class I was reaching the ce l l surface (Figure 3 and Table 1), the poss ib i l i t y remained that the failure to detect K b or D b reflected the sensitivity of the F A C S assay. Cytotoxic T cell assays were, therefore, used to increase the resolution of the analysis of heavy chain expression. To undertake such studies on C M T . 6 4 cells, the V S V N (52-59) peptide ( K b restricted) and the Influenza A N P peptide ( D b restricted), were used. Figure 6 shows that, when incubated for 1 hour wi th l | i M of virus peptide, uninduced C M T . 6 4 cells are able to present both these peptides to the relevant virus specific C T L . These results clearly demonstrate that Class 53 I heavy chain is transported to the cel l surface but at greatly reduced levels compared to normal cel ls . A s few as 200 M H C complexes is sufficient to trigger C T L lysis (184). In these experiments, exogenous p2m present in the foetal calf serum served to facilitate the assembly of stable Class I complexes with the added peptide. These experiments demonstrate that uninduced C M T . 6 4 cells can act as efficient antigen presenting cells when Class I complex formation is achieved by the addition of exogenous factors. 3.2.5 e x p r e s s i o n a n d v i r u s p r e s e n t a t i o n To address the contribution of p2m to the C M T . 6 4 cel l deficiency, in i so la t ion from the other effects of I F N - y treatment, its leve l o f production was increased v ia a vaccinia virus based expression system (Figure 7). Introduction of exogenous copies of the human p2m gene, under control of the vaccinia virus early promoter, successfully restored the levels of synthesis to those seen in IFN-y treated cells (Figure 7. inset). Consistent with earlier findings (Figure 6), the addition of V S V peptide to C M T . 6 4 ( C M T + p ) rendered them susceptible to C T L lys is . Figure 7 shows that the elevation of P2111 expression in V S V infected C M T . 6 4 cells ( C M T + V b 2 - V S V ) did not increase recognition by C T L above the background level seen for uninfected ( C M T + - - ) , V S V infected, p2m untreated cells ( C M T + V - V S V ) or p 2 m treated cells ( C M T + V - V b 2 ) These results indicate that the low expression level of p2m is not the primary deficiency in C M T . 6 4 cells. 5 4 Figure 6: Presentation of viral peptides by C M T . 6 4 ce l l s C M T . 6 4 cells were incubated with l u M of either V S V N 52-59 peptide or Inf l .A N P peptide for 1 hour at 3 7 ° C in the presence of exogenous p 2 m . These cells were then used as targets for lysis by the relevant virus specific C T L (against Infl .A or V S V ) . Control cells for each C T L , C M T . 6 4 incubated in the absence of peptide, indicate the amount of non-specific lysis occurring during the assay. 5 5 CMT 64 +/- VSV; Superinfection with Vac, or Vacb2 % sp. rel. i I T 1 12.5 2 5 50 1 00 E:T Figure 7: Presentation of VSV virus by CMT.64 cells expressing elevated levels of (3 2 m The level of expression of (32m by C M T . 6 4 cells was increased by infection with vaccinia virus containing the (32m gene. Expression levels obtained after v a c c i n i a infect ion (Vb2) are comparable to those a c h i e v e d after I F N - y treatment ( I F N ) , as d e t e r m i n e d by immunprecipitation (inset). Untreated C M T . 6 4 cells express little p 2 m (-). The abi l i ty of elevated ( 3 2 m expression to restore V S V v i rus presentation is shown in the main graph. C M T . 6 4 cells incubated wi th V S V N 52-59 peptide (CMT+p) are efficiently lysed by virus specific C T L . C M T . 6 4 cells infected with V S V ( C M T + V - V S V ) , V b 2 ( C M T + V - V b 2 ) , or V S V and V b 2 ( C M T + V b 2 - V S V ) were lysed as ineff ic ient ly as uninfected C M T . 6 4 cells ( C M T + - ) . 5 6 3.2.6 P e p t i d e g e n e r a t i o n a n d de l i ve ry in C M T . 6 4 c e l l s Current evidence suggests that the peptides expressed in association wi th Class I are predominant ly generated in the cy top lasm and subsequently transported into the lumen of the E R where they bind heavy chain/p^m complexes. A s the T A P 1 and T A P 2 genes have been impl i ca ted in this process, it was of interest to determine their expression profiles in C M T . 6 4 cells. The Northern blotting analysis of total cellular R N A extracted from C M T . 6 4 cells is shown in Figure 8 A . Each lane contains 20 | ig of R N A . The two genes, T A P 1 and T A P 2 , are undetectable in uninduced C M T . 6 4 cells (lane 1). Both these genes are induced by treatment with IFN-y for 48 hours (lane 2). A time course of I F N - y induction of the T A P 2 gene (Figure 8B) reveals that these genes are induced to high levels of expression after 12 hours of IFN-y treatment. Th i s t imecourse of expression c lose ly paral lels that of induction of cel l surface Class I expression in C M T . 6 4 cells. This finding provides indirect support for a role for these genes in Class I transport in C M T . 6 4 . In addition to the T A P genes, two components of the proteasome, L M P 2 and L M P 7 , are absent from uninduced C M T . 6 4 cells but are expressed after IFN-y treatment (185). These results indicate that C M T . 6 4 cells lack expression of genes implicated in both the peptide generation and peptide transport steps required for Class I complex assembly. A s it had been suggested by other studies that L M P 2 and 5 7 Figure 8: Expression of TAP1 and TAP2 genes by C M T . 6 4 cells A ) Northern blotting analysis was performed using 20p.g of total cellular R N A , isolated from either C M T . 6 4 cells (-) or C M T . 6 4 cells treated for 48 hours with I F N - y (+). Filters were probed with random primed T A P 1 (upper panel) or T A P 2 (middle panel) D N A . A n actin probe (lower panel) was used to determine the equality of loading. Hybr id ized filters were exposed to f i lm for 24 hours. B ) The timecourse of T A P 2 gene induction by IFN-y was determined by Northern analysis. C M T . 6 4 cells were incubated with 200U/ml of IFN-y for the time shown (lanes 2 and 3). Lane 1 contains uninduced cells (0 h). 10 \ig of total cel lular R N A were loaded for each sample. Random primed T A P 2 D N A was used to detect the presence of transcripts. The filter was exposed to f i lm for 48 hours . 5 8 T A P - 1 T A P A C T I N 5 9 B Oh 12h 24h 6 0 L M P 7 were not required for antigen presentation in a variety o f systems (57-59), the abil i ty of T A P gene expression to restore v i ra l presentation by C M T . 6 4 cells was first determined. To accomplish this, the rat T A P 1 and 2 genes, under the control of the p -ac t in promoter , were introduced into C M T . 6 4 cells . Clones were selected in i t i a l ly by resistance to neomyc in , bestowed by incorpora t ion of the T A P conta in ing p l a smid . These clones were subsequently screened by Northern analysis for expression of the exogenous T A P genes (Figure 9). Clones expressing high levels of both introduced genes were selected for subsequent studies. The ability of the T A P 1 and 2 genes to restore Class I transport, in the absence of I F N - y induct ion of other components, was determined by F A C S analysis (Table 2). Three transfectant clones, r l 2 - 1 2 , r l2 -21 and r 12-23, were analyzed and compared to parental C M T . 6 4 cells. In a l l three cases, the transfection of the T A P genes resulted in only a slight increase in D b expression at the cel l surface. The size of increase varied among the clones, being highest in clone r 12-21. In the case of K b , however, only a very slight increase in expression was observed in any of the clones. This general pattern of results was observed in several independent experiments, although the absolute numbers obtained in each differed. To address the possibil i ty that the lack of increase in K b was the result of the low expression of the K b gene (Figure 5, A and B ) , clone r l 2 -12 was infected with vaccinia virus constructs containing the K b 6 1 4 7 9 11 12 16 17 21 22 23 Figure 9: T A P 1 and TAP2 expression by C M T . 6 4 transfectants C M T . 6 4 cells were transfected with 10(ig of both rat T A P 1 and T A P 2 expressing plasmids. Clones were selected for resistance to neomycin. Resistant clones were then analyzed by Northern blot t ing for the expression of the introduced genes using random primed T A P 1 (upper panel) or T A P 2 (lower panel) D N A . lOjxg of total cellular R N A from each clone was loaded. The numbers shown are the designation of each clone used as reference in the text. Filters were exposed to f i lm for 24 hours. 6 2 Table 2: Surface expression of MHC Class I complexes by CMT r!2 clones C e l l Line Db Kb C M T . 6 4 1 2 7 r l 2 - 1 2 2 6 5 r l 2 - 2 1 4 8 7 r l 2 - 2 3 1 9 4 F A C S analysis of C M T . 6 4 and the r l 2 clones was carried out using mAbs to K b (Y-3) and D b (28.14.8s). A l l A b incubations were performed at 4 ° C . The results are expressed as are mean linear fluorescence. and f^m genes (Table 3). The increase in K b expression mediated by the vaccinia vector resulted in higher surface expression on both C M T . 6 4 and r l2 -12 cells. However, the increase was almost two fold greater for the T A P transfected clone. The addition of the $2m vector did not significantly increase the level of expression compared to that achieved with K b alone. The specificity of the increases observed is indicated by the lack of effect of any of the treatments on D b expression levels. The introduction of T A P 1 and 2 into C M T . 6 4 cells, therefore, is sufficient to a l low increased transport of both D b and K b complexes but does not restore normal levels of Class I expression. The induction of cel l surface Class I expression by IFN-y correlated with the restored abi l i ty of C M T . 6 4 cells to present v i ra l antigens. To determine i f the increased Class I transport observed i n the iT2 transfectants (Table 2) s imi l a r ly restored antigen processing, v i r a l presentation assays were carried out using infected r l 2 clones as targets. Clone r l 2 - 1 2 was found to be representative in these assays and results from the use of this clone are presented. C M T . 6 4 cells and the r l 2 clone were infected with either V S V virus ( M O I of 10 for 12 hours) or Influenza A virus ( 2 0 0 H A uni ts /10 6 cells for 3 hours), then incubated with virus specific C T L in a 4 hour 5 1 C r release assay. Figures 10A and 10B show that in the case of both viruses, efficient presentation was achieved by the T A P transfected clone. Untransfected C M T . 6 4 cells were, as before, unable to present either virus. The block in antigen presentation in uninduced C M T . 6 4 cells can, therefore, be bypassed by transfection of the T A P genes. 6 4 Table 3: Effect of increased K b and (32m expression in CMT.64 on cell surface Class I levels N o Treatment v v - K b v v - K b + vv-p^m Cel l T inp Db K b Db Kb Db K b C M T . 6 4 4 3 1 0 5 1 7 6 3 r l 2 - 1 2 2 9 0 1 8 123 1 5 117 C M T . 6 4 cells and clone r l 2 - 1 2 were infected wi th vacc in i a virus constructs containing the K b and p 2 m genes. The effect of increased K b ( v v - K b ) or K b and p 2 m ( v v - K b + vv-p 2 m) was measured using mAbs to D b (28.14.8s) and K b (Y-3) . A l l A b incubations were at 4QC. 6 5 A 40-,— 3 0 H cn 55 ^ 20-^  10. 10 B 100 E:T CMT.64 rl2-12 100 E:T Figure 10: Effect of TAP1 and 2 expression on viral presentation by C M T . 6 4 C M T . 6 4 cells and clone r l2 -12 were infected with either V S V virus for 12 hours (figure A ) or Influenza A virus for 3 hours (figure B) . Infected cells were used as targets in a 4 hour 5 1 C r release assay in the presence of v i rus spec i f ic C T L . The results shown us ing r l 2 - 1 2 were representat ive o f those obtained us ing a l l r l 2 c lones tested. 6 6 Importantly, these results demonstrate that uninduced C M T . 6 4 cells are capable of generating the relevant viral peptides. 3.2.7 NK cell lysis of CMT.64 cells A role for T A P dependent peptide transport in mediating resistance to lysis by N K cells has been suggested (136,137). It was therefore of interest to determine the effect of T A P expression in C M T . 6 4 on recognition and k i l l i n g by N K cells. To address this issue, it was first necessary to establish an N K k i l l i n g assay which reproduced the results indicating a role for T A P in N K cell resistance. The R M A - S system was used for this purpose. This ce l l line, which lacks the T A P 2 protein is efficiently k i l l ed by N K cells, compared to the wild-type R M A cel l l ine. Transfection of T A P 2 into R M A - S has been reported to confer resistance to N K k i l l i n g . Figure 11 shows that the assay conditions used were capable of reproducing these findings. The N K cells used were a bulk population isolated from the spleens of C57B1/6 ( H - 2 b ) mice, injected 24 hours previously wi th 200u.g of the interferon inducer, poly T C . The target cells i n the assay were C M T . 6 4 , R M A , R M A - S and R M A - S transfected with the rat T A P 2 c D N A (RMA-S.mtp2) , described in chapter 4. A s predicted, R M A - S were efficiently k i l l ed during the incubation with the N K cells, while R M A cells were largely resistant to such lysis. The restoration of T A P 2 expression by transfection reduced the k i l l i n g of R M A - S cells to levels s imi lar to those for R M A . Under these conditions C M T . 6 4 was efficiently lysed. 67 R M A R M A - S . m t p 2 R M A - S C M T . 6 4 10 100 1000 E:T ratio Figure 11: N K lysis of R M A , R M A - S and RMA-S.mtp2 The abili ty of N K cells isolated from poly I:C treated C57B1/6 mice to lyse R M A , R M A - S and R M A - S cells transfected with the rat T A P 2 gene ( R M A - S . m t p 2 ) is shown. The results were obtained in a 6 hour 5 1 C r release assay. 68 Having established the conditions of the N K lysis assay, the effect of T A P expression in C M T . 6 4 on recognition and k i l l i ng was addressed. It can be seen from Figure 12A that similar k i l l i n g of C M T . 6 4 and the three r l 2 clones was observed. Al though the slightly reduced lysis of clone r l 2 - 1 2 was reproducible, the general pattern of k i l l i n g observed suggests that the presence of the T A P proteins has little effect on N K mediated lysis of C M T . 6 4 cells. This suggests that the level of Class I transport achieved in the T A P transfectants was insufficient to bestow resistance to N K k i l l i n g . One possible explanation of this inabi l i ty of T A P 1 and 2 to protect C M T . 6 4 cells was that the transport of endogenous self peptides was not efficient, resulting in too few Class I complexes being expressed at the ce l l surface. A s results suggest that in the mouse N K system random peptides are sufficient to inhibit k i l l i ng (186), C M T . 6 4 cells were incubated with the V S V N 52-59 peptide and/or the Inf l .A N P peptide to stabilize a l l ce l l surface K b and D b respectively. These experiments used 50u.M of each peptide, fifty fo ld greater amounts than required for efficient virus specific lysis (Figure 6). Despite the high amounts of peptide, N K lysis of C M T . 6 4 cells was unaffected, even in the presence of both the D b and K b binding peptides (Figure 13A). As IFN-y has been shown to induce high levels of Class I expression (Figure 3), the effect of such treatment on N K lysis was determined (Figure 13B). 6 9 The ability of C M T . 6 4 cells and the r l 2 clones to act as targets for lysis by N K cells from C57B1/6 mice was tested. The lysis shown was the result of a 6 hour 5 1 C r release assay. 7 0 A 30-, 25-^ oo 20- | 'oo J 1 5 J 10 -5 -0 10 B CMT.64 C M T + /VSV N ... - • — C M T + Infl .A N P - A — C M T + V S V N , Inf l .A N P E:T 100 R M A C M T . 6 4 C M T . 6 4 + I N F - Y 000 E:T Figure 13: Effect of peptide and IFN-y treatment on N K killing of CMT.64 cells. A ) C M T . 6 4 cells were incubated with 5 0 ( i M of either V S V N 52-59 peptide ( + V S V N ) , In f l .A N P peptide (+Infl.A N P ) or both ( + V S V N , Infl .A) for 1 hour prior to incubation with N K cells from C57B1/6 mice. B) C M T . 6 4 cells were either left untreated or treated with INF-y for 48 hours prior to incubation with N K cells. For both assays the effector and target cells were incubated together for 5 hours. 7 1 Despite the increased expression of Class I, confirmed by F A C S analysis (data not shown), recognition and k i l l ing of C M T . 6 4 cells by N K cells was unchanged f o l l o w i n g I F N - y treatment. The resistance of R M A cells to k i l l i n g demonstrates that the lys is observed was specific and was consistent with the expected N K activity. 7 2 3.3 DISCUSSION C M T . 6 4 cells express greatly reduced amounts of Class I on their cel l surface. H igh levels of expression of both K b and D b can be restored by treatment of the cells with IFN-y . This increased surface expression is accompanied by successful maturation and transport of Class I heavy chains . I F N - y treatment also restores the abili ty of C M T . 6 4 cells to present viral antigens to specific C T L . The deficiency in functional Class I expression is not the result of the decreased level of Class I heavy chain synthesis observed in these cells as C M T . 6 4 cells pulsed with v i ra l peptides form sufficient amounts of both Class I complexes to trigger C T L lysis. Al though p 2 m expression is also decreased in C M T . 6 4 cells , this does not play a direct role in the antigen presentation deficiency of these cel ls . Increased (32m expression did not lead to lysis of v i ra l ly infected cells. C M T . 6 4 cells are deficient in the expression of both the T A P 1 and 2 genes and the L M P 2 and 7 genes encoded in the M H C Class II region of the genome. Treatment of the cells with I F N - y induces expression of a l l these genes. Transfection of the T A P 1 and 2 genes partially restores surface Class I expression and is sufficient to al low the cells to present v i ra l antigens. The inabi l i ty to restore high levels of Class I expression at the ce l l surface is l ike ly the result of l imi t ing amounts of Class I heavy chain. Although T A P 1 and 2 gene expression mediates resistance of R M A - S cells to lysis by N K cel ls , the T A P transfected C M T . 6 4 cells were efficient targets of N K k i l l i n g . Indeed, even peptide pu l sed or I F N - y treated C M T . 6 4 re ta ined their susceptibility to N K lysis. 7 3 3.3.1 The nature of the CMT.64 def ic iency The abil i ty to restore functional Class 1 expression by treatment with I F N - y , shown here for C M T . 