MHC Proteins and Antigen Processing ImmunolRes 1992;11:125-132
John J. Monaco Departmentof Microbiologyand Immunology,MedicalCollegeof Virginia, VirginiaCommonwealth University,Richmond,Va., USA
Major Histocompatibility Complex-LinkedTransport Proteins and Antigen Processing
*oeoaoo,oo~174176176176176
oooloootot*oJo*loooloooQooa.oo~t.oo.oootooooo.ooo~
Key Words Tap-1 Tap-2
Abstract
Antigen processing Peptide transport ATP-binding cassette Endoptasmic reticulum
The major histocompatibility complexes of mice, rats and humans each contain a pair of related genes, Tap-I and Tap-2, that encode members of a large superfamily of proteins having similar structure and function. The TAP-1 (previously called HAM1 in the mouse) and TAP-2 (HAM2) proteins each contain 6-8 predicted membrane-spanning a helices, and a cytoplasmic domain containing a putative ATP-binding site. Recent evidence suggests that a functional TAP-I/TAP-2 heterodimer is required for efficient presentation of antigens to CD8+ cytotoxic T cells. This heterodimer resides in the membrane of the endoplasmic reticulum (ER), and probably functions to transport peptides (produced in the cytoplasm) into the ER lumen for binding to MHC class I molecules. ,oo****~
T cells in general are incapable of recognizing and responding to foreign antigens in their native form. Instead, the T-cell receptor is engaged only by a complex composed of a 'processed' fragment of the foreign antigen bound to a self major histocompatibility molecule. Although this recognition takes place at the surface of the antigen-presenting cell (APC), the fragmentation of foreign antigen and subsequent binding to major histocompatibility complex (MHC) molecules generally occurs intracellularly. Thus, 'antigen pro-
cessing' is defined as those events required for the conversion of native antigen into the relevant peptide fragments, and all the subsequent binding and transport events required for the processed antigen/MHC complex to appear at the cell surface. Although the requirement for antigen processing in MHC class-II-restricted responses has been known for some time [ 1], it has only relatively recently been appreciated that antigen processing is also required for MHC classI-restricted (cytolytic) T-cell responses [2], It
John J. Monaco, PhD Department of Microbiology and Immunology Medical College of Virginia, Virginia Commonwealth University, Box 678 MCV Station, Richmond, VA 23298-0678 (USA)
9 S. Karger AG, Basd 0257-277X/92/ 0112-012552.75/0
4-
Fig. 1. The two antigen-processing pathways. In the endogenous pathway (left), the antigen derived from the cytoplasm is transported into the lumen of the endoplasmic reticulum (ER), where it associates with newly synthesized MHC class-I molecules. The class-I complex is then routed through the Golgi complex (G) and out to the plasma membrane (PM) via normal vesicular transport. In the exogenous pathway (right), MHC classII molecules are synthesized in the ER and travel to the Golgi via vesicular transport. MHC class-II molecules are probably unable to bind peptides in the ER or Golgi due to the presence of invariant chain (not shown). En route to the cell surface, the class-II molecules are routed through an endosomal compartment, where invariant chain dissociates and peptides derived from endocytosed proteins are available for binding.
is now clear that the antigen-processing pathways (and, hence, the source of antigen) for class-I- and class-II-restricted responses are different [3-5]. Figure 1 illustrates these two pathways. Most (or all) antigens destined for the class-I pathway originate in the cytoplasm, having been produced within the cell, e.g., viral antigen or antigen derived from other intracellular parasites. Thus, the class-I pathway of antigen processing is often called the 'endogenous' pathway. In the class-II (exogenous) pathway, antigen taken up by endocytosis from the exterior of the cell is presumed to remain in the lumen of cytoplasmic
126
Monaco
vesicles, where it is degraded. Vesicular transantigenic peptides and MHC class-II molecules. Endogenously produced antigens that lack classical signal sequences [6, 7] for transport into the endoplasmic reticulum (ER) are nonetheless efficiently presented to class-Irestricted T cells [2]. Such antigens (or peptide fragments derived from them) must therefore be transported across the ER membrane via another mechanism. Recent evidence suggests that MHC-encoded products mediate this transport process.
Putative Peptide Transporters and Antigen Processing
H-2
,
,~
ItAM1
K
Pb
':~
'11 .......
A
It
mtp]
mtp2
DPB
RING4 }'3 PSF1
RT1
I-ILA
t~"5'
[ ] = Class-I genes
HAM2
Ob
Ab
Aa
Eb
Eb2
Ea
Bfl2
Bfl
B~
,Off
Dfl2
D~
I
I
I!'
RINGll Y1 PSF2
= Class-II genes
DOB
DQB DQA DRB1 DRB3 DRA
~=Transponergenes
Fig. 2. Genetic maps of the proximal portion of the mouse (H-2), rat (RT1), and human (HLA) MHCs. Class-I, class-II, and transporter genes from each species are depicted. Each gene shown is aligned vertically with its closest interspecies homologue. The gene order indicated is correct, but not all known genes are shown, and the maps are not drawn to scale.
Class-I-Deficient Mutant Cell Lines
Several cell lines have been described [812] that exhibit defective expression of class-I molecules at the cell surface. However, the class-I-structural genes (and the 132-microglobulin gene) in these cells are wild-type in sequence and are transcribed and translated normally. Moreover, class-I surface expression can be rescued by incubating the ceils at reduced temperature, or by incubation with peptides that bind to the appropriate class-Iallelic form [13-15]. Peptides that do not bind to the class-I molecules are without effect. The defect in these cells is apparently at the level of providing the class-I molecules with appropriate peptides. In the absence of peptide, the class-I molecule is unstable and is rapidly lost from the cell surface. Reducing the temperature, or adding exogenous peptides that can bind to these 'empty' class-I molecules results in their stabilization and increases their residence time, and hence total level, at the cell surface. This model would be consistent with either a defect in the ability to generate the appropriate peptides from native antigen, or to transport such peptides into the lumen of the ER where they can bind to class-I molecules.
Several of the class-I-deficient mutants were selected by treatment with anti-MHC class-II antibodies, and are known to carry deletions in the class-II region of the MHC [ 10]. Mapping of the deletional endpoints in these mutants, and in other class-II deletion mutants that do not manifest the class-I-deficient phenotype [16], narrowed the potential position of the defective gene(s) to the region between HLA-DP and HLA-DO, equivalent to the region between tab and Ob in the murine MHC.
