Functional Role of C-Terminal Sequence Elements in the Transporter Associated with Antigen Processing
http://www.100md.com
免疫学杂志 2005年第1期
Abstract
TAP delivers antigenic peptides into the endoplasmic reticulum (ER) that are subsequently bound by MHC class I molecules. TAP consists of two subunits (TAP1 and TAP2), each with a transmembrane (TMD) and a nucleotide-binding (NBD) domain. The two TAP-NBDs have distinct biochemical properties and control different steps during the peptide translocation process. We noted previously that the nonhomologous C-terminal tails of rat TAP1 and TAP2 determine the distinct functions of TAP-NBD1 and -NBD2. To identify the sequence elements responsible for the asymmetrical NBD function, we constructed chimeric rat TAP variants in which we systematically exchanged sequence regions of different length between the two TAP-NBDs. Our fine-mapping studies demonstrate that a nonhomologous region containing the 6/10-loop in conjunction with the downstream switch region is directly responsible for the functional separation of the TAP-NBDs. The 6/10-loop determines the nonsynonymous nucleotide binding of NBD1 and NBD2, whereas the switch region seems to play a critical role in regulating the functional cross-talk between the structural domains of TAP. Based on our findings, we postulate that these two sequence elements build a minimal functional unit that controls the asymmetry of the two TAP-NBDs.
Introduction
Major histocompatibility complex class I molecules are expressed on the cell surface of all nucleated cells and present peptides derived from intracellular proteins to CTLs. CTLs monitor MHC class I molecules for peptides derived from nonself proteins and eliminate infected cells. Peptides presented by MHC class I molecules are generated predominantly in the cytoplasm by the proteasome (1). Transport of antigenic peptides from the cytoplasm into the endoplasmic reticulum (ER)3 by TAP for loading onto MHC class I molecules is a key step in the Ag presentation pathway (1). TAP is a member of the ATP-binding cassette (ABC) transporter family whose members translocate different types of substrates across cellular membranes using energy from ATP hydrolysis (2). ABC transporters have a modular structure comprising two nucleotide-binding domains (NBDs) and two transmembrane domains (TMDs). Each of the two subunits of the heterodimeric TAP, TAP1 and TAP2, has an N-terminal TMD and a C-terminal NBD. The paired TMDs of TAP form the peptide-binding site and the pore through which peptide substrates are translocated, whereas the NBDs provide the energy required for peptide transport by ATP hydrolysis (3).
Major advances have recently been made in the analysis of how peptides are transported by TAP (4). The binding and hydrolysis of nucleotides are believed to power the transport process by inducing conformational changes in the NBDs that are transferred to the TMDs, and to cause the binding and movement of peptide across the membrane (5, 6). Analysis of the transport cycle of TAP led to working models in which the NBDs bind and hydrolyze nucleotides in an alternating and strongly interdependent manner (4, 5). Several groups have noted that the sequence-related NBDs of TAP1 and TAP2 fulfil distinct functions in the transport cycle of TAP (5, 6, 7, 8), and the functional asymmetry of the NBDs is an intrinsic property of these structural domains (9, 10).
We previously showed that TAP1 has a much higher ATP-binding capacity than TAP2 (9). The distinct nucleotide binding and function of the TAP-NBDs are mediated by the nonhomologous C-terminal tails of the two TAP chains and not by the core NBDs containing the conserved sequence motifs of the ATP-binding cassette (11). The functional importance of the C-terminal tails of the NBDs has been shown for several different members of the ABC transporter family (12, 13, 14, 15, 16). For example, the P-glycoprotein (12) can only tolerate small C-terminal deletions. In addition, results reported for the ABC protein sulfonylurea receptor (SUR) suggest that residues in the 11 strand of the C-terminal tail control the nucleotide-binding and catalytic functions of the NBD2 (16). Most interestingly, four different splicing isoforms of the TAP-related ABC half-transporter TAPL have recently been reported with heterogeneity of the C-terminal region, suggesting that the function of this ABC transporter may be modulated by the C-terminal tail (17).
An important contribution to understanding the transport mechanism of TAP would be the identification of sequence elements in the C-terminal tails responsible for the distinctive biochemical and functional properties of the NBDs. In this paper we report the analysis of a series of chimeric rat TAP chains in which we have systematically exchanged sequence stretches between the two NBDs. Our fine-mapping studies demonstrate that the sequence corresponding to the region between 6 and 10 (6/10-loop) and the downstream switch region (18) build a minimal functional unit within the nonhomologous C-terminal tails that determines the different properties of TAP-NBD1 and -NBD2. In particular, our data show that the asymmetrical sequence of the 6/10-loop controls the distinct nucleotide-binding activity of NBD1 and NBD2 in the resting state of TAP, whereas the differences in the switch region appear to be critical for proper transduction of conformational signals between the structural domains of TAP.
This is the first report in which the functions of the 6/10-loop and the switch region of TAP have been structurally defined. Our experimental findings provide new insights into the sequence requirements for the functional divergence of the TAP-NBDs and suggest that molecular regulatory mechanisms in TAP differ from the postulated C-terminal regulation of the ABC-protein SUR (16).
Materials and Methods
Cell lines and cell culture
T2 is a human lymphoblastoid cell line that lacks both TAP genes and expresses only the HLA-A2 and -B5 class I molecules (19). Transfectants of T2 cells containing wild-type rat TAPa (20) and TAP chimeras were cultured in IMDM (Invitrogen Life Technologies) supplemented with 10% FCS (BioWhittaker) and 1 mg/ml G418 (PAA).
Cloning and expression of chimeric TAP chains
The 2.6-kb EcoRI fragments containing full-length cDNA from rat TAP1 (21) and the chimeric chain 1N2 (9), respectively, were cloned into the multiple cloning site of pBluescript KS+ (Stratagene). For the 1G2 construct, the QuikChange site-directed mutagenesis procedure (Stratagene) was used to create a ScaI site in TAP1 at position 2016 and in 1N2 at position 2092 (position 1 is the A of the first AUG). For TAP1 we used the complementary primers 5'-GCACCCACCTGCAGTACTTGGAGAGAGGAGG-3' and 5'-CCTCCTCTCTCCAAGTACTGCAGGTGGGTGC-3', and for 1N2 we used the complementary primers 5'-GCACGACCAGTACTGGGATGAGCAGG-3' and 5'-CCTGCTCATCCCAGTACTGGTCGTGC3'. The chimeric TAP construct 1G2 was created by ligation of the 1.65-kb ScaI fragment of TAP1 to the 3.96-kb ScaI fragment of 1N2. Constructs encoding the TAP chain chimeras 1F2, 1E2, and 1D2 were created by a similar mutagenesis procedure using the TAP1 and 1N2 cDNAs as templates. ScaI sites were introduced in TAP1 at position 2066 (for 1F2), 2014 (for 1E2), and 1977 (for 1D2), respectively. In 1N2 the corresponding positions for the ScaI sites were 2051 (for 1F2), 1998 (for 1E2), and 1975 (for 1D2). For the mutagenesis approach we used the TAP1 mutagenesis primers 5'-GCGGGCCCACCAAGTACTCTTCCTCAAAGAAGGC-3' and 5'-GCCTTCTTTGAGGAAGAGTACTTGGTGGGCCCGC-3' (for 1F2), 5'-CCGAGTGGGCCTCTAGTACTGTTCTTCTGATCACC-3' and 5'-GGTGATCAGAAGAACAGTACTAGAGGCCCACTCGG-3' (for 1E2), and 5'-CCAGCTACGGGTCCAGTACTTCCTGTATGAGAGCC-3' and 5'-GGCTCTCATACAGGAAG TACTGGACCCGTAGCTGG-3' (for 1D2). In the case of the 1N2 template the complementary mutagenesis primers were 5'-GCTGACCAAGTACTGGTGCTCAAGC-3' and 5'-GCTTGAGCACC AGTACTTGGTCAGC-3' (for 1F2), 5'-CGCAGGAGGACAGTACTATGCTGGTCATTGC-3' and 5'-GCAATGACCAGCATAGTACTTGCCTCCTGCG-3' (for 1E2), and 5'-GCAGGCCCTTCAG TACTGGAGATCGCAGG-3' and 5'-CCTGCGATCTCCAGTACTGAAGGGCCTGC-3' (for 1D2). After ligation of the ScaI fragments of TAP1 (1.65, 1.7, 1.76, and 1.79 kb) with the ScaI fragments of 1N2 (3.96, 3.92, 3.87, and 3.85), we restored the original amino acid sequence by an additional site-directed mutagenesis step using the complementary primers 5'-GGAGCACGACCAGCTCATGGAGAGAGGAGG-3' and 5'-CCTCCTCTCTCCATGAGCTGGTCGTGCTCC-3' (for 1G2), 5'-GCTGACCAAGTTCTCTTCCTCAAAGAAGGC-3' and 5'-GCCTTCTTTGAGGAAGAGAACTTGGTCAGC-3' (for 1F2), 5'-CGCAGGAGGACAGGACGGTT CTTCTGATCACC-3' and 5'-GGTGATCAGAAGAACCGTCCTGTCCTCCTGCG-3' (for 1E2), and 5'-GCAGGCCCTTCAGCGGCTCCTGTATGAGAGCC-3' and 5'-GGCTCTCATACAGGAGCCGCTGAAGGGACTGC-3' (for 1D2). The chimera 1G2 encoded residues 1–505 and 706–725 of TAP1 and residues 494–685 of TAP2; the chimera 1F2 contained residues 1–505 and 690–725 of TAP1 and residues 494–672 of TAP2; the chimera 1E2 encoded residues 1–505 and 671–725 of TAP1 and residues 494–653 of TAP2; the chimera 1D2 comprised residues 1–505 and 660–725 of TAP1 and residues 494–646 of TAP2. The chimeric constructs 1N2 (9) and 1F2 (see above) were used to create the 1H2 construct. ScaI sites were created in 1N2 (position 2092) and 1F2 (position 2083) by site-directed mutagenesis with specific primers following the procedure described above. The resulting construct contained residues 1–505 and 690–705 of TAP1 and residues 494–672 and 686–703 of TAP2. To construct variant 1L2 we used the cDNA of 1D2 and 1N2 (9) as templates for site-directed mutagenesis. In 1D2 we introduced a ScaI site a position 2010 using the primers 5'-CCGAGTGGGCCTCTAGTACTGTTCTTCTGATCACC-3' and 5'-GGTGATCAGAAGAACAGTACTAGAGGCCCACTCGG-3', and in 1N2 (9) at position 1998 using the primers 5'-CGCAGGAGGACAGTACTATGCTGGTCATTGC-3' and 5'-GCAATGACCA GCATAGTACTTGCCTCCTGCG-3'. To create the construct encoding 1L2, the 3.87-kb fragment of 1D2 and the 1.75-kb fragment of 1N2 were ligated, and the original sequence was restored using the primers 5'-CCGAGTGGGCCTCTCGGACGATGCTGGTCATTGC-3' and 5'-GCAATAACCAGCATCGTCCGAGAGGCCCACTCGG-3'. The resulting construct encoded residues 1–505 and 660–670 of TAP1 and residues 494–646 and 654–703 of TAP2. The cDNAs of 1D2 and 1N2 were also used for the construct encoding 1S2. In this study we introduced a ScaI site in 1D2 at position 2063 with the primers 5'-GCGGGCCCACCAAGTACTCTTCCTCAAAGAAGGC-3' and 5'-GCCTTCTTTGAGGAAGAGTACTTGGTGGGCCCGC-3'. In 1N2 the ScaI site was created at position 2051 using the primers 5'-GCTGACCAAGTACTGGTGCTCAAGC-3' and 5'-GCTTGAGCACCAGTACTTGGTCAGC-3'. To create the 1S2 construct the 3.9-kb fragment of 1D2 and the 1.5-kb fragment of 1N2 were ligated, and the original sequence was restored with the primers 5'-GCGGGCCCACCACATCCTGGTGCACAAGC-3' and 5'-GCTTGTGCACCAGGATGTGGTGGGCCCGC-3'. The resulting construct encoded residues 1–505 and 660–689 of TAP1 and residues 494–646 and 672–703 of TAP2. All TAP constructs were cloned into the EcoRI site of pHApr1neo (22), and mutated NBD regions were sequenced in both directions. All chimeric TAP constructs were transfected into T2 cells by electroporation using a gene pulser (Bio-Rad) at 270 V and 500 μF. After selection with G418 (1 mg/ml) for 4–6 wk, stable transfectants were subcloned and screened for TAP chain expression by Western blotting.
Antibodies
116/5 is a polyclonal rabbit antiserum specific for the C terminus of rat TAP2 chains (20). D90 is a polyclonal rabbit antiserum recognizing the C terminus of rat TAP1 chains (23). MAC 394 is a monoclonal mouse anti-rat TAP2a Ab (24) derived from immunization with the recombinant histidine-tagged cytoplasmic domain of rat TAP2a. MAC 394 recognizes the polymorphic residues at positions 538 and 539 in the core NBD of rat TAP2a (11). 1p3 is a polyclonal rabbit antiserum that binds epitopes in or between the predicted TMDs of TAP1 (25). 4E is a conformation-dependent mouse mAb that recognizes an epitope common to all HLA-B and -C Ags (26).
Immunoprecipitation, Western blotting, and flow cytometry
For immunoisolation experiments, 1 x 107 cells were washed twice in ice-cold PBS (1.7 mM KH2PO4, 10 mM Na2HPO4, 140 mM NaCl, and 2.7 mM KCl, pH 7.5) before solubilization in lysis buffer (PBS (pH 7.5) containing 1% Triton X-100 (Sigma-Aldrich)). Immunoprecipitations with anti-TAP2 (116/5) and anti-TAP1 (1p3) were performed as described previously (24). Immunoprecipitates were washed with PBS/1% Triton X-100 and eluted with 100 mM Tris-HCl (pH 9) containing 0.5% SDS. Samples were analyzed by Western blotting in conjunction with a specific primary antiserum. Bands were visualized with HRP-conjugated secondary Abs (donkey anti-rabbit IgG-HRP; Amersham Biosciences) and ECL substrate (Amersham Biosciences). Flow cytometry experiments using the anti-HLA-B5-specific mAb 4E were performed as described previously (27).
Transport assay, peptide cross-linking, and nucleotide-binding assays
For transport assays, 2 x 106 cells were permeabilized with streptolysin O (2 U/ml; Murex). After washing with PBS, 0.5 μM radioiodinated peptide S8 (TVDNKTRYR, in the single-letter amino acid code), 10 mM ATP, and incubation buffer (50 mM HEPES (pH 7.5), 250 mM sucrose, 150 mM CH3COOK, 5 mM (CH3COO)2Mg · 4H2O, 1 mM DTT, 1 mM Pefabloc (Roche), and 1.8 μg/ml aprotinin (Sigma-Aldrich)) were added and incubated for 10 min at 37°C. After lysis with 20 mM Tris-HCl (pH 7.5), 500 mM NaCl, and 0.1% Nonidet P-40 (Sigma-Aldrich), transported glycosylated peptides were isolated with Con A-Sepharose (BD Biosciences) and quantitated by gamma counting (28). For peptide cross-linking, permeabilized cells were incubated with 1 μM radioiodinated and N-hydroxysuccinamide-4-azidobenzoate (HSAB)-conjugated peptide S8. Cross-linking was induced by irradiation with a UV lamp at 254 nm for 5 min on ice. Cells were lysed by adding 1% Triton X-100 in PBS. The nucleotide- binding assay was performed as described previously (27). All nucleotide binding experiments were performed with N6-coupled ATP- and ADP-agarose (Sigma-Aldrich). To analyze peptide-induced ATP binding, CHAPS-solubilized cell membranes were pretreated with increasing concentrations of S8 peptide (10, 1, 0.1, and 0.01 μM) for 30 min. Afterward the lysates were incubated with ATP-agarose for 45 min. After washing, the bound proteins were eluted with SDS loading buffer.