6 4 cells (Figures 1, 2 and 3), has been o b s e r v e d w i t h a v a r i e t y o f t u m o u r c e l l l i n e s (59,147,148,163,172,180,181). Several different genes known to be involved in the assembly and transport of Class I molecules are induced by IFN-y , therefore this abil i ty does not reveal the identity of the primary deficiency of these cell lines. The ability to induce C M T . 6 4 cells to express Class I reveals that a l l the genes necessary for successful assembly and transport of these complexes are present and functional in the ce l l line, indicating that the deficiency was not the result of a deletion of a relevant region of the genome. The integrity of the necessary chromosomal regions in C M T . 6 4 cells is shared by other S C L C ce l l l ines (164). This feature distinguishes these cel ls from Class I deficient tumour cells, such as Daudi , in which required genes are non-func t iona l . It has been shown that reduced Class I expression can be the result of either reduced transport of the complexes to the ce l l surface or the expression of unstable complexes which dissociate upon arr iving at the ce l l surface (28). Unstable complexes can be stabilized by incubation of the cells in the presence of peptide and P2 m > increasing the amount of Class I detectable at the ce l l surface. Incubation of C M T . 6 4 cells under such conditions had little effect on the expression of the Class I epitopes recognized by the 28.14.8s and Y - 3 antibodies (Table 1). T h i s demonstrates that the failure of C M T . 6 4 to express Class I is the result 7 4 of inefficient transport of these complexes. Consistent with this f inding, a large amount of both D b and K b molecules are retained in the E R of untreated C M T . 6 4 cells (Figure 5), suggesting that they do not obtain the conformation required to a l low release from this compartment. The progressive loss of signal observed is probably the result of degradation of the incorrectly folded proteins (27,183) The frequency wi th which down regulation of Class I expression is observed on tumour cells has led to the suggestion that this phenotype allows such cells to evade detection by the immune response. Inherent in this model is the assumption that reduced Class I expression is indicat ive of an inab i l i ty to present endogenous antigens eff icient ly. This has only been directly shown for a small number of tumour ce l l lines (59,148,163,172). The correlation between the deficiency in Class I expression and ineff ic iency of v i r a l presentation is demonstrated by C M T . 6 4 cells (Figure 4). This result indicates that the reduction in Class I level has functional ramifications for the ce l l l ine . The inab i l i ty to present v i ra l peptides provides strong, though indirect, evidence that the abi l i ty of these cells to present tumour antigens may also be inhibited, in agreement with the model for tumour ce l l evasion of the immune system. 3.3.1.1 Relevance of decreased Class I heavy chain and 82m expression Formation of Class I complexes requires the presence of heavy chain, (32m and peptide within the E R . IFN-y treatment of C M T . 6 4 cells resulted 7 5 in greatly increased synthesis of both heavy chains and [32m (Figure 5). Although both components were present before IFN-y induction, the low level of expression of these components may provide an explanation for the reduced surface Class I expression. However, the ability of C M T . 6 4 cells incubated with v i ra l peptides to be recognized by specific C T L showed that the reduced amount of heavy cha in produced was sufficient to mediate antigen presentation (Figure 6). Th is argues against the role of this component in the primary deficiency of these cel ls . This result contrasts those obtained from other mouse tumour models. In the case of several tumour cel l lines, the increase in heavy cha in synthesis was suff ic ient to restore ant igen presentat ion (156,161,162). The difference between these latter results and those seen with C M T . 6 4 cells suggests that C M T . 6 4 has other deficiencies in addit ion to the reduced Class I heavy chain synthesis. The apparent discrepancy between the effect of exogenous peptide on stabilization of surface Class I (Table 1) and the C T L k i l l i n g of peptide pulsed cells (Figure 6) most probably reflects the sensi t ivi ty o f the respective assays. Increasing the expression levels of f ^ m similar ly failed to restore the abi l i ty of C M T . 6 4 cells to present v i ra l antigens fo l lowing infection (Figure 7). The analysis of the effect of heavy chain and (32m expression on antigen presentation suggests that the reduced synthesis of these components may be a result of, but is not the primary reason for, the C M T . 6 4 antigen presentation deficient phenotype. Importantly, the lysis of C M T . 6 4 cells incubated in the presence of vira l peptides (Figures 6 and 7), indicates that the inability to present antigen can be restored by 7 6 influencing only Class I expression. This suggests that the reduced Class I expression and the lack of antigen presentation are related, and that any additional factors required for efficient recognition and lysis of C M T . 6 4 cells were present. 3.3.1.2 Implications of the block in peptide transport The inabi l i ty of synthesised heavy chain and (32m to assemble into complexes (Figure 5) implicated the peptide component of the M H C class 1 complex. It is now thought that the peptides present in the Class I complex- are p r imar i ly generated in the cytoplasm and subsequently transported into the lumen of the E R . The failure to bind peptide with heavy chain and p2m may, therefore, reflect a lack of their generation in the cytoplasm or an inabil i ty to transport them to the E R . C M T . 6 4 cells are deficient in proteins involved in both these processes. The T A P 1 and 2 genes, implicated in the transport of peptide, are not expressed in untreated C M T . 6 4 cel ls but are induced by I F N - y (Figure 8 A ) . In addition, expression of the genes coding for the proteasome components L M P 2 and 7 is s imilarly only detectable after IFN-y treatment (data not shown). These four genes are located together in the Class II region of the M H C and their expression appears to be co-ordinately regulated (187). The time course of induction of the T A P 2 gene by IFN-y (Figure 8B) correlates closely with the increased Class I expression on C M T . 6 4 cells. As the other three genes l ike ly fol low similar induction kinetics, this provides support for the role of these genes in Class I assembly in C M T . 6 4 cells. 77 The ability of C M T . 6 4 cells transfected with T A P 1 and 2 to present viral antigens (Figure 3), in the absence of the IFN-y induced expression of other ce l lu la r genes, provides strong evidence that the lack of expression of these proteins is responsible for the C M T . 6 4 phenotype and, therefore, that the blockage in Class I assembly occurs at the level of transport of peptides from the cytoplasm to the E R . Despite this absence of T A P gene expression, C M T . 6 4 cells transport sufficient amounts of heavy chains to mediate presentation of exogenous peptide (Figure 6). This indicates that there is either T A P independent peptide delivery to the E R or that some heavy chain escapes retention in the E R and is expressed at the ce l l surface devoid of peptide. There are precedents for both these explanations. The T A P 1 and 2 deficient mutant T2 expresses a significant amount of H L A - A 2 at its surface. This complex contains peptides derived from signal sequences, which are, therefore, not dependent on the T A P transporter for t ranslocat ion (116,117). In support of the model of escape of heavy chain, it has been shown that i n mice lacking the [3 2m gene, and therefore incapable of correctly assembling Class I molecules, a low amount of Class I heavy chain reaches the ce l l surface (188). One or both of these mechanisms may account for the low level of heavy chain at the surface of C M T . 6 4 cells . The expression of T A P 1 and 2 in C M T . 6 4 cells does not, however, completely restore normal Class I expression (Table 2). The amount of D b heavy chain detected on the surface of the r l 2 clones is higher than on untransfected C M T . 6 4 cel ls . However, this increased expression is slight compared to that achieved fol lowing IFN-y treatment of the cells 7 8 ' (Figure 3). In the case of the K b molecule there is only a very slight increase in the r l 2 clones. As IFN-y induces increased expression of not just the T A P genes, but also heavy chain and p2tn (Figure 5), the failure to restore normal Class I expression levels may be the result of l imi t ing amounts of these latter components. The increased amount of K b complex detected after elevation of K b heavy chain and p2m synthesis supports this explanat ion (Table 3). The greater increase in D b compared to K b mediated by T A P gene expression (Table 2) may be due to the specif ic i ty of the antibodies used to detect the complexes . 28.14.8s ( D b ) is conformation independent and therefore detects al l D b heavy chain expressed at the cel l surface, whether present in an intact Class I complex or as free heavy chain. The conformation specific Y - 3 ( K b ) antibody, on the other hand, only recognizes K b molecules present in a Class I complex. 3.3.1.3 Proteasome subunit requirements The abi l i ty of the r l 2 clones to eff iciently present v i r a l antigens fo l lowing infection indicates that expression of L M P 2 and 7 is not required for this process. This f inding agrees with previous reports from other mutant ce l l lines (57,79). A s role for these proteasome components in peptide generation has been indicated (60-62,66), these proteins may not be irrelevant for a l l antigen presentation, but rather are not required in the specific cases analyzed in these studies. Indeed, the lack of expression of the L M P 2 and 7 genes in the r l 2 clones may be a contributing factor i n the failure to restore high levels of Class I surface expression. The introduction of these genes into C M T . 6 4 and the 7 9 r 12 clones may provide more information about the role of these components in Class I assembly and presentation. 3.3.2 Comparison of CMT.64 with other described antigen processing deficient carcinomas The systematic characterization of the nature of the C M T . 6 4 deficiency described here allows the cel l line to be placed amongst a small , but g r o w i n g , f ami ly of T A P deficient carc inoma cel ls (154,163,177-179,189). The previous studies have described tumour cells as either T A P 1 deficient (154,177-179,189) or T A P 1 and 2 deficient (163). It must be noted that in the former of these studies, the presence of T A P 2 was not investigated and, therefore, the possibi l i ty remains that these cells were deficient in both T A P proteins. W h i l e a correlation exists between the absence of T A P protein and reduced Class I expression i n the large majority of the cells examined, the studies did not directly assess the l i n k between these observations. However , s imi la r to the C M T . 6 4 cel ls , I F N - y treatment restored Class I expression of several human lung carcinomas (163). The studies described here on C M T . 6 4 cells, therefore, represent the first direct demonstration that the loss of T A P gene expression is sufficient to inhibit antigen presentation by tumour cel ls , and thus provide functional evidence for the proposed evasion of the immune response via such a mechanism. A s the reduced synthesis of P2m and heavy chain observed in C M T . 6 4 cells is common to many other S C L C cell lines, the deficiency of T A P mediated peptide transport may also be relevant to the other ce l l l ines. This study of C M T . 6 4 is the first to directly assess the role of the T A P proteins in the 8 0 S C L C phenotype. The similarities of the C M T . 6 4 phenotype and the T A P deficient human lung carcinomas (163) are interest ing. In both cases there was coordinate expression of the T A P 1 and 2 and L M P 2 and 7 genes, as indicated by the lack of detection of any of the genes before IFN-y induction and the s imilar levels of expression fo l lowing the treatment. In a l l cases tested, the cells were capable of presenting exogenously added viral peptides to C T L , indicating that a low level of Class I heavy chain transport was occurring. The extent of these similari t ies in the phenotype of these cells is such that it strongly suggests that a common mechanism may be responsible. It is , therefore, possible that the phenotype of these ce l l lines reflects the shared S C L C origin. However, as the exact ce l l g iv ing rise to these carcinomas is unknown, this possibility can, as yet, not be addressed. 3.3.3 Possible underlying causes of the CMT.64 phenotype The studies on the C M T . 6 4 ce l l l ine have shown that the antigen presentation deficiency is pr imari ly the result of the failure to express the T A P genes (Figure 10). However, it is clear that restoration of T A P expression is not sufficient to obtain normal Class I expression (Table 2). A s discussed above, this may reflect the reduced expression of heavy chain and p ^ m and/or absence of expression of L M P 2 and 7. The s imilar i ty between C M T . 6 4 and the human lung carcinoma ce l l lines argues against this phenotype being the result of several independent gene mutations within the cells. A simpler explanation would be the loss 8 1 of activity of a transcription factor involved in the expression of all these genes. A precedent for this explanation is provided by the mouse embryonic carcinoma cell line F9. The genes for Class I heavy chain and (3 2m are t ranscript ional ly silent in the F9 ce l l s . A s many normal embryonic cells share this transcriptional regulation of the Class I genes, the Class I phenotype of F9 cells is thought to reflect that of the original cel l that gave rise to the tumour cell line (176). The expression of the T A P and L M P genes encoded wi th in the M H C Class II region appears to be coordinately regulated. Studies have indicated that expression of both T A P 1 and L M P 2 is driven from a b i -di rec t ional promoter element located between the two genes (187). Consensus transcription factor binding sites have been identified wi th in this region, namely N F - K B , S p l - G C and the interferon response element I R S E (187,190). It has been speculated from the lack of sites elsewhere in the vic ini ty , that transcription of al l four genes may be regulated by this central element (190). Interestingly, a l l these elements are also located in the enhancer regions of (3 2m and heavy chain (191). The funct ional i ty o f the I R S E i n a l l the promoters has been c lear ly demonstrated in these studies on C M T . 6 4 . The possible role of the other factors in the C M T . 6 4 phenotype is worthy of further study. It is, of course, possible that a currently unknown factor(s) is invo lved in the C M T . 6 4 deficiency. 3.3.4 CMT.64 cells and NK recognition Whi le the relevance of T A P gene expression to antigen presentation by 8 2 C M T . 6 4 cells is clear, it does not appear to play a role in recognition and k i l l i ng of the cells by N K cells (Figure 12). This failure of T A P gene expression to confer resistance to N K lysis differs from the findings using the R M A - S cel l line. R M A - S cells lack the T A P 2 protein and are efficiently ki l led by N K cells compared to the T A P 1 and 2 positive R M A cells (Figure 11). Expression of a functional T A P 2 gene reduces lysis of R M A - S cells to levels similar to R M A cells, indicating that a functional T A P transporter is necessary for resistance to N K lys is . It has been shown previously that incubating R M A - S cells with exogenous peptide was sufficient to render them resistant to N K lysis (186). This effect was observed with a wide variety of peptides, indicating that the effect was not specific on the presence of a particular peptide at the cel l surface. Even when incubated wi th high concentrations of both D b and K b binding peptides, C M T . 6 4 cells remained sensitive to N K lysis (Figure 13A). This indicates that additional factors are required to account for the relationship between N K resistance and sensitivity of C M T . 6 4 cells . The results, however, clearly indicate that the expression of sufficient Class I complexes to mediate C T L recognition of C M T . 6 4 cells, is not sufficient to protect them from N K lysis. This finding is extended by the demonstration that even IFN-y treated C M T . 6 4 cells are efficiently k i l l ed by N K cells (Figure 13B). The lack of protection of C M T . 6 4 cells by IFN-y induced Class I expression contrasts results obtained using the Y A C - 1 ce l l l ine. In the case of Y A C - 1 the relevance of Class I for the IFN-y induced protection is clearly shown by the continued sensitivity of Class I deficient derivatives, even after treatment. In this regard, C M T . 6 4 cells resemble another murine carcinoma ce l l l ines in which elevated Class I levels s imi la r ly fai led to bestow protection from N K lysis 83 (143,192,193). Interestingly, this latter cel l line was also derived from a lung carcinoma, whereas the more common targets in N K assays, including R M A - S and Y A C - 1 , are of lymphoid or igin. The phenotype of the human lung carcinomas, described above, with regard to N K k i l l i n g has not been reported. It w i l l be of interest to determine i f they again show a response similar to C M T . 6 4 . The in vivo relevance of N K recognition of Class I deficient tumour cells has been demonstrated (83,137,141). In the case of both R M A - S and an E L - 4 derivative, restored Class I expression was accompanied by a failure to be efficiently cleared when injected into mice. This result obviously contradicts the findings that elevated H-2 expression led to reduced growth and metastasis of tumour cells (154,177,178). This apparent discrepancy has yet to be explained, although such factors as cel l type and localisation may be important. The influence of ce l l type is demonstrated by the in vitro results comparing C M T . 6 4 and R M A - S recognition by N K cells. Whereas the sensitivity of R M A - S to N K lysis is a reflection of only the Class I expression level, C M T . 6 4 remain sensitive independent of Class I levels. This indicates a fundamental difference between the two tumour ce l l phenotypes with regard to N K lysis . This d i f f e r e n c e m a y r e f l e c t the r e l a t i v e i m p o r t a n c e o f th i s immunosurve i l lance mechanism in the detection and eradicat ion of these different tumour types. It is an intriguing possibi l i ty that i f the reduced Class I expression of C M T . 