MHC-Encoded Transport Proteins
Four groups [16-19] simultaneously reported the cloning of genes in this region that encode members of a family of adenosine triphosphate (ATP)-dependent transport proteins. Figure 2 shows the location, transcriptional orientation, and nomenclature used for these genes, and the published sequences from mouse, rat and human are aligned in figure 3. It is clear from these data that genes in equivalent chromosomal positions in the three species are also most closely related in sequence. Table 1 shows pairwise compari-
127
I
HAMI m~pl ~ING4 PSF2
HAMI mtpl RING4 PSF2
MLPGIFSLLVPE.VPLLRVWVVGLSRWAILGLGVRGVLGVT MAAHAWPTAALLLLLVDWLLLRPV
...........
MA$SRCPAPRGCRCLPGASLAWLGTVLLBLADWVLLRTA--R
. .......
....... AGAHGWLAALQPL
A ....................
9 ......
.........
....... TAL ...... A ....... V-W--AC---RA-VGSKSEN---Q MRLPDLRPWTS-LLVDAA--WLLQGP-GTLLPQ--PGLWLE-TLRLGGLWGLLK-LRG-LGFVGT
......
K--
vAALSLALPGLALFRKLAAWGTLREGDSAGL•YWNSRPDAFAISYVAALPAAALWHKFGSLWApSGNRDAGDMLCRMLGFLG•KKRRLYL•LVLLILSCL ....
G .......
S ....
S---AL
....
N ....
n ....
L---VL
..............
L-GF
.....
HKG ............
DS--G--H
...........
A----G .......... E-IS--APGSA--TR--H-G-H-T--W--A ........... L .... ,-,-,,,-,,,-,--'--'''---'-'---'--'LLLPLCLAT•-TVSLRALVA-AS-APPARVASAPW-WLLVGYGAAGLSWSLW-VLSPP-AQEKEQDQVNNKVLMW-L-KLSR-DLPL-VAAFFF-V-AV-
HAMI mtpl RING4 PSF2
GEMA~PF~TGR~TDWILQDKTV~SFTRNIWLMSILT~ASTALEFA$DGIYNITMGHM~GRV~RE~FRAVLRQETGFFL~NpAGSITSRVTEDTAN~CESI . . . . . . . . . . . . . . . . . . . . . . . . . .A. . . . .A . . .M. . . . .C. . . . . . . . . .V . . . . . .G . . . . . . . . . . . . . . .S. . . . .G. . . . . . . . . . H.... T
HAMI mtpl RING4 PSF2
SD~LSLLLWYLGRALCLL~FMFWGSPYLTLVTLINLPLLFLLPKKLGKVHQSLAVKVQE~LAKSTQVALEALSAMPTVRSFANEEGEAQKFRQKLEEMKT
HAM1 mtpl RING4 PSF2
LNKKEALAY•AEVWTTSV•GMLLKVGILYLGGQLV•RGTVSSGNLV•FVL•QLQFTQAVQVLLSLYPSMQKAVGS•EKIFEYLDRTPCSPLSGSLAPSNM
HAM1 mtpl RING4 .~-2
KGL~EFQDVSFAYPNQPKVQVLQGLTFTLHPGTVTALV~NGSGKSTVAALL~NLYQPTGGQLLLDGQRLVQYDHHYLHTQ~AA~GQEPLLFGRSFRENI
psr2
Q-v-~ . . . . . . . . . .
HAM1 mtpl RING4
AYGLNRTPTMEEITA~AvESGAHDFISGFPQGYDTE~GETGNQLSGGQRQAV~IRKPLLLILDD~TSALDAGNQLRVQRLLYESPKRASRTVLL
HAMJ mtpl RING4
---R
...........
L .......
GSADT
....
LT .........
AV---VG
.....
N ....
V-SHLQG---G
.......
S
E--QQ-QT-N-M
.......
STLSD-L
--TL--HYS--vI-ILGG-FDPHA-ASA-FF-CLFSFG-SLSAGCRG-cFT~--SRINL-I~EQL-SSL---DL---QETKT-ELN--LSS--TLMSNWL
--K-N-F
......
G ....
A--I---F---V---LS
........
RR ....
Y ..............
-EN---F .... V-G .... GI--L-----VS---M .... T .......... V--WY-L-E-Q-R PLNANV--RS-VKVVG-YG--LSI--R---LS-LHM-FTIAAE-VYNTR--E~rLREI-DAV-RAG--VR--VGGLQ
..........
T ....
--Q---V--AVNS
M ...................
....
~ ...........
I .....
V--A
.................
TS-A
.......
T .....
T ...........................
......
R--E M ......
....
S---I
I ......
E--S-I--RV
Q ......
P
....................... E--Q-I-..... GA--H-VCRYKEA--QCRQ
S--A
......................
..................
L--
RC-P--L-T-LHL
-YWRRD-ERALYLLIRR-LHLGVQMLM-SC-L-QMQD-ELTQ-S-L--MI--ESVGSY--T-VYI-GD-LSN--AA--V-S-M--Q-NL-SP-T---TTL
.... E---Q
K .......... ..........
.....
H-N
...........
R-D-L
~--~-E~g-
.........
Y--K R--E
...........
............................
KV ....
EP ...............................
..................................
KP-P--E-R---R
T ..................................
R-D~--K . . . . . . a--r . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
.... T ............ M ............................................ .... TQK ........ A--K .... S .... L ........ D-A-S ....................
.........
EP-~E ....... a--~
QV ....
LQ---
..... ~ - ~ - -
V---EKPIS--~-C---S--~ . . . . - ~ - ~ - ~ - -
R ...........................
EW .......
::::::::::::::::::::::::::::::::::::::::::::::::: CV ............
NS--Q-EQ
......
E-Y--S---
ITQQLSLAEQAHHILFLREGSVGEQGTHLQLMKRGGCYRAMVEALAAPAD
......... ---H---V---D
R ....... K .... C ......... .....
E ...... S ........
