CHARACTERIZATION OF INTERACTIONS IN THE TAP FAMILY OF HALF
ABC TRANSPORTERS.
By
Dennis Brian Leveson-Gower
B.Sc, The University of Victoria, 1999.
A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF
THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
in
THE FACULTY OF GRADUATE STUDIES
(Biochemistry and Molecular Biology)
The University of British Columbia
April 2005
© Dennis Brian Leveson-Gower, 2005 ABSTRACT
The transporter associated with antigen processing (TAP) is an ATP-binding cassette
(ABC) protein which transports peptides for presentation to the immune system. TAP is
composed of two half transporters, TAP1 (ABCB2) and TAP2 (ABCB3) which require
heterodimerization for function. In humans, the TAP gene family consists of TAP 1, TAP2,
and TAPL (ABCB9). While the TAP1-TAP2 complex is well-characterized, the
dimerization state and function of TAPL are unknown. To identify interactions within the
human TAP family, I adapted the dihydrofolate reductase protein-fragment complementation
assay (DHFR PCA) to study the human genes coding for the half ABC transporters. This
assay has been used for the study of membrane-bound proteins in vivo (Remy, I., Wilson, I.
A., and Michnick, S. W. (1999) Science 283(5404), 990-3). With this method, in vivo
TAP1-TAP2 heterodimerization was confirmed, no homodimerizations were detected with
TAP1 or TAP2, and TAPL did not show any interaction with TAP1 or TAP2. However, I
found strong evidence that TAPL forms homodimers. These results provide evidence of a
novel homomeric TAPL interaction and demonstrate that the DHFR PCA will be of general utility in studies of half ABC transporter interactions in vivo. By using an insect-cell microsomal transport assay, I found the first direct evidence that TAPL can transport peptides. Two classical TAP substrates, RRYQNSTEL and RYWANATRST were transported by TAPL. Kinetic characterization of the transport of RYWANATRST indicated
that TAPL follows Michaelis-Menton kinetics with an apparent Km of 3.2 ± 0.4 nM. TABLE OF CONTENTS
Abstract ii Table of contents iii List of tables v List of figures vi List of Abbreviations viii Acknowledgements x I Introduction 1 1.1 — ABC transporters 1 1.2 — Half ABC transporters: a need for dimerization 6 1.3 — Methods for detecting interactions between half ABC trasnporters 7 1.4 — Thesis objectives 9 Bibliography 11 II Application of the DHFR PC A to TAP family of half ABC Transporters 17 2.1 - Introduction: A need to assay for interactions in the TAP family 17 2.2 - Materials and Methods 21 2.2.1 — Creation of DHFR PCA vectors 21 2.2.2 — Cell lines and culture 29 2.2.3 — DHFR survival assay 29 2.2.4 —FACS 30 2.3 - Results 31 2.3.1 — DHFR PCA survival-selection in nucleotide-free media 31 2.3.2 — FACS analysis of clones 32 2.4 - Discussion 35 Bibliography 37 III Utility of the DHFR PCA for the study of ABC Transporters 42 3.1 - Introduction 42 3.2 - Material and methods 46 3.2.1 — Cell lines and culture 46 3.2.2 — Western blots 46 3.2.3 — RNA purification and reverse transcription 47 3.2.4 — PCR of false positive transcripts 48 3.2.5 — MHC Class I expression assay 48 3.2.6 —FACS 49 3.3-Results 50 3.3.1 — Western blot analysis of protein expression in colonies 50 3.3.2 — Characterization of transcripts expressed in false positive colonies 51 3.3.3 — Effect of DHFR fusion on TAP1 and TAP2 function 53 3.4 - Discussion 56 Bibliography 58 IV Peptide transport by TAPL 59 4.1 - Introduction 59 4.1.1— Peptide transport by TAP 1/2 59 4.1.2 — Why TAPL may be a peptide transporter 64 4.2 - Materials and Methods 68 4.2.1 — Creation of a TAPL expression vector for insect cells 68 4.2.2 — Maintenance, transfection, and selection of insect cells 68 4.2.3 — Western blots 69 4.2.4 — Microsome preparation 70 4.2.5 — Peptide transport assay 71 4.3 - Results 74 4.3.1 — Creation of a TAPL-expressing cell line and preparation of microsomes 74 4.3.2 — Peptide transport assay 75 4.4 - Discussion 81 Bibliography 83 V Overall Discussion and Future prospects 89 5.1 - Discussion 89 5.2 - Future prospects 94 5.2.1 — Peptide specificity of TAPL 94 5.2.2 — Localization of TAPL homodimers 95 5.2.3 — Function in immune tolerance 96 5.2.4 — Detection of interactions within the White family of half ABC transporters... 96 Bibliography 98 LIST OF TABLES
Table 2.2.1: Primers used for the sub-cloning of TAP family cDNAs into DHFR PCA vectors LIST OF FIGURES
Figure 1.1.1: X-ray crystallography structure of Escherichia coli MsbA 3
Figure 1.1.2: Potential model for lipid A transport by Escherichia coli MsbA 5
Figure 2.2.1: Crystal structure of DHFR 22
Figure 2.2.2: Sub-cloning of zip-DHFR positive controls 23
Figure 2.2.3: Creation of parental DHFR vectors with 10 amino acid linkers 24
Figure 2.2.4: Modifying DHFR parental vectors to contain 20 aa linkers 25
Figure 2.2.5: Maps of DHFR constructs 28
Figure 2.3.1: Colony counts from survival assay 32
Figure 2.3.2: Relative increase in fluorescence between non-transfected cells
and colonies from survival-selection 34
Figure 3.1.1: Possible origins of false positive colonies 43
Figure 3.1.2: MHC class I expression pathway 45
Figure 3.3.1: Western blot analysis of clones from the DHFR survival assay 51
Figure 3.3.2: RT-PCR detection of transcripts expressed in colonies from
the survival assay 53
Figure 3.3.3: FACS of MHC class I expression 55
Figure 4.1.1: Schematic model of TAP 61
Figure 4.1.2: Substrate recognition motif and substrate binding pocket of
human TAP 63
Figure 4.1.3: Homology between peptide binding regions of TAP1, TAP2, and TAPL 65
Figure 4.1.4: A working model for phagosomal cross-presentation 66
Figure 4.2.1: Determining direction of peptide transport 73 Figure 4.3.1: Selection of TAPL-His expressing High Five cells 74
Figure 4.3.2: ATP-dependent transport of RRYQNSTEL by TAPL 76
Figure 4.3.3: Peptide transport assay with peptides FAPGNYPAL and RYWAN ATRST 77
Figure 4.3.4: Transport of radioactive 0.39 nM RYWANATRST as a function of time 79
Figure 4.3.5: Lineweaver-Burk plot of initial velocity as a function of substrate concentration for TAPL's transport of RYWANATRST by TAPL 80
Figure 5.1.1: Proposed functions of TAPL 93 LIST OF ABBREVIATIONS
ABCB2 ATP-binding cassette sub-family B transporter 2
ABCB3 ATP-binding cassette sub-family B transporter 3
ABCB9 ATP-binding cassette sub-family B transporter 9 bp base pair(s)
CFTR Cystic fibrosis transmembrane conductance regulator
CHO Chinese Hamster ovary
CMV cyto-megalo virus
DHFR PCA dihydrofolate reductase protein-fragment complementation assay
ER endoplasmic reticulum
FACS fluorescence-activated cell sorting
FBS Fetal Bovine Serum
FRET fluorescence resonance energy transfer
HT hypoxanthine and thymidine
MHC Major Histocompatibility Complex
NBD Nucleotide binding domain
Ova Chicken egg ovalbumin
PBS phosphate buffered saline
PCR polymerase chain reaction
RE restriction endonuclease
RT reverse transcription
TAP transporter associated with antigen processing
TAP1 transporter associated with antigen processing 1 also known as ABCB2 TAP2 transporter associated with antigen processing 2 also known as ABCB3
TAPL TAP-Like protein also known as ABCB9
TMD Transmembrane domain
TM Transmembrane helix
SDS Sodium Dodecyl Sulfate
UTR untranslated region
VSV vesicular stomatitis virus. ACKNOWLEDGEMENTS
I would like to thank my supervisor, Dr. Victor Ling, for giving me the opportunity to work in his lab and for constantly challenging me. He taught me how to think for myself and has given me a great foundation for doing science.
It was a great pleasure to work with the members of the Ling lab and I am grateful for all their help and friendship.
I thank my committee members, Dr. Ross MacGillivray and Dr. Natalie Strynadka, for their suggestions, encouragement and support. Without them, I am sure my Ph.D. would have taken much longer.
To my parents, I thank them for always believing in me and letting me find my own way.
Finally, I thank my wife, Ji-Yeon Hong, for giving so much love and support during the tribulations of graduate school. She is always there to lift me up and keep me going. I Introduction
1.1 — ABC transporters
The largest known family of transmembrane proteins is the superfamily of ATP-binding cassette (ABC) transport systems which are found in prokaryotes, eukaryotes and archeabacteria (9,17). Some of the first ABC transporters discovered in eukaryotes were: P- gycoprotein, a multidrug transporter that is over-expressed in tumors resistant to chemotherapy (4); the products of the white and brown genes in Drosophila melanogaster, which transport pigments (13); and the cystic fibrosis transmembrane conductance regulator
(CFTR) which functions as a chloride channel (31). Currently there are 49 members of the human family of ABC transporters annotated in the human genome, several of which have been implicated in human ailments (2,11,12,42).