6 4 cells represents the phenotype of the ce l l of or ig in , then N K k i l l i n g would have to invo lve different parameters in order to avoid k i l l ing of the normal ce l l . In this regard, it w i l l be interesting to determine the N K sensitivity of other normal Class 8 4 I negative cells. Beyond emphasizing the importance of cell type in N K cell k i l l ing , these results using C M T . 6 4 cells as targets do not reveal many insights into the nature of N K recognition events. However, the system may prove useful in the e luc ida t ion of mechanisms of Class I independent recognition by N K cells. The use of bulk populations of N K cells, while relevant for addressing the issue of immunosurve i l lance , w i l l not provide a clear picture of ce l l interactions. To determine the nature of C M T . 6 4 recognition by N K cells, purified N K cell sub-populations w i l l be needed . 8 5 4.1 INTRODUCTION W i t h few exceptions, the studies on T A P mediated peptide transport described in the introduction have concluded that both sub-units of the transporter are strictly required for peptide translocation. W h i l e this may be general ly true wi th in the parameters of many of these transport systems, close inspection of the accumulated data reveals several inconsistencies with this model. 4.1.1 Evidence for non-heterodimeric forms of the TAP transporter 4.1.1.1 Human cell studies The T A P 1 and 2 deficient cel l lines T2 and .174 have proved invaluable for study of the role and mechanism of T A P based transport. The pr imary advantage of the ce l l l ine is that it provides an absolute baseline for T A P transport, al lowing the relative contribution of the T A P sub-units, ind iv idua l ly as we l l as together, to be determined. U s i n g a vacc in ia vector to drive expression of an inf luenza matr ix protein epitope, it was observed that transfection of T A P 1 alone into T2 cells yields a 3-fold increase i n C T L recognition compared to parental T 2 cel ls . A 2-fold increase was seen in T A P 2 transfected T2 (79). The combination of T A P 1 and 2 yielded a 12-fold increase over background i n these experiments. The same study also revealed a detectable increase in H L A - A 2 and H L A - B 5 expression in T2 cells transfected with T A P 2 alone. Interestingly, studies on another human mutant ce l l l ine, .134, which lacks only the T A P 1 gene, also reveals an ability to present 86 influenza A peptides at levels above background. The background level of k i l l i n g in this assay is provided by .174 cells (80). Transfection of T A P 1 into .134 restored high levels of recognition. The s imilar i ty in presentation patterns observed in the T2 cells transfected with T A P 1 and .134 cells is strongly supportive of an effect of T A P 1 or 2 alone in translocation of peptides. It is worth noting, however, that transfection of ind iv idua l sub-units does not have a detectable effect on antigen presentation using other virus systems (90,194). Further evidence against a strict heterodimer requirement comes from studies on N K cell recognition of T2 cells. As discussed in chapter 3, N K recognition and k i l l i ng is most efficient towards cells expressing little or no ce l l surface class I. In such a system, T2 cells represent an excellent target, with high levels of k i l l i n g observed. Whi l e transfection of T A P 1 and 2 reduced k i l l i ng of T2 by N K cells to basal levels, the transfection of either gene individual ly also had a significant effect, reducing k i l l i n g by approximately 30% (136). S imi la r results were obtained using .174 cells . The above data, obtained from the use of T2 and .174 cells, consistently demonstrate that the most efficient transport of peptides is obtained when both T A P 1 and 2 are present. However, the studies also suggest that under some conditions, the requirement for heterodimer formation is circumvented, apparently by functional activity of single sub-units. The potential activity of individual sub-units is supported by structural studies of the T A P transporter. Co-immunoprecipitation of T A P proteins, from T2 cells infected with a T A P expressing vaccinia virus, consistently 87 yielded higher amounts of T A P 1 than T A P 2 (90). This suggests that not a l l T A P proteins present are in the form of a heterodimer. In addition, peptide transport at 3 7 ° C was observed using microsomes prepared from T A P 1 - / - mouse cells. The ability of peptide to bind to T A P 1 has also been reported (106), although no such activity was seen in the presence of only T A P 2 . 4.1.1.2 Mouse cell studies In contrast to studies using human mutants, analysis of T A P transport in mouse cells has been complicated by the lack of a cel l model deficient in both T A P genes. A s described earlier, many investigations have u t i l ized the T A P 2 deficient ce l l l ine, R M A - S . This ce l l l ine expresses reduced levels of ce l l surface class I and, in several assay systems, fai led to present endogenous peptides to C T L (85-87). However , the antigen presentat ion def ic iency in these ce l l s is not comple te . Presentation of V S V virus to specific C T L by R M A - S has been observed (88,195,196). In H - 2 D cells, V S V recognition is mediated pr imari ly by a single immunodominant epitope, N52-59, derived from the N protein (150,151). This has al lowed transfection of the N protein to sensitize cells to lysis by V S V specific C T L . These studies demonstrate that presentation by R M A - S is not a result of a peculiarity of virus infection (196). Al though in i t ia l ly undetected (86,87,196), Influenza A virus was also found to be presented by R M A - S cells (88,195). The requirement of this presentation for longer incubation periods or higher virus M O I , than was necessary for the T A P 1 and 2 positive R M A cells, explains the init ial failure to observe any k i l l ing of infected R M A - S cells. 8 8 The presentation of V S V and In f l .A viruses by R M A - S cells has frequently been attributed to an alternative, T A P independent pathway of antigen presentation. As such a pathway is feasible, it is interesting to compare presentation of a third virus, Sendai, by various antigen presentation deficient cells . This virus is presented as efficiently by R M A - S as by R M A (88). However, unlike V S V and Influenza A , it is also presented by T2 cells transfected with the relevant H-2 class I molecule (126-128). Thus, in this case, the suggestion of an alternative pathway of presentation in R M A and R M A - S is strongly supported by studies using T2. The lack of such support in the case of V S V and Influenza A presentation casts considerable doubt over the relevance of such a pathway in the presentation of these viruses by R M A - S . 4.1.2 R a t i o n a l e for th is s t u d y The extensive investigation into the role of T A P 1 and 2 in peptide transport and Class I assembly has revealed that both sub-units are required for maximal transporter act ivi ty. However, there are several results obtained from these studies that do not appear to fit the exclus ive ly heterodimeric model of the T A P transporter. The clearest examples of these come from the study of the mouse cel l line R M A - S . Interpretation of the presentation by R M A - S has been hindered by the lack of a mouse cel l lacking both T A P proteins. In this study, the role of T A P 1 in peptide transport in R M A - S cells is assessed direct ly v ia reduction in its expression l eve l , achieved by in t roduct ion of an antisense construct. 8 9 In addition, the demonstration that the antigen presentation deficiency in C M T . 6 4 cells can be complemented by transfection of the T A P genes suggests that this ce l l l ine may provide the mouse model system required to d i rec t ly address the relevance of al ternative peptide transport pathways for antigen presentation. The potential function of ind iv idua l T A P sub-units is investigated via transfection of the T A P genes ind iv idua l ly into the cel l l ine. Peptide transport is detected v ia surface Class I expression and presentation of endogenously synthesized antigens. 9 0 4.2 RESULTS 4.2.1 Confirmation of the phenotype of the RMA-S cell line The R M A - S cell line has played a central role in the elucidation of T A P based mechanisms of peptide transport. As it was the goal of this study to further investigate these mechanisms through the comparison of R M A - S and C M T . 6 4 cells , it was essential to demonstrate that the phenotype of the ce l l line observed in our assays correlated with that described i n the previous studies. The level of ce l l surface Class I expression by R M A and R M A - S cells is shown in Table 4. The levels of K D expressed at the ce l l surface was measured, using Y - 3 antibody, fo l lowing a 1 hour incubation at either 4 ° C or 3 7 ° C i n serum-free medium. The highest level of expression on R M A - S cells, obtained at the 4 ° C , was 10 fold lower than that detected on R M A cells. Whi le the level of expression remained constant on R M A cells at both temperatures, incubation at 3 7 ° C caused a large decrease in K D expression by R M A - S cells . The reduced expression and instabil i ty of Class I molecules on R M A - S cells has been attributed to the lack of a functional T A P 2 protein in these cells. To determine i f the results shown in Table 4 were similarly due to this single defect, R M A - S cells were transfected with an expression vector containing the rat T A P 2 c D N A . Transfected cells were selected by resistance to neomycin. The expression levels of on 4 selected clones, called S-mtp2.1 to 2.4, is shown on Table 5. 9 1 Table 4: Comparison of K b expression by RMA and RMA-S C e l l Line 40C 370C R M A 2 1 3 0 2 1 3 2 R M A - S 2 2 3 3 2 C e l l surface K b expression was measured by F A C S analysis using the conformation specific Y - 3 m A b . A b incubations were carr ied out at either 4°C, to detect a l l K b complexes, or 3 7 ° C , where only stable complexes are present. Incubations were carried out i n the absence of serum. Results are expressed as mean linear f luorescence. Va lues obtained for controls incubated with only the secondary A b have been subt rac ted . 9 2 Table 5: Partial restoration of RMA-S surface K D expression by TAP2 Cell Line RMA RMA-S S-mtp 2.1 S-mtp 2.2 S-mtp 2.3 S-mtp 2.4 K b 2176 275 827 812 825 730 C e l l surface K b expression was measured on R M A , R M A - S and R M A - S clones transfected wi th the rat T A P 2 gene", using the m A b Y - 3 . A l l incubations were carried out at 4 ° C in the absence of serum. Results are expressed as linear mean fluorescence. 9 3 The antibody incubations, using Y - 3 antibody, were carried out at 4 ° C . It is clear that the transfection of a functional T A P 2 gene part ial ly restored the level of Class I expression on R M A - S cells . However the levels obtained were significantly lower than those of R M A . This failure to completely complement the deficiency in R M A - S cells is in agreement with previous studies (85-88,137). The deficiency in R M A - S cells impairs their ability to present antigen. This defect, however, is not complete as the ce l l l ine is capable of presenting some viruses (88,195,196). The ability of R M A , R M A - S and R M A - S . m t p 2 cells to present V S V is shown in Figure 14. The cells were infected with virus (moi of 10) for 12 hours prior to incubation in the presence of V S V specific C T L . In agreement with the previous reports, both R M A and R M A - S were capable of presenting V S V peptides. Under the conditions of this assay the level of presentation by R M A was repeatedly higher than that of R M A - S . The presence of the transfected T A P 2 gene increased the level of recognition of infected R M A - S cells . Consistent with the results of F A C S analysis (Table 5), the functional T A P 2 gene did not restore presentation to wild-type levels. The results descr ibed above con f i rm the p rev ious ly de termined phenotype of the R M A - S cel l line and thus validate the assay conditions used for the comparison between this ce l l line and C M T . 6 4 . It was shown in chapter 3 that C M T . 6 4 cells were incapable of presenting V S V or Infl .A to specific C T L . It was, therefore, 9 4 R M A R M A - S R M A - S . m t p 2 0.1 1 10 100 E:T Figure 14: V S V presentation by R M A , R M A - S and R M A - S . m t p 2 The abil i ty of R M A , R M A - S and R M A - S . m t p 2 to present V S V to virus specific C T L was determined in a 4 hour 5 1 C r release assay. The target cells were incubated for 12 hours with V S V at an M O I of 10. Fo l lowing labeling the cells were incubated with splenocytes from V S V infected mice . 9 5 50-i 4 0 -.2 en 3 0 -2 0 -10 -0 i 1 100 E:T R M A R M A - S C M T . 6 4 ioo E:T Figure 15: Comparison of virus presentation by R M A , R M A - S and CMT.64 Target cells were infected with either V S V for 12 hours at an M O I of 10. (graph A ) or Inf l .A for 3 hours with 1000HA units/10 6 cells (graph B ) . Fo l lowing loading with 5 1 C r the cells were incubated with virus specific C T L for 4 hours. 9 6 important to determine i f this lack of presentation was observed under conditions where there was efficient recognition of R M A - S cells. Figure 15 shows that this was indeed the case. The cells were infected with either V S V (Figure 15A) or Infl .A (Figure 15B) and again, in both cases, the general pattern of R M A presenting more efficiently than R M A - S is observed. However, there is no recognition of infected C M T . 6 4 cel ls , even at the highest effector:target ratio. 4.2.2 Investigation of the functional role of TAP1 in RMA-S R M A - S cells are deficient in only one sub-unit of the T A P transporter, the T A P 2 protein. In an attempt to address the potential role for the remaining T A P 1 sub-unit in virus presentation, an anti-sense knockout approach was used. R M A - S cells were transfected with an expression vector containing the rat T A P 1 c D N A cloned in the reverse orientation. Transcription of the c D N A is driven by a P-actin promoter, a l lowing high levels of message to be obtained. Three neomycin resistant clones ( p t m l . l to 1.3) were selected, then analyzed for expression of their endogenous T A P 1 protein by Western blotting (Figure 16). The negative and positive controls for T A P 1 expression are provided by total ce l l lysates from C M T . 6 4 and C M T . 6 4 transfected with the rat T A P 1 gene respectively. Comparison of T A P 1 levels between the three ptm clones and the parental R M A - S reveals sl ightly decreased amounts i n clones p t m l . l and ptm 1.2 but markedly reduced levels in clone ptm 1.3. 9 7 Figure 16: T A P 1 levels are reduced in R M A - S clones expressing an antisense T A P 1 construct. Western blot analysis of whole cel l lysates from R M A - S and the ptm 1 clones was carried out using an a n t i - T A P l antiserum. The specificity of the antisera is demonstrated by the detection of T A P 1 in a C M T clone expressing m t p l . N o corresponding band was detected in untransfected C M T . 6 4 cells. Table 6 : Class I expression levels on RMA-S.ptml clones C e l l L ine Db K b ptm 1.1 2 7 7 0 ptm 1.2 4 0 ND ptm 1.3 8 3 2 The amount of Class I expressed at the cel l surface of three antisense T A P 1 clones was measured by flow cytometry using 28.14.8s ( D b ) and Y - 3 ( K b ) . The values shown are given as a percentage of those obtained for untransfected R M A - S cells. ( N D = not done). A l l antibody incubations were carried out at 4 ° C 9 9 To determine i f this reduced T A P 1 express ion had func t iona l ramifications, the surface Class I expression was measured on R M A - S and the ptm clones (Table 6). The level of Class I on the ptm clones is expressed as a percentage of that seen on R M A - S cells . A significant reduction in the amount of both D D and K D was observed for a l l clones. Consistent with the levels of T A P 1 expression (Figure 16) clone ptm 1.3 showed the most profound reduction in Class I. The antibody used to detect D D in this experiment, 28.14.8s, is not conformation specific and therefore recognizes a l l ce l l surface D u , not just correctly assembled complexes . The reduct ion i n D D levels seen in the ptm clones is therefore indicative of reduced transport of the heavy chain to the ce l l surface rather than of expression of more unstable complexes. Despite the reduction of Class I expression achieved fo l lowing transfection of the T A P 1 antisense message, no reduction in V S V presentation was achieved for any of the p tml clones (data not shown). 4.2.3 Analysis of the function of TAP sub-units in CMT.64 cells The antisense experiments described above were suggestive of, but d id not conclusively demonstrate, a functional role for T A P 1 expressed in the absence of T A P 2 . C M T . 6 4 cells provided a ce l l model to more rigorously investigate the functionality of ind iv idua l T A P sub-units. It was shown in chapter 3 that C M T . 6 4 cells are functionally deficient in T A P transport and that this deficiency correlates with failure to detect T A P m R N A by Northern analysis of total cellular R N A . However , in order to confidently assess the 100 1 2 3 4 TAP-1 TAP-2 Figure 17: R T - P C R analysis of T A P expression by C M T . 6 4 cells 1 u.g of total cellular R N A was subject to R T - P C R was using T A P specific primers. After completion of the reaction, half of the product mix was subsequent ly b lo t ted onto n y l o n membrane and probed us ing independent T A P oligonucleotides labeled wi th 3 2 P . The order of the samples is: lane 1, no R N A ; lane 2, Schneider cells; lane 3, IFN-y induced C M T . 6 4 ; lane 4, uninduced C M T . 6 4 . The figure shown was obtained from a 20 hour autoradiography. 101 activity of single T A P proteins it was necessary to analyze C M T . 6 4 cells for T A P expression using more sensitive techniques. For this reason, reverse transcription polymerase chain reaction ( R T - P C R ) was carried out using primers specific for mouse T A P 1 and T A P 2 . The results of this analysis are shown in Figure 17. R N A message for T A P 1 and 2 was only amplified from R N A from IFN-y induced C M T . 6 4 cells (lane 3). The level of signal detected in uninduced C M T . 6 4 (lane 4) was never greater than that obtained from Schneider cells (lane 2). Schneider cells were chosen as a negative control as they have been demonstrated to be T A P deficient- in in vitro transport assays (102). A second control containing no R N A (lane 1) was included to show the contribution of the initial R N A to the hybridizat ion signal. This result confirms, at greater resolution, the definit ion of C M T . 