For photolabeling of TAP with radiolabeled 8-azido-ATP, membranes of cells were prepared and resuspended in 250 mM sucrose, 50 mM KCl, 2 mM MgCl2, 2 mM EGTA, and 10 mM Tris-HCl (pH 6.8) (27). Membranes corresponding to 3 x 106 cells in a final volume of 100 μl were incubated with 2 μM 8-azido-[-32P]ATP (ICN Biomedicals) for 5 min at 4°C. Cross-linking was induced by irradiation with a UV lamp at 254 nm for 5 min at 4°C.
Preparation of microsomal membranes
Microsomes from 1 x 108 T2 cells expressing chimeric and wild-type TAP proteins were generated by sucrose gradient fractionation (29). Cells were washed twice with ice-cold PBS, resuspended in 10 ml of 10 mM Tris-HCl (pH 7.5) with protease inhibitor mixture (Complete Protease Inhibitor; Roche), and incubated on ice for 10 min. The lysed cells were then homogenized and centrifuged at 800 x g for 5 min at 4°C. The resulting supernatants were resuspended in 5 ml of 1.3 M sucrose buffer (20 mM HEPES (pH 7.5), 25 mM CH3COOK, 5 mM (CH3COO)2Mg · 4H2O, 1 mM DTT, and protease inhibitor mix) and centrifuged again at 800 x g at 4°C for 10 min. The supernatants were then centrifuged at 68,000 x g at 4°C for 2 h, and the membrane pellets were resuspended in 800 μl of 0.25 M sucrose buffer. Subsequently, 5.6 ml of 2.5 M sucrose gradient buffer was added, and the suspension was overlaid carefully with 2.9 ml of 2 M and 2.9 ml of 1.3 M sucrose buffer. Approximately 800 μl of 0.25 M sucrose buffer was carefully loaded on the top of the gradient. The sucrose gradient was centrifuged at 100,000 x g for 16 h at 4°C. The microsomes were collected at the interface between the 2 and 1.3 M sucrose buffer, diluted in 20 mM HEPES buffer (20 mM HEPES (pH 7.5), 25 mM CH3COOK, 5 mM (CH3COO)2Mg · 4H2O, 1 mM DTT, and protease inhibitor mixture), homogenized, and centrifuged at 68,000 x g at 4°C for 1 h. The microsomal pellets were resuspended in 200 μl of 20 mM HEPES buffer. Finally, aliquots of 30–50 μl were snap-frozen in liquid nitrogen and stored at –80°C.
Chemical cross-linking
Membranes from 2 x 106 cells were resuspended in 400 μl of PBS containing 1% digitonin at 4°C, and the homobifunctional cross-linker ethylene glycol bissuccinimidyl succinate (EGS; Pierce) was added to a final concentration of 3 mM. The primary targets for EGS are -amino groups of solvent-exposed lysine residues. After incubation for 30 min at 4°C, the lysed membranes were washed with PBS containing 1% digitonin. The nonameric TAP-binding peptide S8 was added to a final concentration of 15 μM for 1 h at 4°C before EGS cross-linking.
Computational homology modeling of the NBD of rat TAP2
Model building was performed by a conventional three-step procedure. Step 1 was alignment. A global multialignment was obtained with the help of vector NTI using a modified version of CLUSTAL V (30). The sequence of human TAP1 (accession no. Q03518) was read directly from the crystal structure (18). Sequence homologies among human TAP1 and rat TAP2 (accession no. X63854) are in the range of 70%. Secondary structure predictions of the rat TAP-NBD2 sequence were performed with PHD (31) and PSIPRED v2.3 (32). Step 2 was model building. An initial model of the rat TAP-NBD2 main chain was built by searching a database of loops obtained from highly resolved protein structures (33). Loop selections were made on the basis of minimal steric interactions with the rest of the model. A full atom model of the rat TAP-NBD2 ectodomain was generated with the help of the backbone-dependent rotamer library implemented in SCWRL 3.0 (34). Step 3 was refinement of the structure model. Refinement of the rat TAP2-models was performed using the GROMACS package (35). The solvated rat TAP-NBD2-models were energy-minimized until they converged using the steepest descent method. A molecular dynamics-simulated annealing simulation (simulation time, 300 ps; temperature, 0–300 K), followed by a second energy minimization, were used to optimize the conformation of the newly introduced loops. During simulated annealing, positional constraints were applied to all main chain atoms, except atoms of newly introduced loops.
Results
Chimeric studies on the C-terminal regulation of asymmetrical nucleotide binding and NBD function in TAP
We have previously shown that the distinctive nucleotide-binding behavior of NBD1 and NBD2 in the resting state of rat TAP (9, 10) depends directly on the divergent C-terminal tails, rather than on the core NBDs situated between the conserved Walker A and B motifs (11). A chimeric rat TAP2 chain, containing the core NBD2 fused to the C-terminal segment of rat TAP1, adopts the ATP-binding behavior of wild-type NBD1, whereas a corresponding rat TAP1 chimera, containing the core NBD1 fused to the C-terminal segment of rat TAP2, shows the characteristic ADP-binding properties of wild-type NBD2 (11). Moreover, a rat TAP chain chimera, designated 1C2, comprising TMD1 (residues 1–505) and the core NBD2 (residues 494–639) with the C terminus of TAP1 (residues 653–725), acquires not only the nucleotide-binding behavior, but also the functional properties of TAP1 (11). Thus, the TAP chain chimera 1C2 provides a suitable experimental system for fine-mapping the regions within the C terminus that are responsible for the distinct nucleotide-binding properties of the TAP-NBDs.
We constructed a series of chimeric rat TAP chains based upon the variant 1C2 in which we dissected the nonhomologous C-terminal tail into four portions by progressively exchanging sequences between the two TAP-NBDs (Fig. 1). Positions of secondary structure elements in the C-terminal tails of rat TAP1 and rat TAP2 (see Fig. 1) were predicted on the basis of the crystal structure of the human TAP-NBD1 (18). The designations 1D2, 1E2, 1F2, and 1G2 indicate four different chimeric TAP chains where NBD1 regions between residues 506–659, 506–670, 506–689, and 506–705 were replaced by the corresponding NBD2 sequence: residues 494–646, 494–653, 494–672, and 494–685, respectively (Fig. 1). Because recent findings have suggested that the 11/12 region plays an important role in determining the nucleotide-binding properties of the ABC protein SUR (16), we created an additional rat TAP chain variant (1H2) from the chimera 1C2 in which the 11/12 sequence of TAP2 (residues 673–685) was replaced by the corresponding sequence stretch of TAP1 (residues 690–705; Fig. 1). All the constructs contained TMD1 as well as core NBD2 and only differed in their C-terminal sequences. After stable transfection of the different TAP chimeras into the TAP-negative human cell line T2, we tested the steady state expression of each TAP chain by immunoblot analysis using different TAP-specific Abs (see Materials and Methods). As shown in Fig. 2A, all chimeric TAP chains were stably expressed at similar levels.
FIGURE 1. Schematic diagram of wild-type and chimeric TAP chains. Sequences of the nonhomologous C-terminal tails of rat TAP-NBD1 and -NBD2 were retrieved from the GenBank database (GenBank accession no. X57523 and X63854) and aligned using the software Vector NTI (InforMax; Invitrogen Life Technologies). , Amino acid residues that are identical in both NBDs; , similar residues. Below, schematic diagrams of the wild-type TAP subunits and the chimeric TAP constructs 1C2, 1D2, 1E2, 1F2, 1G2, 1H2, and 1N2 are shown. The location of the switch region (switch, residues 671–689 in rat TAP1, residues 654–672 in rat TAP2) as well as the different positions of -helices 6 (residues 653–659 in rat TAP1, residues 640–646 in rat TAP2), 7 (residues 680–684 in rat TAP1, residues 663–667 in rat TAP2), 8 (residues 703–709 in rat TAP1, residues 683–689 in rat TAP2), 9 (residues 712–717 in rat TAP1, residues 692–697 in rat TAP2), -sheets 10 (residues 671–676 in rat TAP1, residues 654–659 in rat TAP2), 11 (residues 688–693 in rat TAP1, residues 671–676 in rat TAP2), 12 (residues 696–701 in rat TAP1, residues 679–782 in rat TAP2), and connecting loop structures, predicted on the basis of the crystal structure of the human TAP-NBD1 (18 ), are shown. TAP1 sequences in the different chimeras are indicated in black, whereas the corresponding sequences of TAP2 are indicated in gray.
FIGURE 2. Stable expression and biochemical characteristics of chimeric TAP variants. A, Expression levels and schematic overview of wild-type and chimeric TAP subunits with altered NBD segments. T2 transfectants were lysed in buffer containing 1% Triton X-100. Cell lysates of 5 x 104 cells were separated by 7.5% SDS-PAGE and blotted onto nitrocellulose membranes. Immunoblots were probed for the different TAP chains using antisera D90 (C-term. NBD1), 116/5 (C-term. NBD2), and Ab MAC 394 (core NBD2). A pictorial overview of wild-type TAP and the different chimeric TAP subunits 1N2, 1C2, 1D2, 1E2, 1F2, 1G2, and 1H2 is indicated at the bottom of the analysis. B, Nucleotide-binding properties of wild-type and chimeric TAP chains. ER membrane fractions of T2 transfectants were resuspended in lysis buffer containing 1% Triton X-100 and incubated with ATP- and ADP-agaroses in the presence of 3 mM MgCl2. After washing the nucleotide-agaroses, bound proteins were eluted with SDS-sample buffer and analyzed by probing Western blots for the C-terminal tail of TAP1 (antiserum D90) or the C-terminal tail of TAP2 (antiserum 116/5). TAP chimeras are indicated schematically on the left.
To investigate whether the progressive exchange of sequence between the two TAP-NBDs has an effect on the nucleotide-binding properties of the chimeric TAP chains, lysates of ER membrane preparations were analyzed by standard nucleotide-binding assays (6, 24) for binding to ATP- and ADP-agarose beads (Fig. 2B). The binding to the different nucleotide-agaroses was compared with that of wild-type TAP subunits, the chimera 1C2 (11), and chimera 1N2, which comprises TMD1 and the entire sequence of NBD2 (9) (Fig. 1). Analysis of TAP1, TAP2, 1C2, and 1N2 (Fig. 2B, left panel) confirmed that the distinct nucleotide binding of the TAP subunits is an inherent property of the NBDs that is determined by the C-terminal tails (11). The variant 1C2 bound in a TAP1-like manner to ATP- as well as ADP-agarose (11), whereas 1N2 interacted only with ADP-agarose, which is a characteristic property of TAP2 (5). Analysis of the other chimeric TAP chains showed that 1D2 bound to ATP- as well as ADP-agarose, whereas the chimeric polypeptides 1E2, 1F2, 1G2, and 1H2 bound only to ADP-agarose (Fig. 2B, right panel).
To explore whether the TAP chain variants can participate as a TAP1-type chain in a functional heterodimeric peptide transporter, we created five different TAP molecules in T2 cells by coexpressing the chimeric variants 1D2, 1E2, 1F2, 1G2, and 1H2 in combination with wild-type TAP2. The expression levels of these TAP variants were comparable to those of wild-type transporters (Fig. 3A, left panel) and showed normal subunit assembly (Fig. 3A, right panel, upper part).
FIGURE 3. Functional properties of chimeric transporters 2-1D2, 2-1E2, 2-1F2, 2-1G2, and 2-1H2. A, Expression levels of wild-type and chimeric transporters (left panel). Cells were lysed in 1% Triton X-100. Lysates of 5 x 104 cells were separated by 7.5% SDS-PAGE and blotted onto nitrocellulose membranes. Immunoblots were probed for TAP chains using antisera D90 (C-term. NBD1) and 116/5 (C-term. NBD2). Complex formation of wild-type and chimeric TAP transporters is shown (right panel, top). Transfected T2 cells were lysed in 1% Triton X-100, and TAP molecules were immunoisolated with anti-TAP2 antisera 116/5 (TAPwt, TAP2-1D2, TAP2-1E2, TAP2-1F2, and TAP2-1G2) or 1p3 (TAP2-1H2). After washing, the bound proteins were eluted with SDS, separated on a 7.5% SDS-gel, and analyzed by probing Western blots for TAP-NBD1 (D90) or TAP-NBD2 (116/5). Peptide-binding properties are shown (right panel, middle). The peptide-binding activity of the TAP variants was analyzed by substrate cross-linking. Microsomal fractions were resuspended in binding buffer and incubated with 1 μM radioiodinated and HSAB-conjugated peptide S8. After cross-linking, membranes were lysed, and TAP was immunoisolated with anti-TAP2 antiserum (116/5). The nucleotide-binding properties of chimeric TAP molecules are shown (right panel, bottom). Membrane fractions of T2 cells expressing wild-type or chimeric transporters were lysed in lysis buffer containing 2 μM radiolabeled 8-azido-ATP. After UV cross-linking, TAP variants were immunoprecipitated with an anti-TAP2 antiserum and separated by SDS-PAGE. B, TAP-mediated peptide transport. Transfected and nontransfected T2 cells were permeabilized with streptolysin O (SLO) and incubated in transport buffer containing 10 mM ATP and radioiodinated peptide S8 for 10 min at 37°C. Bar graphs show the recovered amount of transported labeled peptides as cpm and represent the average values of experiments performed in duplicate. C, Surface expression of MHC class I molecules. Cells were incubated with mAb 4E that recognizes HLA-B5, followed by FITC-labeled secondary Ab. Surface expression of HLA-B5 was detected by flow cytometry (). Mean fluorescence intensity values are indicated. Background staining was determined by incubating only with secondary Ab (). TAP variants are shown schematically.
For functional characterization, the different transporter variants were first analyzed for their peptide- and ATP-binding behaviors (Fig. 3A, right panel, middle and bottom parts). Binding of peptide and nucleotide was analyzed by photolabeling experiments using radioactively labeled and UV-cross-linkable reagents (5). Although all chimeric transporter variants had the capacity to bind peptides, only the variant TAP2-1D2 showed a pattern of ATP binding comparable to that of the functional wild-type transporter (5). In line with the nucleotide-binding assays shown in Fig. 2B, the ATP cross-linking experiments showed detectable ATP labeling for TAP1 in the wild-type transporter and for 1D2 in the chimeric transporter TAP2-1D2, whereas all other transporter variants showed no detectable ATP cross-linking. We measured the peptide transport activities of the different chimeric transporters in standard transport assays (28) using radioiodinated model peptides (5). As shown in Fig. 3B, only TAP2-1D2 showed a significant transport activity compared with wild-type TAP and the transporter variant 2-1C2. The transport efficiency of TAP2-1D2 was 40–50% that of wild-type TAP and 70% that of TAP2-1C2. Thus, like the chimera 1C2 (11), the chimeric polypeptide 1D2 can functionally substitute for wild-type TAP1, albeit inefficiently. In contrast, no detectable peptide transport above that observed for the nontransfected T2 cell line was seen for variants TAP2-1E2, TAP2-1F2, TAP2-1G2, and TAP2-1H2 (Fig. 3B). The different peptide transport activities of the chimeric TAP molecules were also reflected in the surface expression of mature HLA-B5 molecules determined by flow cytometry (Fig. 3C).
Taken together, our data show that the C-terminal sequence information from TAP1 present in the chimera 1D2 confers the nucleotide-binding and functional characteristics of NBD1 upon NBD2, whereas the C-terminal amino acids of NBD1 present in the chimeras 1E2, 1F2, 1G2, and 1H2 do not support an NBD1-like function. Thus, in contrast to the ABC protein SUR (16), our findings clearly demonstrate that in the case of TAP, a region upstream of the 11/12-strands must be responsible for the distinct nucleotide-binding properties of TAP1 and TAP2.