ABC transporters all contain a highly conserved ATP-binding cassette or nucleotide binding domain (NDB) which consists of a Walker A (G-X2-G-X-G-K-S/T-T/S-X4-
hydrophobic) and a Walker B (R-X-hydrophobic2-X2-P/T/S/A-X-hydrophobic4-D-E-A/P/C-
T-S/T/A-A/G-hydrophobic-D) motif, which are similar to those originally described by
Walker (41). There is also a C motif or "signature motif with the sequence (hydrophobic-S-
X-G-Q-R/K-Q-R-hydrophobic-X-hydrophobic-A) (38) which lies upstream of the Walker B motif. These three conserved motifs are always found in the same order: a Walker A at the
N-terminal end of the (NDB), a stretch of 100-150 amino acids, the signature motif, and then the Walker B motif. Structures of NDBs indicate that the Walker A motif, also known as the
P loop, binds to the a- and fi-phosphates of di- and tri-nucleotides bound at the nucleotide binding site (15). The function of the Walker B motif, which forms a 13-strand, is less clear, but it appears to be involved in binding the Mg2"1" ion via the O's of a serine and glutamine
(Mg2"1" is a required cofactor of ABC-ATPases and binds ATP 13- and y-phosphate O's) (19,
20).
Precisely how ABC transporters are able to translocate substrates across membranes is still unclear. It is believed that substrate binding induces a conformational change in the transmembrane domains (TMDs) which initiates ATP hydrolysis (18). It is also known that
ATP hydrolysis at both NBDs of a transporter are required in an "alternating catalytic cycle" mechanism (33, 34, 36). Further insight into the structure and possibly the mechanism of mammalian ABC transporters comes from the structure of MsbA (Figure 1.1.1 (8)). This bacterial transporter is more closely related to the mammalian P-glycoproteins than any other bacterial ABC transporter (8). Figure 1.1.1: X-ray crystallography structure of Escherichia coli MsbA. (A) View of dimer looking into the chamber opening. The transmembrane domain, NBD, and intracellular domain are colored red, cyan, and dark blue, respectively. Transmembrane a-helices are marked and the connecting loops are shown in green. A model of lipid A (not in the crystal structure) is shown to the right, embedded in the lower bilayer leaflet. Solid and dotted green lines represent the boundaries of the membrane bilayer leaflets. Dotted cyan lines indicate the approximate location of the disordered region in the NBD. (B) View of Eco-msbA from extracellular side, perpendicular to the membrane with model of lipid A. Transporter dimensions are labeled and images were rendered using BOBSCRIPT and RASTER 3D. Figure reproduced from Chang et al. (8). From their crystal structure of MsbA, Chang et al. propose a general mechanism for transport by ABC transporters (Figure 1.1.2). In this model, the extensive contact of the intracellular domain with the two NBDs and the transmembrane domain serves to transfer the energy generated from ATP hydrolysis at the NBDs to tertiary re-arrangements of the transmembrane a-helices. When lipid A binds, conformational changes in the transmembrane domain cause the intracellular domain to start nucleotide hydrolysis by the
NBD. ATP hydrolysis causes a conformational shift which brings the two NBDs together.
The charges in the chamber at the inner membrane leaflet now make an energetically unfavourable microenvironment for a hydrophobic substrate causing it to "flip" from the inner to the outer membrane leaflet. After the substrate flips, a signal is sent which causes the NBDs to separate resulting in the extrusion of the substrate from the outer leaflet.
While the model from Chang et al. (8) is interesting, there are still unanswered questions regarding the mechanisms of ABC transporters. Since no crystal structure has been obtained in the presence of ligand, the location and nature of the substrate binding site in a 3- dimensional architecture is unknown. Furthermore, it is also unclear how transporters such as MDR1 and TAP are able to transport such a wide variety of substrates (1, 35). Substrate binding and recruitment
Chamber opening and substrate expulsion Chamber closure and substrate flip-flop
Figure 1.1.2: Potential model for lipid A transport by Escherichia coli MsbA. Stages 1 to 3 begin at top and proceeds clockwise. See text for details. (1) Lipid A binding, triggering of ATP hydrolysis, and recruitment of substrate to chamber. (2) Closure of the chamber and translocation of lipid A. Interaction between the two NBDs is possible. (3) Opening of the chamber, movement of TM2/TM5, release of lipid A to the outer bilayer leaflet, and nucleotide exchange. A small yellow rectangle and a green circle denote the hydrophobic tails and sugar head groups of lipid A, respectively. The transmembrane domain (TM), intracellular domain (ICD), and nucleotide-binding domain (NBD) are labeled. The cell membrane is represented as a set of two horizontal lines separated by a dash to indicate the separation of bilayer leaflets. Blue regions indicate positive charge lining the chamber, and purple regions represent the intracellular domain. The gray region on the outer membrane side of the chamber is hydrophobic. Red and black arrows show the movement of substrate and the structural changes of MsbA, respectively. Figure reproduced from Chang et al. (8). 1.2 — Half ABC Transporters: A Need for Dimerization
Based on experimental evidence, it appears that ABC transporters require, at a minimum,
two ATP-binding cassettes and two transmembrane domains (TMs, consisting of 6 or more
transmembrane helices) to be functional (17). Mutational studies with both yeast STE6
protein and P-glycoprotein have shown that ABC transporters must contain two nucleotide binding domains (NBDs) to function (3, 6). Functional analysis of MalFGK2 from E. coli
confirmed this requirement (10). Thus, it is not surprising that "half ABC transporters,
which as so-called as they contain one membrane-integral domain (typically consisting of 6
or more transmembrane helices) and one ABC subunit, function as either heterodimers or
homodimers. Genetic evidence of half ABC transporter dimerization can be found between
transporters such as: members of the white family in Drosophila; Pxal and Pxa2 in
Saccharomyces cerevisiae; LmrA in Lactococcus lactis; and ABCG5 and ABCG8 in humans
(5,13,16,23, 26,28,29, 37,40). Physical evidence of half ABC transporter interactions has
been found for transporters such as: members of the peroxisomal family in humans (via the
yeast two-hybrid system); ABCG2 in humans (via co-immunoprecipitation); and YvcC in
Bacillus subtilis (via cryo-electron microscopy) (7,22,24). The transporter associated with
antigen processing (TAP) is another well-characterized heterodimer (for discussion see
Chapter 2). 1.3 — Methods for detecting interactions between half ABC trasnporters.
As half ABC transporters must form dimers in order to function, the composition of the dimers must be determined before any functional studies may be considered. Although there is some genetic evidence and physical evidence which strongly supports that half ABC transporters must form dimers to function, in vivo evidence of interactions is weak. Thus, a goal of this thesis was to find an appropriate assay for identifying which half ABC transporters form in vivo interactions. Such an assay would need to work for transporters localized to any cellular membrane or compartment. It would also be beneficial if the assay allowed the proteins to remain functional and did not require the proteins to be over- expressed. These criteria immediately eliminate common methods for finding protein- protein interactions such as the yeast two hybrid assay, which requires nuclear localization of the hybrid proteins (14). Other, more modern assays, however, offer the hope of fulfilling these criteria.
The split-ubiquitin system is a sensitive assay for the detection of in vivo interactions (21,
39). In this assay, ubiquitin is split into two fragments, which are then fused to two different proteins of interest. One of these ubiquitin fragments is also fused to a transcription factor.
If the proteins of interest interact, they will re-form the ubiquitin molecule which is then cleaved by a ubiquitin-specific protease. However, this assay has the disadvantage of requiring the ubiquitin fragments be localized to the cytosol, as this is the intracellular location of the ubiquitin-specific protease. Another assay which is suitable for in vivo study of membrane protein interactions is based on B-galactosidase complementation (32). Unlike the split-ubiquitin system, this assay will work in any membrane compartment. B-galactosidase is split into two fragments, which are then fused to two proteins of interest. If the proteins interact, they will cause the re• assembly of the B-galactosidase, which can then be monitored by the cleavage of a luminescent substrate. The disadvantage of this assay is that the two B-galactosidase subunits, even if weakly associating, are always capable of associating to some extent. This will result in a constant background caused by the spontaneously assembling fragments.
Yet another method for detecting protein-protein interactions is fluorescence resonance energy transfer (FRET) (25). In this method, two different fluorescent proteins are fused to the ends of two proteins of interest. If the proteins interact, a dipole-dipole interaction between the neighboring fluorescent molecules occurs, which can be monitored. FRET is extremely sensitive, but it requires the two attached fluorescent proteins to come within 3-6 nm of each other—interactions outside of this range will not be detected. Furthermore, the degree of over-expression of the fusion proteins must be controlled as over-expression of the proteins will result in a strong fluorescent background. These aspects of the FRET assay have limited its utility.
Perhaps the most promising of the in vivo methods suitable for detecting protein-protein interactions is the dihydrofolate reductase protein-fragment complementation assay (DHFR
PCA) (27). This assay was shown to be suitable for the study of membrane-bound proteins in vivo when it was used to detect ligand-induced conformation change in the EPO receptor (30). Briefly, murine DHFR mRNA is divided into complementary N- and C- terminal fragments and added to the C-termini of two different half ABC transporters genes. If the two half ABC transporters dimerize, they will induce the folding of the N- and C- fragments into an active DHFR molecule. Active DHFR is then be measured by: (1) increased resistance to methotrexate (an inhibitor of DHFR), (2) survival in nucleotide-free media (in
DFHR -/- cells), or (3) increased trapping of fluorescent-labeled methotrexate. The applicability of this assay to all membrane proteins in any organelle, the multiple detection methods for interactions, and the absence of spontaneous reassembly of the DFHR fragments, make this assay ideal for the study of half ABC transporters.
1.4 — Thesis objectives
The first goal of my thesis was to determine the applicability and utility of the dihydrofolate reductase protein-fragment complementation assay (DHFR-PCA) for the study of half ABC transporters. As detailed above, I hypothesized that this method would be suitable to the study of half ABC transporters. Initially the assay would be applied to the well-characterized heterodimer of TAP 1 and TAP2 as a proof of principle. Then I would move on to look for other interactions among the three members of the TAP family, TAP1,
TAP2, and TAPL. This would determine the dimerization state of TAPL, but will also address whether proposed TAP1 and TAP2 homodimers can be detected by the DHFR PCA.