S-
EG-AIR-G---Q---EKK---W---Q-P-DAPE
PSF2
Fig. 3. Alignment of predicted amino acid sequences of MHC-linked transport proteins. Amino acids are shown using a single letter code. Gaps introduced to maximize sequence alignment are depicted with dots. Dashes indicate amino acid identity to the HAM1 sequence. For the HAM1 sequence, putative transmembrane spanning segments are denoted by black bars above the sequence, and ATP-binding too-
tifs are boxed. Residues shared by HAM2 and PSF2, but not present in members of the HAMI family are enclosed in shaded boxes. The origin of the sequences shown is indicated in table [. The NH2-terminal 73 amino acids of the HAM t sequence (in italics) were derived from genomic DNA sequencing, and are assumed to be present based on homology to the other cDNA sequences.
sons of the level of amino acid identity and similarity. The nomenclature for the mouse genes (HAM1 and HAM2) wilt be used to refer to the homologous genes in all three species for the sake of clarity in the remainder of the discussion. Amino acid identity of the HAM1 protein from the different species ranges form 72 to 89 %. The identity of HAM1 and HAM2 proteins is 38-40%. The most conserved parts of the sequences of HAM1
and HAM2 are in the cytoplasmic domains near the two ATP-binding motifs. However, the two molecules are significantly related throughout the cytoplasmic domain, with the exception of the carboxy-terminal 64 amino acids, which diverge completely in sequence. As is the case for most other members of this transporter superfamily, the HAM1 and HAM2 transmembrane domains are only marginally related in terms of sequence iden-
128
Monaco
Putative Peptide Transporters and Antigen Processing
Table 1. Sequence similarities among MHC-linked transport proteins ~
HAMI mtpl RING4 HAM2 PSF2
Reference
HAMI
mtpl
RING4
HAM2
PSF2
17 18 19 _2 25
89 73 38 39
94 72 38 40
83 83 39 40
60 60 62 75
61 63 63 86 -
1 Percent amino acid similarity (above diagonal) and identity (below diagonal) as determined by the computer program BESTFIT [31 ]. 2 Attaya M, Monaco J J: unpublished data; partial sequence published by Monaco et al. [ 17].
tity (although the hydrophobicity of the transmembrane segments is conserved). Sequence comparisons between the HAM1 and HAM2 genes and other members of the transporter family have been published elsewhere [ 17].
Transport Protein Structure The proteins in this family are characterized by a four-domain structure, two homologous domains each containing from 6 to 8 potential transmembrane spanning ct helices, and two 'cytoplasmic' domains each containing a pair of sequence motifs which form a putative ATP-binding domain. Interestingly, these four domains of the functional molecule may be encoded by anywhere from 1 to 4 different genes, depending on the specific transport system involved [20]. For example, the mammalian multidrug resistance (P-glycoprotein) and cystic fibrosis transmembrane regulator contain all four domains on a single polypeptide chain, whereas the Salmonella typhimurium oligopeptide permease has each of the four domains on separate polypeptides. The Salmonella histidine transporter also contains four separate polypeptides, but the two cytoplasmic domains are identical to one another (products of the hisP gene); the two
transmembrane domains are nonidentical (hisQ and hisM). Intermediate between these two extremes are the Escherichia coli hlyB and Bordetella pertussis cyaB gene products, which each contain one transmembrane and one cytoplasmic domain (and which presumably function as homodimers), and the E. coli ribose transporter, where one gene (rbsA) encodes the two (nonidentical) cytoplasmic domains on a single polypeptide, and two genes each code for one of the transmembrane domains. Given the above information, it is possible that the HAM1 and HAM2 gene products function as two separate homodimers, or as a single heterodimer, or even as two separate heterodimers, each paired with the product of an unknown gene or genes. Given that a defect in either HAM1 or HAM2 alone generates a class-I-deficient phenotype as severe as that found when both genes are deleted (see below), the most likely model is that of a HAM1/HAA42 heterodimer. However, the fact that the class-I mutants apparently are capable of presenting some class-I-restricted antigens to T cells [M. Bevan, personal commun.], raises the possibility that all three forms exist, with the heterodimer responsible for the majority of transport, and one or both of the homodimers functioning indepen-
129
dently to transport a restricted subset of antigens (peptides). Isolation of the corresponding proteins will be required to resolve this issue.
Transporter Specificity An interesting unresolved issue is the specificity of the presumed transport process. MHC class-I molecules prefer to bind peptides of defined length and sequence [21-24]. Did the linkage of the transporter genes to the MHC evolve in order to facilitate co-evolution of this specificity? An implication of coevolution of class-I sequence specificity and transporter specificity is that the transporter should be polymorphic, with different alleles in linkage disequilibrium with particular MHC class-I alleles. Although a thorough study of polymorphism of the HAM genes has not been done, restriction fragment length polymorphisms around these genes are relatively rare (unpublished observations), and the amino acid sequence of the HAM1 gene derived from two MHC mismatched cell lines is 100% identical [25]. These observations argue against the possibility of peptide sequence specificity for the transporter. However, this structure may still have specificity for peptide length, since all allelic forms of MHC class-I antigens examined to date exhibit similar preferences (8-10 amino acids) for peptide length.
Evidence for a Transport Defect in Class-I-Deficient Mutants
vious deletion corrects the class-I defect [26]. However, the same cloned gene does not correct the defect in a mutant with an approximately 1,000-kb deletion (containing both the transporter genes). These data suggest that two (or more) genes in this region are required in the class-I antigen-processing pathway. Consistent with this idea is that the HAM1 gene does not correct the defect in the RMA-S cell line (unpublished observations). An obvious candidate for the second gene is HAM2, preliminary evidence suggests that the defect in RMA-S may involve the HAM2 gene, but to date we have achieved only partial restoration of the defect in these cells with cloned HAM2 cDNA. Evidence for a transport defect in the classI-processing defective T2 cell line has been obtained using a minigene encoding a 12 amino acid peptide derived from influenza virus matrix protein. Transfection of this minigene into either parental (T1) or mutant (T2) cells results in presentation of the peptide to class-I-restricted T cells when the peptide sequence is preceded by a classical signal peptide [27]. This indicates that the flu peptide can be presented by either cell if introduced into the lumen of the ER via the normal secretory pathway. Transfection of the same influenza minigene lacking the signal sequence into T1 (wild-type) cells results in presentation of the peptide to class-I-restricted T cells, indicating that peptides produced in the cytoplasm can indeed be transported into the ER in wild-type cells. However, introduction of the same gene into T2 mutant cells does not result in presentation ot the peptide, indicating that T2 cells are defective in this transport function.
Direct evidence for the involvement of the HAM1, and possibly also the HAM2, genes in the class-I pathway has recently been obtained. Introduction of a cloned copy of HAM1 cDNA into a mutant carrying no ob-
130
Monaco
Putative Peptide Transporters and Antigen Processing
A Model for the Class-I Antigen-Processing Pathway In summary, the HAM1 and HAM2 genes probably encode the two subunits of a heterodimer that resides in the ER membrane and transports peptides from the cytoplasm into the lumen of the ER, where they are bound by newly synthesized class-I molecules. (Any unbound peptides are probably degraded in the ER.) Although the source of the peptides is unknown, recent evidence suggests
q g t g q
ooo
ool~
9 1 7 69
9 ~149o~
~ 1 7 6 1 7 ~6 1 7 6 1~ 7 1 6 71 67 16 7~ 61 17 76 61 7 6 1 o7o Q6 9
G9
9
that genes in the MHC may also be involved in the generation of such peptides [28, 29]. At least two of the subunits of a large cytoplasmic protease (called the LMP complex) are encoded by genes positioned immediately upstream (centromeric) of the HAM1 and HAM2 genes. These two structures alone (the transporter heterodimer and the LMP complex) may perform all of the immune-specific functions involved in antigen processing for class-I-restricted T-cell responses [30].