6/10-loop and switch region build a minimal functional unit that controls asymmetry of the two TAP-NBDs
The differences in nucleotide-binding and peptide transport functions of 1D2 and 1E2 (Figs. 2 and 3) suggest that the sequence stretch between the secondary structure elements 6 and 10 (6/10-loop, residues 660–670 in rat TAP1, and residues 647–653 in rat TAP2; see Fig. 1) could play a crucial role in controlling the biochemical and functional behaviors of the TAP-NBDs. Sequence alignment of rat TAP-NBD1 and rat TAP-NBD2 (Fig. 1) shows that this NBD segment is strikingly different in sequence (RLLYESPEWAS vs TWRSQED, in the single-letter amino acid code) and in length (11 residues vs seven residues) between rat TAP1 and TAP2. The same is also true for the TAP-NBDs of other species, e.g., mouse, hamster, and human. Three dimensional modeling of the rat TAP-NBD2 primary sequence using the NBD1 of human TAP (18) as a structural template (Fig. 4A) suggests that the 6/10-loop of TAP2 is also structurally divergent (RMS deviation, 5.5 ?) from that found in TAP1.
FIGURE 4. Construction, expression, and nucleotide-binding properties of TAP chimeras 1L2 and 1S2. A, Homology modeling of TAP-NBD2 (indicated in gray) based on the crystal structure of TAP-NBD1 (18 ) (indicated in black). The overlay shows a close-up of NBD structures between the Walker B motif and the switch region created with Swiss-PdbViewer (http://expasy.org/spdbv). The positions of both elements and the 6/10-loop are indicated. Arrowheads mark the points where the sequences were exchanged in chimeras 1L2 and 1S2. B, Schematic diagram of wild-type TAP chains and the two chimeric constructs (1L2 and 1S2) in which the 6/10-loop (residues 660–670 in rat TAP1 and residues 647–653 in rat TAP2) and the switch region (switch, residues 671–689 in rat TAP1, residues 654–672 in rat TAP2) were exchanged between NBD1 and NBD2. C, Expression levels of wild-type and chimeric TAP subunits with exchanged NBD segments (left panel). T2 transfectants were lysed in buffer containing 1% Triton X-100. Cell lysates of 5 x 104 cells were separated by 7.5% SDS-PAGE and blotted onto nitrocellulose membranes. Immunoblots were probed for the different TAP chains using antisera D90 (C-term. NBD1), 116/5 (C-term. NBD2), and Ab MAC 394 (core NBD2). A pictorial overview of the wild-type TAP and the chimeric TAP subunits is indicated at the bottom of the analysis. Nucleotide-binding properties of wild-type and chimeric TAP chains are shown (right panel). Membrane fractions of T2 cells expressing chimeras 1L2 and 1S2 were resuspended in lysis buffer containing 1% Triton X-100 and incubated with different nucleotide-agaroses. After extensive washing, bound proteins were eluted with SDS-sample buffer and analyzed by probing Western blots for the C-terminal tails of TAP1 (D90) or TAP2 (116/5). TAP chimeras are indicated schematically on the left.
To investigate whether the nature of the 6/10-loop controls the distinctive nucleotide-binding behavior of the TAP chains, we constructed a 1D2-related chimera (1L2) in which we exchanged only the sequence of the 6/10-loop between NBD1 and NBD2 (residues 660–670 in NBD1, residues 647–653 in NBD2; Fig. 4B). In addition, we created a chimeric construct (1S2) in which the sequence exchanged between NBD1 and NBD2 extended from the 6/10-loop to include the nonhomologous residues of the downstream switch region (residues 660–689 in TAP1, residues 647–672 in TAP2; Fig. 4B). The switch region is thought to play a crucial role in the conformational signal transduction of ABC transporter domains (36). After stable transfection of both chimeric TAP constructs into T2 cells (Fig. 4C, left panel), we compared the nucleotide-binding capacities of 1L2 and 1S2 with those of the wild-type TAP chains by performing nucleotide-binding assays with ATP- and ADP-agaroses. As can be seen from the analysis in Fig. 4C (right panel), both chimeras 1L2 and 1S2 appear to have the same ATP- and ADP-binding behavior as wild-type TAP1, indicating that the sequence information of the 6/10-loop of TAP1 alone has the capacity to confer NBD1-like nucleotide-binding properties upon NBD2.
To analyze whether the chimeric TAP chains 1L2 and 1S2 acquire the functional properties of wild-type TAP1, we created two TAP variants in T2 cells by coexpressing 1L2 or 1S2 with wild-type TAP2. Expression levels of two variants, TAP2-1L2 and TAP2-1S2, were comparable to those of wild-type TAP (Fig. 5A, left panel) and showed a balanced subunit assembly (Fig. 5A, right panel, upper part). Photo-cross-linking of peptide substrates and 8-azido-ATP was performed on membrane preparations of the T2 transfectants and assessed for labeling of TAP polypeptides as described above (see Fig. 3). For the resting states of both chimeric transporters TAP2-1L2 and TAP2-1S2, we observed peptide- (Fig. 5A, right panel, middle part) and ATP-binding properties (Fig. 5A, right panel, bottom part) indistinguishable from those of the functional wild-type transporter molecules. Furthermore, peptide transport assays (Fig. 5B, left panel) revealed that peptide translocation activity was clearly detectable for TAP2-1S2 (35–40% of wild-type TAP), consistent with the elevated HLA-B5 surface expression data for the corresponding T2 transfectant (Fig. 5B, right panel). In contrast, however, the variant TAP2-1L2 was completely transport-inactive (Fig. 5B, left and right panels). In view of the different biochemical and functional behaviors of 1D2 and 1E2 (Figs. 2 and 3), our experiments on TAP2-1L2 (Fig. 5) suggest that the 6/10-loop of TAP1 is necessary, but clearly not sufficient, to confer TAP-NBD1-like function upon TAP-NBD2. Furthermore, the functional phenotype of the chimera 1S2 (Fig. 5) and the nonfunctional phenotype of the chimera 1E2 (Figs. 2 and 3) indicate that the switch region of TAP-NBD1 alone is also not sufficient to alter the functional character of NBD2.
FIGURE 5. Functional characterization of chimeric transporter variants 2-1L2 and 2-1S2. A, Expression levels of wild-type and chimeric transporters (left panel). Cells were lysed in 1% Triton X-100. Lysates of 5 x 104 cells were separated by 7.5% SDS-PAGE and blotted onto nitrocellulose membranes. Immunoblots were probed for TAP chains using antisera D90 (C-term. NBD1) and 116/5 (C-term. NBD2). Complex formation of wild-type and chimeric TAP transporters (right panel, top). Transfected T2 cells were lysed in 1% Triton X-100, and TAP molecules were immunoisolated with anti-TAP2 antisera 116/5 (wild-type TAP) or 1p3 (TAP2-1L2 and TAP2-1S2). After washing, the bound proteins were eluted with SDS, separated on a 7.5% SDS-gel, and analyzed by probing Western blots for TAP-NBD1 (D90) or TAP-NBD2 (116/5). Peptide-binding properties (right panel, middle). The peptide-binding activities of the TAP variants were analyzed by substrate cross-linking. Microsomal fractions were resuspended in binding buffer and incubated with 1 μM radioiodinated and HSAB-conjugated peptide S8. After cross-linking, membranes were lysed, and TAP was immunoisolated with anti-TAP2 antiserum (116/5). The nucleotide-binding properties of chimeric TAP molecules are shown (right panel, bottom). Membrane fractions of T2 cells expressing wild-type or chimeric transporters were lysed in lysis buffer containing 2 μM radiolabeled 8-azido-ATP. After UV cross-linking, TAP variants were immunoprecipitated with an anti-TAP2 antiserum and separated by SDS-PAGE. B, TAP-mediated peptide transport (left panel). Transfected and nontransfected T2 cells were permeabilized with streptolysin O and incubated in transport buffer containing 10 mM ATP and radioiodinated peptide S8 for 10 min at 37°C. Bar graphs show the recovered amount of transported labeled peptides as cpm and represent the average values of experiments performed in duplicate. Surface expression of MHC class I molecules (right panel). Cells were incubated with mAb 4E that recognizes HLA-B5, followed by FITC-labeled secondary Ab (). Mean fluorescence intensity values are indicated. Background staining was determined by incubating only with secondary Ab (). TAP variants are shown schematically.
We conclude that the nonconserved 6/10-loop and the switch region build a minimal functional unit that controls different aspects of the distinctive behavior of the TAP-NBDs. In the resting state of TAP, sequence differences in the 6/10-loop determine the nonsynonymous nucleotide binding of NBD1 and NBD2, whereas the two different switch regions are probably required to initiate the translocation process upon nucleotide and peptide binding.
Asymmetrical sequence character of the switch region is critical for the functional cross-talk between the TAP domains
In line with our data presented above (Figs. 4 and 5), studies of different ABC transporters have suggested that the switch region is not implicated in nucleotide binding, but could play an important role in transducing information about different conformational states between the structural domains (36). In the case of TAP, it has been demonstrated that peptide binding has an allosteric effect on the ATP-binding capacity (6) and stimulates ATPase activity in a functional peptide transporter (37). The differences in the transport activity of TAP2-1L2 and TAP2-1S2 (Fig. 5) suggest that the nonhomologous sequences of the switch regions in TAP1 and TAP2 might be important for the correct functional interplay between the TAP domains upon peptide binding. To investigate this, we first compared the intrinsic ATP-binding capacities of the chimeras TAP2-1L2 and TAP2-1S2 with that of the wild-type transporter by affinity chromatography using ATP-agarose and elution of bound polypeptides with increasing concentrations of free MgATP (0–10 mM). As shown in Fig. 6A, both chimeric transporters were as efficiently eluted by MgATP as was the wild-type transporter, with a 70% release at 1 mM MgATP. This shows that both chimeric transporter variants have intrinsic ATP affinities similar to that of wild-type TAP. Based on the observation by Karttunen et al. (6) that peptide binding to wild-type TAP stimulates the ATP-binding capacity of the NBDs, we compared the ATP-binding behaviors of wild-type and the chimeric transporters by nucleotide affinity chromatography in the presence of increasing concentrations of peptide substrate. In confirmation of previous findings (6), the wild-type transporter showed a clear peptide-induced ATP binding with a 2- to 3-fold increase at 10 μM S8 peptide (TVDNKTRYR, in the single-letter amino acid code). A similar increase in ATP-binding behavior could be observed for the transport active variant TAP2-1S2. In contrast, however, in the case of the inactive variant TAP2-1L2, no peptide-induced ATP binding was observed (Fig. 6B), suggesting that the block of peptide transport of TAP2-1L2 is due to a defect in the functional interplay between peptide- and nucleotide-binding sites.
FIGURE 6. Functional cross-talk between the structural TAP domains requires the asymmetrical sequence character of the switch regions. A, Purified membrane fractions of T2 transfectants expressing wild-type TAP or chimeric transporters (TAP2-1L2 and TAP2-1S2) were resuspended in a Tris-buffered lysis buffer containing 2% CHAPS and incubated with ATP-agarose in the presence of 3 mM MgCl2. Bound TAP chains were eluted with increasing concentrations (0–10 mM) of MgATP. The nucleotide matrix and the eluted fractions were analyzed by probing Western blots for TAP1 (D90) and TAP2 (116/5). B, Lysed cell membranes of T2(TAPwt), T2(2-1L2), and T2(2-1S2) were incubated with ATP-agaroses in the presence of 3 mM MgCl2 and increasing concentrations (0–10 μM) of the TAP peptide substrate S8. Bound proteins were eluted with SDS-sample buffer and analyzed by probing Western blots for TAP1 (D90) and TAP2 (116/5). The bands of both TAP subunits in the ECL-fluorographs of A and B were quantified by densitometric scanning, and the peak integrals obtained were plotted in arbitrary units. C, Peptide-dependent cross-linking of TAP. Detergent-lysed membranes from transfected T2 cells were incubated in the presence or the absence of 15 μM S8 peptide for 1 h at 4°C before chemical cross-linking with EGS (left panel). Cross-linked membrane lysates were resolved on an SDS gel and immunoblotted with the antisera against the C termini of TAP1 (D90) and TAP2 (116/5). The EGS-cross-linked products of TAP (indicated by arrowhead) were quantified by densitometric analysis (right panel).
It has been previously demonstrated that peptide-induced conformational changes in TAP can be visualized by the cross-linking agent EGS (38). Peptide binding leads to conformational changes in the solvent-exposed portions of TAP and facilitates intramolecular cross-linking between TAP1 and TAP2 (6, 38, 39). To determine whether the variants TAP2-1L2 and TAP2-1S2 differ from each other in their conformational behavior upon peptide binding, cell membranes of T2 transfectants expressing wild-type TAP, TAP2-1L2, and TAP2-1S2 were incubated with EGS in the presence and the absence of S8 peptide (Fig. 6C). In the absence of S8-peptide, the same kind of limited TAP cross-linking could be observed for all transporter variants (Fig. 6C). However, in contrast to wild-type TAP and variant TAP2-1S2, no peptide-enhanced formation of the TAP cross-link product could be observed for variant TAP2-1L2. This suggests that the interdomain cross-talk is compromised in this chimeric transporter variant.
Discussion
The two NBDs of the antigenic peptide transporter TAP have different nucleotide-binding properties and seem to control different steps in the ATP-driven peptide transport cycle (4, 5, 6, 8). Several findings suggest that the catalytic function of TAP-NBD2 controls the peptide-binding properties of TAP, whereas ATP-hydrolysis in TAP-NBD1 accompanies the translocation of peptide from the cytosol into the lumen of the ER (6, 8, 9). Our previous studies have demonstrated that the distinct nucleotide-binding properties and functions of the TAP-NBDs are mainly determined by the nonhomologous C-terminal tails of the two TAP chains (11).
To gain insight into how the C-terminal tails regulate the activity of antigenic peptide transporter TAP, we have searched for discrete sequence elements that determine the distinct biochemical behaviors of NBD1 and NBD2. We approached this problem by systematically exchanging the sequences of TAP1 and TAP2 to create different chimeric TAP chains (Figs. 1 and 4A). We have been able to identify two sequence elements in the nonhomologous C-terminal tails, namely the 6/10-loop and the switch region, that are of critical importance for the distinct functions of the two TAP-NBDs.
The analysis of the chimeric TAP chains 1D2, 1E2, 1F2, and 1G2 (Figs. 2 and 3) and the results obtained for chimeras 1L2 and 1S2 (Figs. 4 and 5) clearly demonstrate that the nonhomologous 6/10-loop is the crucial determinant that controls the distinct ATP-binding behaviors of the two TAP-NBDs. Structural analysis of various ABC transporters and their NBDs (18, 40, 41, 42) showed that the 6/10-loop lies adjacent to the switch region and is in close proximity to the Walker B motif (18, 40, 41). It has been previously shown for the TAP-related P-glycoprotein that a region in NBD1 containing the sequence between 6 and 10 cannot be substituted by the corresponding sequence stretch of NBD2 without affecting transport function (43). Thus, this NBD region seems to be important for molecular processes that determine the functional separation of the two NBDs of P-glycoprotein (44, 45). Furthermore, residues in the 6/10-loop region also seem to be involved in controlling the proper NBD function of bacterial ABC transporters (46). However, there is no indication from structural or biochemical studies that the 6/10-loop region makes direct contact with bound ATP and/or ADP (18, 40, 41, 42). Thus, it is tempting to speculate that the conformational constraints imposed by the different 6/10-loops might affect the access of nucleotides to the TAP-NBDs. The 6/10-loops could participate in structural deviations of the two TAP-NBDs that are critical for the different nucleotide-binding capacities of TAP-NBD1 and -NBD2. In support of this idea, the biochemical studies of several groups have suggested that NBD1 and NBD2 of TAP have similar nucleotide-binding affinities, but differ in their ATP accessibility (6, 47). As mentioned above, a characteristic feature of the 6/10-loop regions in TAP-NBD1 and -NBD2 is their striking difference in sequence and length (Fig. 1). This is not only true for the rat TAP subunits, but also for the TAP chains of other species. Although caution is required when assessing the significance of three-dimensional structure modeling, our data for TAP-NBD2 (Fig. 4A), based on the crystal structure of the human TAP-NBD1 (18), indicate that the 6/10-loop may be one of the most significant structural differences between the two TAP-NBDs. This could reflect a structural adaptation to the requirements of nonsynonymous ATP binding in TAP1 and TAP2. Most interestingly, Karttunen et al. (6) have provided experimental evidence that in the resting state of TAP, the conformation of TAP-NBD2 is in a structurally more closed state, whereas the nucleotide-binding site of TAP1 seems to be open and solvent-accessible. In line with this, preliminary studies by our own group indicate that the 6/10-loop contributes to the regulation of nucleotide binding in a conformational, rather than a sequence-specific, manner (R. M. Leonhardt and M. R. Knittler, unpublished observation). Comparison of crystal structures of the NBDs of prokaryotic ABC transporters has revealed that the sequence between 6 and 10 connects two structurally mobile regions, the D-loop and the switch region, that seem to control NBD conformation in a dynamic and nucleotide-dependent manner (42). To elucidate the structure/function relationship of the 6/10-loops in TAP, information on the tertiary structure of both TAP-NBDs will be essential. Attempts to purify and solve the structure of TAP-NBD2 are being conducted by our laboratory.