6 4 as a T A P 1 and 2 deficient ce l l l ine, and supports the use of this ce l l line in the study of single T A P sub-unit function. Clones expressing either T A P 1 ( r l ) or T A P 2 (r2) indiv idual ly , as de termined by N o r t h e r n ana lys i s , were selected f o l l o w i n g transfection of the relevant T A P c D N A (Figure 18). Three representative clones for each T A P , along with three double transfectants described in chapter 3, were used in the subsequent studies. It was shown earlier that expression of both T A P sub-units resulted in only a slight, but detectable, increase in cell surface Class I (Table 2). To determine the influence of ind iv idua l T A P sub-units on ce l l surface Class I expression, F A C S analysis was carried out on r l , r2 1 0 2 Figure 18: T A P mRNA expression by C M T r l and r2 transfectants C M T . 6 4 cells were transfected wi th the rat T A P 1 or 2 genes as described in materials and methods. Clones selected by their resistance to n e o m y c i n were then further analyzed for express ion of the introduced T A P gene by Northern blotting. 10|xg of total cellular R N A of each sample was run per lane. The gel was blotted and then hybridized using random prime labeled T A P 1 (figure A ) or T A P 2 (Figure B) D N A . The lane numbering refers to the identification number given to each clone. The figures are the result of 16 hour autoradiography. Clones 1-1, 1-4, 1-10, 2-1, 2-4, and 2-10 were selected for use in the subsequent s tudies . 103 1 4 6 9 1 0 11 104 Table 7: Surface Class I expression by the CMT TAP transfectants C e l l Line Db Kb C M T . 6 4 1 8 0 r 1 -1 6 7 3 6 r l - 4 164 6 4 r l - 1 0 6 8 3 6 r 2 - l 1 0 3 r 2 - 4 3 6 2 6 r 2 - 1 0 3 9 4 8 r l 2 - 1 2 4 9 9 r l 2 - 2 1 103 2 2 r l 2 - 2 3 3 6 7 F l o w cytometry analysis of K b and D b expression by C M T . 6 4 and the C M T T A P clones was carried out using mAbs 28.14.8S ( D b ) and Y - 3 ( K b ) . A l l A b incubations were performed at 4 ° C . The values are expressed as mean linear fluorescence fol lowing subtraction of negative controls. 105 and r l , 2 clones (Table 7). The results obtained for C M T . 6 4 and the C M T r l 2 clones was consistent with those described earlier. In the case of the C M T clones containing T A P 1 ( r l - 1 , r l - 4 and r l -10 ) , there was a detectable increase in both Class I complexes. Although there was considerable c lona l var ia t ion, the increase in Db was of a s imilar magnitude to that for the r l 2 clones. However, in the case of K b the increase was, in al l cases, greater than that seen for the r l 2 clones. For C M T cells containing T A P 2 ( r2 - l , r2-4 and r2-10), the increase in Class I expression was generally less than that of the other transfectant types. Indeed, in the case of clone 2-1 there does not appear to be an increase over the parental C M T . 6 4 cells . The absolute numbers obtained from different F A C S experiments were variable but the general pattern shown on Table 7 was observed throughout. The increase in Class I levels shown in Table 7 are most easi ly explained by an abili ty of individual T A P proteins to transport peptides into the E R , where they can bind heavy chain and (32m, a l l o w i n g transport of the complex to the ce l l surface. To determine that the effects observed were indeed the result of this established pathway, the transport of Class I heavy chain out of the E R was studied by pulse-chase analysis. Comparison of heavy chain processing between C M T . 6 4 and clone r l - 4 is shown in Figure 19. In C M T . 6 4 there is no maturation of heavy chain to the higher molecular weight form characteristic of transport to the G o l g i , but rather a loss of signal, presumably due to degradation within the E R , is observed. In the presence of the T A P 1 protein however, the processing events are detectable after 2 hours in the case of Db , and 106 Figure 19: Pulse-chase analysis of K b and D b from C M T . 6 4 and T A P 1 transfected C M T (clone rl-4) cells. Transport of K b and Db molecules to the cel l surface occurs in the T A P 1 transfected C M T cel ls (clone r-4), as indicated by the increase in molecular weight of the heavy chain during oligosaccharide side chain process ing . In untransfected C M T . 6 4 cel ls no process ing can be observed, indicating retention within the E R or cis G o l g i . 107 1 hour for K b . There is a good correlation between the results reported on R M A - S cells and the general pattern of Class I transport and expression seen in the C M T . 6 4 cells transfected wi th a single T A P gene. To extend this compar ison, the abi l i ty of the C M T transfectants to present v i ra l antigens was analyzed. As in chapter 3, V S V and Infl .A were used, both of which are presented by R M A - S cells. In Figure 20A the presentation of V S V by a representative T A P 1 clone (r l -4) , a T A P 2 clone (r2-10) and C M T . 6 4 is compared directly. As has been shown before, there is no presentation by C M T . 6 4 cells. Upon expression of either T A P gene, however, there is a significant increase in recognition of infected cells by V S V specific C T L . The efficiency of recognition is greater with T A P 1 than wi th T A P 2 at a l l effector:target ratios. A l l target cells were incubated for 12 hours with V S V at an moi of 10. In the case of Inf l .A presentation (Figure 20B) , a similar pattern is observed. There is no recognition of infected C M T . 6 4 cells but efficient recognition of both T A P 1 and T A P 2 transfectants. A g a i n , there is better presentation of v i ra l peptides by r l clone. The three target cel l lines were incubated with Inf l .A for 3 hours prior to a 4 hour release assay in the presence of In f l .A specific C T L . It is noticeable that the pattern of v i ra l antigen presentation seen with C M T . 6 4 and the T A P clones parallels the Class I expression levels shown in Table 7. T o further investigate the eff ic iency of v i r a l presentation by the different forms of T A P transporter, a titration analysis of Influenza A 108 A E:T C M T . 6 4 r 1-4 r 2 - 1 0 ioo E:T F i g u r e 20 : V i r u s presenta t ion by C M T r l and r2 c l ones . The ability of C M T r l and C M T r 2 clones to present V S V (figure A ) and Inf l .A (figure B ) was compared to the parental C M T . 6 4 ce l l l ine . The results shown using clones r l - 4 and r2-10 are representative of each class of transfectant. The target cells were incubated wi th either V S V for 12 hours ( M O I of 10) or Inf l .A for 4 hours (200HA un i t s / lO 6 cells) prior to a 4 hour release assay in the presence of virus specific C T L . 109 Figure 21: Comparison of Infl.A presentation efficiency by C M T . T A P clones. The ability of the C M T T A P clones to present Inf l .A when infected with a range of virus concentrations was determined. The target cells used were C M T . 6 4 (upper left panel), C M T 1-4 (lower left), C M T 2 - 1 0 (upper right), and C M T 1 2 - 1 2 (lower right) For each concentration used the virus infection was for 3 hours prior to incubation of the target cells with Inf l .A specific C T L . K e y indicates number of H A uni ts /10 5 cells used for infections. 1 10 presentation was carried out (Figure 21). The target cells, C M T . 6 4 and the representative T A P clones ( r l - 4 , r2-10 and r 12 -12), were incubated with a range of concentrations, from 0 to 5 0 0 H A uni ts /10 6 cells, of Inf l .A virus for 3 hours. Recognition of the cells by specific C T L was then measured in a standard 4 hour 5 1 C r release assay. A s expected, there was no presentation by C M T . 6 4 ce l l s , even when incubated with virus at a concentration of 500 H A units/10 5 cells. There was efficient presentation by the r l clone ( C M T 1 - 4 ) and the r2 clone ( C M T 2 - 1 0 ) , with the greater effect being seen for the r l clone. In both cases vi ra l presentation was achieved at the two highest concentrations, the two lower concentrations yie lding essentially background levels of lysis. The results obtained for the T A P 1 and 2 positive clone ( C M T 1 2 -12) again show efficient presentation of the two higher concentrations of v i rus but l i t t le recogni t ion of cel ls infected wi th the lower concentrations. The specificity of the recognition in this assay is clearly demonstrated by the complete absence of k i l l i n g of uninfected target cells (all panels). The abi l i ty of ind iv idua l T A P proteins to transport immuno log ica l l y relevant v i ra l peptides has been demonstrated directly (Figures 20 and 21). The increase in ce l l surface Class I expression seen in the T A P transfectants (Table 7) suggested that these transporter forms were also able to transport endogenous self peptides. To determine i f the increase in Class I expression observed was funct ional ly relevant for c e l l recognition, the ability of the T A P clones to present antigen to allogeneic C T L was measured (Figure 22). In addition to the three representative clones ( r l -4 , r2-10 and r l2-12) , C M T . 6 4 and C M T . 6 4 1 1 1 80 CMT.64 C M T + IFN-y C M T 1-4 C M T 2-10 C M T 12-12 E:T Figure 22: Allogeneic peptide presentation by CMT.64 and the C M T . T A P clones The abili ty of the various C M T cel l clones to act as allogeneic targets was examined using H - 2 d a n t i - H - 2 b C T L obtained from B a l b / C mice immun ized wi th C57B1/6 splenocytes. Target cel ls were untreated except for C M T . I F N - y which was incubated with 200units/ml of mouse I F N - y for 24 hours prior to the 4 hour release assay. 1 12 induced with IFN-y were included in the assay. The allogeneic C T L used in this study were isolated from Balb/c (H-2 d ) immunized with splenocytes from C57B1/6 mice ( H - 2 b ) . As expected, from their far greater level of surface Class I expression, IFN-y induced C M T . 6 4 cells were the most efficiently recognized targets. Uninduced C M T . 6 4 cells proved to be poor targets, with only slight recognition achieved at the highest E : T ratios. The pattern of presentation by the T A P clones was s imilar to that observed in the Influenza titration assay (Figure 21). Clone r l2 -12 was the most efficiently recognized T A P clone, followed by r l - 4 and then r2-10. These results demonstrate that the different forms of the T A P transporter are indeed capable of mediating transport of endogenous peptides. 1 13 4.3 D I S C U S S I O N The absolute requirement for both T A P 1 and T A P 2 is currently the basis of a l l proposed T A P mediated transport models. The results obtained from the study of a variety antigen processing mutants, however, have not been entirely consistent with this proposed model. M o s t notably, there is disagreement between the commonly studied human mutant cel l line, T2 , which lacks both T A P genes, and the mouse line, R M A - S , which lacks only T A P 2 . According to the strict heterodimer model the absence of T A P 2 should lead to the same transport deficient phenotype as results from the delet ion of both genes. However , comparison between these two cell lines reveals that this is not the case; whereas T2 is unable to present V S V and Infl .A viruses to specific C T L , both these viruses are eff icient ly presented by R M A - S ce l l s . Th is presentation by R M A - S cel ls is poor ly understood and has been attributed to a general leakiness in the primary deficiency. In this study, C M T . 6 4 cells, rigorously shown to be T A P 1 and 2 deficient, were used to re-evaluate the nature of the transport observed in R M A - S cells. C M T . 6 4 cells expressing individual T A P sub-units were created to allow comparison of the peptide transport by T A P structures other than the proposed heterodimer. Measurement of peptide transport by several criteria revealed that indiv idual T A P proteins are capable of forming functional peptide transporters. These results provide an alternative explanation for the R M A - S phenotype and have implicat ions for other proposed T A P independent mechanisms of peptide transport. 1 14 4.3.1 E f f e c t of T A P 2 o n P e p t i d e T r a n s p o r t in R M A - S C e l l s The greater efficiency of peptide transport when both T A P 1 and 2 are present is c lear ly demonstrated by studies in a l l mode l systems. Comparison of ce l l surface Class I expression by R M A and R M A - S reveals that the loss of the T A P 2 protein leads to a large reduction in the amount of both K b and D b complexes reaching the surface (Table 4). The effect is even greater when the stability of the complexes at the ce l l surface is measured. Under more stringent condit ions, in this case incubation in the absence of exogenous P2m at 37°C, the majority of the complexes expressed at the surface of R M A - S cells dissociate (Table 4). The reason for this dissociation at the cel l surface is thought to be the lack of peptides, leading to expression of so-called "empty" complexes (28,196,197) . It has been d i f f i cu l t to determine whether these complexes are indeed empty or whether they contain peptides with low affinity for the binding groove (197). Wi th in the context of the T A P heterodimer model, such peptides could be present in the E R as a result of T A P independent pathways, such as s ignal sequence-mediated translocation. The presence of peptides within the E R , capable of binding to the Class I heavy chains is supported by the low level of stable complexes expressed on R M A - S cells (Table 4). The abi l i ty of these complexes to withstand the incubation in the absence of serum indicates that they may contain high affinity binding peptide. A l t h o u g h the amount is sl ightly higher in this study than previously reported (197), the presence of such stable complexes is consistently observed. The direct role of the T A P 2 deficiency in the reduced Class I expression 115 is demonstrated by complementation of the deficiency by transfection of R M A - S cells with a functional T A P 2 gene (Table 5) . The introduction of T A P 2 greatly increased the amount of Class I at the ce l l surface. Interestingly, however, the expression levels were not restored to those seen in the non-selected R M A cells. This result is in agreement with several published studies and is not s imply the result of either the levels of T A P expression obtained or the introduction of the rat T A P 2 gene rather than the mouse gene (85-88,137). Indeed, studies in which both mouse T A P 1 and 2 were transfected into R M A - S cells also failed to achieve a complete restoration of Class I expression (86). These results strongly suggest that the cel l line contains other mutations in addition to the point mutation in the T A P 2 gene. The inabi l i ty of introduced T A P 2 to completely restore the normal phenotype is also observed in ant igen presentat ion studies (Figure 14) i n w h i c h the R M A - S transfectants containing T A P 2 (S-mtp2) proved better at presenting the v i r a l epitopes than R M A - S cells, but did not reach the level seen with R M A cells . This agrees with the published report (88). The differing abi l i ty of R M A and R M A - S to present V S V is in agreement with the findings of Hosken and Bevan, who found that a higher virus M O I was required for presentation by R M A - S (196). However, studies by another group fa i led to detect a difference between the c e l l l ines unless inhibitors of protein synthesis or secretion were included in the assay (88,195). Differences to account for this discrepancy are not obvious but may reflect such factors as the virus stock used. Interestingly, one of the published studies failed to detect In f l .A presentation by R M A - S (196), while the other showed, with the same virus strain, that it s imply r equ i red a longer v i r a l incuba t ion per iod (88). Th i s apparent 1 16 contradiction emphasizes the sensitivity of the assays to the particular conditions used. This sensitivity may be relevant to many of the studied systems. Despite these differences, the main point of the R M A - S studies is consistent; that the absence of T A P 2 in R M A - S cells causes reduced eff iciency of peptide transport but does not completely prevent the presentation of some vira l epitopes. The lack of a variant of R M A that lacks both T A P genes, however, prevents the role of T A P 1 in the presentation observed from being addressed. 4.3.2 The Role of TAP1 in Peptide Transport in RMA-S Cells The abi l i ty of R M A - S cells to present V S V and Influenza A is not consistent wi th the heterodimer mode l of T A P transport. Th is is emphasized by the fact that the most common explanation put forward to date has involved an i l l-defined leakiness in the antigen processing de f i c i ency (196) . The on ly other def ined pathway of pept ide translocation into the E R , that mediated by signal sequence, would not appear to be relevant for either of these viruses as the pr imary epitopes are contained within cytosolic proteins (195). The inabi l i ty of T 2 cells to present either of these viruses supports this contention. However, it is not possible to draw direct comparisons between the T2 and R M A - S systems given their distinct origins, and species or tissue differences could account for the variation in antigen presentation. In this context, the inabil i ty of C M T . 6 4 to present virus, under conditions in which infected R M A - S cells were efficiently k i l l ed by V S V specific C T L (Figure 15), has important implications. This result makes two main points: First, the ability of R M A - S cells to present V S V and Infl .A is not 1 17 simply the result of the viral maturation pathway in mouse cells. This correlates with the findings from the T2 system and with R M A - S cells tranfected with the V S V N gene (196). Second, i f the presentation by R M A - S is the result of a T A P independent pathway, this pathway is not common to a l l mouse cel ls . These observations, coupled wi th the demonstration that introduction of the T A P genes into C M T . 6 4 cells is suff icient to restore virus presentation, suggest that the peptide transport occurring in R M A - S cells may indeed be TAP-media ted . The poss ib i l i ty that the T A P 1 protein present in R M A - S cel ls may be sufficient to al low transport of v i ra l peptides has not been addressed p r e v i o u s l y . 