Finally, we hypothesize that TAPL will dimerize to form a functional transporter and, due to its sequence similarity to TAP1 and TAP2 and some preliminary data from another lab (discussed in Chapter 4), that it may be a peptide transporter, Thus, the three objectives of this thesis were:
1. To apply the dihydrofolate protein-fragment complementation assay (DHFR PCA) to
the study of half ABC transporter interactions in vivo.
2. To find novel interactions between human half ABC transporters.
3. To determine the function of the novel half ABC transporter interactions.
The first two objectives are addressed in Chapter 2 when the DHFR PCA was applied to the TAP family of half ABC transporters. In Chapter 3,1 further evaluated the utility of the
DHFR PCA and commented on its applicability to half ABC transporters in general. Finally, in Chapter 4,1 hypothesized that TAPL homodimers could be functioning as peptide transporters and tested this hypothesis with a microsomal transport assay. Bibliography
1. Abele, R., and R. Tampe. 2004. The ABCs of immunology: structure and function
of TAP, the transporter associated with antigen processing. Physiology 19:216-24.
2. Allikmets, R., B. Gerrard, A. Hutchinson, and M. Dean. 1996. Characterization of
the human ABC superfamily: isolation and mapping of 21 new genes using the
expressed sequence tags database. Hum Mol Genet 5:1649-55.
3. Azzaria, M., E. Schurr, and P. Gros. 1989. Discrete mutations introduced in the
predicted nucleotide-binding sites of the mdrl gene abolish its ability to confer
multidrug resistance. Molecular & Cellular Biology. 9:5289-97.
4. Bell, D. R., J. H. Gerlach, N. Kartner, R. N. Buick, and V. Ling. 1985. Detection
of P-glycoprotein in ovarian cancer: a molecular marker associated with multidrug
resistance. J Clin Oncol 3:311-5.
5. Berge, K. E., H. Tian, G. A. Graf, L. Yu, N. V. Grishin, J. Schultz, P.
Kwiterovich, B. Shan, R. Barnes, and H. H. Hobbs. 2000. Accumulation of dietary
cholesterol in sitosterolemia caused by mutations in adjacent ABC
transporters.[comment]. Science. 290:1771-5.
6. Berkower, C, and S. Michaelis. 1991. Mutational analysis of the yeast a-factor
transporter STE6, a member of the ATP binding cassette (ABC) protein superfamily.
EMBO Journal. 10:3777-85.
7. Chami, M., E. Steinfels, C. Orelle, J. M. Jault, A. Di Pietro, J. L. Rigaud, and S.
Marco. 2002. Three-dimensional structure by cryo-electron microscopy of YvcC, an
homodimeric ATP-binding cassette transporter from Bacillus subtilis. Journal of
Molecular Biology. 315:1075-85. 8. Chang, G., and C. B. Roth. 2001. Structure of MsbA from E. coli: a homolog of the
multidrug resistance ATP binding cassette (ABC) transporters.[comment]. Science.
293:1793-800.
9. Childs, S., and V. Ling. 1994. The MDR superfamily of genes and its biological
implications. Important Adv Oncol:21-36.
10. Davidson, A. L., and S. Sharma. 1997. Mutation of a single MalK subunit severely
impairs maltose transport activity in Escherichia coli. Journal of Bacteriology.
179:5458-64.
11. Dean, M., Y. Hamon, and G. Chimini. 2001. The human ATP-binding cassette
(ABC) transporter superfamily. J Lipid Res 42:1007-17.
12. Dean, M., A. Rzhetsky, and R. Allikmets. 2001. The human ATP-binding cassette
(ABC) transporter superfamily. Genome Res 11:1156-66.
13. Dreesen, T. D., D. H. Johnson, and S. Henikoff. 1988. The brown protein of
Drosophila melanogaster is similar to the white protein and to components of active
transport complexes. Molecular & Cellular Biology. 8:5206-15.
14. Fields, S., and O. Song. 1989. A novel genetic system to detect protein-protein
interactions. Nature 340:245-6.
15. Gaudet, R., and D. C. Wiley. 2001. Structure of the ABC ATPase domain of human
TAP1, the transporter associated with antigen processing. EMBO Journal. 20:4964-
72.
16. Hettema, E. H., C. W. van Roermund, B. Distel, M. van den Berg, C. Vilela, C.
Rodrigues-Pousada, R. J. Wanders, and H. F. Tabak. 1996. The ABC transporter proteins Patl and Pat2 are required for import of long-chain fatty acids into
peroxisomes of Saccharomyces cerevisiae. EMBO Journal. 15:3813-22.
17. Higgins, C. F. 1992. ABC transporters: from microorganisms to man. Annual
Review of Cell Biology. 8:67-113.
18. Higgins, C. F. 2001. ABC transporters: physiology, structure and mechanism--an
overview. Research in Microbiology. 152:205-10.
19. Hopfner, K. P., A. Karcher, D. S. Shin, L. Craig, L. M. Arthur, J. P. Carney,
and J. A. Tainer. 2000. Structural biology of Rad50 ATPase: ATP-driven
conformational control in DNA double-strand break repair and the ABC-ATPase
superfamily. Cell 101:789-800.
20. Hung, L. W., I. X. Wang, K. Nikaido, P. Q. Liu, G. F. Ames, and S. H. Kim. 1998.
Crystal structure of the ATP-binding subunit of an ABC transporter. Nature 396:703-
7.
21. Johnsson, N., and A. Varshavsky. 1994. Split ubiquitin as a sensor of protein
interactions in vivo. Proceedings of the National Academy of Sciences of the United
States of America 91:10340-4.
22. Kage, K., S. Tsukahara, T. Sugiyama, S. Asada, E. Ishikawa, T. Tsuruo, and Y.
Sugimoto. 2002. Dominant-negative inhibition of breast cancer resistance protein as
drug efflux pump through the inhibition of S-S dependent homodimerization.
International Journal of Cancer. 97:626-30.
23. Lee, M. H., K. Lu, S. Hazard, H. Yu, S. Shulenin, H. Hidaka, H. Kojima, R.
Allikmets, N. Sakuma, R. Pegoraro, A. K. Srivastava, G. Salen, M. Dean, and S. B. Patel. 2001. Identification of a gene, ABCG5, important in the regulation of
dietary cholesterol absorption. Nature Genetics. 27:79-83.
24. Liu, L. X., K. Janvier, V. Berteaux-Lecellier, N. Cartier, R. Benarous, and P.
Aubourg. 1999. Homo- and heterodimerization of peroxisomal ATP-binding cassette
half-transporters. Journal of Biological Chemistry. 274:32738-43.
25. Majoul, I., M. Straub, R. Duden, S. W. Hell, and H. D. Soling. 2002. Fluorescence
resonance energy transfer analysis of protein-protein interactions in single living cells
by multifocal multiphoton microscopy. Journal of Biotechnology 82:267-77.
26. O'Hare, K., C. Murphy, R. Levis, and G. M. Rubin. 1984. DNA sequence of the
white locus of Drosophila melanogaster. Journal of Molecular Biology. 180:437-55.
27. Pelletier, J. N., F. X. Campbell-Valois, and S. W. Michnick. 1998.
Oligomerization domain-directed reassembly of active dihydrofolate reductase from
rationally designed fragments. Proceedings of the National Academy of Sciences of
the United States of America 95:12141-6.
28. Pepling, M., and S. M. Mount. 1990. Sequence of a cDNA from the Drosophila
melanogaster white gene. Nucleic Acids Research. 18:1633.
29. Poelarends, G. J., P. Mazurkiewicz, and W. N. Konings. 2002. Multidrug
transporters and antibiotic resistance in Lactococcus lactis. Biochimica et Biophysica
Acta. 1555:1-7.
30. Remy, I., I. A. Wilson, and S. W. Michnick. 1999. Erythropoietin receptor
activation by a ligand-induced conformation change. Science 283:990-3.
31. Riordan, J. R., J. M. Rommens, B. Kerem, N. Alon, R. Rozmahel, Z. Grzelczak,
J. Zielenski, S. Lok, N. Plavsic, J. L. Chou, and et al. 1989. Identification of the cystic fibrosis gene: cloning and characterization of complementary DNA. Science
245:1066-73.
32. Rossi, F., C. A. Charlton, and H. M. Blau. 1997. Monitoring protein-protein
interactions in intact eukaryotic cells by beta-galactosidase complementation.
Proceedings of the National Academy of Sciences of the United States of America
94:8405-10.
33. Sauna, Z. E., and 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. Journal
of Biological Chemistry 276:11653-61.
34. Sauna, Z. E., and S. V. Ambudkar. 2000. Evidence for a requirement for ATP
hydrolysis at two distinct steps during a single turnover of the catalytic cycle of
human P-glycoprotein. Proceedings of the National Academy of Sciences of the
United States of America 97:2515-20.
35. Sauna, Z. E., M. M. Smith, M. Muller, K. M. Kerr, and S. V. Ambudkar. 2001.
The mechanism of action of multidrug-resistance-linked P-glycoprotein. Journal of
Bioenergetics & Biomembranes 33:481-91.
36. Senior, A. E., M. K. al-Shawi, and I. L. Urbatsch. 1995. The catalytic cycle of P-
glycoprotein. FEBS Letters. 377:285-9.
37. Shani, N., P. A. Watkins, and D. Valle. 1995. PXA1, a possible Saccharomyces
cerevisiae ortholog of the human adrenoleukodystrophy gene. Proceedings of the
National Academy of Sciences of the United States of America. 92:6012-6. 38. Shyamala, V., V. Baichwal, E. Beall, and G. Ames. 1991. Structure-function
analysis of the histidine permease and comparison with cystic fibrosis mutations.