Recent studies on the ABC protein SUR have identified a sequence in the 11/12 region of the C-terminal tails that determines the different nucleotide-binding and functional properties of that protein and its subtypes (16). It is thought that the functional role of the NBDs of the ABC protein SUR differs from that of ABC transporters such as TAP, which, in contrast to SUR, use nucleotides to pump substrates across cellular membranes (48). In line with this idea, our present findings suggest that the regulation of NBD function by the C terminus of SUR and TAP is based on a different molecular mechanism. The studies shown in Figs. 2B and 3A indicate that sequence differences in the 11/12 regions of the TAP chains make no contribution to the distinctive nucleotide-binding behaviors of the TAP-NBDs.
The crucial finding of our work on the TAP chain chimeras 1L2 and 1S2 (Figs. 4 and 5) is that the functional separation of the TAP-NBDs is essentially controlled by two adjacent sequence elements: the 6/10-loop and the switch region. Our experiments shown in Figs. 5 and 6 suggest that sequence differences in the switch region are not responsible for the distinct nucleotide-binding behaviors of TAP1 and TAP2 (Fig. 5), but seem to be required for the correct transfer of conformational signals between peptide-binding site and NBDs (Fig. 6). Karttunen et al. (6) have previously demonstrated that peptide binding to TAP induces conformational changes in the NBDs, favoring the binding of ATP. Similar observations of substrate-enhanced ATP binding have been made for other ABC transporters (49, 50). The functional TAP variant 2-1S2 shows the same peptide-enhanced ATP binding as wild-type TAP, whereas almost no peptide effect was observed for TAP2-1L2 (Fig. 6B), containing two identical NBD2 switch regions. A comparison of the TAP variants 2-1L2 and 2-1S2 (Fig. 6) indicates that peptide-induced ATP binding depends on the functional behavior of the TAP1-like subunit. This supports the hypothesis that ATP binding and hydrolysis by TAP-NBD1 accompany the transport cycle after peptide binding to TAP (4, 11). Because TAP2-1L2 can bind peptide substrates (Fig. 5A) and has normal ATP-binding behavior in the absence of additional substrates (Fig. 6A), it is likely to be folded correctly. It is reasonable to assume that the defect in the peptide-enhanced ATP binding in TAP2-1L2 results from a block in the conformational signaling between the structural domains of TAP. In support of this view, we could show that peptide binding to wild-type TAP and TAP2-1S2, but not to the variant TAP2-1L2, brings solvent-exposed portions of the TAP domains together and facilitates EGS cross-linking of the subunits (Fig. 6C). Although the exact sites of EGS cross-linking in TAP remain to be determined, our findings presented in Fig. 6 suggest that the presence of distinct switch regions of TAP1 and TAP2 is indispensable for the normal functional interplay of TAP domains and subunits.
It is well established that substrate binding by TMDs of ABC transporters triggers conformational rearrangements in the NBDs (51). However, the mechanisms of interdomain communication are still speculative. Unfortunately, because none of the present ABC transporter structures (52, 53) was determined with a bound substrate, they provide only limited information on the conformational interplay between substrate and nucleotide-binding sites. However, the existing data for bacterial ABC transporters indicate that a reorientation of the NBDs is required to stabilize the binding of ATP (51). The mutual conformational cross-talk between the TMDs and the ATP-binding sites seems to be mediated and controlled by the -helical subdomains of the NBDs (54), which contain the conserved and conformationally mobile Q-loop. Recent findings provide evidence that the structural changes in the -helical subdomain are coupled to a repositioning of sequences (42) that are probably involved in the formation of the NBD dimer interface (53). These sequence elements include the D loop and the switch region. Thus, the enhancement of ATP binding upon peptide binding could result from stabilization of the TAP structure in a particular conformation or, more specifically, could be due to altered NBD-NBD interactions that enhance nucleotide binding. In view of the phenotype of TAP2-1L2, it is interesting to note that mutations in the switch region of the bacterial ABC-transporter HisQMP2 affect not only the catalytic activity, but also the normal interdomain communication between paired NBDs (55).
It is notable that TAP2 contains the consensus sequence of the switch region, whereas TAP1 has a glutamine in place of the conserved histidine found in most pro- and eukaryotic ABC transporters. However, despite our experimental efforts, to date we have been unable to identify the discrete amino acid residues responsible for the functional nonequivalence of the switch regions in rat TAP1 and rat TAP2, suggesting that the asymmetrical character of the switch regions might be not confined to a single nonhomologous residue. Clearly, additional studies are needed to characterize the precise role(s) of the switch region in the interdomain signaling of the two TAP chains.
Our previous (11) and present experiments (Figs. 4–6) demonstrate that sequence differences within the core NBDs are not essential for the functional asymmetry of the TAP-NBDs. However, this does not exclude the possibility that differences between the sequence elements of the ATP-binding cassettes could participate to some extent in the nonsynonymous functions of TAP1 and TAP2. A clear and important finding of our present work is that functional separation of TAP-NBD1 and TAP-NBD2 depends on more than one distinct sequence element in the NBDs. Recent experiments on the TAP-related tandem ABC transporter multidrug-resistant protein 1 (MRP1) (56) suggest that sequence differences at the C-terminal end of the Walker B motifs (aspartate or glutamate, respectively, which are also present in TAP1 (LILLD) and TAP2 (LILLE), contribute to the functional divergence of the NBDs. In this context it should be noted that MRP1 (49, 57) is not only similar to TAP in its asymmetrical NBD function, but also has 6/10-loops of different length and similar signature motifs and switch regions. It will be interesting to investigate whether the NBD sequences of MRP1 can functionally substitute for sequences in the core NBDs and the C-terminal tails of TAP.
Acknowledgments
We are indebted to Prof. Dr. Jonathan Howard for his helpful comments on the manuscript. We thank Dr. Robert Wilson for critical reading the manuscript, and Dr. Frank Momburg for the gift of the anti-TAP1 antiserum 1p3.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was supported by the Land Nordrhein-Westfalen through the University of Cologne and the Deutsche Forschungsgemeinschaft (Grant KN541-1/1).
2 Address correspondence and reprint requests to Dr. Michael R. Knittler, Institute for Genetics, University of Cologne, Zülpicher Strasse 47, 50674 Cologne, Germany. E-mail address: knittler@uni-koeln.de
3 Abbreviations used in this paper: ER, endoplasmic reticulum; ABC, ATP-binding cassette; EGS, ethylene glycol bissuccinimidyl succinate; NBD, nucleotide-binding domain; TMD, transmembrane domain; SUR, sulfonylurea receptor; HSAB, N-hydroxysuccinamide-4-azidobenzoate; MRP, multidrug-resistant protein.
Received for publication July 21, 2004. Accepted for publication October 22, 2004.
References
Yewdell, J. W., E. Reits, J. Neefjes. 2003. Making sense of mass destruction: quantitating MHC class I antigen presentation. Nat. Rev. Immunol. 3:952
Momburg, F., G. J. H?mmerling. 1998. Generation and TAP-mediated transport of peptides for major histocompatibility complex class I molecules. Adv. Immunol. 68:192.
Lankat-Buttgereit, B., R. Tampé. 1999. The transporter associated with antigen processing TAP: structure and function. FEBS Lett. 464:108.
van Endert, P., L. Saveanu, E. Hewitt, P. J. Lehner. 2002. Powering the peptide pump: TAP crosstalk with energetic nucleotides. Trends Biochem. Sci. 27:454
Alberts, P., O. Daumke, E. D. Deverson, J. C. Howard, M. R. Knittler. 2001. Distinct functional properties of the TAP subunits coordinate the nucleotide dependent transport cycle. Curr. Biol. 11:242.
Karttunen, J. T., P. J. Lehner, S. S. Gupta, E. W. Hewitt, P. Cresswell. 2001. Distinct functions and cooperative interaction of the subunits of the transporter associated with antigen processing (TAP). Proc. Natl. Acad. Sci. USA 98:7431.
Hewitt, E. W., S. S. Gupta, P. J. Lehner. 2001. The human cytomegalovirus gene product US6 inhibits ATP binding by TAP. EMBO J. 20:387.
Saveanu, L., S. Daniel, P. M. van Endert. 2001. Distinct functions of the ATP binding cassettes of transporters associated with antigen processing: a mutational analysis of Walker A and B sequences. J. Biol. Chem. 276:22107
Daumke, O., M. R. Knittler. 2001. Functional asymmetry of the ATP-binding-cassettes of the ABC transporter TAP is determined by intrinsic properties of the nucleotide binding domains. Eur. J. Biochem. 268:4776.
Arora, S., P. E. Lapinski, M. Raghavan. 2001. Use of chimeric proteins to investigate the role of transporter associated with antigen processing (TAP) structural domains in peptide binding and translocation. Proc. Natl. Acad. Sci. USA 98:7241.
Bouabe, H., M. R. Knittler. 2003. The distinct nucleotide binding states of the transporter associated with antigen processing (TAP) are regulated by the nonhomologous C-terminal tails of TAP1 and TAP2. Eur. J. Biochem. 270:4531
Currier, S. J., K. Ueda, M. C. Willingham, I. Pastan, M. M. Gottesman. 1989. Deletion and insertion mutants of the multidrug transporter. J. Biol. Chem. 264:14376.
Haardt, M., M. Benharouga, D. Lechardeur, N. Kartner, G. L. Lukacs. 1999. C-terminal truncations destabilize the cystic fibrosis transmembrane conductance regulator without impairing its biogenesis: a novel class of mutation. J. Biol. Chem. 274:21873.
Sharma, N., A. Crane, J. P. t. Clement, G. Gonzalez, A. P. Babenko, J. Bryan, L. Aguilar-Bryan. 1999. The C-terminus of SUR1 is required for trafficking of KATP channels. J. Biol. Chem. 274:20628.
Booth, C. L., L. Pulaski, M. M. Gottesman, I. Pastan. 2000. Analysis of the properties of the N-terminal nucleotide-binding domain of human P-glycoprotein. Biochemistry 39:5518
Matsushita, K., K. Kinoshita, T. Matsuoka, A. Fujita, T. Fujikado, Y. Tano, H. Nakamura, Y. Kurachi. 2002. Intramolecular interaction of SUR2 subtypes for intracellular ADP-induced differential control of K(ATP) channels. Circ. Res. 90:554.
Yamaguchi, Y., H. Iseoka, A. Kobayashi, M. Maeda. 2004. The carboxyl terminal sequence of rat transporter associated with antigen processing (TAP)-like (ABCB9) is heterogeneous due to splicing of its mRNA. Biol. Pharm. Bull. 27:100
Gaudet, R., D. C. Wiley. 2001. Structure of the ABC ATPase domain of human TAP1, the transporter associated with antigen processing. EMBO J. 20:4964
DeMars, R., C. C. Chang, S. Shaw, P. J. Reitnauer, P. M. Sondel. 1984. Homozygous deletions that simultaneously eliminate expressions of class I and class II antigens of EBV-tansformed B-lymphoblastoid cells. I. Reduced proliferance responses of autologous and allogeneic T cells that have decreased expression of class II antigens. Hum. Immunol. 11:77.
Momburg, F., V. Ortiz-Navarrete, J. Neefjes, E. Goulmy, Y. van de Wal, H. Spits, S. J. Powis, G. W. Butcher, J. C. Howard, P. Walden, et al 1992. Proteasome subunits encoded by the major histocompatibility complex are not essential for antigen presentation. Nature 360:174.
Deverson, E. V., I. R. Gow, W. J. Coadwell, J. J. Monaco, G. W. Butcher, J. C. Howard. 1990. MHC class II region encoding proteins related to the multidrug resistance family of transmembrane transporters. Nature 348:738.
Gunning, P., J. Leavitt, G. Muscat, S. Y. Ng, L. Kedes. 1987. A human -actin expression vector system directs high-level accumulation of antisense transcripts. Proc. Natl. Acad. Sci. USA 84:4831.
Powis, S. J., A. R. Townsend, E. V. Deverson, J. Bastin, G. W. Butcher, J. C. Howard. 1991. Restoration of antigen presentation to the mutant cell line RMA-S by an MHC-linked transporter. Nature 354:528.
Knittler, M. R., K. Gülow, A. Seelig, J. C. Howard. 1998. MHC class I molecules compete in the endoplasmic reticulum for access to transporter associated with antigen processing. J. Immunol. 161:5967.
Nijenhuis, M., S. Schmitt, E. A. Armandola, R. Obst, J. Brunner, G. J. H?mmerling. 1996. Identification of a contact region for peptide on the TAP1 chain of the transporter associated with antigen processing. J. Immunol. 156:2186.
Lindahl, E., B. Hess, D. van der Spoel. 2001. GROMACS 3.0: a package for molecular simulation and trajectory analysis. J. Mol. Mod. 7:306.
Schneider, E., S. Hunke. 1998. ATP-binding-cassette (ABC) transport systems: functional and structural aspects of the ATP-hydrolyzing subunits/domains. FEMS Microbiol. Rev. 22:1.
Chen, M., R. Abele, R. Tampé. 2003. Peptides induce ATP hydrolysis at both subunits of the transporter associated with antigen processing. J. Biol. Chem. 278:29686
Lacaille, V. G., M. J. Androlewicz. 1998. Herpes simplex virus inhibitor ICP 47 destabilizes the transporter associated with antigen processing (TAP) heterodimer. J. Biol. Chem. 273:17386.
Antoniou, A. N., S. Ford, E. S. Pilley, N. Blake, S. J. Powis. 2002. Interactions formed by individually expressed TAP1 and TAP2 polypeptide subunits. Immunology 106:182.
Hung, L.-W., I. X. Wang, K. Nikaido, P.-Q. Liu, G. Ferro-Luzzi Ames, S.-H. Kim. 1998. Crystal structure of the ATP-binding subunit of an ABC-transporter. Nature 396:703.
Diederichs, K., J. Diez, G. Greller, C. Muller, J. Breed, C. Schnell, C. Vonrhein, W. Boos, W. Welte. 2000. Crystal structure of MalK, the ATPase subunit of the trehalose/maltose ABC transporter of the archaeon Thermococcus litoralis. EMBO J. 19:5951.
Karpowich, N., O. Martsinkevich, L. Millen, Y. R. Yuan, P. L. Dai, K. MacVey, P. J. Thomas, J. F. Hunt. 2001. Crystal structures of the MJ1267 ATP binding cassette reveal an induced-fit effect at the ATPase active site of an ABC transporter. Structure 9:571.
Beaudet, L., P. Gros. 1995. Functional dissection of P-glycoprotein nucleotide-binding domains in chimeric and mutant proteins. Modulation of drug resistance profiles. J. Biol. Chem. 270:17159.
Hrycyna, C. A., M. Ramachandra, U. A. Germann, W. C. Cheng, I. Pastan, M. M. Gottesmann. 1999. Both ATP sites of human P-glycoprotein are essential but not symmetric. Biochemistry 38:13887.