4.3.2.1 Role of TAP1 in peptide transport in RMA-S In order to directly assess the contribution of T A P 1 to peptide transport in R M A - S cells, T A P 1 expression levels were reduced by introducing an antisense message of the gene. A recent report successfully used a similar approach to create an R M A - S - l i k e phenotype in R M A cells by expressing a T A P 2 antisense message (198). In this study, three clones containing the antisense construct were obtained and Western analysis revealed that each had reduced levels of T A P 1 compared to the parental R M A - S cells (Figure 16). When cel l surface Class I expression by the clones was measured, there appeared to be a close correlation between the degree of reduction of T A P 1 and the level of expression of Class I (Table 6). Even in the case of clone ptm 1.3, which exhibited the most profound 1 18 reduction in Class I expression fo l lowing introduction of the antisense construct, no reduction in efficiency of V S V presentation was observed (data not shown). This f inding is perhaps not surprising given the results obtained with the T A P 1 and 2 transfected C M T . 6 4 cells . These cells efficiently presented V S V but, when measured by F A C S , expressed little K b complex, the relevant heavy chain for V S V presentation. The level of Class I expressed in the presence the T A P 1 antisense message was significantly higher than that seen on the C M T r l 2 clones. Whi l e these results are the first to show a relationship between T A P 1 expression and peptide transport in R M A - S cells , they provide only circumstantial evidence in support of a functional role for the T A P 1 protein. U n t i l a complete inhibi t ion of T A P 1 synthesis is achieved in R M A - S cells , this approach cannot provide definitive evidence that the T A P 1 protein is mediating peptide transport. This interpretation of the results is based on the assumption that T A P 2 is not expressed in R M A - S c e l l s . It has been f r e q u e n t l y s h o w n by W e s t e r n and immunoprecipi ta t ion analyses that this protein is absent from these ce l l s . It is the contention of the " leaky" explanat ion of R M A - S presentation that some translation of the T A P 2 gene continues despite the presence of the stop codon (196). Whi le there is no evidence for this occurring, it is difficult to completely rule out. If such a situation exists, then c l ea r ly the peptide transport occur r ing i n R M A - S is not attributable to the T A P 1 sub-unit alone. The acceptance that this unsupported explanation has found, coupled with the demonstration of a role for T A P 1 in the Class I expression by R M A - S (Table 6), underline the importance of directly addressing the function of ind iv idua l T A P 1 19 sub-units in mouse cells. 4.3.3 C M T . 6 4 C e l l s a s a M o d e l for T A P T r a n s p o r t e r A c t i v i t y 4.3.3.1 CMT.64 provide a TAP1 and 2 deficient background The demonstration that the C M T . 6 4 cells are unable to present v i ra l peptides to C T L suggested a possible alternative strategy to analyze the funct ional role of ind iv idua l T A P proteins. A l though the profound presentation defect was indicative of a lack of the T A P proteins, it was cri t ical that this description held up to the most sensitive analysis. For this reason RT-PCR/hyb r id i za t i on analysis was carried out (Figure 17). This technique failed to detect either T A P gene message in uninduced C M T . 6 4 cells (Figure 17, lane 4). A s expected, T A P message was readily detected i n IFN-y induced C M T . 6 4 cells (Figure 17, lane 3). W h i l e it is arguable whether the antigen presentation or P C R analysis represents the most sensitive readout for the presence of the T A P proteins, the combinat ion of these two approaches provides the strongest evidence possible that expression of these genes is indeed absent in this cel l line. The importance of these results is highlighted by the induc ib i l i ty of expression of the genes by IFN-y. In such a situation, where the genes are present and can be functionally expressed under certain conditions, it is impossible to state with absolute certainty that they are completely absent in the uninduced cells. As there are no mouse cel l lines in which there is a deletion of the genes, as is the case for the human lines T2 and .174, the detailed analysis of the C M T . 6 4 l ine provides an alternative T A P deficient ce l l model in wh ich to study peptide 120 t ranspor t . 4.3.3.2 Individually Expressed TAP Sub-units can Transport Peptides The generation of single T A P expressing C M T clones (Figure 18) al lowed interesting comparisons to be made with established T A P mutants. Mos t notably, the r l clones represent an analogous situation to the T A P 2 deficient R M A - S cells, and the r2 clones resemble cells from the T A P l - / _ mouse. F A C S analysis revealed that for al l three categories of clones, r l , r2 and r l 2 , there was a detectable increase in ce l l surface Class I expression (Table 7). This finding is consistent with those obtained from the other systems. R M A - S cells (Table 4) and T A P W - cells (95,199) express a reduced but detectable level of both Class I complexes. The greater level of Class I complexes on both these cel ls , compared to C M T r l or r2 clones, is l ike ly explained by the higher levels of both heavy chain and p2m expression by R M A - S and T A P 1 _ / - cel ls . The abil i ty of T A P 2 expressed alone in C M T . 6 4 cells to increase Class I expression provides a possible explanation for the above background Class I expression detected on cells obtained from the T A P 1 deficient mouse (95,199). Importantly, the interpretation of the results from the T A P 1 _ / - mouse is subject to the same constraints as those from R M A - S cells due to the lack of a double T A P deficient control. If the transfected T A P sub-unit was indeed mediating peptide transport into the E R in C M T . 6 4 cells, the increase in cell surface Class I expression would be expected to be accompanied by an increase in maturation and 121 transport of the heavy chains from the E R . It is clear from Figure 19 that this was indeed the case. This demonstrates that the increase of Class I at the cell surface was the result of increased transport from the E R and not due to stabilization of complexes already at the cel l surface. Once again the results from the T A P 1 transfected C M T clone ( C M T 1-4) are consistent wi th those obtained from R M A - S , p rov id ing further evidence in support of functional T A P 1 transporters. Another interesting f inding comes from the comparison of Class I expression between the single T A P clones, especially r l , and the r l 2 clones. The results from other model systems, inc luding R M A - S . m t p 2 described above, suggested that the presence of both T A P genes would lead to greatly increased expression of both Class I complexes over that seen with T A P 1 alone. In the C M T system this is clearly not the case (Table 7). Whi l e the level of D b is comparable between the two sets of clones, K b expression is always lower in the r l 2 clones than the r l clones. This suggests that there is a difference between the peptide transport media ted by T A P 1 alone compared to T A P 1 / T A P 2 transporters, a l though whether this difference is quant i ta t ive or quali tat ive is unclear from this study. It is possible that peptides transported by the T A P heterodimer are more poorly suited for binding to K b than those transported by the T A P 1 homodimer. It is clear from the abili ty of C M T r l 2 clones to present the K b restricted V S V epitope that this does not apply to al l K b binding peptides. This effect may be more apparent in C M T . 6 4 cells than R M A - S . m t p 2 cells due to the more l imi t ing amount of heavy chain available for peptide binding in the former cells . 122 The abili ty of both the r l and r2 clones to present vira l peptides to specific C T L was essential i f the model of functional single T A P sub-units was correct. As it has been shown above that C M T . 6 4 cells, unlike R M A - S , were unable to present either V S V or Infl .A even at high virus M O I , these viruses were used to determine this ability. The transfection of either T A P 1 or T A P 2 was sufficient to restore presentation of both viruses by C M T cells (Figures 7 A and 7B) . In both virus systems, the presentation achieved by T A P 1 was greater than that in the presence of only T A P 2 . The ab i l i ty of both T A P sub-units to mediate v i r a l presentation indiv idual ly argues strongly against the possibi l i ty that the effects seen are due to undetectable amounts of the endogenous T A P proteins. If this were the case, then the results show that both endogenous T A P proteins must be expressed at levels sufficient to form functional transporters wi th the introduced T A P sub-unit. G i v e n the high eff iciency of T A P sub-unit assembly (90), it seems extremely unlikely that C M T . 6 4 cells infected with large amounts of virus would be completely unable to present any peptides. 4.3.3.3 Comparison of different TAP transporters The different efficiency of peptide transport in the presence of distinct T A P sub-units is demonstrated by the ability of the C M T transfectants to present Inf l .A virus after infection with a range of virus M O I (Figure 21). Both T A P ! ( C M T 1-4) and T A P 1 and 2 ( C M T 1 2 - 1 2 ) transporters were more efficient than T A P 2 alone ( C M T 2 - 1 0 ) . However , the presence of both T A P sub-units together did not greatly improve the peptide transport efficiency compared to T A P 1 alone. This difference 123 between the two transporters is more subtle than observed for R M A and R M A - S , where the presence of both T A P proteins in R M A cells results in significantly more efficient Inf l .A presentation than seen in the R M A - S cells (88). A possible explanation for this is, again, the l im i t i ng amount of heavy chain present in C M T . 6 4 cells wh ich w i l l require less peptide to reach maximal b inding . This explanat ion is supported by the almost identical lysis seen for both r l and r l 2 clones at the highest M O I , which suggests that the l imi t of presentation has been reached. Taken together these results demonstrate that the different forms of T A P transporter are capable of transporting v i ra l peptides, but do so with differing efficiency. A similar pattern of activity is seen when the transport of endogenous "se l f" peptides, rather than v i r a l peptides, is measured us ing al loreactive C T L (Figure 22). The greater presentation of al logeneic peptides by I F N - y induced C M T . 6 4 cells compared to the r l 2 clone (CMT12-12) is expected, as a result of the increased Class I expression. The abil i ty of C M T 12-12 to act as a better target for allo-specific C T L than C M T 1 - 4 apparently contradicts the f ind ing that the latter expressed higher levels of cell surface Class I (Table 7). However, it has been shown that C M T 1 2 - 1 2 efficiently presents V S V despite the low level of expression of K b by this clone. This indicates that functional quantities of peptide can be presented despite the inabi l i ty of F A C S analysis to detect Class I complexes. Considering this, it is feasible that C M T 1 2 - 1 2 presents higher amounts of the relevant allogeneic peptides, despite the lower Class I expression. The abili ty of both R M A - S and T A P 1 ' / " ce l ls to present al logeneic peptides has been reported 1 24 (92,95,97,200). In each case this presentation has been attributed to signal sequence mediated translocat ion. As the relevance of this pathway to func t iona l peptide presentat ion has recent ly been questioned (124), these results from the C M T . 6 4 based system provide an a l ternat ive exp lana t ion for the poo r ly unders tood transport obse rved . 4.3.4 P o s s i b l e e x p l a n a t i o n s for T A P s u b - u n i t t r a n s p o r t e f f i c i e n c y It is clear from the results discussed above that T A P transporters other than the recognized heterodimer are capable of mediat ing peptide transport. The consistent finding that T A P 1 was more efficient at such transport than T A P 2 can possibly be explained by recent findings from in vitro assays. Whi le it has been shown that both sub-units are capable of b ind ing A T P (90,102), apparently a requirement for transport, peptide binding in the presence of T A P 2 alone has not been observed (105,107). Whi l e the abili ty of T A P 1 to bind peptide in the absence of T A P 2 is controversial, some binding has been observed (106). It has been clearly shown that T A P heterodimers bind peptide eff ic ient ly (98,105-107). In addition to this difference in peptide binding capacity T A P 1 and 2 also differ in their ability to bind directly to heavy chain (201-203). It has been suggested that this b inding may faci l i tate efficient transfer of peptide from the T A P complex to the Class I molecule (203). The inabi l i ty of T A P 2 to form this connection may render it less efficient at del ivering peptides to the Class I complex. Thus T A P 1 and 2 differ in at least two ways that would appear to be relevant for peptide transport and Class I complex formation. In both 125 cases T A P 1 is l i ke ly to be more efficient when present alone, and therefore these differences may account for the greater efficiency of transport mediated by the T A P 1 transporter. However , it is noteworthy that in all the in vitro experiments the T A P heterodimer has been the most efficient complex, as is the case in the C M T . 6 4 system. 4.3.5 C o n t r a d i c t i o n s w i t h o t h e r p e p t i d e t r a n s p o r t s y s t e m s The demonstration of functional transport by T A P sub-units raises the question of why this phenomenon has not been observed in other studies. In the case of R M A - S , I would argue that it has, but that it was interpreted differently. However , it is undeniable that many other systems have been studied in which it would be expected that this result would be observed. There are several possible explanations for why this has not been the case. The two ce l l lines in wh ich this transport has been most clearly observed, C M T . 6 4 and R M A - S , are derived from mouse, although the C M T . 6 4 system u t i l i zed rat T A P genes. Few of the other frequently used ce l l models share this or ig in and therefore the phenomenon may be species speci f ic . Th i s is obviously not true of the T A P 1 _ / - mouse. In this case, however, while it is abundantly clear from the phenotype of the mouse that loss of the T A P 1 gene has profound effects on peptide transport and Class I mediated events, a few observations have been made which may be explained by peptide transport by the remaining T A P 2 protein. These include ce l l surface Class I expression, allopeptide presentation and selection of C D 8 + T cells (95-97,199). Given that T A P 2 is clearly the least efficient transporter (Figures 7,8 and 9), it is perhaps not 126 surprising that the level of transport observed in this system is low and that, to date, the abili ty of cells from the T A P 1 _ / - mouse to present either V S V or Infl .A has not been reported. The most studied T A P deficient ce l l line is the human hybrid, T2 . As discussed previously there have been several results obtained using this ce l l line that indicate functional peptide transport in the presence of i nd iv idua l T A P sub-units. These observations have invo lved both endogenous " s e l f peptides (79,136) and viral peptides (79). Whi l e the abili ty of individual T A P sub-units to mediate V S V presentation by T2 has not been analyzed, the low level presentation of the Influenza matrix protein epitope by T 2 - T A P 1 has been observed (79). This correlates with the ability of the T A P 1 deficient human cel l line .134 to present In f l .A , albeit ineff ic ient ly . It may transpire that the above studies fortuitously invo lved two viruses in which this phenomenon could be observed, or the reduced efficiency of the T A P sub-units may account for the lack of presentation of other viruses by T 2 . Further studies involv ing higher titres or longer incubation periods may reveal p resen ta t ion . The use of in vitro systems has been instrumental in defining many characteristics of T A P transport and has, without exception, been used to support the heterodimeric form of the transporter. These studies, however, share one important characteristic: peptide transport in T2 and/or R M A - S is used to represent the basal level of transport. Whi le it may be true that these cells do transport peptide less efficiently, both translocate peptides into the E R . In the case of T2 this transport is 127 mediated primari ly by signal sequences and allows expression of H L A -A 2 heavy chain complexes (116,117). As has been extensively shown for R M A - S , this cel l line is capable of transporting both " s e l f peptides and virus derived peptides to levels sufficient for triggering C T L lysis. It follows that the baseline of peptide transport in the in vitro assays lies above the level required for functional peptide mediated events. The inab i l i t y o f these systems to detect transport by T A P sub-units , therefore, is most l ikely a reflection of their sensitivity, which is clearly much less than the C T L readout used in this study of C M T . 6 4 cells. In summary, the results described here demonstrate that i nd iv idua l ly expressed T A P sub-units are capable of transporting peptides into the E R for binding to Class I complexes. The different forms of the T A P transporter exhibi t different efficiencies which may be explained by identified properties of the T A P transport system. The results provide an explanation for the antigen presentation by the T A P 2 deficient R M A -S ce l l l ine and suggest alternative pathways to account for the peptide transport in T A P l - / _ cells. In addition these results suggest the widely used in vitro assays for measuring peptide transport may be of insufficient sensit ivity to detect al l immunolog ica l ly relevant peptide t ranspor t . 128 5. GENERAL CONCLUSIONS 5.1 CMT.64 cells and evasion of the immune response C M T . 6 4 cells, isolated from a murine small cell lung carcinoma, exhibit a profound deficiency in M H C Class I restricted antigen presentation (59,147,148). The detailed description of the C M T . 6 4 ce l l phenotype given in this thesis provides an explanation of causes of this deficiency. The most important f inding is the demonstration that the failure to express the genes encoding T A P 1 and T A P 2 was sufficient to prevent antigen presentation by these ce l l s . In addit ion to this def iciency, C M T . 