9 1 4 9 1 4 9 o o1 9 4 9 o 19 4 ~9 1 7 6
9 1 4 9 9 1 4 9I 9
9 1 4 9 1 4 o9 Q 9
. 4
9
9 1 7 6 ~176a
9 1 4 9o o
9 1 4 9 9 to
9 il
9 olo
o*
o o o o o o o
References 1 Ziegler K, Unanue ER: Identification of a macrophage antigen-processing event required for I-regionrestricted antigen presentation to T Iymphocytes. J Immunol 1981 :t 27: 1869-1875. 2 Townsend A, Bodmer H: Antigen recognition by class I restricted T lymphocytes. Annu Rev lmmunol 1989;7:601-624. 3 Brae9 T J, Braciale VL: Antigen presentation: Structural themes and functional variations. Immunol Today 1991;12:124-129. 4 Yewdell JW, Bennick JR: The binary logic of antigen processing and presentation to T cells. Cell 1990;62: 203-206. 5 Germain RN: The ins and outs of antigen processing and presentation. Nature 1986;322:687-689. 6 Gierasch LM: Signal sequences. Biochemistry 1989;28:923-930. 7 von Heijne G: Signal sequences: The limits of variation. J Mol Biol 1985; 184:99-105, 8 Kg.rre K, Ljunggren HG, Piontek G, Kiessling R: Selective rejection of H-2-deficient lymphoma variants suggests alternative immune defence strategy. Nature 1986;319: 675-678.
9 Townsend A, Ohldn C, Bastin J, Ljunggren HG, Foster L, K~irre K: Association of class I major histocompatbility heavy and light chains induced by viral peptides. Nature 1989;340:443-448. 10 DeMars R, Rudersdorf R, Chang C, Petersen J, Strandtmann J, Kom N, Sidwell B, Orr HT: Mutations that impair a posttranscriptional step in expression of HLA-A and -B antigens. Proc Natl Acad Sci USA 1985; 82:8183-8187. 11 Cerundolo V, Alexander J, Anderson K, Lamb C, Cresswell P, McMichael A, Gotch F, Townsend A: Presentation of viral antigen controlled by a gene in the major histocompatibility complex. Nature 1990;345: 449-452. 12 Hosken NA, Bevan MJ: Defective presentation of endogenous antigen by a cell line expressing class I molecules. Science 1990;248:367-370. 13 Townsend A, Elliot T, Cerundolo V, Foster L, Barber B, Tse A: Assembly of MHC class I molecules analyzed in vitro. Cell 1990;62:285-295. 14 Ljunggren HG, Stam NJ, (~hl6n C, Neefjes J J, H6glund P, Heemels MT, Bastin J, Schumacher TNM, Townsend A, K~irre K, Ploegh HL: Empty MHC class I molecules come out in the cold. Nature 1990;346: 476-480.
15 Schumacher TNM, Heemels MT, Neefjes J J, Kast WM, Melief CJM, Ploegh HL: Direct binding of peptide to empty MHC class I molecules on intact cells and in vitro. Cell 1990;62:563-567. I6 Spies T, Bresnahan M, Bahrain S, Arnold D, Blanck G, Mellins E, Pious D, DeMars R: A gene in the human major histocompatibility complex class II region controlling the class I antigen processing pathway. Nature 1990;348:744-747. 17 Monaco J J, Cho S, Attaya M: Transport protein genes in the murine MHC: Possible implications for antigen processing. Science 1990;250: 1723-1726. 18 Deverson EV, Gow IR, Coadwell WJ, Monaco J J, Butcher GW, Howard JC: MHC class II region encoding proteins related to the multidrug resistance family oftransmembrane transporters. Nature 1990;348:738-741. 19 Trowsdale J, Hanson I, Mockridge I, Beck S, Townsend A, Kelly A: Sequences encoded in the class II region of the MHC related to the 'ABC' superfamily of transporters. Nature 1990;348:741-744.
131
20 Hyde SC, Emsley P, Hartshorn M J, Mimmack MM, Gileadi U, Pearse SR, Gallagher MP, Gill DR, Hubbard RE, Higgins CF: Structural model of ATP-binding proteins associated with cystic fibrosis, multidrug resistance and bacterial transport. Nature 1990;346:362-365. 2i Falk K, R6tzschke O, Rammensee HG: Cellular peptide composition governed by major histocompatibility complex class 1 molecules. Nature 1990;348:248-251. 22 R6tschke O, Falk K, Deres K, Sehild H, Norda M, Metzger J, Jung G, Rammensee HG: Isolation and analysis of naturally processed viral peptides as recognized by cytotoxic T cells. Nature 1990;348:252-254. 23 Van Bleek GM, Nathenson SG: Isolation of an endogenously processed immunodominant viral peptide from the class I H-2Kb molecule. Nature 1990;348:213-216.
24 Elliott T, Cerundolo V, Elvin J, Townsend A: Peptide-induced conformational change of the class I heavy chain. Nature 1991 ;351:402406. 25 Bahram S, Arnold D, Bresnahan M, Strominger JL, Spies T: A second peptide transporter gene in the human MHC class II region. Proc NatI Acad Sci USA 1991, in press. 26 Spies T, DeMars R: Restored expression of major histocompatibility class I molecules by gene transfer of a putative peptide transporter. Nature 1991;351:323-324. 27 Anderson K, Cresswell P, Gammon M, Hermes J, WiUiamson A, Zweerink H: Endogenously synthesized peptide with an endoplasmic reticulum signal sequence sensitizes antigen processing mutant cells to class I-restricted cell-mediated lysis. J Exp Med 1991;174:489-492.
28 Brown MG, DriscoU J, Monaco J J: Structural and serological similarity of the MHC-linked LMP and proteasome (multicatalytic proteinase) complexes. Nature 1991;353:355357. 29 Martinez CK, Monaco I J: A major histocompatibility complex-linked LMP gene: HomoIogy to proteasome subunits. Nature 199t;353: 664-667. 30 Monaco J J: A model of antigen processing for MHC class I-restricted responses. Immunol Today 1992; 13:173-179 31 Devereux J, Haeberli P, Smithies O: A comprehensive set of sequence analysis programs for the VAX. Nucleic Acids Res 1984; 12:387-395.