Sauna, Z. E., S. V. Ambudkar. 2001. Characterization of the catalytic cycle of ATP hydrolysis by human P-glycoprotein: the two ATP hydrolysis events in a single catalytic cycle are kinetically similar but affect different functional outcomes. J. Biol. Chem. 276:11653.(Sarah Ehses, Ralf M. Leon)
TAP delivers antigenic peptides into the endoplasmic reticulum (ER) that are subsequently bound by MHC class I molecules. TAP consists of two subunits (TAP1 and TAP2), each with a transmembrane (TMD) and a nucleotide-binding (NBD) domain. The two TAP-NBDs have distinct biochemical properties and control different steps during the peptide translocation process. We noted previously that the nonhomologous C-terminal tails of rat TAP1 and TAP2 determine the distinct functions of TAP-NBD1 and -NBD2. To identify the sequence elements responsible for the asymmetrical NBD function, we constructed chimeric rat TAP variants in which we systematically exchanged sequence regions of different length between the two TAP-NBDs. Our fine-mapping studies demonstrate that a nonhomologous region containing the 6/10-loop in conjunction with the downstream switch region is directly responsible for the functional separation of the TAP-NBDs. The 6/10-loop determines the nonsynonymous nucleotide binding of NBD1 and NBD2, whereas the switch region seems to play a critical role in regulating the functional cross-talk between the structural domains of TAP. Based on our findings, we postulate that these two sequence elements build a minimal functional unit that controls the asymmetry of the two TAP-NBDs.
Introduction
Major histocompatibility complex class I molecules are expressed on the cell surface of all nucleated cells and present peptides derived from intracellular proteins to CTLs. CTLs monitor MHC class I molecules for peptides derived from nonself proteins and eliminate infected cells. Peptides presented by MHC class I molecules are generated predominantly in the cytoplasm by the proteasome (1). Transport of antigenic peptides from the cytoplasm into the endoplasmic reticulum (ER)3 by TAP for loading onto MHC class I molecules is a key step in the Ag presentation pathway (1). TAP is a member of the ATP-binding cassette (ABC) transporter family whose members translocate different types of substrates across cellular membranes using energy from ATP hydrolysis (2). ABC transporters have a modular structure comprising two nucleotide-binding domains (NBDs) and two transmembrane domains (TMDs). Each of the two subunits of the heterodimeric TAP, TAP1 and TAP2, has an N-terminal TMD and a C-terminal NBD. The paired TMDs of TAP form the peptide-binding site and the pore through which peptide substrates are translocated, whereas the NBDs provide the energy required for peptide transport by ATP hydrolysis (3).
Major advances have recently been made in the analysis of how peptides are transported by TAP (4). The binding and hydrolysis of nucleotides are believed to power the transport process by inducing conformational changes in the NBDs that are transferred to the TMDs, and to cause the binding and movement of peptide across the membrane (5, 6). Analysis of the transport cycle of TAP led to working models in which the NBDs bind and hydrolyze nucleotides in an alternating and strongly interdependent manner (4, 5). Several groups have noted that the sequence-related NBDs of TAP1 and TAP2 fulfil distinct functions in the transport cycle of TAP (5, 6, 7, 8), and the functional asymmetry of the NBDs is an intrinsic property of these structural domains (9, 10).
We previously showed that TAP1 has a much higher ATP-binding capacity than TAP2 (9). The distinct nucleotide binding and function of the TAP-NBDs are mediated by the nonhomologous C-terminal tails of the two TAP chains and not by the core NBDs containing the conserved sequence motifs of the ATP-binding cassette (11). The functional importance of the C-terminal tails of the NBDs has been shown for several different members of the ABC transporter family (12, 13, 14, 15, 16). For example, the P-glycoprotein (12) can only tolerate small C-terminal deletions. In addition, results reported for the ABC protein sulfonylurea receptor (SUR) suggest that residues in the 11 strand of the C-terminal tail control the nucleotide-binding and catalytic functions of the NBD2 (16). Most interestingly, four different splicing isoforms of the TAP-related ABC half-transporter TAPL have recently been reported with heterogeneity of the C-terminal region, suggesting that the function of this ABC transporter may be modulated by the C-terminal tail (17).
An important contribution to understanding the transport mechanism of TAP would be the identification of sequence elements in the C-terminal tails responsible for the distinctive biochemical and functional properties of the NBDs. In this paper we report the analysis of a series of chimeric rat TAP chains in which we have systematically exchanged sequence stretches between the two NBDs. Our fine-mapping studies demonstrate that the sequence corresponding to the region between 6 and 10 (6/10-loop) and the downstream switch region (18) build a minimal functional unit within the nonhomologous C-terminal tails that determines the different properties of TAP-NBD1 and -NBD2. In particular, our data show that the asymmetrical sequence of the 6/10-loop controls the distinct nucleotide-binding activity of NBD1 and NBD2 in the resting state of TAP, whereas the differences in the switch region appear to be critical for proper transduction of conformational signals between the structural domains of TAP.
This is the first report in which the functions of the 6/10-loop and the switch region of TAP have been structurally defined. Our experimental findings provide new insights into the sequence requirements for the functional divergence of the TAP-NBDs and suggest that molecular regulatory mechanisms in TAP differ from the postulated C-terminal regulation of the ABC-protein SUR (16).
Materials and Methods
Cell lines and cell culture
T2 is a human lymphoblastoid cell line that lacks both TAP genes and expresses only the HLA-A2 and -B5 class I molecules (19). Transfectants of T2 cells containing wild-type rat TAPa (20) and TAP chimeras were cultured in IMDM (Invitrogen Life Technologies) supplemented with 10% FCS (BioWhittaker) and 1 mg/ml G418 (PAA).
Cloning and expression of chimeric TAP chains
The 2.6-kb EcoRI fragments containing full-length cDNA from rat TAP1 (21) and the chimeric chain 1N2 (9), respectively, were cloned into the multiple cloning site of pBluescript KS+ (Stratagene). For the 1G2 construct, the QuikChange site-directed mutagenesis procedure (Stratagene) was used to create a ScaI site in TAP1 at position 2016 and in 1N2 at position 2092 (position 1 is the A of the first AUG). For TAP1 we used the complementary primers 5'-GCACCCACCTGCAGTACTTGGAGAGAGGAGG-3' and 5'-CCTCCTCTCTCCAAGTACTGCAGGTGGGTGC-3', and for 1N2 we used the complementary primers 5'-GCACGACCAGTACTGGGATGAGCAGG-3' and 5'-CCTGCTCATCCCAGTACTGGTCGTGC3'. The chimeric TAP construct 1G2 was created by ligation of the 1.65-kb ScaI fragment of TAP1 to the 3.96-kb ScaI fragment of 1N2. Constructs encoding the TAP chain chimeras 1F2, 1E2, and 1D2 were created by a similar mutagenesis procedure using the TAP1 and 1N2 cDNAs as templates. ScaI sites were introduced in TAP1 at position 2066 (for 1F2), 2014 (for 1E2), and 1977 (for 1D2), respectively. In 1N2 the corresponding positions for the ScaI sites were 2051 (for 1F2), 1998 (for 1E2), and 1975 (for 1D2). For the mutagenesis approach we used the TAP1 mutagenesis primers 5'-GCGGGCCCACCAAGTACTCTTCCTCAAAGAAGGC-3' and 5'-GCCTTCTTTGAGGAAGAGTACTTGGTGGGCCCGC-3' (for 1F2), 5'-CCGAGTGGGCCTCTAGTACTGTTCTTCTGATCACC-3' and 5'-GGTGATCAGAAGAACAGTACTAGAGGCCCACTCGG-3' (for 1E2), and 5'-CCAGCTACGGGTCCAGTACTTCCTGTATGAGAGCC-3' and 5'-GGCTCTCATACAGGAAG TACTGGACCCGTAGCTGG-3' (for 1D2). In the case of the 1N2 template the complementary mutagenesis primers were 5'-GCTGACCAAGTACTGGTGCTCAAGC-3' and 5'-GCTTGAGCACC AGTACTTGGTCAGC-3' (for 1F2), 5'-CGCAGGAGGACAGTACTATGCTGGTCATTGC-3' and 5'-GCAATGACCAGCATAGTACTTGCCTCCTGCG-3' (for 1E2), and 5'-GCAGGCCCTTCAG TACTGGAGATCGCAGG-3' and 5'-CCTGCGATCTCCAGTACTGAAGGGCCTGC-3' (for 1D2). After ligation of the ScaI fragments of TAP1 (1.65, 1.7, 1.76, and 1.79 kb) with the ScaI fragments of 1N2 (3.96, 3.92, 3.87, and 3.85), we restored the original amino acid sequence by an additional site-directed mutagenesis step using the complementary primers 5'-GGAGCACGACCAGCTCATGGAGAGAGGAGG-3' and 5'-CCTCCTCTCTCCATGAGCTGGTCGTGCTCC-3' (for 1G2), 5'-GCTGACCAAGTTCTCTTCCTCAAAGAAGGC-3' and 5'-GCCTTCTTTGAGGAAGAGAACTTGGTCAGC-3' (for 1F2), 5'-CGCAGGAGGACAGGACGGTT CTTCTGATCACC-3' and 5'-GGTGATCAGAAGAACCGTCCTGTCCTCCTGCG-3' (for 1E2), and 5'-GCAGGCCCTTCAGCGGCTCCTGTATGAGAGCC-3' and 5'-GGCTCTCATACAGGAGCCGCTGAAGGGACTGC-3' (for 1D2). The chimera 1G2 encoded residues 1–505 and 706–725 of TAP1 and residues 494–685 of TAP2; the chimera 1F2 contained residues 1–505 and 690–725 of TAP1 and residues 494–672 of TAP2; the chimera 1E2 encoded residues 1–505 and 671–725 of TAP1 and residues 494–653 of TAP2; the chimera 1D2 comprised residues 1–505 and 660–725 of TAP1 and residues 494–646 of TAP2. The chimeric constructs 1N2 (9) and 1F2 (see above) were used to create the 1H2 construct. ScaI sites were created in 1N2 (position 2092) and 1F2 (position 2083) by site-directed mutagenesis with specific primers following the procedure described above. The resulting construct contained residues 1–505 and 690–705 of TAP1 and residues 494–672 and 686–703 of TAP2. To construct variant 1L2 we used the cDNA of 1D2 and 1N2 (9) as templates for site-directed mutagenesis. In 1D2 we introduced a ScaI site a position 2010 using the primers 5'-CCGAGTGGGCCTCTAGTACTGTTCTTCTGATCACC-3' and 5'-GGTGATCAGAAGAACAGTACTAGAGGCCCACTCGG-3', and in 1N2 (9) at position 1998 using the primers 5'-CGCAGGAGGACAGTACTATGCTGGTCATTGC-3' and 5'-GCAATGACCA GCATAGTACTTGCCTCCTGCG-3'. To create the construct encoding 1L2, the 3.87-kb fragment of 1D2 and the 1.75-kb fragment of 1N2 were ligated, and the original sequence was restored using the primers 5'-CCGAGTGGGCCTCTCGGACGATGCTGGTCATTGC-3' and 5'-GCAATAACCAGCATCGTCCGAGAGGCCCACTCGG-3'. The resulting construct encoded residues 1–505 and 660–670 of TAP1 and residues 494–646 and 654–703 of TAP2. The cDNAs of 1D2 and 1N2 were also used for the construct encoding 1S2. In this study we introduced a ScaI site in 1D2 at position 2063 with the primers 5'-GCGGGCCCACCAAGTACTCTTCCTCAAAGAAGGC-3' and 5'-GCCTTCTTTGAGGAAGAGTACTTGGTGGGCCCGC-3'. In 1N2 the ScaI site was created at position 2051 using the primers 5'-GCTGACCAAGTACTGGTGCTCAAGC-3' and 5'-GCTTGAGCACCAGTACTTGGTCAGC-3'. To create the 1S2 construct the 3.9-kb fragment of 1D2 and the 1.5-kb fragment of 1N2 were ligated, and the original sequence was restored with the primers 5'-GCGGGCCCACCACATCCTGGTGCACAAGC-3' and 5'-GCTTGTGCACCAGGATGTGGTGGGCCCGC-3'. The resulting construct encoded residues 1–505 and 660–689 of TAP1 and residues 494–646 and 672–703 of TAP2. All TAP constructs were cloned into the EcoRI site of pHApr1neo (22), and mutated NBD regions were sequenced in both directions. All chimeric TAP constructs were transfected into T2 cells by electroporation using a gene pulser (Bio-Rad) at 270 V and 500 μF. After selection with G418 (1 mg/ml) for 4–6 wk, stable transfectants were subcloned and screened for TAP chain expression by Western blotting.
Antibodies
116/5 is a polyclonal rabbit antiserum specific for the C terminus of rat TAP2 chains (20). D90 is a polyclonal rabbit antiserum recognizing the C terminus of rat TAP1 chains (23). MAC 394 is a monoclonal mouse anti-rat TAP2a Ab (24) derived from immunization with the recombinant histidine-tagged cytoplasmic domain of rat TAP2a. MAC 394 recognizes the polymorphic residues at positions 538 and 539 in the core NBD of rat TAP2a (11). 1p3 is a polyclonal rabbit antiserum that binds epitopes in or between the predicted TMDs of TAP1 (25). 4E is a conformation-dependent mouse mAb that recognizes an epitope common to all HLA-B and -C Ags (26).
Immunoprecipitation, Western blotting, and flow cytometry
For immunoisolation experiments, 1 x 107 cells were washed twice in ice-cold PBS (1.7 mM KH2PO4, 10 mM Na2HPO4, 140 mM NaCl, and 2.7 mM KCl, pH 7.5) before solubilization in lysis buffer (PBS (pH 7.5) containing 1% Triton X-100 (Sigma-Aldrich)). Immunoprecipitations with anti-TAP2 (116/5) and anti-TAP1 (1p3) were performed as described previously (24). Immunoprecipitates were washed with PBS/1% Triton X-100 and eluted with 100 mM Tris-HCl (pH 9) containing 0.5% SDS. Samples were analyzed by Western blotting in conjunction with a specific primary antiserum. Bands were visualized with HRP-conjugated secondary Abs (donkey anti-rabbit IgG-HRP; Amersham Biosciences) and ECL substrate (Amersham Biosciences). Flow cytometry experiments using the anti-HLA-B5-specific mAb 4E were performed as described previously (27).
Transport assay, peptide cross-linking, and nucleotide-binding assays
For transport assays, 2 x 106 cells were permeabilized with streptolysin O (2 U/ml; Murex). After washing with PBS, 0.5 μM radioiodinated peptide S8 (TVDNKTRYR, in the single-letter amino acid code), 10 mM ATP, and incubation buffer (50 mM HEPES (pH 7.5), 250 mM sucrose, 150 mM CH3COOK, 5 mM (CH3COO)2Mg · 4H2O, 1 mM DTT, 1 mM Pefabloc (Roche), and 1.8 μg/ml aprotinin (Sigma-Aldrich)) were added and incubated for 10 min at 37°C. After lysis with 20 mM Tris-HCl (pH 7.5), 500 mM NaCl, and 0.1% Nonidet P-40 (Sigma-Aldrich), transported glycosylated peptides were isolated with Con A-Sepharose (BD Biosciences) and quantitated by gamma counting (28). For peptide cross-linking, permeabilized cells were incubated with 1 μM radioiodinated and N-hydroxysuccinamide-4-azidobenzoate (HSAB)-conjugated peptide S8. Cross-linking was induced by irradiation with a UV lamp at 254 nm for 5 min on ice. Cells were lysed by adding 1% Triton X-100 in PBS. The nucleotide- binding assay was performed as described previously (27). All nucleotide binding experiments were performed with N6-coupled ATP- and ADP-agarose (Sigma-Aldrich). To analyze peptide-induced ATP binding, CHAPS-solubilized cell membranes were pretreated with increasing concentrations of S8 peptide (10, 1, 0.1, and 0.01 μM) for 30 min. Afterward the lysates were incubated with ATP-agarose for 45 min. After washing, the bound proteins were eluted with SDS loading buffer.