Journal of Biological Chemistry 266:18714-18719.
39. Stagljar, I., C. Korostensky, N. Johnsson, and S. te Heesen. 1998. A genetic
system based on split-ubiquitin for the analysis of interactions between membrane
proteins in vivo. Proceedings of the National Academy of Sciences of the United
States of America 95:5187-92.
40. Tearle, R. G., J. M. Belote, M. McKeown, B. S. Baker, and A. J. Howells. 1989.
Cloning and characterization of the scarlet gene of Drosophila melanogaster.
Genetics. 122:595-606.
41. Walker, J. E., M. Saraste, M. J. Runswick, and N. J. Gay. 1982. Distantly related
sequences in the alpha- and beta-subunits of ATP synthase, myosin, kinases and other
ATP-requiring enzymes and a common nucleotide binding fold. Embo J 1:945-51.
42. Yabuuchi, H., S. Takayanagi, K. Yoshinaga, N. Taniguchi, H. Aburatani, and T.
Ishikawa. 2002. ABCC13, an unusual truncated ABC transporter, is highly
expressed in fetal human liver. Biochem Biophys Res Commun 299:410-7. II Application of the DHFR PCA to TAP family of
half ABC Transporters1
2.1 - Introduction: A need to assay for interactions in the TAP family
Perhaps the best characterized half ABC transporters are the ones which form the transporter associated with antigen processing (TAP). TAP translocates peptides from the cytoplasm into the lumen of the endoplasmic reticulum (ER) where they are bound by Major
Histocompatibility Complex (MHC) class I molecules for presentation to the immune system
(reviewed in (1)). This transporter is a heterodimer consisting of TAP 1 (ABCB2) and TAP2
(ABCB3). Genetic evidence that TAP1 and TAP2 form heterodimers comes from studies of human mutant cell lines which found that expression of both genes was needed for antigen processing (3, 19, 25). Co-immunoprecipitation studies indicate that TAP1 and TAP2 are found as a complex in the ER membrane (7,24). This TAP heterodimer can function without any additional factor of the immune system (12,28). A direct physical interaction has also been demonstrated between TAP1 and TAP2 when photo-reactive peptide analogues labeled both TAP1 and TAP2, suggesting that the peptide-binding site of TAP is formed by amino acids from both TAP1 and TAP2 (13). Crosslinking experiments and gel filtration analysis suggest that TAP1 and TAP2 form a functional heterodimer with a stoichiometry of
1:1 (10,12,16). Transmission electron microscopy has since confirmed that TAP 1 and
TAP2 form a single heterodimeric complex (29).
1 A version of this chapter has been published in: Dennis B. Leveson-Gower, Stephen W. Michnick, and Victor Ling. Detection of TAP Family Dimerizations by an In Vivo Assay in Mammalian Cells. Biochemistry. 2004 Nov 9; 43(44): 14257-64. The third and final member of the TAP family is the TAP-Like (TAPL, ABCB9) protein, which shares 38 and 40% amino acid sequence identity with TAP1 and TAP2, respectively.
There is some debate over whether TAPL is localized to lysosomes or the endoplasmic reticulum, where TAP1 and TAP2 are localized (8, 9, 30). Kobayashi et al. (8, 9) found that when TAPL fused to green fluorescent protein was transfected into Cos-1 cells, it had similar staining patterns to PDI and Brefeldin A-BODIPY (two ER markers) and Lysotracker (a lysosome marker); as the authors believe that the staining pattern of TAPL more closely matches the staining of the ER markers, they conclude that TAPL is primarily retained on the
ER membrane. In contrast to these results, Zhang et al. (30) found that when native TAPL was transfected into SKOV cells, the staining of TAPL correlated poorly with an ER marker
(calnexin), correlated well with the staining pattern of an endosome marker (protein transferrin receptor), and correlated very well with the staining pattern of a lysosome marker
(LAMP2); from these results, the authors concluded that TAPL resides primarily in endo- lysosomes. High expression of TAPL was found in testis, and moderate expression was found in brain, spinal cord, and thyroid (30). Staining patterns with anti-TAPL antibody indicate that it is expressed in the Sertoli cells of the seminiferous tubules (30), which form part of the blood-testis barrier separating spermatogonia from spermatocytes and spermatids.
Although its function is unknown, the high degree of amino acid sequence identity between
TAPL, TAP1 and TAP2 suggests that TAPL may be a peptide transporter that may dimerize with TAP 1 orTAP2. While it is well known that TAP1 forms a heterodimer with TAP2, there is also some evidence that TAP1 may form homodimers (5,11). When rat TAP1 was introduced into the murine small lung carcinoma cell line CMT.64 (a line that arose spontaneously in a C57BL/6 mouse with deficiencies in MHC class I surface expression and endogenous antigen presentation that can be reversed with interferon-y treatment), the cells gained the ability to be recognized by specific cytotoxic T-lymphocytes when infected with vesicular stomatitis virus (VSV) (5). Furthermore, the introduction of rat TAP1 caused VSV peptides to bind to putative lumenal ER proteins, suggesting that they had been transported by TAP1 alone (11). Their finding that proper immune function could be restored by introduction of rat TAP1 alone led these authors to conclude that TAP1 forms homodimers.
In addition to the question of TAP 1 homodimers, it is not known if TAP 1 and/or TAP2 may interact with TAPL, or if TAPL or TAP2 forms homodimers. While a TAP2 homodimer has been proposed (2), there is no conclusive biochemical evidence for such a pairing. As the dimerization state of TAPL has not been examined, it is possible that it may interact with TAP1, TAP2, or form homodimers. It may be that these transporters behave like White, Brown, and Scarlet in Drosophila, where White pairs with either Brown or
Scarlet to transport different substrates (4,14,18, 26). Therefore, there is a need to assay which interactions occur in vivo between all the members of the TAP family.
To study interactions amongst TAP family members, I adapted the dihydrofolate reductase protein-fragment complementation assay (DHFR PCA) to this family. This assay was shown to be suitable for the study of membrane-bound proteins in vivo when it was used to detect ligand-induced conformation change in the erythropoietin receptor (22). The
DHFR PCA is a direct way of determining if a physical interaction exists between two proteins. This method was chosen as it can detect protein-protein interactions embedded in
various subcellular membranes regardless of which membranes they are in, can work in
mammalian cells, and does not rely on indirect evidence of interaction. Using this approach,
I sought to confirm, in vivo, the interaction between TAP1 and TAP2, and to assay for other
interactions between TAP1, TAP2, and TAPL. 2.2 - Materials and Methods
2.2.1 — Creation of DHFR PCA vectors
As shown in a crystal structure of DHFR (Figure *ABC*), DHFR can be divided into three structural fragments, Fl, F2, and F3. Leucine zipper DHFR constructs, designated Z-
F[l,2] (for a leucine zipper with the N-terminal half of DHFR) and Z-F[3] (for a leucine zipper with a C-terminal half of DHFR) were gifts from Dr. Steven Michnick (Universite de
Montreal, Montreal, QC, Canada) (17). As outlined in Figure 2.2.2, The coding sequences for leucine zipper-Fragment 1/2 and leucine zipper-fragment 3 were released from the plasmids Z-F[l,2] and Z-F[3] with appropriate restriction endonucleases, gel purified, and ligated into digested, dephosphorylated, and gel-purified pcDNA3 (neomycin resistance gene, cyto-megalo virus (CMV) promoter, from Invitrogen, Burlington, Ontario, Canada) and pcDNA3.1MUT (Zeocin resistance gene, CMV promoter, from Invitrogen with the Nhel restriction site removed by site directed mutagenesis by Dr. Douglas Hogue) vectors, respectively. The resulting vectors were named Zip-Fl/2 and Zip-F3. Figure 2.2.1: Crystal structure of DHFR. Fragments F[l] and F [2] are shown in blue and cyan, and F[3] is shown in green. Arrows indicate location of fusion to oligomerizing proteins. Figure reproduced from Pelletier et al. (17). Digestion
I Zip |(GGGGS)x2| DHFR F1/21
Zip-F1/2
Digestion & dephosphorylation
Digestion
1 GGGGS)x2 DHFR F3 Zip |(GGGGS)x2|DHFRF3~| || Ligation Zip |(GGGGS)x2|c En PCDN/.3 2? jmultiple cloning site I """^ "j | ""^ Zip-F3
pcDNA3 1MUTt.A' , pcDNA31MUT
Digestion & dephosphorylation
Figure 2.2.2: Sub-cloning of zip-DHFR positive controls. For details, see text above.
pD3-F3 vector, a pcDNA3.1MUT vector with F3 fragment of DHFR inserted, was created by Dr. Douglas Hogue. As outlined in Figure 2.2.3, pD3-Fl/2 vector, was created by first performing PCR on the Zip-Fl/2 vector with primers DHFR12F (5'-GTAGCGGCC
GCTAGCGGTGGCGGTGGCTCTGGAGGTGGTGGG-3') and DHFR12R (5'-ACTA
GTTCTAGATTAGGTACCCAATTCCGGTTG-3') for 5 cycles at 52 °C followed by 25 cycles at 63 °C. DHFR12F contains Notl and Nhel restriction sites followed by a sequence complementary to the 5' end of Fl/2. DHFR12R is a reverse primer containing aXbal restriction site followed by a sequence complementary to the 3' end of Fl/2. The resulting 24 PCR product and pcDNA3 vector (Invitrogen) were then digested with Notl and Xbal. Next, the digested PCR product was gel purified, the digested pcDNA3 vector was dephosphorylated and gel-purified, and the two were ligated together.
PCR & Digestion
Zip |(GGGGS)x2 DHFR FI^T*^ |(GGGGS)x2| DHFR F1/2
Digestion & • pcDNA3 1MV i dephosphorylation pD3-F3 (from D.Hogue)
Figure 2.2.3: Creation of parental DHFR vectors with 10 amino acid linkers. For details, see text above.