132
Monaco
Putative Peptide Transporters and Antigen Processing
John J. Monaco Departmentof Microbiologyand Immunology,MedicalCollegeof Virginia, VirginiaCommonwealth University,Richmond,Va., USA
Major Histocompatibility Complex-LinkedTransport Proteins and Antigen Processing
*oeoaoo,oo~174176176176176
oooloootot*oJo*loooloooQooa.oo~t.oo.oootooooo.ooo~
Key Words Tap-1 Tap-2
Abstract
Antigen processing Peptide transport ATP-binding cassette Endoptasmic reticulum
The major histocompatibility complexes of mice, rats and humans each contain a pair of related genes, Tap-I and Tap-2, that encode members of a large superfamily of proteins having similar structure and function. The TAP-1 (previously called HAM1 in the mouse) and TAP-2 (HAM2) proteins each contain 6-8 predicted membrane-spanning a helices, and a cytoplasmic domain containing a putative ATP-binding site. Recent evidence suggests that a functional TAP-I/TAP-2 heterodimer is required for efficient presentation of antigens to CD8+ cytotoxic T cells. This heterodimer resides in the membrane of the endoplasmic reticulum (ER), and probably functions to transport peptides (produced in the cytoplasm) into the ER lumen for binding to MHC class I molecules. ,oo****~
T cells in general are incapable of recognizing and responding to foreign antigens in their native form. Instead, the T-cell receptor is engaged only by a complex composed of a 'processed' fragment of the foreign antigen bound to a self major histocompatibility molecule. Although this recognition takes place at the surface of the antigen-presenting cell (APC), the fragmentation of foreign antigen and subsequent binding to major histocompatibility complex (MHC) molecules generally occurs intracellularly. Thus, 'antigen pro-
cessing' is defined as those events required for the conversion of native antigen into the relevant peptide fragments, and all the subsequent binding and transport events required for the processed antigen/MHC complex to appear at the cell surface. Although the requirement for antigen processing in MHC class-II-restricted responses has been known for some time [ 1], it has only relatively recently been appreciated that antigen processing is also required for MHC classI-restricted (cytolytic) T-cell responses [2], It
John J. Monaco, PhD Department of Microbiology and Immunology Medical College of Virginia, Virginia Commonwealth University, Box 678 MCV Station, Richmond, VA 23298-0678 (USA)
9 S. Karger AG, Basd 0257-277X/92/ 0112-012552.75/0
4-
Fig. 1. The two antigen-processing pathways. In the endogenous pathway (left), the antigen derived from the cytoplasm is transported into the lumen of the endoplasmic reticulum (ER), where it associates with newly synthesized MHC class-I molecules. The class-I complex is then routed through the Golgi complex (G) and out to the plasma membrane (PM) via normal vesicular transport. In the exogenous pathway (right), MHC classII molecules are synthesized in the ER and travel to the Golgi via vesicular transport. MHC class-II molecules are probably unable to bind peptides in the ER or Golgi due to the presence of invariant chain (not shown). En route to the cell surface, the class-II molecules are routed through an endosomal compartment, where invariant chain dissociates and peptides derived from endocytosed proteins are available for binding.
is now clear that the antigen-processing pathways (and, hence, the source of antigen) for class-I- and class-II-restricted responses are different [3-5]. Figure 1 illustrates these two pathways. Most (or all) antigens destined for the class-I pathway originate in the cytoplasm, having been produced within the cell, e.g., viral antigen or antigen derived from other intracellular parasites. Thus, the class-I pathway of antigen processing is often called the 'endogenous' pathway. In the class-II (exogenous) pathway, antigen taken up by endocytosis from the exterior of the cell is presumed to remain in the lumen of cytoplasmic
126
Monaco
vesicles, where it is degraded. Vesicular transantigenic peptides and MHC class-II molecules. Endogenously produced antigens that lack classical signal sequences [6, 7] for transport into the endoplasmic reticulum (ER) are nonetheless efficiently presented to class-Irestricted T cells [2]. Such antigens (or peptide fragments derived from them) must therefore be transported across the ER membrane via another mechanism. Recent evidence suggests that MHC-encoded products mediate this transport process.
Putative Peptide Transporters and Antigen Processing
H-2
,
,~
ItAM1
K
Pb
':~
'11 .......
A
It
mtp]
mtp2
DPB
RING4 }'3 PSF1
RT1
I-ILA
t~"5'
[ ] = Class-I genes
HAM2
Ob
Ab
Aa
Eb
Eb2
Ea
Bfl2
Bfl
B~
,Off
Dfl2
D~
I
I
I!'
RINGll Y1 PSF2
= Class-II genes
DOB
DQB DQA DRB1 DRB3 DRA
~=Transponergenes
Fig. 2. Genetic maps of the proximal portion of the mouse (H-2), rat (RT1), and human (HLA) MHCs. Class-I, class-II, and transporter genes from each species are depicted. Each gene shown is aligned vertically with its closest interspecies homologue. The gene order indicated is correct, but not all known genes are shown, and the maps are not drawn to scale.
Class-I-Deficient Mutant Cell Lines
Several cell lines have been described [812] that exhibit defective expression of class-I molecules at the cell surface. However, the class-I-structural genes (and the 132-microglobulin gene) in these cells are wild-type in sequence and are transcribed and translated normally. Moreover, class-I surface expression can be rescued by incubating the ceils at reduced temperature, or by incubation with peptides that bind to the appropriate class-Iallelic form [13-15]. Peptides that do not bind to the class-I molecules are without effect. The defect in these cells is apparently at the level of providing the class-I molecules with appropriate peptides. In the absence of peptide, the class-I molecule is unstable and is rapidly lost from the cell surface. Reducing the temperature, or adding exogenous peptides that can bind to these 'empty' class-I molecules results in their stabilization and increases their residence time, and hence total level, at the cell surface. This model would be consistent with either a defect in the ability to generate the appropriate peptides from native antigen, or to transport such peptides into the lumen of the ER where they can bind to class-I molecules.
Several of the class-I-deficient mutants were selected by treatment with anti-MHC class-II antibodies, and are known to carry deletions in the class-II region of the MHC [ 10]. Mapping of the deletional endpoints in these mutants, and in other class-II deletion mutants that do not manifest the class-I-deficient phenotype [16], narrowed the potential position of the defective gene(s) to the region between HLA-DP and HLA-DO, equivalent to the region between tab and Ob in the murine MHC.