For photolabeling of TAP with radiolabeled 8-azido-ATP, membranes of cells were prepared and resuspended in 250 mM sucrose, 50 mM KCl, 2 mM MgCl2, 2 mM EGTA, and 10 mM Tris-HCl (pH 6.8) (27). Membranes corresponding to 3 x 106 cells in a final volume of 100 μl were incubated with 2 μM 8-azido-[-32P]ATP (ICN Biomedicals) for 5 min at 4°C. Cross-linking was induced by irradiation with a UV lamp at 254 nm for 5 min at 4°C.
Preparation of microsomal membranes
Microsomes from 1 x 108 T2 cells expressing chimeric and wild-type TAP proteins were generated by sucrose gradient fractionation (29). Cells were washed twice with ice-cold PBS, resuspended in 10 ml of 10 mM Tris-HCl (pH 7.5) with protease inhibitor mixture (Complete Protease Inhibitor; Roche), and incubated on ice for 10 min. The lysed cells were then homogenized and centrifuged at 800 x g for 5 min at 4°C. The resulting supernatants were resuspended in 5 ml of 1.3 M sucrose buffer (20 mM HEPES (pH 7.5), 25 mM CH3COOK, 5 mM (CH3COO)2Mg · 4H2O, 1 mM DTT, and protease inhibitor mix) and centrifuged again at 800 x g at 4°C for 10 min. The supernatants were then centrifuged at 68,000 x g at 4°C for 2 h, and the membrane pellets were resuspended in 800 μl of 0.25 M sucrose buffer. Subsequently, 5.6 ml of 2.5 M sucrose gradient buffer was added, and the suspension was overlaid carefully with 2.9 ml of 2 M and 2.9 ml of 1.3 M sucrose buffer. Approximately 800 μl of 0.25 M sucrose buffer was carefully loaded on the top of the gradient. The sucrose gradient was centrifuged at 100,000 x g for 16 h at 4°C. The microsomes were collected at the interface between the 2 and 1.3 M sucrose buffer, diluted in 20 mM HEPES buffer (20 mM HEPES (pH 7.5), 25 mM CH3COOK, 5 mM (CH3COO)2Mg · 4H2O, 1 mM DTT, and protease inhibitor mixture), homogenized, and centrifuged at 68,000 x g at 4°C for 1 h. The microsomal pellets were resuspended in 200 μl of 20 mM HEPES buffer. Finally, aliquots of 30–50 μl were snap-frozen in liquid nitrogen and stored at –80°C.
Chemical cross-linking
Membranes from 2 x 106 cells were resuspended in 400 μl of PBS containing 1% digitonin at 4°C, and the homobifunctional cross-linker ethylene glycol bissuccinimidyl succinate (EGS; Pierce) was added to a final concentration of 3 mM. The primary targets for EGS are -amino groups of solvent-exposed lysine residues. After incubation for 30 min at 4°C, the lysed membranes were washed with PBS containing 1% digitonin. The nonameric TAP-binding peptide S8 was added to a final concentration of 15 μM for 1 h at 4°C before EGS cross-linking.
Computational homology modeling of the NBD of rat TAP2
Model building was performed by a conventional three-step procedure. Step 1 was alignment. A global multialignment was obtained with the help of vector NTI using a modified version of CLUSTAL V (30). The sequence of human TAP1 (accession no. Q03518) was read directly from the crystal structure (18). Sequence homologies among human TAP1 and rat TAP2 (accession no. X63854) are in the range of 70%. Secondary structure predictions of the rat TAP-NBD2 sequence were performed with PHD (31) and PSIPRED v2.3 (32). Step 2 was model building. An initial model of the rat TAP-NBD2 main chain was built by searching a database of loops obtained from highly resolved protein structures (33). Loop selections were made on the basis of minimal steric interactions with the rest of the model. A full atom model of the rat TAP-NBD2 ectodomain was generated with the help of the backbone-dependent rotamer library implemented in SCWRL 3.0 (34). Step 3 was refinement of the structure model. Refinement of the rat TAP2-models was performed using the GROMACS package (35). The solvated rat TAP-NBD2-models were energy-minimized until they converged using the steepest descent method. A molecular dynamics-simulated annealing simulation (simulation time, 300 ps; temperature, 0–300 K), followed by a second energy minimization, were used to optimize the conformation of the newly introduced loops. During simulated annealing, positional constraints were applied to all main chain atoms, except atoms of newly introduced loops.
Results
Chimeric studies on the C-terminal regulation of asymmetrical nucleotide binding and NBD function in TAP
We have previously shown that the distinctive nucleotide-binding behavior of NBD1 and NBD2 in the resting state of rat TAP (9, 10) depends directly on the divergent C-terminal tails, rather than on the core NBDs situated between the conserved Walker A and B motifs (11). A chimeric rat TAP2 chain, containing the core NBD2 fused to the C-terminal segment of rat TAP1, adopts the ATP-binding behavior of wild-type NBD1, whereas a corresponding rat TAP1 chimera, containing the core NBD1 fused to the C-terminal segment of rat TAP2, shows the characteristic ADP-binding properties of wild-type NBD2 (11). Moreover, a rat TAP chain chimera, designated 1C2, comprising TMD1 (residues 1–505) and the core NBD2 (residues 494–639) with the C terminus of TAP1 (residues 653–725), acquires not only the nucleotide-binding behavior, but also the functional properties of TAP1 (11). Thus, the TAP chain chimera 1C2 provides a suitable experimental system for fine-mapping the regions within the C terminus that are responsible for the distinct nucleotide-binding properties of the TAP-NBDs.
We constructed a series of chimeric rat TAP chains based upon the variant 1C2 in which we dissected the nonhomologous C-terminal tail into four portions by progressively exchanging sequences between the two TAP-NBDs (Fig. 1). Positions of secondary structure elements in the C-terminal tails of rat TAP1 and rat TAP2 (see Fig. 1) were predicted on the basis of the crystal structure of the human TAP-NBD1 (18). The designations 1D2, 1E2, 1F2, and 1G2 indicate four different chimeric TAP chains where NBD1 regions between residues 506–659, 506–670, 506–689, and 506–705 were replaced by the corresponding NBD2 sequence: residues 494–646, 494–653, 494–672, and 494–685, respectively (Fig. 1). Because recent findings have suggested that the 11/12 region plays an important role in determining the nucleotide-binding properties of the ABC protein SUR (16), we created an additional rat TAP chain variant (1H2) from the chimera 1C2 in which the 11/12 sequence of TAP2 (residues 673–685) was replaced by the corresponding sequence stretch of TAP1 (residues 690–705; Fig. 1). All the constructs contained TMD1 as well as core NBD2 and only differed in their C-terminal sequences. After stable transfection of the different TAP chimeras into the TAP-negative human cell line T2, we tested the steady state expression of each TAP chain by immunoblot analysis using different TAP-specific Abs (see Materials and Methods). As shown in Fig. 2A, all chimeric TAP chains were stably expressed at similar levels.
FIGURE 1. Schematic diagram of wild-type and chimeric TAP chains. Sequences of the nonhomologous C-terminal tails of rat TAP-NBD1 and -NBD2 were retrieved from the GenBank database (GenBank accession no. X57523 and X63854) and aligned using the software Vector NTI (InforMax; Invitrogen Life Technologies). , Amino acid residues that are identical in both NBDs; , similar residues. Below, schematic diagrams of the wild-type TAP subunits and the chimeric TAP constructs 1C2, 1D2, 1E2, 1F2, 1G2, 1H2, and 1N2 are shown. The location of the switch region (switch, residues 671–689 in rat TAP1, residues 654–672 in rat TAP2) as well as the different positions of -helices 6 (residues 653–659 in rat TAP1, residues 640–646 in rat TAP2), 7 (residues 680–684 in rat TAP1, residues 663–667 in rat TAP2), 8 (residues 703–709 in rat TAP1, residues 683–689 in rat TAP2), 9 (residues 712–717 in rat TAP1, residues 692–697 in rat TAP2), -sheets 10 (residues 671–676 in rat TAP1, residues 654–659 in rat TAP2), 11 (residues 688–693 in rat TAP1, residues 671–676 in rat TAP2), 12 (residues 696–701 in rat TAP1, residues 679–782 in rat TAP2), and connecting loop structures, predicted on the basis of the crystal structure of the human TAP-NBD1 (18 ), are shown. TAP1 sequences in the different chimeras are indicated in black, whereas the corresponding sequences of TAP2 are indicated in gray.
FIGURE 2. Stable expression and biochemical characteristics of chimeric TAP variants. A, Expression levels and schematic overview of wild-type and chimeric TAP subunits with altered NBD segments. T2 transfectants were lysed in buffer containing 1% Triton X-100. Cell lysates of 5 x 104 cells were separated by 7.5% SDS-PAGE and blotted onto nitrocellulose membranes. Immunoblots were probed for the different TAP chains using antisera D90 (C-term. NBD1), 116/5 (C-term. NBD2), and Ab MAC 394 (core NBD2). A pictorial overview of wild-type TAP and the different chimeric TAP subunits 1N2, 1C2, 1D2, 1E2, 1F2, 1G2, and 1H2 is indicated at the bottom of the analysis. B, Nucleotide-binding properties of wild-type and chimeric TAP chains. ER membrane fractions of T2 transfectants were resuspended in lysis buffer containing 1% Triton X-100 and incubated with ATP- and ADP-agaroses in the presence of 3 mM MgCl2. After washing the nucleotide-agaroses, bound proteins were eluted with SDS-sample buffer and analyzed by probing Western blots for the C-terminal tail of TAP1 (antiserum D90) or the C-terminal tail of TAP2 (antiserum 116/5). TAP chimeras are indicated schematically on the left.
To investigate whether the progressive exchange of sequence between the two TAP-NBDs has an effect on the nucleotide-binding properties of the chimeric TAP chains, lysates of ER membrane preparations were analyzed by standard nucleotide-binding assays (6, 24) for binding to ATP- and ADP-agarose beads (Fig. 2B). The binding to the different nucleotide-agaroses was compared with that of wild-type TAP subunits, the chimera 1C2 (11), and chimera 1N2, which comprises TMD1 and the entire sequence of NBD2 (9) (Fig. 1). Analysis of TAP1, TAP2, 1C2, and 1N2 (Fig. 2B, left panel) confirmed that the distinct nucleotide binding of the TAP subunits is an inherent property of the NBDs that is determined by the C-terminal tails (11). The variant 1C2 bound in a TAP1-like manner to ATP- as well as ADP-agarose (11), whereas 1N2 interacted only with ADP-agarose, which is a characteristic property of TAP2 (5). Analysis of the other chimeric TAP chains showed that 1D2 bound to ATP- as well as ADP-agarose, whereas the chimeric polypeptides 1E2, 1F2, 1G2, and 1H2 bound only to ADP-agarose (Fig. 2B, right panel).
To explore whether the TAP chain variants can participate as a TAP1-type chain in a functional heterodimeric peptide transporter, we created five different TAP molecules in T2 cells by coexpressing the chimeric variants 1D2, 1E2, 1F2, 1G2, and 1H2 in combination with wild-type TAP2. The expression levels of these TAP variants were comparable to those of wild-type transporters (Fig. 3A, left panel) and showed normal subunit assembly (Fig. 3A, right panel, upper part).
FIGURE 3. Functional properties of chimeric transporters 2-1D2, 2-1E2, 2-1F2, 2-1G2, and 2-1H2. A, Expression levels of wild-type and chimeric transporters (left panel). Cells were lysed in 1% Triton X-100. Lysates of 5 x 104 cells were separated by 7.5% SDS-PAGE and blotted onto nitrocellulose membranes. Immunoblots were probed for TAP chains using antisera D90 (C-term. NBD1) and 116/5 (C-term. NBD2). Complex formation of wild-type and chimeric TAP transporters is shown (right panel, top). Transfected T2 cells were lysed in 1% Triton X-100, and TAP molecules were immunoisolated with anti-TAP2 antisera 116/5 (TAPwt, TAP2-1D2, TAP2-1E2, TAP2-1F2, and TAP2-1G2) or 1p3 (TAP2-1H2). After washing, the bound proteins were eluted with SDS, separated on a 7.5% SDS-gel, and analyzed by probing Western blots for TAP-NBD1 (D90) or TAP-NBD2 (116/5). Peptide-binding properties are shown (right panel, middle). The peptide-binding activity of the TAP variants was analyzed by substrate cross-linking. Microsomal fractions were resuspended in binding buffer and incubated with 1 μM radioiodinated and HSAB-conjugated peptide S8. After cross-linking, membranes were lysed, and TAP was immunoisolated with anti-TAP2 antiserum (116/5). The nucleotide-binding properties of chimeric TAP molecules are shown (right panel, bottom). Membrane fractions of T2 cells expressing wild-type or chimeric transporters were lysed in lysis buffer containing 2 μM radiolabeled 8-azido-ATP. After UV cross-linking, TAP variants were immunoprecipitated with an anti-TAP2 antiserum and separated by SDS-PAGE. B, TAP-mediated peptide transport. Transfected and nontransfected T2 cells were permeabilized with streptolysin O (SLO) and incubated in transport buffer containing 10 mM ATP and radioiodinated peptide S8 for 10 min at 37°C. Bar graphs show the recovered amount of transported labeled peptides as cpm and represent the average values of experiments performed in duplicate. C, Surface expression of MHC class I molecules. Cells were incubated with mAb 4E that recognizes HLA-B5, followed by FITC-labeled secondary Ab. Surface expression of HLA-B5 was detected by flow cytometry (). Mean fluorescence intensity values are indicated. Background staining was determined by incubating only with secondary Ab (). TAP variants are shown schematically.
For functional characterization, the different transporter variants were first analyzed for their peptide- and ATP-binding behaviors (Fig. 3A, right panel, middle and bottom parts). Binding of peptide and nucleotide was analyzed by photolabeling experiments using radioactively labeled and UV-cross-linkable reagents (5). Although all chimeric transporter variants had the capacity to bind peptides, only the variant TAP2-1D2 showed a pattern of ATP binding comparable to that of the functional wild-type transporter (5). In line with the nucleotide-binding assays shown in Fig. 2B, the ATP cross-linking experiments showed detectable ATP labeling for TAP1 in the wild-type transporter and for 1D2 in the chimeric transporter TAP2-1D2, whereas all other transporter variants showed no detectable ATP cross-linking. We measured the peptide transport activities of the different chimeric transporters in standard transport assays (28) using radioiodinated model peptides (5). As shown in Fig. 3B, only TAP2-1D2 showed a significant transport activity compared with wild-type TAP and the transporter variant 2-1C2. The transport efficiency of TAP2-1D2 was 40–50% that of wild-type TAP and 70% that of TAP2-1C2. Thus, like the chimera 1C2 (11), the chimeric polypeptide 1D2 can functionally substitute for wild-type TAP1, albeit inefficiently. In contrast, no detectable peptide transport above that observed for the nontransfected T2 cell line was seen for variants TAP2-1E2, TAP2-1F2, TAP2-1G2, and TAP2-1H2 (Fig. 3B). The different peptide transport activities of the chimeric TAP molecules were also reflected in the surface expression of mature HLA-B5 molecules determined by flow cytometry (Fig. 3C).
Taken together, our data show that the C-terminal sequence information from TAP1 present in the chimera 1D2 confers the nucleotide-binding and functional characteristics of NBD1 upon NBD2, whereas the C-terminal amino acids of NBD1 present in the chimeras 1E2, 1F2, 1G2, and 1H2 do not support an NBD1-like function. Thus, in contrast to the ABC protein SUR (16), our findings clearly demonstrate that in the case of TAP, a region upstream of the 11/12-strands must be responsible for the distinct nucleotide-binding properties of TAP1 and TAP2.