6 4 cells also express reduced levels of Class I heavy chain and (32m. This reduction in the level of the Class I complexes, although not alone sufficient to account for the antigen presentation def ic iency, may provide important clues to the under ly ing cause of the C M T . 6 4 phenotype. The role of M H C Class I expression in the lysis of C M T . 6 4 cells by N K cells indicates that these cells are recognized in a M H C -independent fashion s imilar to several other sol id tumour-derived ce l l types. This information may contribute to the growing understanding of the mechanisms of N K cel l recognition and the role of these cells in i m m u n o s u r v e i l l a n c e . A reduction in expression of M H C Class I is frequently observed on the surface of tumour cells . This reduced expression may have important impl ica t ions for tumour progression by undermining the immune response against the ce l l s (156 ,161 ,162) . S e v e r a l mechan i sms responsible for the altered expression of Class I have been described in 129 detail . There has also been a recent accumulation of circumstant ial evidence suggesting that decreased expression of the T A P 1 and T A P 2 genes represents such a mechanism (154,175,177-179). The f inding presented in this thesis, that restoration of T A P gene expression was sufficient to restore antigen presentation by C M T . 6 4 cells , is the first direct demonstration that decreased T A P expression is funct ional ly relevant to carcinoma cel ls . Al though not directly addressed here, it seems probable that this f inding extends to many other tumour cells exhibi t ing a similar T A P deficiency. This possibi l i ty, however, remains to be addressed. If a role for decreased T A P gene expression is indeed a common situation, then it may have implications for cancer therapy, as inducing T A P gene expression may restore the abi l i ty of the immune system to detect the tumour cells. As shown here, this restoration can be ach ieved by in ter feron-^ treatment or by the in t roduct ion of exogenous T A P genes under different transcription control The antigen presentation studies used in this study invo lved v i ra l and allogeneic antigens. These provide an excellent and reproducible system i n w h i c h to analyze the cont r ibu t ion of the different c e l l u l a r components. However, these systems do not address the effect of T A P expression on the presentation of tumour antigens. Clea r ly it is these latter antigens that w i l l mediate recognition of the tumour cells by the immune system in vivo. In preliminary investigations, the expression of T A P genes in C M T . 6 4 resulted in increased efficiency of clearance of the ce l l s f rom syngeneic C57B1/6 mice (Judie A l i m o n t i , pe rsona l communicat ion). This observation provides a further indicat ion of the relevance of this mechanism to immunosurveil lance. The frequency of 130 the T A P deficient phenotype, especially amongst lung carcinoma cells (163), make these f indings exc i t ing and worthy of much further investigation to determine their ful l potential. Whi le the introduction of T A P genes into C M T . 6 4 cells had a significant effect on presentation of antigens to C T L , it did not alter the sensitivity of C M T . 6 4 cells to lysis by N K cells. Indeed, none of the approaches to restoring Class I expression prevented N K lysis of C M T . 6 4 . This result contrasts those seen for many other tumour ce l l types, most notably leukemia cells, and indicates that the N K cel l recognition involves other molecules on the surface of the C M T . 6 4 . The Class I independent lysis of C M T . 6 4 parallels that seen for other tumour cells derived from sol id tumours (143,192,193). The nature of the target structures on these cells remains to be determined. The use of the defined C M T . 6 4 system may we l l contribute to the understanding of this mechanism. W h i l e the abil i ty to restore antigen presentation by the expression of the T A P genes in C M T . 6 4 was an important finding, it seems l ikely that this does not represent the source of the overall defect in these cells . The reduced expression of both Class I heavy chain and (32m, in addition to the lack of T A P 1 and T A P 2 , suggest that there may be a general transcription abnormality in these cells. The identification of the nature of this primary deficiency represents a significant challenge. Analys is of the relevant promoter elements invo lved suggests possible candidate factors, but, as yet, this remains purely speculative. The possibil i ty that a l l of the components of the Class I complex are to some degree coordinately regulated, is an exciting one. Such a situation would seem 13 1 to make sense intui t ively, but while the regulation of heavy chain and p2m expression is we l l understood, a l ink with T A P expression has not been observed. C M T . 6 4 cells provide an ideal model in wh ich to investigate this possibi l i ty . 5.2 R o l e of a l ternat ive T A P t r a n s p o r t e r s in M H C C l a s s I a s s e m b l y In recent years there has been a tremendous advance in our understanding of the mechanisms of peptide transport into the E R . These studies have revealed the T A P transporter to be the primary mechanism i n v o l v e d i n this process. To date, however , on ly the heterodimeric form of the transporter has been shown to be functional. The descript ion of the C M T . 6 4 phenotype provided an idea l model system in wh ich to study the contr ibution of T A P dependent and independent transport of peptides for Class I assembly. These studies provided information that revealed that previously unidentif ied forms of the T A P transporter were functional in peptide transport. The introduction of either the T A P 1 or T A P 2 subunits into C M T . 6 4 was sufficient to restore M H C Class I transport and the presentation of both vi ra l and allogeneic antigens. This result correlates we l l with the effect of decreasing T A P 1 expression in R M A - S cells , which resulted in a reduction of surface Class I expression. Al though functional peptide transport by T A P 1 is consistent with a l l observations made using the T A P 2 deficient R M A - S ce l l line, this activity has not previously been assessed. These studies, using two independent cel l lines, provide strong evidence for T A P based peptide transport i n the absence of the heterodimeric structure. Previous studies have not detected such 132 ac t iv i t y . These findings also have implications for the interpretation of results obtained using in vitro peptide transport systems, systems which have supported the exc lu s ive ly heterodimeric model of T A P transport (90,94,98-102). A s no act iv i ty of single T A P subunits has been measured in these systems, this suggests that the sensitivity of the in vitro assays is s ign i f i can t ly lower than required to detect a l l immunologica l ly relevant peptide transport. It w i l l be of considerable interest to determine the reason for this reduced sensitivity. W h i l e it may s imply be a ref lec t ion of the detection assay itself, other parameters may be involved . One intr iguing poss ibi l i ty , suggested by comparison of the different assay systems, is that the delivery of the peptides to the T A P transporter is an important step in the overal l pathway of peptide translocation. In the phys io log ica l situation, the generation of peptide fragments may be int imately l i n k e d to the transporter complex. Dissociat ion of these components, as occurs in the in vitro assays, may reduce the efficiency of the process. A second possible explanation involves the identity of the peptides studied; the spec i f ic i ty or ef f ic iency of the alternative T A P transporters may determine which peptides are transported. Clear ly , a direct comparison of specific peptide transport by the transporters w o u l d address this possibil i ty. Whi le further study is required to address these issues, it is clear that the demonstration of T A P subunit activity in C M T . 6 4 cells has raised several questions relat ing to the nature o f the transporter complex and to the systems currently used to address its activity. 133 It is important to emphasize that, while T A P subunits are functional, the heterodimeric form of the transporter is clearly the most efficient complex. Results from a variety of independent systems, including the T A P 1 - / - mouse model (92,94) and R M A - S (88,195,196) and T2 cel l lines (79,80), have provided ample evidence that this is the case. For this reason it w i l l be important to address the physiological relevance of the alternative forms of the T A P transporter. Given the apparent coordinate regulation of T A P gene expression, it is possible that conditions in vivo do not favour the formation of non-heterodimeric forms of the complex, or their reduced eff ic iency may render their ove ra l l cont r ibu t ion irrelevant. These are questions that can be addressed clearly in the C M T . 6 4 cel l system. 134 6.N0MENCLATURE A B C A T P - b i n d i n g cassette A T P adenosine triphosphate p 2 m b e t a - 2 - m i c r o g l o b u l i n B S A bovine serum albumin cim class I modifier C T L cytotoxic T lymphocyte DEPC d ie thy l pyrocarbonate D M S O d i m e t h y l s u l p h o x i d e D N A deoxyribonucleic acid ER endoplasmic re t iculum F A C S fluorescence activated ce l l sorting FD f a r aday FITC fluorescein isothiocyanate GITC guanid in ium isothiocyanate H I V human immunodefic iency virus I F N i n t e r f e r o n I i invariant chain I n f l . A influenza A virus i.p. i n t r a p e r i t o n e u m IRSE interferon response element k D a k i l o d a l t o n L M P low molecular weight protein M H C major histocompatibi l i ty complex M O I mul t ip l ic i ty of infection N F A no first antibody N K natural k i l l e r P A G E polyacrylamide gel electrophoresis PBS phosphate buffered saline PCR polymerase chain reaction P M S F p h e n y l m e t h y l s u l f o n y l f l u o r i d e RT reverse transcriptase SDS sodium dodecyl sulphate s l g surface immunog lobu l in T A P transporter associated with antigen presentation TCID tissue culture infection dose TcR T cel l receptor U u n i t V S V vesicular stomatitis virus V V vaccin ia virus 135 7. R E F E R E N C E S 1. K l e i n , J . 1982. Immunology: The science of se l f -nonse l f discr iminat ion. Wiley and Sons press 2. Landsteiner, K . 1962. The specificity of serological reactions. rev.ed.Dover,New York 3. Burnett, F . M . 1959. The c lonal selection theory of acquired immunity. Cambridge University Press,London. 4. Davis, M . M . and P.J. Bjorkman. 1988. T-cel l receptor genes and T-cel l recognition. Nature 334:395. 5. Townsend, A . , J. Rothbard, F . M . Gotch, G . Bahadur, D . Wraith and A . J . M c M i c h a e l . 1986. The epitopes of Influenza nucleoprotein recognized by cytotoxic T lymphocytes can be defined with short synthetic peptides. Cell 44:959. 6. Gorer, P. 1936. The detection of differences in mouse erythrocytes by the employment of immune sera. Br.J.Exp.Pathol 17:42-50. 7. Snell , G . D . 1948. Methods for the study of histocompatibility genes. J.Genet. 49:87-108. 8. S t roynowski , I. 1990. Molecules related to the class I major histocompatibil i ty complex antigens. Annu.Rev.Immunol. 8:501-530. 9. Zinkernagel, R . M . and P . C . Doherty. 1974. Restriction of in vitro T cel l -mediated cy to tox ic i ty i n lymphocy t i c chor iomeningi t i s w i t h i n a syngeneic or semiallogeneic system. Nature 248:701-702. 10. Janeway, C A . 1992. The T ce l l receptor as a mult icomponent signaling machine: C D 4 / C D 8 coreceptors and C D 4 5 in T ce l l activation. Annu.Rev.Immunol. 10:645. 11. Germain, R. and D . H . Margulies. 1993. The biochemistry and ce l l b io logy of antigen procesing and presentation. Annu.Rev .Immunol. 1 1 : 4 0 3 - 4 5 0 . 12. Daar, A . , S. Fuggle, J . Fabre, A . T ing and P. Mor r i s . 1984. The detailed distr ibution of H L A - A , B and C antigens in normal human organs. Transplantation 38:287-293. 136 13. Joly, E . and M . B . A . Oldstone. 1992. Neuronal cells are deficient in loading peptides onto M H C Class I molecules. Neuron 8:1185-1190. 14. Roby, K . F . , K . Fei , Y . Yang and J.S. Hunt. 1994. Expression of H L A class Il-associated peptide transporter and proteasome genes in human placentas and trophoblast cell lines. Immunol. 83:444-448. 15. Cresswell , P. 1994. Assembly, Transport and Function of M H C Class II molecules. Annu.Rev.Immunol. 12:259-293. 16. Roche, P . A . , M . S . Marks and P. Cresswell. 1991. Formation of a nine-subunit complex by H L A Class II glycoproteins and the invariant chain. Nature 354:392-394. 17. Peters, P.J., J.J. Neefjes, V . Oorschot, H . L . Pleogh and H.J . Geuze. 1991. Segregation of M H C class II molecules from class I molecules in the Golg i complex for transport to lysosomal compartments. Nature 349:669-672. 18. Bakke , O. and B . Dobberstein. 1990. M H C class Il-associated invariant chain contains a sorting signal for endosomal compartments. Cell 63:707-715. 19. Kropshofer, H . , A . B . Vogt and G . Hammerl ing. 1995. Structural features of the invariant chain fragment C L I P control l ing rapid release f rom H L A - D R molecu les and i n h i b i t i o n of pept ide b i n d i n g . Proc.Natl.Acad.Sci.USA 92:8313. 20. Roche, P . A . and P. Cresswell . 1990. Invariant chain association with H L A - D R molecules inhibi ts immunogenic peptide b ind ing . Nature 3 4 5 : 6 1 5 - 6 1 9 . 21. Teyton, L . , D . O'Sullivan, P .W. Dickson, V . Lotteau, A . Sette, P. Fink and P . A . Peterson. 1990. Invariant chain distinguishes between the exogenous and endogenous antigen presentation pathways. Nature 3 4 8 : 3 9 - 4 1 . 22. Riberdy, J . M . , J .R. Newcomb, M . J . Surman, J . A . Barbosa and P. Cresswel l . 1992. H L A - D R molecules from an antigen-processing mutant ce l l line are associated with invariant chain peptides. Nature 360:474-4 7 7 . 23. Sloan, V . S . , P. Cameron, G . Porter, M . Gammon, M . Amaya, E . Mel l ins 137 and D . M . Zaller. 1995. Mediat ion by H L A - D M of dissociation of peptides from H L A - D R . Nature 375:802-805. 24. Brodsky, F . M . and L . E . Guagliardi. 1991. The cell biology of antigen processing and presentation. Annu.Rev.Immunol. 9:707-744. 25. Goodfel low, P . N . , E . A . Jones, V . van Heyningen, E . Solomon, M . Bobrow, M . Miggiano and W . F . Bodmer. 1975. The (32-microglobulin gene is on chromosome 15 and not in the H L A region. Nature 254:267-269. 26. Fraser, J .D. , H . A l l e n , R . A . Flavel l and J .L . Strominger. 1987. C e l l surface expression of H - 2 D b requires N- l inked glycans. Immunogen. 2 6 : 3 1 - 3 5 . 27. Hayes, B . K . , F. Esquivel , J.R. Bennink, J .W. Yewde l l and A . V a r k i . 1995. Structure of N- l inked oligosaccharides of M H C Class I molecules from cells deficient in the antigenic peptide transporter. J.Immunol. 1 5 5 : 3 7 8 0 - 3 7 8 7 . 28. Townsend, A . , C . Ohlen, J. Bastin, H . - G . Ljunggren, L . Foster and K . Karre. 1989. Associat ion of Class I major histocompatibili ty heavy and light chains induced by viral peptides. Nature 340:443-448. 29. E l l io t , T., V . Cerundolo, J . E l v i n and A . Townsend. 1991. Peptide-induced conformational change of the class I heavy chain . Nature 3 5 1 : 4 0 2 - 4 0 6 . 30. Degen, E . , M . Cohen-Doyle and D . B . W i l l i a m s . 1992. Eff ic ient d i ssoc ia t ion of the p88 chaperone from major h i s t ocompa t ib i l i t y complex Class I molecules requires both (^ -mic rog lobu l in and peptide. J.Exp.Med. 175:1653-1661. 31. Nobner, E . and P. Parham. 1995. Species-specific differences in chaperone interaction of human and mouse M H C class I molecules . J.Exp.Med. 181:327-337. 32. Jackson, M R . , M . Cohen-Doyle, P. Peterson and D . B . Wi l l i ams . 1994. Regula t ion of M H C Class I transport by the molecular chaperone, Calnexin (p88, IP90). Science 263:384-387. 33. Rajagopalan, S. and M . B . Brenner. 1994. C a l n e x i n retains unassembled major histocompatibili ty complex Class I free heavy chains 138 in the endoplasmic reticulum. J.Exp.Med. 180:407-412. 34. Bjorkman, P.J. , M . A . Saper, B . Samraoui, W . S . Bennet, J .L . Strominger and D . C . W i l e y . 1987. Structure of the human class I histocompatibility antigen, H L A - A 2 . Nature 329:506. 35. Matsumura, M . , D . H . Fremont, P . A . Peterson and L A . Wi l son . 1992. Emerging principles for the recognition of peptide antigens by M H C Class I molecules. Science 257:927-934. 36. Bouvier, M . and D . C . Wi ley . 1994. Importance of peptide amino and carboxyl termini to the stability of M H C Class I molecules. Science 2 6 5 : 3 9 8 - 4 0 2 . 37. Fremont, D . H . , M . Matsumura, E . A . Stura, P . A . Peterson and L A . W i l s o n . 1992. Crystal structures of two vira l peptides in complex with murine M H C Class I H - 2 K b . Science 257:919-927. 38. Zhang, W . , A . C M . Young , M . Imarai, S .G. Nathenson and J .C . Sacchet t in i . 1992. Crys ta l structure of the major h i s tocompat ib i l i ty complex Class I H - 2 K b molecule containing a single v i r a l peptide: Impl ica t ions for peptide b ind ing and T - c e l l receptor recogni t ion . Proc.Natl.Acad.Sci.USA 89:8403-8407. 39. Latron, F . , L . Pazmany, J . Morr i son , R . Moots , M . A . Saper, A . M c M i c h a e l and J . L . Strominger. 1992. A cr i t i ca l role for conserved residues in the cleft of H L A - A 2 in presentation of a nonapeptide to T cells. Science 257:964-967. 40. Chen, Y . , J. Sidney, S. Southwood, A . L . C o x , K . Sakaguchi, R . A . Henderson, E . Appel la , D . F . Hunt, A . Sette and V . H . Engelhard. 1994. Naturally processed peptides longer than nine amino acid residues bind to Class I M H C molecule H L A - A 2 . 1 with high affinity and in different conformations. J.Immunol. 152:2874-2881. 41. Brown, J .H . , T .S . Jardetzky, J .C. Gorga, L . J . Stern, R . G . Urban, J .L . Strominger and D . C . W i l e y . 1993. Three dimensional structure of the human histocompatibili ty antigen H L A - D R 1 . Nature 364:33. 42. Rammensee, H - G . 1995. Chemistry of peptides associated with M H C Class I and Class II molecules. Curr.Opin.Immunol. 7:85-96. 43. Rammensee, H - G . , T. Friede and S. Stevanovic. 1995. M H C ligands 139 and peptide motifs: first listing. Immunogen. 41:178-228. 44. Goldberg, A . L . and K . L . Rock. 1992. Proteolysis, proteasomes and antigen presentation. Nature 357:375-379. 45. Monaco, J.J. and H . O . McDevit t . 