To create parental DHFR vectors with 20 amino acid linkers, 10 amino acid linkers were
inserted into pD3-Fl/2 and pD3-F3 to create pD3-Fl/2-L and pD3-F3-L, respectively (Figure
2.2.4). First, 10 uL each of 0.25 ug / uL oligonucleotides NotLP5 (5'-
GGCCGCTGGTGGCGGTGGCTCTGGAGGTGGTGGGTCCTT-3') andNotLP3 (5'-
GGCCAAGGACCCACCACCTCCAGAGCCACCGCCACCAGC-3') were mixed and boiled for 1 minute and then left to cool to room temperature; this created a double-stranded
oligonucleotide containing the coding sequence for a 10 amino acid flexible linker
(GGGGSGGGGS) flanked by Notl site overhangs. Next, the double-stranded oligomer was
ligated into digested pD3-Fl/2 and pD3-F3 to create pD3-Fl/2-L and pD3-F3-L, respectively. The double-stranded insert was designed so that only the 5'-JVofI site would be regenerated in the resulting vectors.
pD3-F1/2-L pD3-F3-L
Figure 2.2.4: Modifying DHFR parental vectors to contain 20 amino acid linkers. For details, see text above.
DHFR PCA vectors for human TAP1, TAP2, and TAPL were created via the same method as the pD3-Fl/2 vector was created (Figure 2.2.3). Briefly, various primer pairs were used to add restriction endonuclease (RE) sites to remove the 3' stop codons of the TAP genes and enable the cloning of them in-frame into the various DHFR PCA vectors. Primers used for each PCR reaction are outlined in Table 2.2.1. Next, the PCR products were digested with the appropriate REs, gel purified, and ligated into a digested, dephosphorylated, and gel-purified plasmid. In total, 12 different DHFR PCA vectors were created for the TAP family (Figure 2.2.5). Original plasmids containing human TAP1 and TAP2 cDNAs were gifts from Dr. Thomas Spies (Fred Hutchinson Cancer Research Center,
Seattle, WA, U.S.A.). Human TAPL cDNA was provided by Dr. Fang Zhang from our lab.
After restriction sites were added to TAP1, TAP2, and TAPL by polymerase chain reaction (PCR) and ligated into the DHFR PCA vectors, all the constructs were sequenced for verification. TAP1 and TAPL inserts were identical to published sequences, but TAP2-
Lio-F3 and TAP2-L20-F3 contained the following changes: I V at amino acid (AA) 378
and a silent mutation at AA 385; TAP2-Lio-Fl,2 and TAP2-L2o-Fl,2 contained these two
changes as well as a R-> K at AA 661. These were considered acceptable because one mutation gave the same amino acid and other two gave amino acids that are similar in size
and character. 3'-RE For Forward Primer 5'-RE Reverse Primer site Creation of site Name Sequence Name Sequence added Vector: added: TAPl-Lio-Fl/2 5' -GCTAGG ATCC ATGGCTAG 5' -GCTAGCGGCCGCTTCTGGA TAPI-L20-FI/2 TAP1BH1 CTCTAGGTGTCCCGCTCCC- BamHl TAPINt GCATCTGCAGGAGCCTGCAC- Notl TAPl-Lio-F3 3' 3'
TAP1-L20-F3 TAP2-Lio-Fl/2 5' -GCTAAG ATCT ATGCGGCT TAP2Bgl2 CCCTGACCTGAGACCCTGG- Bgia Notl TAP2-L20-F1/2 3' 5'-GCTAGCGGCCGCGTCCATC TAP2Nt AGCCGCTGCTGAACCAGGCG- TAP2-Lio-F3 5' -GCTAGGTACC ATGCGGCT 3' TAP2KnI CCCTGACCTGAGACCCTGG- Kpnl Notl
TAP2-L20-F3 3'
TAPL-Lio-Fl/2 5' -GCTAGG ATCC ATGCGGCT 5' -GCTAGCGGCCGCGGCCTTG TAPL-L20-FI/2 CL015 BamHl CL013 Notl GTGG AAGGCG-3' TGACTGCCG-3'
TAPL-L10-F3
TAPL-L20-F3 Table 2.2.1: Primers used for the sub-cloning of TAP family cDNAs into DHFR PCA vectors. Each primer contains a RE site at its 5' end followed by a sequence complementary to either the 5' end of the TAP cDNA or the antisense of the 3' end of the TAP cDNA. As a result, PCRs were performed at a lower annealing temperature (46 °C) for the first 5 cycles because only a portion of the primer is binding to the template. After the first 5 cycles, the PCR product generated could serve as the template for future cycles, so the annealing temperature was increased (60 °C) to ensure specific annealing. Figure 2.2.5: Maps of DHFR constructs. Constructs from left to right are: (A) TAPl-Lio-
Fl/2, (B) TAP2-L10-Fl/2, (C) TAPL-Li0-Fl/2, (D) TAPl-L20-Fl/2, (E) TAP2-L20-Fl/2, (F)
TAPL-L20-Fl/2, (G) TAP1-L10-F3, (H) TAP2-Li0-F3, (I) TAPL-Li0-F3, (J) TAP1-L20- F3,
(K) TAP2-L20-F3, (L) TAPL-L20-F3, (M) Zip-Fl/2, and (N) Zip-F3. Flexible linker consists of repeating units of the amino acids GGGGS. *Note: A BgRl site on the 5' end of TAP2-L-
Fl/2 was used to clone into the BamHl site of pcDNA3 thus eliminating both sites. 2.2.2 — Cell lines and culture
The Chinese hamster ovary (CHO) cell line CHO DUKX-B11, a DHFR-deficient cell line originally characterized by L. Chasin (6,27), was maintained in Minimum Essential alpha Medium with 292 mg / L L-glutamine, without ribonucleosides and deoxyribonucleosides (Invitrogen), and supplemented with 105 units Penicillin-Streptomycin
(Invitrogen) per mL, 105 uM sodium hypoxanthine, 16.8 uM thymidine (from hypoxanthine and thymidine (HT) supplement—Invitrogen), and 10.5 % dialyzed Fetal Bovine Serum
(FBS, Hyclone, Logan, UT, U.S.A.). Nucleotide-free medium is identical to the above medium except that it does not contain the HT supplement. All cell lines were grown at 37
°C in a 5 % CO2 atmosphere in a humidified incubator.
2.2.3 — DHFR survival assay
To detect interactions between members of the TAP family, TAP genes are fused to complementary N- and C-terminal fragments (fragments Fl/2 and F3) of the DHFR gene. If the two half ABC transporters dimerize, they will induce the folding of Fl/2 and F3 fragments into an active DHFR molecule. In DHFR-deficient cells, introduction of active
DHFR will enable the cells to survive in nucleotide-free media. Leucine zipper vectors Zip-
Fl/2 and Zip-F3 were co-transfected as positive controls. TAPI-L20-FI/2, TAPI-L20-F3,
TAP2-L20-F1/2, TAP2-L20-F3, TAPL-L20-FI/2, and TAPL-L20-F3, were then paired in different combinations to determine which interactions occur. For example, a TAP1-TAP2 interaction was assayed by co-transfecting TAPI-L20-FI/2 and TAP2-L20-F3 into DHFR- deficient CHO DUKX-B11 cells and selecting for reconstituted DHFR with nucleotide-free medium. Negative controls were leucine zipper DHFR constructs paired with TAP-DHFR constructs, which should not dimerize. Plasmids used in transfections were prepared from
TOPI OF E. coli (Invitrogen) with a QIAfilter Plasmid Midi Kit (Qiagen Inc., Mississauga,
Ontario, Canada). CHO DUKX-B11 cells (4 x 105) were plated on 15 cm plates (176 cm2) and co-transfected with pairs of plasmids using Polyfect (Qiagen) following the manufacturer's instructions. To select for interacting pairs, the medium was exchanged 48 hours later for nucleotide-free medium. Colonies were counted after 10 days of selection.
2.2.4 —FACS
For detection of DHFR, approximately 8 x 10s cells were incubated in media containing
10 uM AlexaFluor-488-Methotrexate (Molecular Probes) for 17 hours. Cells were de- stained by washing twice with 2 mL PBS and incubating them in 2 mL media without dye for 2 hours. Then cells were trypsinized, suspended as single cells, collected by centrifugation at 500 x g, and re-suspended in 1 mL cold PBS with 10 % FBS. Analysis was carried out with 488 nm excitation light and a 525 nm emission filter on a Coulter Epics Elite
ESP (Beckman Coulter Canada Inc., Mississauga, ON, Canada). 2.3 - Results
2.3.1 — DHFR PCA survival-selection in nucleotide-free media
Interactions between members of the TAP family were assayed by co-transfecting TAP-
DHFR constructs with a 20 amino-acid flexible linker between the C-termini of TAP and the
N-terminal of Fl/2 or F3 (Figure 2.2.5). The linker was increased to 20 amino acids because early experiments with a 10 amino acid linker did not give significant colony counts above those of negative controls. DHFR constructs with leucine zippers, short proteins that dimerize, were co-transfected as positive controls. Leucine zipper-DHFR constructs paired with TAP-DHFR constructs served as negative controls as they are not expected to interact. As indicated in Figure 2.3.1, the standard deviation was quite high between trials of the assay; however, there is a significant and consistent difference between the colony counts of positive and negative controls. Positive control leucine zipper pairs produced 12.8 colonies per 4 x 105 cells. The known TAP1-TAP2 heterodimers yielded an average of 4.4 colonies. TAPL-TAPL produced a mean of 7.1 colonies, suggesting that TAPL forms homodimers. Negative controls (TAPs with leucine zippers) produced between 0.1 and 0.3 colonies. TAP1-TAP1; TAP 1-TAPL, TAPL-TAP2, and TAP2-TAP2 produced 0.3, 0.9,
0.1 and 0.4 colonies, respectively. As these values were not significantly higher than those of the negative controls, interactions between these pairs were not indicated. Surviving
colonies were subsequently cross-validated for the presence of functional DHFR by FACS
analysis (see below). The presence of colonies on negative control plates, which indicates that a background of nucleotide-independent false positive colonies can occur, was
investigated as well (see Chapter 3). Z = Leucine zipper T1 = TAP1
15- T2 = TAP2 TL = TAPL
0) U in 10
m .2 5
1 0 4 -1 Z/Z T1/T2 TL/TL T1/T1 T1/TL TL/T2 T2/T2 T1/Z T2/Z TL/Z
Figure 2.3.1: Colony counts from survival assay. Pairs of plasmids were co-transfected into DHFR-deficient CHO cells. The media was exchanged 48 hours later for nucleotide- free media to select for interacting pairs. Colonies were counted after 10 days of selection.