6/10-loop and switch region build a minimal functional unit that controls asymmetry of the two TAP-NBDs
The differences in nucleotide-binding and peptide transport functions of 1D2 and 1E2 (Figs. 2 and 3) suggest that the sequence stretch between the secondary structure elements 6 and 10 (6/10-loop, residues 660–670 in rat TAP1, and residues 647–653 in rat TAP2; see Fig. 1) could play a crucial role in controlling the biochemical and functional behaviors of the TAP-NBDs. Sequence alignment of rat TAP-NBD1 and rat TAP-NBD2 (Fig. 1) shows that this NBD segment is strikingly different in sequence (RLLYESPEWAS vs TWRSQED, in the single-letter amino acid code) and in length (11 residues vs seven residues) between rat TAP1 and TAP2. The same is also true for the TAP-NBDs of other species, e.g., mouse, hamster, and human. Three dimensional modeling of the rat TAP-NBD2 primary sequence using the NBD1 of human TAP (18) as a structural template (Fig. 4A) suggests that the 6/10-loop of TAP2 is also structurally divergent (RMS deviation, 5.5 ?) from that found in TAP1.
FIGURE 4. Construction, expression, and nucleotide-binding properties of TAP chimeras 1L2 and 1S2. A, Homology modeling of TAP-NBD2 (indicated in gray) based on the crystal structure of TAP-NBD1 (18 ) (indicated in black). The overlay shows a close-up of NBD structures between the Walker B motif and the switch region created with Swiss-PdbViewer (http://expasy.org/spdbv). The positions of both elements and the 6/10-loop are indicated. Arrowheads mark the points where the sequences were exchanged in chimeras 1L2 and 1S2. B, Schematic diagram of wild-type TAP chains and the two chimeric constructs (1L2 and 1S2) in which the 6/10-loop (residues 660–670 in rat TAP1 and residues 647–653 in rat TAP2) and the switch region (switch, residues 671–689 in rat TAP1, residues 654–672 in rat TAP2) were exchanged between NBD1 and NBD2. C, Expression levels of wild-type and chimeric TAP subunits with exchanged NBD segments (left panel). T2 transfectants were lysed in buffer containing 1% Triton X-100. Cell lysates of 5 x 104 cells were separated by 7.5% SDS-PAGE and blotted onto nitrocellulose membranes. Immunoblots were probed for the different TAP chains using antisera D90 (C-term. NBD1), 116/5 (C-term. NBD2), and Ab MAC 394 (core NBD2). A pictorial overview of the wild-type TAP and the chimeric TAP subunits is indicated at the bottom of the analysis. Nucleotide-binding properties of wild-type and chimeric TAP chains are shown (right panel). Membrane fractions of T2 cells expressing chimeras 1L2 and 1S2 were resuspended in lysis buffer containing 1% Triton X-100 and incubated with different nucleotide-agaroses. After extensive washing, bound proteins were eluted with SDS-sample buffer and analyzed by probing Western blots for the C-terminal tails of TAP1 (D90) or TAP2 (116/5). TAP chimeras are indicated schematically on the left.
To investigate whether the nature of the 6/10-loop controls the distinctive nucleotide-binding behavior of the TAP chains, we constructed a 1D2-related chimera (1L2) in which we exchanged only the sequence of the 6/10-loop between NBD1 and NBD2 (residues 660–670 in NBD1, residues 647–653 in NBD2; Fig. 4B). In addition, we created a chimeric construct (1S2) in which the sequence exchanged between NBD1 and NBD2 extended from the 6/10-loop to include the nonhomologous residues of the downstream switch region (residues 660–689 in TAP1, residues 647–672 in TAP2; Fig. 4B). The switch region is thought to play a crucial role in the conformational signal transduction of ABC transporter domains (36). After stable transfection of both chimeric TAP constructs into T2 cells (Fig. 4C, left panel), we compared the nucleotide-binding capacities of 1L2 and 1S2 with those of the wild-type TAP chains by performing nucleotide-binding assays with ATP- and ADP-agaroses. As can be seen from the analysis in Fig. 4C (right panel), both chimeras 1L2 and 1S2 appear to have the same ATP- and ADP-binding behavior as wild-type TAP1, indicating that the sequence information of the 6/10-loop of TAP1 alone has the capacity to confer NBD1-like nucleotide-binding properties upon NBD2.
To analyze whether the chimeric TAP chains 1L2 and 1S2 acquire the functional properties of wild-type TAP1, we created two TAP variants in T2 cells by coexpressing 1L2 or 1S2 with wild-type TAP2. Expression levels of two variants, TAP2-1L2 and TAP2-1S2, were comparable to those of wild-type TAP (Fig. 5A, left panel) and showed a balanced subunit assembly (Fig. 5A, right panel, upper part). Photo-cross-linking of peptide substrates and 8-azido-ATP was performed on membrane preparations of the T2 transfectants and assessed for labeling of TAP polypeptides as described above (see Fig. 3). For the resting states of both chimeric transporters TAP2-1L2 and TAP2-1S2, we observed peptide- (Fig. 5A, right panel, middle part) and ATP-binding properties (Fig. 5A, right panel, bottom part) indistinguishable from those of the functional wild-type transporter molecules. Furthermore, peptide transport assays (Fig. 5B, left panel) revealed that peptide translocation activity was clearly detectable for TAP2-1S2 (35–40% of wild-type TAP), consistent with the elevated HLA-B5 surface expression data for the corresponding T2 transfectant (Fig. 5B, right panel). In contrast, however, the variant TAP2-1L2 was completely transport-inactive (Fig. 5B, left and right panels). In view of the different biochemical and functional behaviors of 1D2 and 1E2 (Figs. 2 and 3), our experiments on TAP2-1L2 (Fig. 5) suggest that the 6/10-loop of TAP1 is necessary, but clearly not sufficient, to confer TAP-NBD1-like function upon TAP-NBD2. Furthermore, the functional phenotype of the chimera 1S2 (Fig. 5) and the nonfunctional phenotype of the chimera 1E2 (Figs. 2 and 3) indicate that the switch region of TAP-NBD1 alone is also not sufficient to alter the functional character of NBD2.
FIGURE 5. Functional characterization of chimeric transporter variants 2-1L2 and 2-1S2. A, Expression levels of wild-type and chimeric transporters (left panel). Cells were lysed in 1% Triton X-100. Lysates of 5 x 104 cells were separated by 7.5% SDS-PAGE and blotted onto nitrocellulose membranes. Immunoblots were probed for TAP chains using antisera D90 (C-term. NBD1) and 116/5 (C-term. NBD2). Complex formation of wild-type and chimeric TAP transporters (right panel, top). Transfected T2 cells were lysed in 1% Triton X-100, and TAP molecules were immunoisolated with anti-TAP2 antisera 116/5 (wild-type TAP) or 1p3 (TAP2-1L2 and TAP2-1S2). After washing, the bound proteins were eluted with SDS, separated on a 7.5% SDS-gel, and analyzed by probing Western blots for TAP-NBD1 (D90) or TAP-NBD2 (116/5). Peptide-binding properties (right panel, middle). The peptide-binding activities of the TAP variants were analyzed by substrate cross-linking. Microsomal fractions were resuspended in binding buffer and incubated with 1 μM radioiodinated and HSAB-conjugated peptide S8. After cross-linking, membranes were lysed, and TAP was immunoisolated with anti-TAP2 antiserum (116/5). The nucleotide-binding properties of chimeric TAP molecules are shown (right panel, bottom). Membrane fractions of T2 cells expressing wild-type or chimeric transporters were lysed in lysis buffer containing 2 μM radiolabeled 8-azido-ATP. After UV cross-linking, TAP variants were immunoprecipitated with an anti-TAP2 antiserum and separated by SDS-PAGE. B, TAP-mediated peptide transport (left panel). Transfected and nontransfected T2 cells were permeabilized with streptolysin O and incubated in transport buffer containing 10 mM ATP and radioiodinated peptide S8 for 10 min at 37°C. Bar graphs show the recovered amount of transported labeled peptides as cpm and represent the average values of experiments performed in duplicate. Surface expression of MHC class I molecules (right panel). Cells were incubated with mAb 4E that recognizes HLA-B5, followed by FITC-labeled secondary Ab (). Mean fluorescence intensity values are indicated. Background staining was determined by incubating only with secondary Ab (). TAP variants are shown schematically.
We conclude that the nonconserved 6/10-loop and the switch region build a minimal functional unit that controls different aspects of the distinctive behavior of the TAP-NBDs. In the resting state of TAP, sequence differences in the 6/10-loop determine the nonsynonymous nucleotide binding of NBD1 and NBD2, whereas the two different switch regions are probably required to initiate the translocation process upon nucleotide and peptide binding.
Asymmetrical sequence character of the switch region is critical for the functional cross-talk between the TAP domains
In line with our data presented above (Figs. 4 and 5), studies of different ABC transporters have suggested that the switch region is not implicated in nucleotide binding, but could play an important role in transducing information about different conformational states between the structural domains (36). In the case of TAP, it has been demonstrated that peptide binding has an allosteric effect on the ATP-binding capacity (6) and stimulates ATPase activity in a functional peptide transporter (37). The differences in the transport activity of TAP2-1L2 and TAP2-1S2 (Fig. 5) suggest that the nonhomologous sequences of the switch regions in TAP1 and TAP2 might be important for the correct functional interplay between the TAP domains upon peptide binding. To investigate this, we first compared the intrinsic ATP-binding capacities of the chimeras TAP2-1L2 and TAP2-1S2 with that of the wild-type transporter by affinity chromatography using ATP-agarose and elution of bound polypeptides with increasing concentrations of free MgATP (0–10 mM). As shown in Fig. 6A, both chimeric transporters were as efficiently eluted by MgATP as was the wild-type transporter, with a 70% release at 1 mM MgATP. This shows that both chimeric transporter variants have intrinsic ATP affinities similar to that of wild-type TAP. Based on the observation by Karttunen et al. (6) that peptide binding to wild-type TAP stimulates the ATP-binding capacity of the NBDs, we compared the ATP-binding behaviors of wild-type and the chimeric transporters by nucleotide affinity chromatography in the presence of increasing concentrations of peptide substrate. In confirmation of previous findings (6), the wild-type transporter showed a clear peptide-induced ATP binding with a 2- to 3-fold increase at 10 μM S8 peptide (TVDNKTRYR, in the single-letter amino acid code). A similar increase in ATP-binding behavior could be observed for the transport active variant TAP2-1S2. In contrast, however, in the case of the inactive variant TAP2-1L2, no peptide-induced ATP binding was observed (Fig. 6B), suggesting that the block of peptide transport of TAP2-1L2 is due to a defect in the functional interplay between peptide- and nucleotide-binding sites.
FIGURE 6. Functional cross-talk between the structural TAP domains requires the asymmetrical sequence character of the switch regions. A, Purified membrane fractions of T2 transfectants expressing wild-type TAP or chimeric transporters (TAP2-1L2 and TAP2-1S2) were resuspended in a Tris-buffered lysis buffer containing 2% CHAPS and incubated with ATP-agarose in the presence of 3 mM MgCl2. Bound TAP chains were eluted with increasing concentrations (0–10 mM) of MgATP. The nucleotide matrix and the eluted fractions were analyzed by probing Western blots for TAP1 (D90) and TAP2 (116/5). B, Lysed cell membranes of T2(TAPwt), T2(2-1L2), and T2(2-1S2) were incubated with ATP-agaroses in the presence of 3 mM MgCl2 and increasing concentrations (0–10 μM) of the TAP peptide substrate S8. Bound proteins were eluted with SDS-sample buffer and analyzed by probing Western blots for TAP1 (D90) and TAP2 (116/5). The bands of both TAP subunits in the ECL-fluorographs of A and B were quantified by densitometric scanning, and the peak integrals obtained were plotted in arbitrary units. C, Peptide-dependent cross-linking of TAP. Detergent-lysed membranes from transfected T2 cells were incubated in the presence or the absence of 15 μM S8 peptide for 1 h at 4°C before chemical cross-linking with EGS (left panel). Cross-linked membrane lysates were resolved on an SDS gel and immunoblotted with the antisera against the C termini of TAP1 (D90) and TAP2 (116/5). The EGS-cross-linked products of TAP (indicated by arrowhead) were quantified by densitometric analysis (right panel).
It has been previously demonstrated that peptide-induced conformational changes in TAP can be visualized by the cross-linking agent EGS (38). Peptide binding leads to conformational changes in the solvent-exposed portions of TAP and facilitates intramolecular cross-linking between TAP1 and TAP2 (6, 38, 39). To determine whether the variants TAP2-1L2 and TAP2-1S2 differ from each other in their conformational behavior upon peptide binding, cell membranes of T2 transfectants expressing wild-type TAP, TAP2-1L2, and TAP2-1S2 were incubated with EGS in the presence and the absence of S8 peptide (Fig. 6C). In the absence of S8-peptide, the same kind of limited TAP cross-linking could be observed for all transporter variants (Fig. 6C). However, in contrast to wild-type TAP and variant TAP2-1S2, no peptide-enhanced formation of the TAP cross-link product could be observed for variant TAP2-1L2. This suggests that the interdomain cross-talk is compromised in this chimeric transporter variant.
Discussion
The two NBDs of the antigenic peptide transporter TAP have different nucleotide-binding properties and seem to control different steps in the ATP-driven peptide transport cycle (4, 5, 6, 8). Several findings suggest that the catalytic function of TAP-NBD2 controls the peptide-binding properties of TAP, whereas ATP-hydrolysis in TAP-NBD1 accompanies the translocation of peptide from the cytosol into the lumen of the ER (6, 8, 9). Our previous studies have demonstrated that the distinct nucleotide-binding properties and functions of the TAP-NBDs are mainly determined by the nonhomologous C-terminal tails of the two TAP chains (11).
To gain insight into how the C-terminal tails regulate the activity of antigenic peptide transporter TAP, we have searched for discrete sequence elements that determine the distinct biochemical behaviors of NBD1 and NBD2. We approached this problem by systematically exchanging the sequences of TAP1 and TAP2 to create different chimeric TAP chains (Figs. 1 and 4A). We have been able to identify two sequence elements in the nonhomologous C-terminal tails, namely the 6/10-loop and the switch region, that are of critical importance for the distinct functions of the two TAP-NBDs.
The analysis of the chimeric TAP chains 1D2, 1E2, 1F2, and 1G2 (Figs. 2 and 3) and the results obtained for chimeras 1L2 and 1S2 (Figs. 4 and 5) clearly demonstrate that the nonhomologous 6/10-loop is the crucial determinant that controls the distinct ATP-binding behaviors of the two TAP-NBDs. Structural analysis of various ABC transporters and their NBDs (18, 40, 41, 42) showed that the 6/10-loop lies adjacent to the switch region and is in close proximity to the Walker B motif (18, 40, 41). It has been previously shown for the TAP-related P-glycoprotein that a region in NBD1 containing the sequence between 6 and 10 cannot be substituted by the corresponding sequence stretch of NBD2 without affecting transport function (43). Thus, this NBD region seems to be important for molecular processes that determine the functional separation of the two NBDs of P-glycoprotein (44, 45). Furthermore, residues in the 6/10-loop region also seem to be involved in controlling the proper NBD function of bacterial ABC transporters (46). However, there is no indication from structural or biochemical studies that the 6/10-loop region makes direct contact with bound ATP and/or ADP (18, 40, 41, 42). Thus, it is tempting to speculate that the conformational constraints imposed by the different 6/10-loops might affect the access of nucleotides to the TAP-NBDs. The 6/10-loops could participate in structural deviations of the two TAP-NBDs that are critical for the different nucleotide-binding capacities of TAP-NBD1 and -NBD2. In support of this idea, the biochemical studies of several groups have suggested that NBD1 and NBD2 of TAP have similar nucleotide-binding affinities, but differ in their ATP accessibility (6, 47). As mentioned above, a characteristic feature of the 6/10-loop regions in TAP-NBD1 and -NBD2 is their striking difference in sequence and length (Fig. 1). This is not only true for the rat TAP subunits, but also for the TAP chains of other species. Although caution is required when assessing the significance of three-dimensional structure modeling, our data for TAP-NBD2 (Fig. 4A), based on the crystal structure of the human TAP-NBD1 (18), indicate that the 6/10-loop may be one of the most significant structural differences between the two TAP-NBDs. This could reflect a structural adaptation to the requirements of nonsynonymous ATP binding in TAP1 and TAP2. Most interestingly, Karttunen et al. (6) have provided experimental evidence that in the resting state of TAP, the conformation of TAP-NBD2 is in a structurally more closed state, whereas the nucleotide-binding site of TAP1 seems to be open and solvent-accessible. In line with this, preliminary studies by our own group indicate that the 6/10-loop contributes to the regulation of nucleotide binding in a conformational, rather than a sequence-specific, manner (R. M. Leonhardt and M. R. Knittler, unpublished observation). Comparison of crystal structures of the NBDs of prokaryotic ABC transporters has revealed that the sequence between 6 and 10 connects two structurally mobile regions, the D-loop and the switch region, that seem to control NBD conformation in a dynamic and nucleotide-dependent manner (42). To elucidate the structure/function relationship of the 6/10-loops in TAP, information on the tertiary structure of both TAP-NBDs will be essential. Attempts to purify and solve the structure of TAP-NBD2 are being conducted by our laboratory.