MHC-Encoded Transport Proteins
Four groups [16-19] simultaneously reported the cloning of genes in this region that encode members of a family of adenosine triphosphate (ATP)-dependent transport proteins. Figure 2 shows the location, transcriptional orientation, and nomenclature used for these genes, and the published sequences from mouse, rat and human are aligned in figure 3. It is clear from these data that genes in equivalent chromosomal positions in the three species are also most closely related in sequence. Table 1 shows pairwise compari-
127
I
HAMI m~pl ~ING4 PSF2
HAMI mtpl RING4 PSF2
MLPGIFSLLVPE.VPLLRVWVVGLSRWAILGLGVRGVLGVT MAAHAWPTAALLLLLVDWLLLRPV
...........
MA$SRCPAPRGCRCLPGASLAWLGTVLLBLADWVLLRTA--R
. .......
....... AGAHGWLAALQPL
A ....................
9 ......
.........
....... TAL ...... A ....... V-W--AC---RA-VGSKSEN---Q MRLPDLRPWTS-LLVDAA--WLLQGP-GTLLPQ--PGLWLE-TLRLGGLWGLLK-LRG-LGFVGT
......
K--
vAALSLALPGLALFRKLAAWGTLREGDSAGL•YWNSRPDAFAISYVAALPAAALWHKFGSLWApSGNRDAGDMLCRMLGFLG•KKRRLYL•LVLLILSCL ....
G .......
S ....
S---AL
....
N ....
n ....
L---VL
..............
L-GF
.....
HKG ............
DS--G--H
...........
A----G .......... E-IS--APGSA--TR--H-G-H-T--W--A ........... L .... ,-,-,,,-,,,-,--'--'''---'-'---'--'LLLPLCLAT•-TVSLRALVA-AS-APPARVASAPW-WLLVGYGAAGLSWSLW-VLSPP-AQEKEQDQVNNKVLMW-L-KLSR-DLPL-VAAFFF-V-AV-
HAMI mtpl RING4 PSF2
GEMA~PF~TGR~TDWILQDKTV~SFTRNIWLMSILT~ASTALEFA$DGIYNITMGHM~GRV~RE~FRAVLRQETGFFL~NpAGSITSRVTEDTAN~CESI . . . . . . . . . . . . . . . . . . . . . . . . . .A. . . . .A . . .M. . . . .C. . . . . . . . . .V . . . . . .G . . . . . . . . . . . . . . .S. . . . .G. . . . . . . . . . H.... T
HAMI mtpl RING4 PSF2
SD~LSLLLWYLGRALCLL~FMFWGSPYLTLVTLINLPLLFLLPKKLGKVHQSLAVKVQE~LAKSTQVALEALSAMPTVRSFANEEGEAQKFRQKLEEMKT
HAM1 mtpl RING4 PSF2
LNKKEALAY•AEVWTTSV•GMLLKVGILYLGGQLV•RGTVSSGNLV•FVL•QLQFTQAVQVLLSLYPSMQKAVGS•EKIFEYLDRTPCSPLSGSLAPSNM
HAM1 mtpl RING4 .~-2
KGL~EFQDVSFAYPNQPKVQVLQGLTFTLHPGTVTALV~NGSGKSTVAALL~NLYQPTGGQLLLDGQRLVQYDHHYLHTQ~AA~GQEPLLFGRSFRENI
psr2
Q-v-~ . . . . . . . . . .
HAM1 mtpl RING4
AYGLNRTPTMEEITA~AvESGAHDFISGFPQGYDTE~GETGNQLSGGQRQAV~IRKPLLLILDD~TSALDAGNQLRVQRLLYESPKRASRTVLL
HAMJ mtpl RING4
---R
...........
L .......
GSADT
....
LT .........
AV---VG
.....
N ....
V-SHLQG---G
.......
S
E--QQ-QT-N-M
.......
STLSD-L
--TL--HYS--vI-ILGG-FDPHA-ASA-FF-CLFSFG-SLSAGCRG-cFT~--SRINL-I~EQL-SSL---DL---QETKT-ELN--LSS--TLMSNWL
--K-N-F
......
G ....
A--I---F---V---LS
........
RR ....
Y ..............
-EN---F .... V-G .... GI--L-----VS---M .... T .......... V--WY-L-E-Q-R PLNANV--RS-VKVVG-YG--LSI--R---LS-LHM-FTIAAE-VYNTR--E~rLREI-DAV-RAG--VR--VGGLQ
..........
T ....
--Q---V--AVNS
M ...................
....
~ ...........
I .....
V--A
.................
TS-A
.......
T .....
T ...........................
......
R--E M ......
....
S---I
I ......
E--S-I--RV
Q ......
P
....................... E--Q-I-..... GA--H-VCRYKEA--QCRQ
S--A
......................
..................
L--
RC-P--L-T-LHL
-YWRRD-ERALYLLIRR-LHLGVQMLM-SC-L-QMQD-ELTQ-S-L--MI--ESVGSY--T-VYI-GD-LSN--AA--V-S-M--Q-NL-SP-T---TTL
.... E---Q
K .......... ..........
.....
H-N
...........
R-D-L
~--~-E~g-
.........
Y--K R--E
...........
............................
KV ....
EP ...............................
..................................
KP-P--E-R---R
T ..................................
R-D~--K . . . . . . a--r . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
.... T ............ M ............................................ .... TQK ........ A--K .... S .... L ........ D-A-S ....................
.........
EP-~E ....... a--~
QV ....
LQ---
..... ~ - ~ - -
V---EKPIS--~-C---S--~ . . . . - ~ - ~ - ~ - -
R ...........................
EW .......
::::::::::::::::::::::::::::::::::::::::::::::::: CV ............
NS--Q-EQ
......
E-Y--S---
ITQQLSLAEQAHHILFLREGSVGEQGTHLQLMKRGGCYRAMVEALAAPAD
......... ---H---V---D
R ....... K .... C ......... .....
E ...... S ........