Bars represent the mean of 10 trials with the standard deviation indicated.
2.3.2 — FACS analysis of clones
As DHFR binds fluorescent methotrexate in a one-to-one ratio with high affinity, an increase in fluorescence would confirm the presence of reconstituted DHFR in the cells.
Cells from nucleotide-free media selection assays were stained with AlexaFluor-488-
Methotrexate and analyzed by FACS with 488 nm excitation light and a 525 nm emission filter. Stained, non-transfected CHO DUKX-B11 cells served as negative controls against which fold increase in fluorescence was calculated. All colonies from the survival assay trapped fluorescent methotrexate, indicating the presence of DHFR. On average, cells were
2.5 to 5 times more fluorescent than control cells. Four representative FACS histograms, one each for TAP1-TAP2 heterodimers, TAPL-TAPL homodimers, leucine zipper pairs (the positive control), and a false positive colony (leucine zipper with TAP1), are shown in Figure
2.3.2. Figure 2.3.2: Relative increase in fluorescence between non-transfected cells and colonies from survival-selection. Unfilled histograms labeled DHFR"7" are non-transfected control cells. Filled peaks represent clones of (A) TAP1-TAP2 heterodimers, (B) TAPL-
TAPL homodimers, (C) leucine zipper positive controls, and (D) a leucine zipper paired with
TAP1 (a false positive colony as these two proteins are not expected to interact). Cells were incubated overnight (17 hours) in AlexaFluor-488-Methotrexate. Cells were de-stained in media without dye for 2 hours. FACS analysis was carried out with 488 nm excitation and
525 nm emission. Zero point for fluorescence is set arbitrarily with each experiment; thus, absolute value comparison cannot be compared between experiments. 2.4 - Discussion
This study is the first to directly assay for in vivo interactions between all members of the
TAP family of human half ABC transporters. I successfully adapted the DHFR PCA to the
TAP family and used it to confirm the interaction of TAP 1 and TAP2 in vivo. While there are many data suggesting that TAP1 and TAP2 form a functional dimer, this is the first time a direct interaction has been shown in intact, living cells. As shown in Figure 2.3.1,1 found that TAPL forms homodimers; however, I was not able to detect any other interactions between TAP1, TAP2, and TAPL. These results provide evidence that the transporters in the
TAP family form exclusive pairs, and that TAP1 does not form homodimers.
The discovery that members of the TAP family form exclusive dimers raises several questions. Since TAP1 only forms a heterodimer with TAP2, it is interesting that its transcript would be expressed at much higher levels than TAP2 in most tissues (20). Also, despite contrary examples such as the White protein in Drosophila where White pairs with either Brown or Scarlet to transport different substrates (4,14,18,26), I did not find any of the TAP proteins to have more than one dimerization partner. If the transporters do not have multiple dimerization states, then why are half transporters not constructed as full transporters? It could be that it is easier for the organism to regulate expression and function by having half transporters, or even that the transport of substrates is regulated by the association and dissociation of half ABC transporters. Another possibility is that encoding the TAP transporter on two separate genes is a way to generate more diversity in a population, and, ultimately, resistance to a wider variety of pathogens. TAP1 homodimerizations were not detected with the DHFR PCA, which is in contrast with previous suggestions that TAP1 may function without TAP2 (5,11). The introduction of rat TAP to small cell lung carcinoma cells (CMT.64 cells) caused them to be recognized by specific CTL when infected with vesicular stomatitis virus (VSV) (5). Furthermore, VSV peptides became bound to putative lumenal ER proteins after exogenous rat TAP1 was introduced (11). While the data from the CMT.64 cell line is intriguing, it has not been replicated in other cell lines or in vitro. Furthermore, CMT.64 cells, which are able to express MHC class I when induced by interferon-gamma, do not appear to have any defect in the TAP1 or TAP2 proteins themselves (15,23). Thus, introduction of surplus TAP1 into these cells may promote the expression or stabilization of TAP2. It may also be that the exogenous TAP1 could contribute to a "leakiness" of the ER membrane in these cells or affect vesicular transport in some way. Thus, while the idea of a TAP1 homodimer is intriguing, I did not evidence of TAP 1 homodimerization. It should be noted, however, that if such interactions occur at a very low level (e.g., less than 25 dimers per cell), they would not be detected by the DHFR PCA (21).
The discovery that TAPL forms homodimers is fundamental to expanding our knowledge about the function of this poorly understood half ABC transporter. The ability to homodimerize suggests that TAPL can function independently of other half ABC transporters. Thus, I can begin studying the function of TAPL, as I know that it homodimerizes to form a "full" ABC transporter. The lack of an association between TAPL and TAP1 or TAP2 is consistent with TAPL residing in lysosomes (30), a location where no other mammalian ABC transporters have been found. Bibliography
1. Abele, R., and R. Tampe. 2004. The ABCs of immunology: structure and function
of TAP, the transporter associated with antigen processing. Physiology 19:216-24.
2. Antoniou, A. N., S. Ford, E. S. Pilley, N. Blake, and S. J. Powis. 2002. Interactions
formed by individually expressed TAP1 and TAP2 polypeptide subunits.
Immunology. 106:182-9.
3. Cerundolo, V., J. Alexander, K. Anderson, C. Lamb, P. Cresswell, A.
McMichael, F. Gotch, and A. Townsend. 1990. Presentation of viral antigen
controlled by a gene in the major histocompatibility complex. Nature 345:449-52.
4. Dreesen, T. D., D. H. Johnson, and S. Henikoff. 1988. The brown protein of
Drosophila melanogaster is similar to the white protein and to components of active
transport complexes. Molecular & Cellular Biology. 8:5206-15.
5. Gabathuler, R., G. Reid, G. Kolaitis, J. Driscoll, and W. A. Jefferies. 1994.
Comparison of cell lines deficient in antigen presentation reveals a functional role for
TAP-1 alone in antigen processing. Journal of Experimental Medicine. 180:1415-25.
6. Graf, L. H., Jr., and L. A. Chasin. 1982. Direct demonstration of genetic alterations
at the dihydrofolate reductase locus after gamma irradiation. Molecular & Cellular
Biology 2:93-6.
7. Kelly, A., S. H. Powis, L. A. Kerr, I. Mockridge, T. Elliott, J. Bastin, B.
Uchanska-Ziegler, A. Ziegler, J. Trowsdale, and A. Townsend. 1992. Assembly
and function of the two ABC transporter proteins encoded in the human major
histocompatibility complex. Nature 355:641-4. 8. Kobayashi, A., M. Kasano, T. Maeda, S. Hori, K. Motojima, M. Suzuki, T.
Fujiwara, E. Takahashi, T. Yabe, K. Tanaka, M. Kasahara, Y. Yamaguchi, and
M. Maeda. 2000. A half-type ABC transporter TAPL is highly conserved between
rodent and man, and the human gene is not responsive to interferon-gamma in
contrast to TAP1 and TAP2. Journal of Biochemistry. 128:711-8.
9. Kobayashi, A., T. Maeda, and M. Maeda. 2004. Membrane Localization of
Transporter Associated with Antigen Processing (TAP)-Like (ABCB9) Visualized in
Vivo with a Fluorescence Protein-Fusion Technique. Biological & Pharmaceutical
Bulletin 27:1916-22.
10. Lacaille, V. G., and M. J. Androlewicz. 1998. Herpes simplex virus inhibitor ICP47
destabilizes the transporter associated with antigen processing (TAP) heterodimer.
Journal of Biological Chemistry. 273:17386-90.
11. Marusina, K., G. Reid, R. Gabathuler, W. Jefferies, and J. J. Monaco. 1997.
Novel peptide-binding proteins and peptide transport in normal and TAP-deficient
microsomes. Biochemistry. 36:856-63.
12. Meyer, T. H., P. M. van Endert, S. Uebel, B. Ehring, and R. Tampe. 1994.
Functional expression and purification of the ABC transporter complex associated
with antigen processing (TAP) in insect cells. FEBS Letters. 351:443-7.
13. Nijenhuis, M., S. Schmitt, E. A. Armandola, R. Obst, J. Brunner, and G. J.
Hammerling. 1996. Identification of a contact region for peptide on the TAP1 chain
of the transporter associated with antigen processing. Journal of Immunology.
156:2186-95. 14. O'Hare, K., C. Murphy, R. Levis, and G. M; Rubin. 1984. DNA sequence of the
white locus of Drosophila melanogaster. Journal of Molecular Biology. 180:437-55.
15. Ortiz-Navarrete, V., A. Seelig, M. Gernold, S. Frentzel, P. M. Kloetzel, and G. J.
Hammerling. 1991. Subunit of the '20S' proteasome (multicatalytic proteinase)
encoded by the major histocompatibility complex. Nature. 353:662-4.
16. Ortmann, B., J. Copeman, P. J. Lehner, B. Sadasivan, J. A. Herberg, A. G.
Grandea, S. R. Riddell, R. Tampe, T. Spies, J. Trowsdale, and P. Cresswell. 1997.