1982. Identification of a fourth class of proteins l inked to the murine major h is tocompat ibi l i ty complex . Proc.Natl.Acad.ScLUSA 79:3001-3005. 46. Waxman, L . , J . M . Fagan and A . L . Goldberg. 1987. Demonstration of two distinct high molecular weight proteases in rabbit reticulocytes, one of which degrades ubiquitin conjugates. J.Biol.Chem. 262:2451. 47. Harding, C . V . , J. France, Rui . Song, J . M . Farah, S. Chatterjee, M . Iqbal and R. S iman . 1995. N o v e l dipeptide aldehydes are proteasome inhibitors and block the M H C - I antigen processing pathway. J.Immunol. 1 5 5 : 1 7 6 7 - 1 7 7 5 . 48. Grant, E . , M . T . Michalek, A . L . Goldberg and K . L . Rock. 1995. Rate of antigen degradation by the ubiquitin-proteasome pathway influences M H C Class I presentation. J.Immunol. 155:3750-3758. 49. Rock, K . L . , C . Gramm, L . Rothstein, K . Clark, R. Stein, Lawrence. Dick , D . Hwang and A . L . Goldberg. 1995. Inhibitors of the proteasome block the degradation of most ce l l proteins and the generation o f peptides presented on M H C Class I molecules. Cell 78:761-771. 50. Michalek, M . T . , E .P . Grant, C . Gramm, A . L . Goldberg and K . L . Rock. 1993. A role for the ubiquitin-dependent proteolytic pathway in M H C class I-restricted antigen presentation. Nature 363:552-554. 51. Sijts, A . J . A . , M . S . Vi l lanueva and E . G . Pamer. 1996. C T L epitope generation is t ightly l inked to the cellular proteolysis of a Listeria monocytogenes antigen. J.Immunol. 156:1497-1503. 52. B r o w n , M . G . , J. D r i s c o l l and J.J. Monaco. 1991. Structural and s e r o l o g i c a l s i m i l a r i t y o f M H C - l i n k e d L M P and p ro teasome (multicatalytic proteinase) complexes. Nature 353:355-357. 53. Glynne, R., S .H . Powis, S. Beck, A . Ke l ly , L - A . Kerr and J. Trowsdale. 1991. A proteasome-related gene between the two A B C transporter loc i in the class II region of the human M H C . Nature 353:357-360. 140 54. Ortiz-Navarette, V . , A . Seelig, M . Gernold, S. Frentzel, P . M . Kloetzel and G . J . H a m m e r l i n g . 1991. Subuni t of the '20S ' proteasome (mul t ica ta ly t ic proteinase) encoded by the major h i s tocompat ib i l i ty complex. Nature 353:662-664. 55. Martinez, C . and J.J. Monaco. 1991. Homolgy of proteasome subunits to a major h is tocompat ib i l i ty complex - l i nked L M P 2 gene. Nature 3 5 3 : 6 6 4 - 6 6 7 . 56. Ke l ly , A . , S .H . Powis, R. Glynne, E . Radley, S. Beck and J. Trowsdale. 1991. Second proteasome-related gene in the human M H C class II region. Nature 353:667-668. 57. Arnold , D . , J. Driscol l , M . Androlewicz, E . Hughes, P. Cresswell and T. Spies. 1992. Proteasome subunits encoded in the M H C are not generally required for the processing of peptides bound by M H C Class I molecules. Nature 360:171-174. 58. Zhou, X . , F . Momburg, T. L i u , U . M . Abde l -Mota l , M . Jondal, G . J . Hammerl ing and H - G . Ljunggren. 1994. Presentation of v i ra l antigens restricted by H - 2 K b or K d in proteasome subunit L M P 2 - and L M P 7 -deficient cells. Eur.J.Immunol. 24:1863-1868. 59. Gabathuler, R. , G . Reid , G . Kolai t i s , J . Dr i sco l l and W . A . Jefferies. 1994. Comparison of ce l l lines deficient in antigen presentation reveals a functional role for T A P - 1 alone in antigen processing. J .Exp .Med. 1 8 0 : 1 4 1 5 - 1 4 2 5 . 60. Fehling, H.J . , W . Swat, C. Laplace, R. Kuhn, K . Rajewsky, U . Mul le r and H . von Boehmer. 1994. M H C Class I expression in mice lacking the proteasome subunit L M P - 7 . Science 265:1234-1237, 61. Van Kaer, L . , P . G . Ashton-Rickadt, M . Eichelberger, M . Gaczynska, K . Nagashima, K . L . Rock, A . L . Goldberg, P .C . Doherty and S. Tonegawa. 1994. Altered peptidase and viral-specif ic T ce l l response in L M P 2 mutant mice. Immunity 1:533-541. 62. Cerundolo, V . , A . K e l l y , T. El l io t , J . Trowsdale and A . Townsend. 1995. Genes encoded in the major histocompatibil i ty complex affecting the generation of peptides for T A P transport. Eur. J .Immunol. 25:554-5 6 2 . 63. Nandi, D . , H . Jiang and J.J. Monaco. 1996. Identification of M E C L - 1 141 ( L M P - 1 0 ) as the third IFN-y- inducible proteasome subunit. J .Immunol. 1 5 6 : 2 3 6 1 - 2 3 6 4 . 64. Yang, Y . , J .B . Waters, K . Fruh and P . A . Peterson. 1992. Proteasomes are regulated by interferon-y: Implicat ions for antigen process ing. Immunol. 89:4928-4932. 65. Ak iyama , K . , K . Yokota , S. Kagawa, N . Shimbara, T. Tamura, H . Ak ioka , H . G . Nothwang, C . Noda, K . Tanaka and A . Ichihara. 1994. c D N A cloning and interferon-y down-regulation of proteasomal subunits X and Y . Science 265:1231-1234. 66. Dr iscol l , J. , M . G . Brown, D . Finley and J.J. Monaco. 1993. M H C - l i n k e d L M P gene products spec i f i ca l ly alter peptidase ac t iv i t ies of the proteasome. Nature 365:262-264. 67. Eisenlohr, L . C . , I. Bacik, J.R. Bennink, K . Bernstein and J .W. Yewde l l . 1992. Express ion of a membrane protease enhances presentation of endogenous antigens to M H C Class I-restricted T lymphocytes . Cell 7 1 : 9 6 3 - 9 7 2 . 68. Snyder, H . L . , J .W. Y e w d e l l and J.R. Bennink. 1994. T r imming of antigenic peptides i n an early secretory compartment. J .Ex.Med. 1 8 0 : 2 3 8 9 - 2 3 9 4 . 69. Salter, R . D . , D . N . Howe l l and P. Cresswell . 1985. Genes regulating H L A Class I ant igen express ion in T - B lymphob las t h y b r i d s . Immunogen. 21:235-246. 70. DeMars, R., C C . Chang, S. Shaw, P.J. Reitnauer and P . M . Sondel. 1984. Homozygous deletions that simultaneously eliminate expression of Class I and Class II antigens of EBV-transformed B-lymphoblas to id cel ls . I. Reduced proliferative responses of autologous and allogeneic T cells to mutant cel ls that have dcreased expression of Class II antigens. Hum.Immunol. 11:575-577. 71. Livingstone, A . M . , S.J. Powis, A . G . Diamond, G . W . Butcher and J . C Howard. 1989. A trans-acting major his tocompat ibi l i ty complex - l i nked gene whose alleles determine gain and loss changes in the antigenic structure of a classical Class I molecule. J.Ex.Med. 170:777-795. 142 72. Deverson, E . V . , I.R. Gow, W.J . Coadwell, J.J. Monaco, G . W . Butcher and J .C . Howard. 1990. M H C Class II region encoding proteins related to the mul t id rug resistance fami ly of transmembrane transporters. Nature 3 4 8 : 7 3 8 - 7 4 1 . 73. Trowsdale, J. , I. Hanson, I. Mockridge, S. Beck, A . Townsend and A . K e l l y . 1990. Sequences encoded in the Class II region of the M H C related to the ' A B C superfamily of transporters. Nature 348:741-744. 74. Spies, T., M . Bresnahan, S. Bahram, D . Arnold, G . Blanck, E . Mel l ins , D . P i ous and R . D e M a r s . 1990. A gene i n the human major histocompatibility complex Class II region controlling the Class I antigen presentation pathway. Nature 348:744-747. 75. Powis, S.H. , I. Mockridge, A . Kel ly , L . - A . Kerr, R. Glynne, U . Gileadi , S. Beck and J. Trowsdale. 1992. Polymorphism in a second A B C transporter gene located w i t h i n the Class II region of the human major histocompatibili ty complex. Proc.Natl.Acad.Sci.USA 89:1463-1467. 76. Bahram, S., D . Arnold , M . Bresnahan, J .L . Strominger and T. Spies. 1991. T w o putative subunits of a peptide pump encoded in the human major histocompatibility complex Class II region. Proc.Natl.Acad.Sci.USA 8 8 : 1 0 0 9 4 - 1 0 0 9 8 . 77. Monaco, J.J., S. Cho and M . Attaya. 1990. Transport protein genes in the murine M H C : possible implications for antigen processing. Science 2 5 0 : 1 7 2 3 - 1 7 2 6 . 78. Higgins, C . F . 1992. A B C transporters: From microorganisms to man. Annu.Rev.Cell Biol. 8:67-113. 79. Momburg , F. , V . Ortiz-Navarrete, J . Neefjes, E . Goulmy, Y . van de W a l , H . Spits, S.J. Powis, G . W . Butcher, J .C. Howard, P. Walden and G.J . H a m m e r l i n g . 1992. Proteasome subunits encoded by the major h is tocompat ib i l i ty complex are not essential for antigen presentation. Nature 360:174-177. 80. Spies, T., V . Cerundolo, M . Colonna, P. Cresswell, A . Townsend and R. DeMars . 1992. Presentation of viral antigen by M H C Class I molecules is dependent on a putative peptide transporter heterodimer. Nature 3 5 5 : 6 4 4 - 6 4 6 . 143 81. Spies, T . and R. D e M a r s . 1991. Restored expression of major his tocompatibi l i ty Class I molecules by gene transfer of a putative peptide transporter. Nature 351:323-324. 82. Ke l ly , A . , S .H. Powis, L . - A . Kerr, I. Mockridge, T. Elliott , J . Bastin, B . Uchanska-Ziegler , A . Ziegler , J . Trowsdale and A . Townsend. 1992. Assembly and function of the two A B C transporter proteins encoded in the human major histocompatibility complex. Nature 355:641-644. 83. Ljunggren, H . - G . , S. Paabo, M . Cochet, G . K l i n g , P. Kour i l sky and K . Karre . 1989. M o l e c u l a r analysis of H-2 deficient l ymphoma l ines : distinct defects in biosythesis and association of M H C Class I heavy chains and (^ -microg lobul in observed in cells with increased sensitvity to N K cell lysis. J.Immunol. 142:2911-2917. 84. Yang, Y . , K . Fruh, J. Chambers, J .B . Waters, L . W u , T. Spies and P . A . Peterson. 1992. Major histocompatibility complex (MHC)-encoded H A M 2 is necessary for antigenic peptide loading onto Class I M H C molecules. J.Biol.Chem. 267:11669-11672. 85. A l d r i c h , C . J . , R. Waltr ip, E . Hermel, M . Attaya, K . F . L indah l , J .J. Monaco and J. Forman. 1992. T cel l recognition of Q a - 1 D antigens on cells lacking a functional Tap-2 transporter. J.Immunol. 149:3773-3777. 86. Attaya, M . , S. Jameson, C K . Martinez, E . Hermel , C . A l d r i c h , J . Forman, K . F . Lindahl , M . J . Bevan and J.J. Monaco. 1992. Ham-2 corrects the Class I antigen-processing defect in R M A - S cells. Nature 355:647-649 . 87. Powis, S.J., A . R . M . Townsend, E . V . Deverson, J. Bastin, G . W . Butcher and J . C Howard. 1991. Restoration of antigen presentation to the mutant cell line R M A - S by an M H C - l i n k e d transporter. Nature 354:528-531. 88. Yewdel l , J .W., F. Esquivel, D . Arnold, T. Spies, L . C Eisenlohr and J.R. Bennink. 1993. Presentation of numerous vira l peptides to mouse major histocompatibil i ty complex ( M H C ) Class I-restricted T lymphocytes is mediated by the human M H C - e n c o d e d transporter or by a hyb r id mouse-human transporter. J.Exp.Med. 177:1785-1790. 89. Powis, S.J., E . V . Deverson, W.J . Coadwell , A . Ciruela, N . S . Huskisson, H . Smith, G . W . Butcher and J . C Howard. 1992. Effect of polymorphism of an M H C - l i n k e d transporter on the peptides assembled in a Class I 144 molecule. Nature 357:211-215. 90. Russ, G . , F. Esquivel , J .W. Yewdel l , P. Cresswell , T. Spies and J.R. Bennink . 1995. Assembly , intracellular loca l iza t ion , and nucleotide binding properties of the human peptide transporters T A P 1 and T A P 2 expressed by recombinant vaccinia viruses. J .Biol.Chem. 270:21312-2 1 3 1 8 . 91. Kleijmeer, M . J . , A . Ke l ly , H . J . Geuze, J .W. Slot, A . Townsend and J. T r o w s d a l e . 1992. L o c a t i o n of M H C - e n c o d e d transporters in the endoplasmic reticulum and c/s-golgi. Nature 357:342-344. 92. V a n Kaer, L . , P . G . Ashton-Rickardt, H . L . Ploegh and S. Tonegawa. 1992. TAP1 mutant mice are deficient in antigen presentation, surface Class I molecules, and CD4"8+ T cells. Cell 71:1205-1214. 93. Ljunggren, H . - G . , L . V a n Kaer, H . L . Ploegh and S. Tonegawa. 1994. A l t e r e d natural k i l l e r c e l l repertoire i n Tap-1 mutant m i c e . Proc.Natl.Acad.Sci. USA 91:6520-6524. 94. Shepherd, J . C , T . N . M . Schumacher, P . G . Ashton-Rickardt, S. Imaeda, H . L . Ploegh, C . A . Janeway Jr. and S. Tonegawa. 1993. TAPl -dependen t peptide translocation in vitro is A T P dependent and peptide selective. Cell 74:577-584. 95. Ljunggren, H . - G . , L . Van Kaer, M . S . Sabatine, H . Auchincloss Jr., S. Tonegawa and H . L . Ploegh. 1995. M H C Class I expression and CD8+ T cell d e v e l o p m e n t i n T A P l / ( 3 2 - m i c r o g l o b u l i n doub le mutant m i c e . Int.Immunol. 7:975-984. • • 96. A ld r i ch , C . J . , H . - G . Ljunggren, L . V a n Kaer, P . G . Ashton-Rickardt, S. Tonegawa and J . Forman. 1994. Pos i t ive select ion of self- and alloreactive C D 8 + T cells in Tap-1 mutant mice. Proc.Natl.Acad.Sci. USA 9 1 : 6 5 2 5 - 6 5 2 8 . 97. Ljunggren, H . - G . , L . V a n Kaer, P . G . Ashton-Rickardt, S. Tonegawa and H . L . Ploegh. 1995. Different ia l react ivi ty of res idual C D 8 + T lymphocytes in T A P and (3 2-microglobulin mutant mice. Eur. J .Immunol. 2 5 : 1 7 4 - 1 7 8 . 98. van Endert, P . M . , R. Tampe, T . H . Meyer, R. Tisch, J.-F. Bach and H . O . M c D e v i t t . 1994. A sequential model for peptide binding and transport 145 by the transporters associated wi th antigen processing. Immunity 1:491-500. 99. Meyer, T . H . , P . M . van Endert, S. Uebel, B . Ehring and R. Tampe. 1994. Funct ional expression and purification of the A B C transporter complex associated with antigen processing (TAP) in insect cells. FEBS Lett. 351:443-447. 100. Androlewicz , M . J . , K . S . Anderson and P. Cresswell . 1993. Evidence that transporters associated with antigen processing translocate a major histocompatibil i ty complex Class I-binding peptide into the endoplasmic re t i cu lum in an ATP-dependent manner. Proc.Natl.Acad.Sci.USA 9 0 : 9 1 3 0 - 9 1 3 4 . 101. M u l l e r , K . M . , C . Ebensperger and R. Tampe. 1994. Nucleot ide b i n d i n g to the h y d r o p h i l i c C- te rmina l domain of the transporter associated with antigen processing ( T A P ) . J.Biol.Chem. 269:14032-1 4 0 3 7 . 102. Wang, K . , K . Fruh, P . A . Peterson and Y . Yang . 1994. Nucleotide b ind ing of the C- te rmina l domains of the major h i s tocompat ib i l i ty complex-encoded tranpsorter expressed in Drosophila melanogaster cells. FEBS Lett. 350:337-341. 103. Cho, S., M . Attaya, M . G . Brown and J.J. Monaco. 1991. A cluster of transcribed sequences between the Pb and Ob genes of the murine major histocompatibility complex. Proc.Natl.Acad.Sci.USA 88:5197-5201. 104. Neefjes, J.J., F . Momburg and G.J . Hammerling. 1993. Selective and A T P - d e p e n d e n t t r ans loca t ion of peptides by the M H C - e n c o d e d transporter. Science 261:169-111. 105. A n d r o l e w i c z , M . J . and P. Cresswel l . 1994. Human transporters associated wi th antigen processing possess a promiscuous peptide-binding site. Immunity 1:7-14. 106. Nijenhuis, M . , S. Schmitt, E . A . Armandola, R. Obst, J. Brunner and G . J . Hammerl ing. 1996. Identification of a contact region for peptide on the T A P 1 chain of the transporter associated with antigen processing. J.Immunol. 156:2186-2195. 107. Andro lewicz , M . J . , B . Ortmann, P . M . van Endert, T. Spies and P. Cresswel l . 1994. Characteristics of peptide and major histocompatibil i ty 146 complex Class I/P2-rnicroglobulin binding to the transporters associated with antigen processing ( T A P 1 and T A P 2 ) . Proc.Natl.Acad. Sci. USA 9 1 : 1 2 7 1 6 - 1 2 7 2 0 . 108. Momburg, F. , E . A . Armandola, M . Post and G . J . Hammerl ing. 1996. Residues in T A P 2 peptide tranpsporters controll ing substrate, specificity. J.Immunol. 156:1756-1763. 109. Momburg , F. , J . Roesle, G . J . Hammerl ing and J.J. Neefjes. 1994. Peptide size selection by the major histocompatability complex-encoded peptide transporter. J.Exp.Med. 179:1613-1623. 110. Powis, S.J., L . L . Young, E . Joly, P.J. Barker, L . Richardson, R.P. Brandt, C . J. Mel i f , J .C. Howard and G . W . Butcher. 1996. The rat cim effect: T A P allele-dependent changes in a Class I M H C anchor motif and evidence against C-terminal trimming of peptides in the E R . Immunity 4:159-165. 111. Heemels, M - T . , T . N . M . Schumacher, K . Wonigei t and H . L . Pleogh. 1993. Peptide translocation by variants of the transporter associated with antigen processing. Science 262:2059-2063. 112. Momburg, F. , J. Roelse, J .C. Howard, G . W . Butcher, G . J . Hammerling and J.J. Neefjes. 1994. Selectivity of M H C - e n c o d e d peptide transporters from human , mouse and rat. Nature 367:648-651. 113. Schumacher, T . N . M . , D . V . Kantersaria, M - T . Heemels, P . G . Ashton-Rickardt, J .C. Shepard, K . Fruh, Y . Yang, P . A . Peterson, S. Tonegawa and H . L . Pleogh. 1994. Peptide length and sequence specificity of the mouse T A P 1 / T A P 2 translocator. J.Ex.Med. 179:533-540. 114. Neefjes, J . , E . Gottfried, J. Roelse, M . Gromme, R . Obst, G . J . Hammerl ing and F . Momburg . 1995. Analys is of the fine specificity of rat, mouse and human T A P peptide transporters. Eur .J .Immunol. 2 5 : 1 1 3 3 - 1 1 3 6 . 115. Gournier, H . , S. Pascolo, C - A . Siegrist, J. Jehan, B . Perarnau, Z . Garcia, T. Rose, J.J. Neefjes and F . A . Lemonnier. 1995. Restriction of self-antigen presentation to cy to ly t i c T lymphocytes by mouse peptide pumps. Eur.J.Immunol. 25:2019-2026. 116. Henderson, R . A . , H . Miche l , K . Sakaguchi, J. Shabanowitz, E . Appel la , D . F. Hunt and V . H . Engelhard. 1992. HLA-A2.1-associa ted peptides from a mutant ce l l l ine: a second pathway of antigen presentation. Science 147 2 5 5 : 1 2 6 4 - 1 2 6 6 . 117. W e i , M . L . and P. Cresswell . 1992. H L A - A 2 molecules in an antigen-processing mutant cel l contain signal sequence-derived peptides. Nature 3 5 6 : 4 4 3 - 4 4 6 . 118. Hammond, S .A. , R . C . Bollinger, T . W . Tobery and"R.F. Sil iciano. 1993. Transporter-independent processing of H I V - 1 envelope protein for recognition by CD8+ T cells. Nature 364:158-161. 119. Hammond, S . A . , R . P . Johnson, S . A . Kalams, B . D . Walke r , M . Takiguchi , J.T. Safrit, R . A . Koup and R . F . Si l ic iano. 1995. A n epitope-selective, transporter associated with antigen presentation ( T A P ) - I n -dependent pathway and a more general T A P 1/2-dependent antigen-process ing pathway a l l o w r ecogn i t i on of the H I V - 1 enve lope glycoprotein by CD8+ C T L . J.Immunol. 154:6140-6156. 120. Anderson, K . , P. Cresswell , M . Gammon, J. Hermes, A . Wi l l i amson and H . Zweer ink . 1991. Endogenously synthesized peptide wi th an endoplasmic re t iculum signal sequence sensitizes antigen processing mutant cel ls to Class I-restricted cel l -mediated lys i s . J.Exp.Med. 1 7 4 : 4 8 9 - 4 9 2 . 121. Restifo, N .P . , I. Bacik, K . R . Irvine, J .W. Yewde l l , B . J . McCabe , R . W . Anderson, L . C . Eisenlohr, S .A. Rosenberg and J.R. Bennink. 1995. Antigen processing i n v ivo and the e l ic i ta t ion of pr imary C T L responses. J.Immunol. 154:4414-4422. 122. Zweerink, H . J . , M . C . Gammon, U . Utz , S . Y . Sauma, T. Harrer, J .C . Hawkins, R .P . Johnson, A . Sirotina, J .D. Hermes, B . D . Walker and W . E . Bidd ison . 