S-
EG-AIR-G---Q---EKK---W---Q-P-DAPE
PSF2
Fig. 3. Alignment of predicted amino acid sequences of MHC-linked transport proteins. Amino acids are shown using a single letter code. Gaps introduced to maximize sequence alignment are depicted with dots. Dashes indicate amino acid identity to the HAM1 sequence. For the HAM1 sequence, putative transmembrane spanning segments are denoted by black bars above the sequence, and ATP-binding too-
tifs are boxed. Residues shared by HAM2 and PSF2, but not present in members of the HAMI family are enclosed in shaded boxes. The origin of the sequences shown is indicated in table [. The NH2-terminal 73 amino acids of the HAM t sequence (in italics) were derived from genomic DNA sequencing, and are assumed to be present based on homology to the other cDNA sequences.
sons of the level of amino acid identity and similarity. The nomenclature for the mouse genes (HAM1 and HAM2) wilt be used to refer to the homologous genes in all three species for the sake of clarity in the remainder of the discussion. Amino acid identity of the HAM1 protein from the different species ranges form 72 to 89 %. The identity of HAM1 and HAM2 proteins is 38-40%. The most conserved parts of the sequences of HAM1
and HAM2 are in the cytoplasmic domains near the two ATP-binding motifs. However, the two molecules are significantly related throughout the cytoplasmic domain, with the exception of the carboxy-terminal 64 amino acids, which diverge completely in sequence. As is the case for most other members of this transporter superfamily, the HAM1 and HAM2 transmembrane domains are only marginally related in terms of sequence iden-
128
Monaco
Putative Peptide Transporters and Antigen Processing
Table 1. Sequence similarities among MHC-linked transport proteins ~
HAMI mtpl RING4 HAM2 PSF2
Reference
HAMI
mtpl
RING4
HAM2
PSF2
17 18 19 _2 25
89 73 38 39
94 72 38 40
83 83 39 40
60 60 62 75
61 63 63 86 -
1 Percent amino acid similarity (above diagonal) and identity (below diagonal) as determined by the computer program BESTFIT [31 ]. 2 Attaya M, Monaco J J: unpublished data; partial sequence published by Monaco et al. [ 17].
tity (although the hydrophobicity of the transmembrane segments is conserved). Sequence comparisons between the HAM1 and HAM2 genes and other members of the transporter family have been published elsewhere [ 17].
Transport Protein Structure The proteins in this family are characterized by a four-domain structure, two homologous domains each containing from 6 to 8 potential transmembrane spanning ct helices, and two 'cytoplasmic' domains each containing a pair of sequence motifs which form a putative ATP-binding domain. Interestingly, these four domains of the functional molecule may be encoded by anywhere from 1 to 4 different genes, depending on the specific transport system involved [20]. For example, the mammalian multidrug resistance (P-glycoprotein) and cystic fibrosis transmembrane regulator contain all four domains on a single polypeptide chain, whereas the Salmonella typhimurium oligopeptide permease has each of the four domains on separate polypeptides. The Salmonella histidine transporter also contains four separate polypeptides, but the two cytoplasmic domains are identical to one another (products of the hisP gene); the two
transmembrane domains are nonidentical (hisQ and hisM). Intermediate between these two extremes are the Escherichia coli hlyB and Bordetella pertussis cyaB gene products, which each contain one transmembrane and one cytoplasmic domain (and which presumably function as homodimers), and the E. coli ribose transporter, where one gene (rbsA) encodes the two (nonidentical) cytoplasmic domains on a single polypeptide, and two genes each code for one of the transmembrane domains. Given the above information, it is possible that the HAM1 and HAM2 gene products function as two separate homodimers, or as a single heterodimer, or even as two separate heterodimers, each paired with the product of an unknown gene or genes. Given that a defect in either HAM1 or HAM2 alone generates a class-I-deficient phenotype as severe as that found when both genes are deleted (see below), the most likely model is that of a HAM1/HAA42 heterodimer. However, the fact that the class-I mutants apparently are capable of presenting some class-I-restricted antigens to T cells [M. Bevan, personal commun.], raises the possibility that all three forms exist, with the heterodimer responsible for the majority of transport, and one or both of the homodimers functioning indepen-
129
dently to transport a restricted subset of antigens (peptides). Isolation of the corresponding proteins will be required to resolve this issue.
Transporter Specificity An interesting unresolved issue is the specificity of the presumed transport process. MHC class-I molecules prefer to bind peptides of defined length and sequence [21-24]. Did the linkage of the transporter genes to the MHC evolve in order to facilitate co-evolution of this specificity? An implication of coevolution of class-I sequence specificity and transporter specificity is that the transporter should be polymorphic, with different alleles in linkage disequilibrium with particular MHC class-I alleles. Although a thorough study of polymorphism of the HAM genes has not been done, restriction fragment length polymorphisms around these genes are relatively rare (unpublished observations), and the amino acid sequence of the HAM1 gene derived from two MHC mismatched cell lines is 100% identical [25]. These observations argue against the possibility of peptide sequence specificity for the transporter. However, this structure may still have specificity for peptide length, since all allelic forms of MHC class-I antigens examined to date exhibit similar preferences (8-10 amino acids) for peptide length.
Evidence for a Transport Defect in Class-I-Deficient Mutants
vious deletion corrects the class-I defect [26]. However, the same cloned gene does not correct the defect in a mutant with an approximately 1,000-kb deletion (containing both the transporter genes). These data suggest that two (or more) genes in this region are required in the class-I antigen-processing pathway. Consistent with this idea is that the HAM1 gene does not correct the defect in the RMA-S cell line (unpublished observations). An obvious candidate for the second gene is HAM2, preliminary evidence suggests that the defect in RMA-S may involve the HAM2 gene, but to date we have achieved only partial restoration of the defect in these cells with cloned HAM2 cDNA. Evidence for a transport defect in the classI-processing defective T2 cell line has been obtained using a minigene encoding a 12 amino acid peptide derived from influenza virus matrix protein. Transfection of this minigene into either parental (T1) or mutant (T2) cells results in presentation of the peptide to class-I-restricted T cells when the peptide sequence is preceded by a classical signal peptide [27]. This indicates that the flu peptide can be presented by either cell if introduced into the lumen of the ER via the normal secretory pathway. Transfection of the same influenza minigene lacking the signal sequence into T1 (wild-type) cells results in presentation of the peptide to class-I-restricted T cells, indicating that peptides produced in the cytoplasm can indeed be transported into the ER in wild-type cells. However, introduction of the same gene into T2 mutant cells does not result in presentation ot the peptide, indicating that T2 cells are defective in this transport function.
Direct evidence for the involvement of the HAM1, and possibly also the HAM2, genes in the class-I pathway has recently been obtained. Introduction of a cloned copy of HAM1 cDNA into a mutant carrying no ob-
130
Monaco
Putative Peptide Transporters and Antigen Processing
A Model for the Class-I Antigen-Processing Pathway In summary, the HAM1 and HAM2 genes probably encode the two subunits of a heterodimer that resides in the ER membrane and transports peptides from the cytoplasm into the lumen of the ER, where they are bound by newly synthesized class-I molecules. (Any unbound peptides are probably degraded in the ER.) Although the source of the peptides is unknown, recent evidence suggests
q g t g q
ooo
ool~
9 1 7 69
9 ~149o~
~ 1 7 6 1 7 ~6 1 7 6 1~ 7 1 6 71 67 16 7~ 61 17 76 61 7 6 1 o7o Q6 9
G9
9
that genes in the MHC may also be involved in the generation of such peptides [28, 29]. At least two of the subunits of a large cytoplasmic protease (called the LMP complex) are encoded by genes positioned immediately upstream (centromeric) of the HAM1 and HAM2 genes. These two structures alone (the transporter heterodimer and the LMP complex) may perform all of the immune-specific functions involved in antigen processing for class-I-restricted T-cell responses [30].