A critical role for tapasin in the assembly and function of multimeric MHC class I-
TAP complexes. Science 277:1306-9.
17. PeUetier, J. N., F. X. CampbeU-Valois, and S. W. Michnick. 1998.
Oligomerization domain-directed reassembly of active dihydrofolate reductase from
rationally designed fragments. Proceedings of the National Academy of Sciences of
the United States of America 95:12141-6.
18. Pepling, M., and S. M. Mount. 1990. Sequence of a cDNA from the Drosophila
melanogaster white gene. Nucleic Acids Research. 18:1633.
19. Powis, S. J., A. R. Townsend, E. V. Deverson, J. Bastin, G. W. Butcher, and J. C.
Howard. 1991. Restoration of antigen presentation to the mutant cell line RMA-S by
an MHC-linked transporter. Nature. 354:528-31.
20. Ralph, S. G. 2002. mRNA expression profiling of the ATP-binding cassette
transporter family. Ph.D. The University of British Columbia, Vancouver.
21. Remy, I., and S. W. Michnick. 1999. Clonal selection and in vivo quantitation of
protein interactions with protein-fragment complementation assays, [erratum appears in Proc Natl Acad Sci U S A 1999 Jun 22;96(13):7610]. Proceedings of the National
Academy of Sciences of the United States of America. 96:5394-9.
22. Remy, I., I. A. Wilson, and S. W. Michnick. 1999. Erythropoietin receptor
activation by a ligand-induced conformation change. Science 283:990-3.
23. Sibille, C, K. Gould, G. Hammerling, and A. Townsend. 1992. A defect in the
presentation of intracellular viral antigens is restored by interferon-gamma in cell
lines with impaired major histocompatibility complex class I assembly. European
Journal of Immunology. 22:433-40.
24. Spies, T., V. Cerundolo, M. Colonna, P. Cresswell, A. Townsend, and R. DeMars.
1992. Presentation of viral antigen by MHC class I molecules is dependent on a
putative peptide transporter heterodimer. Nature 355:644-6.
25. Spies, T., and R. DeMars. 1991. Restored expression of major histocompatibility
class I molecules by gene transfer of a putative peptide transporter.[see comment].
Nature 351:323-4.
26. Tearle, R. G., J. M. Belote, M. McKeown, B. S. Baker, and A. J. Howells. 1989.
Cloning and characterization of the scarlet gene of Drosophila melanogaster.
Genetics. 122:595-606.
27. Urlaub, G., and L. A. Chasin. 1980. Isolation of Chinese hamster cell mutants
deficient in dihydrofolate reductase activity. Proceedings of the National Academy of
Sciences of the United States of America 77:4216-20.
28. Urlinger, S., K. Kuchler, T. H. Meyer, S. Uebel, and R. Tampe. 1997. Intracellular
location, complex formation, and function of the transporter associated with antigen
processing in yeast. European Journal of Biochemistry. 245:266-272. 29. Velarde, G., R. C. Ford, M. F. Rosenberg, and S. J. Powis. 2001. Three-
dimensional structure of transporter associated with antigen processing (TAP)
obtained by single Particle image analysis. Journal of Biological Chemistry.
276:46054-63.
30. Zhang, F., W. Zhang, L. Liu, C. L. Fisher, D. Hui, S. Childs, K. Dorovini-Zis,
and V. Ling. 2000. Characterization of ABCB9, an ATP binding cassette protein
associated with lysosomes. Journal of Biological Chemistry. 275:23287-94. III Utility of the DHFR PCA for the study of ABC transporters1
3.1 - Introduction
Following the successful application of the DHFR PCA to the TAP family of half ABC transporters, I decided to evaluate the general utility for this assay in more detail. Although meaningful conclusions could be drawn regarding which interactions could occur in vivo, the presence of colonies on negative control plates was unexpected and puzzling. Thus, I investigated how these false positive colonies could occur and how I could accurately identify false positive colonies in the future. Furthermore, I wanted to determine whether the fusion of DHFR to the C-terminus of the half ABC transporters inhibited their ability to function.
To determine how false positive colonies could occur, one must first speculate on what possible events could lead to the generation of these colonies. It has already been demonstrated that it is impossible for the DHFR fragments to interact without being fused to oligomerizing proteins or domains (3). Thus, there are two probable origins of false positive colonies (Figure 3.1.1). One possibility is that the CHO cell line CHO DUKX-B11 cell line is reverting to a wild-type phenotype by somehow re-expressing its endogenous DHFR. This is possible because when the CHO DUKX-B11 deficient cell line was originally created and characterized (6), the nature of the knock-down was not precisely defined. The DUKX-B11
A version of this chapter has been published in: Dennis B. Leveson-Gower, Stephen W. Michnick, and Victor Ling. Detection of TAP Family Dimerizations by an In Vivo Assay in Mammalian Cells. Biochemistry. 2004 Nov 9; 43(44): 14257-64. cell line was created by selecting against DHFR activity in the presence of the mutagen ethyl methanesulfonate and then y-rays. The loss of DHFR catalytic activity was used as indication that functional DHFR was not present. The other possibility is that the murine
DHFR fragments Fl/2 and F3 (that were transfected into the CHO cells on two separate plasmids) have recombined with each other or with some endogenous hamster DHFR sequences. This could create a new functional DHFR molecule that would enable survival in nucleotide free media (Figure 2.3.1) and trap fluorescent methotrexate (Figure 2.3.2), as was observed.
1. DHFR negative cells have reverted to a wild-type phenotype
2. A recombination event has occurred.
Figure 3.1.1: Possible origins of false positive colonies. For details, see text above.
My second objective was to determine whether the fusion of DHFR to the C-terminus of
the half ABC transporters inhibited their ability to function. If the fusion of either DHFR
fragments Fl/2 or F3 does not inhibit the activity of ABC transporters, it means that the assay could also be applied to srrucmre-function studies, including whether function is regulated by the association / disassociation of half ABCs or if inhibitors of function work by disrupting a functional dimer. As illustrated in Figure 3.1.2, the function of the TAP1-TAP2 heterodimer is to translocate peptides from the cytoplasm into the lumen of the endoplasmic reticulum (ER), where they are assembled into Major Histocompatibility Complex (MHC) class I molecules for presentation to the immune system (reviewed in (1)). To determine if
TAPI-L20-FI/2 and TAP2-L20-F3 DHFR constructs could function, I transfected them into
TAP1- and TAP2- deficient cells, respectively. If the introduced TAPl-L20-Fl/2 and TAP2-
L20-F3 were able to restore MHC class I expression in these cell lines, it would indicate that the modified TAP proteins are still able to function. This method has also been used by others as an indication that modified TAPs were able to function in the antigen processing machinery (2). MHC class I
Figure 3.1.2: MHC class I expression pathway. Antigens processed by the proteosome are transported by TAP, assembled into MHC class I, and presented on the cell surface. 3.2 - Material and methods
3.2.1 — Cell lines and culture
BRE-169 (TAP 1-deficient), STF1-169 (TAP2-deficient), and STF1-169/TAP1.2
(TAP2-complemented STF1-169) cell lines were all gifts from Dr. Henri de la Salle
(Etablissement Francais du Sang - Alsace, Strasbourg Cedex, France). STF-169 and BRE-
169 cell lines were made from STF1 and BRE are TAP2 and TAP1 deficient primary skin
fibroblast cell lines, respectively, which were transfected with t and T antigen of SV40, and human telomerase (unpublished). Each line was cultured in Dulbecco's Modified Eagle
Medium with 4.5 g / L glucose, 584 mg / L L-glutamine, and 110 mg / L sodium pyruvate
and pyridoxine hydrochloride (Invitrogen), supplemented with 105 units Penicillin-
Streptomycin (Invitrogen) per mL, and 10.5 % dialyzed Fetal Bovine Serum (Hyclone). All
cells lines were grown at 37 °C in a 5 % C02 atmosphere in a humidified incubator.
3.2.2 — Western blots
Crude membranes were prepared by suspending approximately 8 x 105 cells in 500 pL phosphate buffered saline (PBS) containing 10 uL of Protease Inhibitor Cocktail
(Pharmingen, San Diego, CA, U.S.A.) and sonicating for 30 seconds on ice. Nuclei and
whole cells were removed by centrifuging at 500 x g for 5 minutes at 4 °C. The supernatant was transferred to a new tube and membranes were pelleted by centrifuging at 48,000 x g for
3 hours at 4 °C. Pellets were solubilized with 100 uL of 2% sodium dodecyl sulfate (SDS) in
PBS without calcium and magnesium (Invitrogen). Protein concentration was determined via the Bicinchoninic Acid Assay (Pierce
Biotechnology, Rockford, IL, U.S.A.). Each sample (8 p.g total protein) was loaded onto a
No vex 4-12 % Bis-Tris gel (Invitrogen) and electrophoresed according to the manufacturer's instructions. Anti-mouse DHFR monoclonal antibody was from Research Diagnostics Inc.
(Flanders NJ, U.S.A.); the identity of the epitope that this antibody binds to is proprietary, but my results indicate it is binding to fragment 3 of DHFR. Anti-TAPl monoclonal antibody # 148.3 was a gift from Dr. Robert Tampe (Johann Wolfgang Goethe-Universitat
Frankfurt am Main, Institute of Biochemistry, Frankfurt, Germany), and anti-TAP2 monoclonal antibody # 429.5 was a gift from Dr. Peter M. Van Endert (Institut National de la
Sante' et de la Recherche Medicale, Paris, France). Anti-TAPL polyclonal antibody IB3B was produced in our laboratory. All blots were visualized with either anti-mouse or anti- rabbit HRP-conjugated antibodies (Jackson ImmunoResearch Laboratories, Inc., West Grove,
PA, U.S.A.) and ECL+ (Amersham Biosciences Corp., Piscataway, NJ, U.S.A.).