1993. Presentation of endogenous peptides to M H C Class I-restricted cytotoxic T lymphocytes in transport deletion mutant T2 cells. J.Immunol. 150:1763-1771. 123. Bacik, I., J .H . Cox, R. Anderson, J .W. Yewdel l and J.R. Bennink. 1994. T A P (transporter associated wi th antigen process ing)- independent presentation of endogenously synthesized peptides is enhanced by endoplasmic re t iculum insertion sequences located at the amino- but not carboxyl-terminus of the peptide. J .Immunol. 152:381-387. 124. Hombach, J . , H . Pircher, S. Tonegawa and R . M . Zinkernagel. 1995. S t r i c t l y t ranspor ter o f an t igen p resen ta t ion ( T A P ) - d e p e n d e n t 148 presentation of an immunodominant cytotoxic T lymphocyte epitope in the signal sequence of a virus protein. J .Exp.Med. 182:1615-1619. 125. Aldr ich , C . J . , A . DeCloux, A . S . Woods, R.J . Cotter, M . J . Soloski and J. Fo rman . 1994. Ident i f icat ion of a TAP-dependen t leader peptide recognized by alloreactive T cells specific for a Class lb antigen. Cell 7 9 : 6 4 9 - 6 5 8 . 126. Zhou, X . , R. Glas, T. L i u , H . - G . Ljunggren and M . Jondal. 1993. An t igen processing mutant T2 cells present v i r a l antigen restricted through H - 2 K b . Eur.J .Immunol. 23:1802-1808. 127. Zhou, X . , T. L i u , L . Franksson, E . Lederer, H . - G . Ljunggren and M . Jondal . 1995. Character izat ion of TAP-independent and brefeldin A -resistant presentation of Sendai virus antigen to C D 8 + cytotoxic T lymphocytes. Scand.J.Immunol. 42:66-75. 128. L i u , T., X . Zhou, C . Orvell , E . Lederer, H . - G . Ljunggren and M . Jondal. 1995. Heat-inactivated Sendai virus can enter mult iple M H C Class I processing pathways and generate cytotoxic T lymphocyte responses in v ivo . J.Immunol. 154:3147-3155. 129. Bevan, M . J . 1995. Antigen presentation to cytotoxic T lymphocytes in v ivo. J.Exp.Med. 182:639-641. 130. Rock , K . L . , L . Rothstein, S. Gamble and C . Fleischacker. 1993. Charac ter iza t ion of antigen-presenting cel ls that present exogenous antigens in association with Class I M H C molecules. J.Immunol. 150:438-4 4 6 . 131. Reis e Sousa, C . and R . N . Germain. 1995. Major histocompatibility complex Class I presentation of peptides der ived f rom soluble exogenous antigen by a subset of cells engaged i n phagocytosis . J.Exp.Med. 182:841-851. 132. Bachmann, M . F . , A . Oxenius, H . Pircher, H . Hengartner, P . A . Ashton-Rickardt , S. Tonegawa and R . M . Zinkernagel . 1995. TAPl - independen t loading of Class I molecules by exogenous viral proteins. Eur.J.Immunol. 2 5 : 1 7 3 9 - 1 7 4 3 . 133. Kovacsovics -Bankowski , M . and K . L . Rock. 1995. A phagosome-to-cytosol pathway for exogenous antigens presented on M H C Class I molecules. Science 267:243-246. 1 4 9 134. Kagi , D . , B . Ledermann, K . Burk i , P. Seiler, B . Odermatt, K . J . Olsen, E . R . Podack, R . M . Zinkernagel and H . Hengartner. 1994. Cyto tox ic i ty mediated by T cells and natural k i l l e r cells is greatly impaired in perforin-deficient mice. Nature 369:31-37. 135. Gumperz, J .E. and P. Parham. 1995. The enigma of the natural ki l ler cell . Nature 378:245-248. 136. Salcedo, M . , F . Momburg , G . J . Hammerl ing and H . - G . Ljunggren. 1994. Resistance to natural ki l ler cel l l iysis conferred by T A P 1 / 2 genes in human antigen-processing mutant cells. J.Immunol. 152:1702-1708. 137. Franksson, L . , E . George, S. Powis, G . Butcher, J . Howard and K . Karre. 1993. Tumorigenic i ty conferred to lymphoma mutant by major h i s t o c o m p a t i b i l i t y complex-encoded transporter gene. J.Exp.Med. 1 7 7 : 2 0 1 - 2 0 5 . 138. Ljunggren, H . - G . , C . Ohlen, P. Hoglund, L . Franksson and K . Karre. 1991. The R M A - S lymphoma mutant; consequences of a peptide loading defect on immunologica l recognition and graft rejection. Int.J.Cancer Supp. 6:38-44. 139. Ljunggren, H . - G . and K . Karre . 1985. Host resistance directed selectively against H-2 deficient lymphoma variants: analysis of the mechanism. J.Exp.Med. 162:1745-1759. 140. Karre, K . , H . - G . Ljunggren, G . Piontek and R. K ie s s l i ng . 1986. Se lec t ive re jec t ion of H-2-def ic ien t l y m p h o m a variants suggests alternative immune defence strategy. Nature 319:675-678. 141. Glas, R. , K . Sturmhofel, G . J . Hammerl ing , K . Karre and H . - G . Ljunggren. 1992. Restorat ion of a tumorigenic phenotype by P 2 -microglobulin transfection to E L - 4 mutant cells. J.Exp.Med. 175:843-846. 142. Sturmhofel, K . and G . J . Hammerl ing. 1990. Reconstitution of H-2 Class I expression by gene transfection decreases suscept ibi l i ty to natural killer cells of an E L 4 Class I loss variant. Eur. J.Immunol. 20:171-177. 143. Pena, J . and R. Solana. 1992. His tocompat ib i l i ty antigens and natural k i l le r susceptibility. Immunol.Res. 11:133-140. 150 144. Yokoyama, W . M . and W . E . Seaman. 1993. The Ly-49 and N K R - P 1 gene families encoding lectin-l ike receptors on natural k i l l e r cel ls : The N K gene complex. Annu.Rev.Immunol. 11:613-635. 145. Y o k o y a m a , W . M . 1995. N a t u r a l k i l l e r c e l l r ecep tors . Curr.Opin.Immunol. 7:110-120. 146. Franks, L . M . , A . W . Carbonell, V . J . Hemmings and P . N . Riddle. 1976. Metas tas iz ing tumours from serum-supplemented and serum-free ce l l lines from a C57B1 mouse lung tumour. Cancer Res. 36:1049-1055. 147. Kla r , D . and G. J . Hammerling. 1989. Induction of assembly of M H C CLass I heavy chains with ( ^ -mic rog lobu l in by interferon-y. EMBO J. 8 : 4 7 5 - 4 8 1 . 148. S ib i l l e , C , K . Gould , G . Hammerling and A . Townsend. 1992. A defect in the presentation of intracellular v i ra l antigens is restored by i n t e r f e r o n - y i n c e l l l ines wi th impaired major h i s tocompat ib i l i ty complex Class I assembly. Eur.J.Immunol. 22:433-440. 149. Jefferies, W . A . , G . Kolai t is and R. Gabathuler. 1993. IFN-y induced recognition of the antigen-processing variant C M T . 6 4 by cytolytic T cells can be replaced by sequential addition of (3 2 microglobul in and antigenic peptides. J.Immunol. 151:2974-2985. 150. Puddington, L . , M . J . Bevan, J . K . Rose and L . Lefrancois. 1986. N protein is the predominant antigen recognized by vesicular stomatitis virus-specific cytotoxic T cells. J.Virol. 60:708-717. 151. V a n Bleek , G . M . and S . G . Nathenson. 1990. Isolat ion of an endogenously processed immunodominant vira l peptide from the Class I H - 2 K b molecule. Nature 348:213-216. 152. Rotzsche, O., K . Falk, K . Deres, H . Schild, M . Norda, J. Metzger, G . Jung and H - G . Rammensee. 1990. Isolation and analysis of naturally occuring viral peptides as recognized by cytotoxic T cells. Nature 348:252-254. 153. Rees, R . C , A . - M . Buckle , K . Gelsthorpe, V . James, C . W . Potter, K . Rogers and G . Jacob. 1988. Loss of polymorphic A and B locus H L A antigens in colon carcinoma. Br.J.Cancer 57:374-377. 154. Keating, P.J. , F . V . Cromme, M . Duggan-Keen, P.J.F. Snijders, J . M . M . 151 Walboomers, R . D . Hunter, P . A . Dyer and P . L . Stern. 1995. Frequency of down-regula t ion of i n d i v i d u a l H L A - A and - B alleles in ce rv i ca l carcinomas in relation to T A P - 1 expression. Br.J.Cancer 405-411. 155. Lopez-Nevot, M . A . , F. Esteban, A . Ferron, J. Gutierrez, M . R . Ol iva , C . Romero, C . Huelin, F. Ruiz-Cabello and F. Garrido. 1988. H L A Class I gene expression on human primary tumours and autologous metastases: demonstration of selective losses of H L A antigens on colorectal, gastric and laryngeal carcinomas. Br.J .Cancer 221-226. 156. Wal l i ch , R., N . Bulbic , G .J . Hammerling, S. Katzav, S. Segal and M . Feldman. 1985. Abroga t ion of metastatic properties of tumour cells byde novo expression of H - 2 K antigens fol lowing H-2 gene transfection. Nature 315:301-305. 157. Sambrook, J., E . F . Fritsch and T. Maniatis. 1989. Molecular cloning: A laborarory manual. Cold Spring Harbour Laborarory Press,Cold Spring Harbour,NY. 158. Marrack , P. and J . Kappler . 1994. Subversion of the immune system by pathogens. Cell 76:323-332. 159. Greenberg, P . D . 1991. Adop t ive T ce l l therapy of tumours: mechanisms operative i n the recognit ion and e l iminat ion of tumour cells. Adv.Immunol. 49:281. 160. van der Bruggen, P., C . Traversasri, P. Chomez, C . Lurquin, E . De Plaen, B . Van der Eynde, A . Kruth and T. Boon. 1991. A gene encoding an antigen recognized by cytoloytic T lymphocytes on a human melanoma. Science 254:1643-1646. 161. Tanaka, K . , K . J . Isselbacher, G . Khoury and G . Jay. 1985. Reversal of oncogenesis by the expression of a major histocompatibi l i ty complex Class I gene. Science 228:26-30. 162. H u i , K . , F . Grosve ld and H . Festenstein. 1984. Reject ion of transplantable A K R leukaemia cells fo l lowing M H C DNA-media t ed ce l l transformation. Nature 311:750-752. 163. Restifo, N .P . , F. Esquivel, Y . Kawakami, J .W. Yewdel l , J.J. Mule , S .A. Rosenberg and J .R . Benn ink . 1993. Identification of human cancers deficient in antigen processing. J.Exp.Med. 177:265-272. 152 164. Doyle, A . , W.J . Martin, K . Funa, A . Gazdar, D . Carney, S.E. Mart in, I. L inno i l a , F. Cuttitta, J. Mulshine, P. Bunn and J. Minna . 1985. Markedly decreased expression of Class I histocompatibility antigens, protein, and m R N A in human small-cell lung cancer. J.Exp.Med. 161:1135-1151. 165. Festenstein, H . and F . Garrido. 1986. M H C antigens and malignancy. Nature 322:502-503. 166. Momburg , F. , A . Ziegler, J. Harpprecht, P. Mol l e r , G . Moldenhauer and G . J . Hammerl ing. 1989. Selective loss of H L A - A or H L A - B antigen expression in colon carcinoma. J .Immunol. 142:352-358. 167. Funa, K . , A . F . Gazdar, J .D. Minna and R.I . Linnoi la . 1986. Paucity of P 2 - m i c r o g l o b u l i n expression on smal l c e l l lung cancer, b ronch ia l ca rc ino ids , and certain other neuroendocrine tumors. Lab.Invest. 5 5 : 1 8 6 - 1 9 2 . 168. Goepel, J.R., R . C . Rees, K . Rogers, C.J . Stoddard, W . E . G . Thomas and L . Shepherd. 1991. Loss of monomorphic and polymorphic H L A antigens in metastatic breast and colon carcinoma. Br.J.Cancer 64:880-883. 169. Redondo, M . , A . Concha, R. Oldivie la , A . Cueto, A . Gonzalez, F. Garr ido and F . Ruiz -Cabe l lo . 1991. Expression of H L A Class I and II antigens in bronchogenic carcinomas: its relationship to cel lular D N A content and cl inical-pathological parameters. Cancer Res. 51:4948-4954. 170. Dammrich, J . , H . K . Muller-Hermelink, A . Mattner, J. Buchwald and S. Z i f f e r . 1990. H i s t o c o m p a t i b i l i t y antigen express ion i n pu lmonary carcinomas as indica t ion of differentiation and of special subtypes. Cancer 65:1942-1954. 171. Wintzer , H . - O . , M . Benzing and S. von Kle i s t . 1990. L a c k i n g prognostic significance of P2-microglobulin, M H C Class I and Class II antigen expression in breast carcinomas. Br.J .Cancer 289-295. 172. Restifo, N . P . , F . Esquivel , A . L . Asher, H . Stotter, R . J . Barth, J .R. Bennink, J.J. M u l e , J .W. Y e w d e l l and S . A . Rosenberg. 1991. Defective presentation of endogenous antigens by a murine sarcoma: implicat ions for the failure of an anti-tumor immune response. J .Immunol. 1 4 7 : 1 4 5 3 - 1 4 5 9 . 173. Ljunggren, H . G . and K . Karre. 1990. In search of the "missing self": 153 M H C molecules and N K recognition. Immunol.To day 11:237-244. 174. Seong, R . H . , C . A . Clayberger, A . M . Krensky and J.R. Parnes. 1988. Rescue of Daudi cel l H L A expression by transfection of the mouse p2-microglobulin gene. J.Ex.Med. 167:288. 175. Lassam, N . and G . Jay. 1989. Suppression of M H C Class I R N A in highly oncogenic cells occurs at the level of transcription ini t ia t ion. J.Immunol. 143:3792-3797. 176. Croce, C M . , A . Linnenbach, K . Huebner, J.R. Parnes, D . H . Margulies, E . A p p e l l a and J . G . Se idman. 1981. C o n t r o l of express ion of h i s t o c o m p a t i b i l i t y antigens (H-2) and P 2 - m i c r o g l o b u l i n i n F 9 teratocarcinoma stem cells. Proc.Natl.Acad.Sci.USA 75:5754-5758. 177. Korkolopoulou, P., L . Kaklamanis, F. Pezzella, A . L . Harris and K . C Gatter. 1996. Loss of antigen-presenting molecules ( M H C Class I and T A P - 1 ) in lung cancer. Br.J.Cancer 73:148-153. 178. Cromme, F . V . , P.F.J , van Bommel, J . M . M . Walboomers, M . P . W . Gallee, P . L . Stern, P. Kenemans, T . J . M . Helmerhorst, M . J . Stukart and C J . L . M . Mei jer . 1994. Differences in M H C and T A P - 1 expression in cervical cancer lymph node metastases as compared with the primary tumours. Br.J.Cancer 69:1176-1181. 179. Kaklamanis , L . , A . Townsend, L A . Doussis-Anagnostopoulou, N . Mor tensen , A . L . Har r i s and K . C . Gatter. 1994. Loss o f major his tocompatibi l i ty complex-encoded transporter associated wi th antigen presentation (TAP) in colorectal cancer. Am.J.Pathol. 145:505-509. 180. Restifo, N . P . , P.J . Spiess, S.E. Karp, J.J. Mule and S .A . Rosenberg. 1992. A nonimmunogenic sarcoma transdiuced wi th the c D N A for interferon-y elicits C D 8 + T cells against the wild-type tumor: correlation with antigen presentation capability. J .Exp.Med. 175:1423-1431. 181. Houghton, A . N . , T . M . Thomson, D . Gross, H . F . Oettgen and L . J . Old . 1984. Surface antigens of melanoma and melanocytes: specif ici ty of induction of la antigens by human y-interferon. J.Exp.Med. 160:255-269. 182. Ljunggren, H . G . , N . J . Stam, O. Ohlen, J.J. Neefjes, P. Hoglund, M . T . Heemels, J. Bastin, T . N . Schumacher, A . Townsend, K . Karre and H . L . Pleogh. 1990. Empty M H C Class I molecules come out in the cold. Nature 154 3 4 6 : 4 7 6 - 4 8 0 . 183. Raposa, G . , H . M . van Santen, R. Leijendekker, H . J . Geuze and H . L . P leogh . 1995. M i s f o l d e d major h is tocompat ib i l i ty complex Class I m o l e c u l e s accumula te in an expanded E R - G o l g i in te rmedia te compartment. J. Cell.Biol. 131:1403-1419. 184. Christinck, E .R . , M . A . Luscher, B . H . Barber and D . B . Wi l l i ams . 1991. Peptide binding to Class I M H C on l iv ing cells and quantitation of complexes required for C T L lysis. Nature 352:67-70. 185. Franksson, L . , M . Petersson, R. Kiess l ing and K . Karre . 1993. Immunizat ion against tumor and minor his tocompatibi l i ty antigens by eluted cel lular peptides loaded on antigen processing defective cel ls . Eur. J.Immunol. 23:2606-2613. 186. Correa, I. and D . H . Raulet. 1995. Binding of diverse peptides to M H C CLass I molecules inhibits target cel l lysis by activated natural k i l l e r cells. Immunity 2:61-71. 187. Wright, K . L . , L . C . White, A . Ke l ly , S. Beck, J. Trowsdale and J . P . - Y . T ing . 1995. Coordinate regulation of the human T A P 1 and L M P 2 genes from a shared bidirectional promoter. J.Exp.Med. 181:1459-1471. 188. B i x , M . and D . Raulet. 1992. Functionally conformed free Class I heavy chains exist on the surface of P2 mic rog lobu l in negative ce l l s . J.Exp.Med. 176:829-838. 189. Cromme, F . V . , J. Airey, M . - T . Heemels, H . L . Ploegh, P.J . Keating, P . L . Stern, C . J . L . M . Meijer and J . M . M . Walboomers. 1994. Loss of transporter protein, encoded by the T A P - 1 gene, is highly correlated wi th loss of H L A expression in cervical carcinomas. J.Exp.Med. 179:335-340. 190. Beck, S., A . K e l l y , E . Radley, F . Khurshid, R .P . Aldreton and J. Trowsdale. 1992. D N A sequence analysis of 66 kb of the human M H C Class II region encoding a cluster of genes for antigen processing. J.Mol.Biol. 228:433-441. 191. David-Watine, B . , A . Israel and P. Kour i l sky . 1990. The regulation and expression of M H C class I genes. Immunol.Today 11:286-292. 192. Dennert, G . , C . Landon, E . M . Lord , D . W . Bahler and J .G . Frelinger. 1988. Lys i s of a lung carcinoma by poly EC-induced natural k i l le r cells 155 is independent of the expression of Class I histocompatibil i ty antigens. J.Immunol. 140:2472-2475. 193. Nishimura, M . L , I. Stroynowski, L . Hood and S. Ostrand-Rosenberg. 1988. H - 2 K b antigen expression has no effect on natural k i l l e r susceptibil i ty and tumourigenicity of a murine hepatoma. J.Immunol. 1 4 1 : 4 4 0 3 - 4 4 0 9 . 194. Lobigs , M . and A . Mullbacher . 1995. Recombinant expression of vaccinia virus-encoded T A P 1 and T A P 2 promotes M H C Class I-restricted antigen presentation in a Syrian hamster cel l l ine. Immunol.Cell Biol. 7 3 : 1 8 1 - 1 8 4 . 195. Esquivel , F. , J. Yewde l l and J. Bennink. 1992. R M A / S cells present endogenously synthesized cy toso l i c proteins to Class I-restr icted cytotoxic T lymphocytes. J.Exp.Med. 175:163-168. 196. Hosken, N . A . and M . J . Bevan . 1992. A n endogenous antigenic peptide bypasses the Class I antigen presentation defect in R M A - S . J.Exp.Med. 175:719-729. 197. Day, P . M . , F. Esquivel , J. Kukszo , J .R. Bennink and J .W. Y e w d e l l . 1995. Effect of T A P on the generation and intracellular trafficking of peptide-receptive major histocompatibi l i ty complex Class I molecules. Immunity 2:137-147. 198. Nair , S .K. , D . Snyder and E . Gi lboa . 1996. C e l l treated with T A P - 2 antisense oligonucleotides are potent antigen-presenting cells in vitro and in vivo. J.Immunol. 156:1772-1780. 199. van Santen, H . M . , A . Woolsey, P . G . Ashton-Rickardt, L . Van Kaer, E . J . Baas, A . Berns, S. Tonegawa and H . L . Ploegh. 1995. Increase in positive se lec t ion of C D 8 + T cells in T A P l - m u t a n t mice by human P 2 -microglobul in transgene. J.Exp.Med. 181:787-792. 200. Ohlen, C , J. Bastin, H . - G . Ljunggren, L . Foster, E . Wolpert, G . Kle in , A . M . Townsend and K . Karre. 1990. Resistance to H-2-restricted but not to a l lo -H-2-spec i f ic graft and cytotoxic T lymphocyte responses in lymphoma mutant. J.Immunol. 145:52-58. 201. Suh, W . - K . , M . F . Cohen-Doyle, K . Fruh, K . Wang, P . A . Peterson and D . B . W i l l i a m s . 1994. Interaction of M H C Class I molecules with the transporter associated with antigen processing. Science 264:1322-1326. 156 202. Ortmann, B . , M . J . Androlewicz and P. Cresswell . 1994. M H C Class I / p 2 - m i c r o g l o b u l i n complexes associate with T A P transporters before peptide binding. Nature 368:864-867. 203. Grandea III, A . G . , M . J . Androlewicz, R.S . AthwaL D . E . Geraghty and T. Spies. 1995. Dependence of peptide binding by M H C Class I molecules on their interaction with T A P . Science 270:105-108. 157 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0088072/manifest

Comment

Related Items