9 1 4 9 1 4 9 o o1 9 4 9 o 19 4 ~9 1 7 6
9 1 4 9 9 1 4 9I 9
9 1 4 9 1 4 o9 Q 9
. 4
9
9 1 7 6 ~176a
9 1 4 9o o
9 1 4 9 9 to
9 il
9 olo
o*
o o o o o o o
References 1 Ziegler K, Unanue ER: Identification of a macrophage antigen-processing event required for I-regionrestricted antigen presentation to T Iymphocytes. J Immunol 1981 :t 27: 1869-1875. 2 Townsend A, Bodmer H: Antigen recognition by class I restricted T lymphocytes. Annu Rev lmmunol 1989;7:601-624. 3 Brae9 T J, Braciale VL: Antigen presentation: Structural themes and functional variations. Immunol Today 1991;12:124-129. 4 Yewdell JW, Bennick JR: The binary logic of antigen processing and presentation to T cells. Cell 1990;62: 203-206. 5 Germain RN: The ins and outs of antigen processing and presentation. Nature 1986;322:687-689. 6 Gierasch LM: Signal sequences. Biochemistry 1989;28:923-930. 7 von Heijne G: Signal sequences: The limits of variation. J Mol Biol 1985; 184:99-105, 8 Kg.rre K, Ljunggren HG, Piontek G, Kiessling R: Selective rejection of H-2-deficient lymphoma variants suggests alternative immune defence strategy. Nature 1986;319: 675-678.
9 Townsend A, Ohldn C, Bastin J, Ljunggren HG, Foster L, K~irre K: Association of class I major histocompatbility heavy and light chains induced by viral peptides. Nature 1989;340:443-448. 10 DeMars R, Rudersdorf R, Chang C, Petersen J, Strandtmann J, Kom N, Sidwell B, Orr HT: Mutations that impair a posttranscriptional step in expression of HLA-A and -B antigens. Proc Natl Acad Sci USA 1985; 82:8183-8187. 11 Cerundolo V, Alexander J, Anderson K, Lamb C, Cresswell P, McMichael A, Gotch F, Townsend A: Presentation of viral antigen controlled by a gene in the major histocompatibility complex. Nature 1990;345: 449-452. 12 Hosken NA, Bevan MJ: Defective presentation of endogenous antigen by a cell line expressing class I molecules. Science 1990;248:367-370. 13 Townsend A, Elliot T, Cerundolo V, Foster L, Barber B, Tse A: Assembly of MHC class I molecules analyzed in vitro. Cell 1990;62:285-295. 14 Ljunggren HG, Stam NJ, (~hl6n C, Neefjes J J, H6glund P, Heemels MT, Bastin J, Schumacher TNM, Townsend A, K~irre K, Ploegh HL: Empty MHC class I molecules come out in the cold. Nature 1990;346: 476-480.
15 Schumacher TNM, Heemels MT, Neefjes J J, Kast WM, Melief CJM, Ploegh HL: Direct binding of peptide to empty MHC class I molecules on intact cells and in vitro. Cell 1990;62:563-567. I6 Spies T, Bresnahan M, Bahrain S, Arnold D, Blanck G, Mellins E, Pious D, DeMars R: A gene in the human major histocompatibility complex class II region controlling the class I antigen processing pathway. Nature 1990;348:744-747. 17 Monaco J J, Cho S, Attaya M: Transport protein genes in the murine MHC: Possible implications for antigen processing. Science 1990;250: 1723-1726. 18 Deverson EV, Gow IR, Coadwell WJ, Monaco J J, Butcher GW, Howard JC: MHC class II region encoding proteins related to the multidrug resistance family oftransmembrane transporters. Nature 1990;348:738-741. 19 Trowsdale J, Hanson I, Mockridge I, Beck S, Townsend A, Kelly A: Sequences encoded in the class II region of the MHC related to the 'ABC' superfamily of transporters. Nature 1990;348:741-744.
131
20 Hyde SC, Emsley P, Hartshorn M J, Mimmack MM, Gileadi U, Pearse SR, Gallagher MP, Gill DR, Hubbard RE, Higgins CF: Structural model of ATP-binding proteins associated with cystic fibrosis, multidrug resistance and bacterial transport. Nature 1990;346:362-365. 2i Falk K, R6tzschke O, Rammensee HG: Cellular peptide composition governed by major histocompatibility complex class 1 molecules. Nature 1990;348:248-251. 22 R6tschke O, Falk K, Deres K, Sehild H, Norda M, Metzger J, Jung G, Rammensee HG: Isolation and analysis of naturally processed viral peptides as recognized by cytotoxic T cells. Nature 1990;348:252-254. 23 Van Bleek GM, Nathenson SG: Isolation of an endogenously processed immunodominant viral peptide from the class I H-2Kb molecule. Nature 1990;348:213-216.
24 Elliott T, Cerundolo V, Elvin J, Townsend A: Peptide-induced conformational change of the class I heavy chain. Nature 1991 ;351:402406. 25 Bahram S, Arnold D, Bresnahan M, Strominger JL, Spies T: A second peptide transporter gene in the human MHC class II region. Proc NatI Acad Sci USA 1991, in press. 26 Spies T, DeMars R: Restored expression of major histocompatibility class I molecules by gene transfer of a putative peptide transporter. Nature 1991;351:323-324. 27 Anderson K, Cresswell P, Gammon M, Hermes J, WiUiamson A, Zweerink H: Endogenously synthesized peptide with an endoplasmic reticulum signal sequence sensitizes antigen processing mutant cells to class I-restricted cell-mediated lysis. J Exp Med 1991;174:489-492.
28 Brown MG, DriscoU J, Monaco J J: Structural and serological similarity of the MHC-linked LMP and proteasome (multicatalytic proteinase) complexes. Nature 1991;353:355357. 29 Martinez CK, Monaco I J: A major histocompatibility complex-linked LMP gene: HomoIogy to proteasome subunits. Nature 199t;353: 664-667. 30 Monaco J J: A model of antigen processing for MHC class I-restricted responses. Immunol Today 1992; 13:173-179 31 Devereux J, Haeberli P, Smithies O: A comprehensive set of sequence analysis programs for the VAX. Nucleic Acids Res 1984; 12:387-395.
132
Monaco
Putative Peptide Transporters and Antigen Processing