Recent studies on the ABC protein SUR have identified a sequence in the 11/12 region of the C-terminal tails that determines the different nucleotide-binding and functional properties of that protein and its subtypes (16). It is thought that the functional role of the NBDs of the ABC protein SUR differs from that of ABC transporters such as TAP, which, in contrast to SUR, use nucleotides to pump substrates across cellular membranes (48). In line with this idea, our present findings suggest that the regulation of NBD function by the C terminus of SUR and TAP is based on a different molecular mechanism. The studies shown in Figs. 2B and 3A indicate that sequence differences in the 11/12 regions of the TAP chains make no contribution to the distinctive nucleotide-binding behaviors of the TAP-NBDs.
The crucial finding of our work on the TAP chain chimeras 1L2 and 1S2 (Figs. 4 and 5) is that the functional separation of the TAP-NBDs is essentially controlled by two adjacent sequence elements: the 6/10-loop and the switch region. Our experiments shown in Figs. 5 and 6 suggest that sequence differences in the switch region are not responsible for the distinct nucleotide-binding behaviors of TAP1 and TAP2 (Fig. 5), but seem to be required for the correct transfer of conformational signals between peptide-binding site and NBDs (Fig. 6). Karttunen et al. (6) have previously demonstrated that peptide binding to TAP induces conformational changes in the NBDs, favoring the binding of ATP. Similar observations of substrate-enhanced ATP binding have been made for other ABC transporters (49, 50). The functional TAP variant 2-1S2 shows the same peptide-enhanced ATP binding as wild-type TAP, whereas almost no peptide effect was observed for TAP2-1L2 (Fig. 6B), containing two identical NBD2 switch regions. A comparison of the TAP variants 2-1L2 and 2-1S2 (Fig. 6) indicates that peptide-induced ATP binding depends on the functional behavior of the TAP1-like subunit. This supports the hypothesis that ATP binding and hydrolysis by TAP-NBD1 accompany the transport cycle after peptide binding to TAP (4, 11). Because TAP2-1L2 can bind peptide substrates (Fig. 5A) and has normal ATP-binding behavior in the absence of additional substrates (Fig. 6A), it is likely to be folded correctly. It is reasonable to assume that the defect in the peptide-enhanced ATP binding in TAP2-1L2 results from a block in the conformational signaling between the structural domains of TAP. In support of this view, we could show that peptide binding to wild-type TAP and TAP2-1S2, but not to the variant TAP2-1L2, brings solvent-exposed portions of the TAP domains together and facilitates EGS cross-linking of the subunits (Fig. 6C). Although the exact sites of EGS cross-linking in TAP remain to be determined, our findings presented in Fig. 6 suggest that the presence of distinct switch regions of TAP1 and TAP2 is indispensable for the normal functional interplay of TAP domains and subunits.
It is well established that substrate binding by TMDs of ABC transporters triggers conformational rearrangements in the NBDs (51). However, the mechanisms of interdomain communication are still speculative. Unfortunately, because none of the present ABC transporter structures (52, 53) was determined with a bound substrate, they provide only limited information on the conformational interplay between substrate and nucleotide-binding sites. However, the existing data for bacterial ABC transporters indicate that a reorientation of the NBDs is required to stabilize the binding of ATP (51). The mutual conformational cross-talk between the TMDs and the ATP-binding sites seems to be mediated and controlled by the -helical subdomains of the NBDs (54), which contain the conserved and conformationally mobile Q-loop. Recent findings provide evidence that the structural changes in the -helical subdomain are coupled to a repositioning of sequences (42) that are probably involved in the formation of the NBD dimer interface (53). These sequence elements include the D loop and the switch region. Thus, the enhancement of ATP binding upon peptide binding could result from stabilization of the TAP structure in a particular conformation or, more specifically, could be due to altered NBD-NBD interactions that enhance nucleotide binding. In view of the phenotype of TAP2-1L2, it is interesting to note that mutations in the switch region of the bacterial ABC-transporter HisQMP2 affect not only the catalytic activity, but also the normal interdomain communication between paired NBDs (55).
It is notable that TAP2 contains the consensus sequence of the switch region, whereas TAP1 has a glutamine in place of the conserved histidine found in most pro- and eukaryotic ABC transporters. However, despite our experimental efforts, to date we have been unable to identify the discrete amino acid residues responsible for the functional nonequivalence of the switch regions in rat TAP1 and rat TAP2, suggesting that the asymmetrical character of the switch regions might be not confined to a single nonhomologous residue. Clearly, additional studies are needed to characterize the precise role(s) of the switch region in the interdomain signaling of the two TAP chains.
Our previous (11) and present experiments (Figs. 4–6) demonstrate that sequence differences within the core NBDs are not essential for the functional asymmetry of the TAP-NBDs. However, this does not exclude the possibility that differences between the sequence elements of the ATP-binding cassettes could participate to some extent in the nonsynonymous functions of TAP1 and TAP2. A clear and important finding of our present work is that functional separation of TAP-NBD1 and TAP-NBD2 depends on more than one distinct sequence element in the NBDs. Recent experiments on the TAP-related tandem ABC transporter multidrug-resistant protein 1 (MRP1) (56) suggest that sequence differences at the C-terminal end of the Walker B motifs (aspartate or glutamate, respectively, which are also present in TAP1 (LILLD) and TAP2 (LILLE), contribute to the functional divergence of the NBDs. In this context it should be noted that MRP1 (49, 57) is not only similar to TAP in its asymmetrical NBD function, but also has 6/10-loops of different length and similar signature motifs and switch regions. It will be interesting to investigate whether the NBD sequences of MRP1 can functionally substitute for sequences in the core NBDs and the C-terminal tails of TAP.
Acknowledgments
We are indebted to Prof. Dr. Jonathan Howard for his helpful comments on the manuscript. We thank Dr. Robert Wilson for critical reading the manuscript, and Dr. Frank Momburg for the gift of the anti-TAP1 antiserum 1p3.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was supported by the Land Nordrhein-Westfalen through the University of Cologne and the Deutsche Forschungsgemeinschaft (Grant KN541-1/1).
2 Address correspondence and reprint requests to Dr. Michael R. Knittler, Institute for Genetics, University of Cologne, Zülpicher Strasse 47, 50674 Cologne, Germany. E-mail address: knittler@uni-koeln.de
3 Abbreviations used in this paper: ER, endoplasmic reticulum; ABC, ATP-binding cassette; EGS, ethylene glycol bissuccinimidyl succinate; NBD, nucleotide-binding domain; TMD, transmembrane domain; SUR, sulfonylurea receptor; HSAB, N-hydroxysuccinamide-4-azidobenzoate; MRP, multidrug-resistant protein.
Received for publication July 21, 2004. Accepted for publication October 22, 2004.
References
Yewdell, J. W., E. Reits, J. Neefjes. 2003. Making sense of mass destruction: quantitating MHC class I antigen presentation. Nat. Rev. Immunol. 3:952
Momburg, F., G. J. H?mmerling. 1998. Generation and TAP-mediated transport of peptides for major histocompatibility complex class I molecules. Adv. Immunol. 68:192.
Lankat-Buttgereit, B., R. Tampé. 1999. The transporter associated with antigen processing TAP: structure and function. FEBS Lett. 464:108.
van Endert, P., L. Saveanu, E. Hewitt, P. J. Lehner. 2002. Powering the peptide pump: TAP crosstalk with energetic nucleotides. Trends Biochem. Sci. 27:454
Alberts, P., O. Daumke, E. D. Deverson, J. C. Howard, M. R. Knittler. 2001. Distinct functional properties of the TAP subunits coordinate the nucleotide dependent transport cycle. Curr. Biol. 11:242.
Karttunen, J. T., P. J. Lehner, S. S. Gupta, E. W. Hewitt, P. Cresswell. 2001. Distinct functions and cooperative interaction of the subunits of the transporter associated with antigen processing (TAP). Proc. Natl. Acad. Sci. USA 98:7431.
Hewitt, E. W., S. S. Gupta, P. J. Lehner. 2001. The human cytomegalovirus gene product US6 inhibits ATP binding by TAP. EMBO J. 20:387.
Saveanu, L., S. Daniel, P. M. van Endert. 2001. Distinct functions of the ATP binding cassettes of transporters associated with antigen processing: a mutational analysis of Walker A and B sequences. J. Biol. Chem. 276:22107
Daumke, O., M. R. Knittler. 2001. Functional asymmetry of the ATP-binding-cassettes of the ABC transporter TAP is determined by intrinsic properties of the nucleotide binding domains. Eur. J. Biochem. 268:4776.
Arora, S., P. E. Lapinski, M. Raghavan. 2001. Use of chimeric proteins to investigate the role of transporter associated with antigen processing (TAP) structural domains in peptide binding and translocation. Proc. Natl. Acad. Sci. USA 98:7241.
Bouabe, H., M. R. Knittler. 2003. The distinct nucleotide binding states of the transporter associated with antigen processing (TAP) are regulated by the nonhomologous C-terminal tails of TAP1 and TAP2. Eur. J. Biochem. 270:4531
Currier, S. J., K. Ueda, M. C. Willingham, I. Pastan, M. M. Gottesman. 1989. Deletion and insertion mutants of the multidrug transporter. J. Biol. Chem. 264:14376.
Haardt, M., M. Benharouga, D. Lechardeur, N. Kartner, G. L. Lukacs. 1999. C-terminal truncations destabilize the cystic fibrosis transmembrane conductance regulator without impairing its biogenesis: a novel class of mutation. J. Biol. Chem. 274:21873.
Sharma, N., A. Crane, J. P. t. Clement, G. Gonzalez, A. P. Babenko, J. Bryan, L. Aguilar-Bryan. 1999. The C-terminus of SUR1 is required for trafficking of KATP channels. J. Biol. Chem. 274:20628.
Booth, C. L., L. Pulaski, M. M. Gottesman, I. Pastan. 2000. Analysis of the properties of the N-terminal nucleotide-binding domain of human P-glycoprotein. Biochemistry 39:5518
Matsushita, K., K. Kinoshita, T. Matsuoka, A. Fujita, T. Fujikado, Y. Tano, H. Nakamura, Y. Kurachi. 2002. Intramolecular interaction of SUR2 subtypes for intracellular ADP-induced differential control of K(ATP) channels. Circ. Res. 90:554.
Yamaguchi, Y., H. Iseoka, A. Kobayashi, M. Maeda. 2004. The carboxyl terminal sequence of rat transporter associated with antigen processing (TAP)-like (ABCB9) is heterogeneous due to splicing of its mRNA. Biol. Pharm. Bull. 27:100
Gaudet, R., D. C. Wiley. 2001. Structure of the ABC ATPase domain of human TAP1, the transporter associated with antigen processing. EMBO J. 20:4964
DeMars, R., C. C. Chang, S. Shaw, P. J. Reitnauer, P. M. Sondel. 1984. Homozygous deletions that simultaneously eliminate expressions of class I and class II antigens of EBV-tansformed B-lymphoblastoid cells. I. Reduced proliferance responses of autologous and allogeneic T cells that have decreased expression of class II antigens. Hum. Immunol. 11:77.
Momburg, F., V. Ortiz-Navarrete, J. Neefjes, E. Goulmy, Y. van de Wal, H. Spits, S. J. Powis, G. W. Butcher, J. C. Howard, P. Walden, et al 1992. Proteasome subunits encoded by the major histocompatibility complex are not essential for antigen presentation. Nature 360:174.
Deverson, E. V., I. R. Gow, W. J. Coadwell, J. J. Monaco, G. W. Butcher, J. C. Howard. 1990. MHC class II region encoding proteins related to the multidrug resistance family of transmembrane transporters. Nature 348:738.
Gunning, P., J. Leavitt, G. Muscat, S. Y. Ng, L. Kedes. 1987. A human -actin expression vector system directs high-level accumulation of antisense transcripts. Proc. Natl. Acad. Sci. USA 84:4831.
Powis, S. J., A. R. Townsend, E. V. Deverson, J. Bastin, G. W. Butcher, J. C. Howard. 1991. Restoration of antigen presentation to the mutant cell line RMA-S by an MHC-linked transporter. Nature 354:528.
Knittler, M. R., K. Gülow, A. Seelig, J. C. Howard. 1998. MHC class I molecules compete in the endoplasmic reticulum for access to transporter associated with antigen processing. J. Immunol. 161:5967.
Nijenhuis, M., S. Schmitt, E. A. Armandola, R. Obst, J. Brunner, G. J. H?mmerling. 1996. Identification of a contact region for peptide on the TAP1 chain of the transporter associated with antigen processing. J. Immunol. 156:2186.
Lindahl, E., B. Hess, D. van der Spoel. 2001. GROMACS 3.0: a package for molecular simulation and trajectory analysis. J. Mol. Mod. 7:306.
Schneider, E., S. Hunke. 1998. ATP-binding-cassette (ABC) transport systems: functional and structural aspects of the ATP-hydrolyzing subunits/domains. FEMS Microbiol. Rev. 22:1.
Chen, M., R. Abele, R. Tampé. 2003. Peptides induce ATP hydrolysis at both subunits of the transporter associated with antigen processing. J. Biol. Chem. 278:29686
Lacaille, V. G., M. J. Androlewicz. 1998. Herpes simplex virus inhibitor ICP 47 destabilizes the transporter associated with antigen processing (TAP) heterodimer. J. Biol. Chem. 273:17386.
Antoniou, A. N., S. Ford, E. S. Pilley, N. Blake, S. J. Powis. 2002. Interactions formed by individually expressed TAP1 and TAP2 polypeptide subunits. Immunology 106:182.
Hung, L.-W., I. X. Wang, K. Nikaido, P.-Q. Liu, G. Ferro-Luzzi Ames, S.-H. Kim. 1998. Crystal structure of the ATP-binding subunit of an ABC-transporter. Nature 396:703.
Diederichs, K., J. Diez, G. Greller, C. Muller, J. Breed, C. Schnell, C. Vonrhein, W. Boos, W. Welte. 2000. Crystal structure of MalK, the ATPase subunit of the trehalose/maltose ABC transporter of the archaeon Thermococcus litoralis. EMBO J. 19:5951.
Karpowich, N., O. Martsinkevich, L. Millen, Y. R. Yuan, P. L. Dai, K. MacVey, P. J. Thomas, J. F. Hunt. 2001. Crystal structures of the MJ1267 ATP binding cassette reveal an induced-fit effect at the ATPase active site of an ABC transporter. Structure 9:571.
Beaudet, L., P. Gros. 1995. Functional dissection of P-glycoprotein nucleotide-binding domains in chimeric and mutant proteins. Modulation of drug resistance profiles. J. Biol. Chem. 270:17159.
Hrycyna, C. A., M. Ramachandra, U. A. Germann, W. C. Cheng, I. Pastan, M. M. Gottesmann. 1999. Both ATP sites of human P-glycoprotein are essential but not symmetric. Biochemistry 38:13887.
Sauna, Z. E., S. V. Ambudkar. 2001. Characterization of the catalytic cycle of ATP hydrolysis by human P-glycoprotein: the two ATP hydrolysis events in a single catalytic cycle are kinetically similar but affect different functional outcomes. J. Biol. Chem. 276:11653.(Sarah Ehses, Ralf M. Leon)