3.2.3 — RNA purification and reverse transcription
Purification of RNA was achieved with the RNeasy Midi Kit (Qiagen) following the manufacturer's instructions. To remove any remaining genomic DNA, RNA samples were then treated with deoxyribonuclease I, as per the manufacturer's instructions, except that the concentration of RNA in the reaction was doubled. Reverse transcription (RT) was performed using the Superscript First-Strand Synthesis System for RT-PCR (Invitrogen) following the manufacturer's instructions using either oligo dT (for amplification of full- length hamster DHFR), or the gene-specific primers MUSB2R (5' -TTCCGGTTGTTCAA TAAGTC-3') for murine DHFR fragment Fl/2, and MUSD2R (5'-CTTCTCGTAGACTT
CAAAC-3') for murine DHFR fragment F3.
3.2.4 — PCR of false positive transcripts
cDNAs generated by reverse transcription with oligo dT were used as templates for all
PCR reactions. To detect full-length hamster DHFR cDNA, the primer DHUSF (5'-CGCG
CCAAACTTGGGGGAAGCACAGCGTAC-3'), complementary to the 5' untranslated region (UTR), and the primer DHDSR (5' -GGAGGAAAGCAGTAGAACTTGAAGTCAA
TC-3'), complementary to the 3' UTR, were used at an annealing temperature of 55 °C. To detect murine DHFR over a hamster DHFR background, murine specific primers MUSA2F
(5'-CGGAGACCTACCCTGGCCTC-3') and MUSB2R were used at 54 °C for Fl/2, and primers MUSC2F (5'-GAATCAACCAGGCCACCTT-3') and MUSD2R were used at 52 °C for F3. Annealing temperatures, which would specifically amplify murine Fl/2 and F3 over endogenous hamster transcript, were determined by temperature gradient PCR (data not shown). Samples were run on No vex 10 % TBE gels (Invitrogen) according to the manufacturer's instructions and stained with SYBR Gold nucleic acid gel stain (Molecular
Probes, Eugene, OR, U.S.A.).
3.2.5 — MHC Class I expression assay
Cells were seeded at a density of 4 x 105 cells per well of a 6-well plate (9.6 cm2). After
18 hours, cells were transfected with either TAPI-L20-FI/2 or TAP2-L20-F3 using Polyfect
(Qiagen) following the manufacturer's instructions. Cells were grown for 2 days before being trypsinized into single-cell suspensions and centrifuged down at 411 x g into wells of a 96-U-well plate. Cells were then re-suspended in 100 pL media with 10 pL of R- phycoerythrin-conjugated W6/32 antibody (DakoCytomation, Mississauga, Ontario, Canada) and incubated at 4 °C for 30 minutes in the dark. Samples were then centrifuged and washed three times with 200 pX of cold PBS (Invitrogen) supplemented with 10 % FBS. After re- suspension in 500 pL of the wash buffer, the cells were analyzed by fluorescence-activated cell sorting (FACS).
3.2.6 —FACS
For detection of MHC class I molecules, stained cells were analyzed with 488 nm excitation light and a 575 nm emission filter on a Coulter Epics Elite ESP (Beckman Coulter
Canada Inc., Mississauga, ON, Canada). 3.3 - Results
3.3.1 — Western blot analysis of protein expression in colonies from the survival assay
To confirm that colony growth from survival assay (Chapter 2) was a result of the genes transfected, protein expression from selected clones was analyzed by Western blot (Figure
3.3.1). Negative control DHFR-deficient CHO-DUKX-B11 cells, which were not transfected, did not show any bands reactive to TAP1- (not shown), TAP2-, TAPL-, or
DHFR-reactive antibodies. When probed with anti-TAP2 antibody, five different TAP1-
TAP2 clones all produced a band of approximately 80 kDa, which is close to the expected size for a 75 kDa TAP2 molecule fused to the 9 kDa F3 fragment of DHFR. When these clones were probed with anti-TAPl antibodies, a similar band was seen (data not shown).
Colonies expected to contain TAPL homodimers also produced a band at approximately 80 kDa, which is also close to the expected size for a 72 kDa TAPL protein fused to the 9 kDa
F3 fragment of DHFR. Leucine-zipper positive control colonies had the expected 16 kDa band for the Zipper-F3 protein. All TAP1-TAP2, and TAPL-TAPL clones were also confirmed by probing with anti-DHFR antibody (data not shown). These results indicate that colonies which grew from the transfection of leucine zipper homodimers, TAP1-TAP2 dimers, and TAPL homodimers, do express the appropriate DHFR fusion proteins. However,
TAP1-TAP1, TAP1-TAPL, and negative control leucine zipper-TAP colonies all expressed a 21 kDa band, which is the predicted size of full-length DHFR. As full-length DHFR was never transfected into any of these cells, these colonies are considered false positives. N (N N IN IN — •75 Q-Q-Q-a-a._Q-Q._Q. > QTQ.Q.Q.5L>CLCLQ.QI 2I<<<<<._•<<<< P P P P N P P '
• |fc antj.TAp2 mAb
"m Q.CLQ_Q-Q._Q-CL0CL Q S_L?_l_Q-L?_iS'L? > o.o.a.a.o.>o.iii i->!Si-i-i-i_Ri_f_i_!_
anti-TAPL pcAb Q. Q_ Q a a a a a | S i p § — MMNrslMPPPl^P Q Q Q Q 5 ,o._,_.a_.d. £ £ £ f . „ _>NNRHNRFPFP 3°-«
anti-DHFR mAb
Figure 3.3.1: Western blot analysis of clones from the DHFR survival assay. Crude lysates were prepared from each clone and run on a 4-12 % Bis-Tris SDS-PAGE. Gels were then transferred to PVDF membrane and probed with anti-TAP2 monoclonal, anti-TAPL polyclonal, or anti-DHFR monoclonal antibodies. Lane labels indicate which two DHFR constructs were used in the generation of each clone, and -/- indicates DHFR-deficient CHO-
DUKX-B11 cells, which were not transfected.
3.3.2 — Characterization of transcripts expressed in false positive colonies
In order to understand the origin of the false positives, mRNA expression levels were examined (Figure 3.3.2). Using cDNA pools prepared from different cell lines selected in the survival assay, specific primer pairs were used to determine whether mouse-specific or hamster-specific sequences were present in the various clones. Primers specific to the 5' and 3' UTRs of hamster DHFR indicated that full-length transcript for hamster DHFR (expected size = 668 base pairs (bp)) was present in all the selected clones and in non-transfected CHO
DUKX-B11 cells (Figure 3.3.2 A). The DNA sequences of the PCR fragments were determined and matched the hamster DHFR sequence except for a C -> G transversion at position 410 (A of ATG =1) resulting in an amino acid change of a threonine to an arginine.
Thus it is apparent that the DUKX-B11 cell line contains full-length mRNA, yet lacks DHFR protein. When the mRNA expression results are combined with the results from the Western analysis, which did not indicate full-length DHFR in positive clones or untransfected cells, they indicate that CHO DUKX-B11 cells are DHFR-deficient due to a defect in the translation of the DHFR transcript.
To determine if the false positive colonies arose by re-expressing endogenous hamster
DHFR, or by recombination of the murine DHFR fragments introduced, I determined which clones contained transcripts with murine DHFR sequences. To detect murine fragments Fl/2 and F3,1 used murine-specific primers at annealing temperatures which would specifically amplify murine Fl/2 and F3 over endogenous hamster transcript. Using these conditions, I found murine Fl/2 fragments (expected size = 225 bp) expressed only in clones from the survival assay which had arisen from the transfection of TAP1-TAP2, TAPL-TAPL, and leucine zipper pairs (Figure 3.3.2 B). Similarly, murine F3 fragments (expected size =172 bp) were also expressed only in clones from TAP1-TAP2, TAPL-TAPL, and leucine zipper pairs (Figure 3.3.2 C). As all the colonies classified as false positives did not express murine
DHFR transcript, the false positive colonies most likely arose from the CHO DUKX-B11 cells re-expressing their endogenous hamster DHFR and not some other genetic event involving the transfected murine DHFR fragments.
CM -J _J 1- OL CL Q. OL • ^ P e 3? E E Nj a h J- Q_ Q. ~^ O. CL Q. N r ^ w < 200 bp 100 bp Murine F3 (173 bp) Figure 3.3.2: RT-PCR detection of transcripts expressed in colonies from the survival assay. (A) Amplification with primers to the 5' and 3' UTR of hamster DHFR. (B) Amplification with primers to the F1/2 fragment of murine DHFR. (C) Amplification with primers to the F3 fragment of murine DHFR. 3.3.3 — Effect of DHFR fusion on TAP1 and TAP2 function Expression of MHC class I molecules on the surface of cells is dependent on peptides being transported by TAP1 and TAP2. When peptide transport is defective, MHC class I heavy chains remain trapped in the ER and, consequently, surface expression of MHC class I is severely reduced. To determine whether TAPl-L20-Fl/2 and TAP2-L20-F3 DHFR constructs could function in antigen presentation, I transfected them into TAP 1-deficient or TAP2-deficient cell lines to determine if they would facilitate MHC class I expression. As these experiments were performed with transient transfection of the genes, we would expect a heterogeneous population due to only a portion of the cells being successfully transfected and varying expression levels of the introduced genes. Expression of MHC class I was detected by fluorescent W6/32 antibody. Figure 3.3.3 B shows that the major peak of positive control wild-type-like (STF1-169/TAP1.2) cells is approximately 50 times more fluorescent than the major peak of TAP2-deficient (STF1-169) cells. When TAP2- J_20-F3 is transfected into the TAP2-deficient cells, a large increase in fluorescence in observed with a significant proportion of the transfectant population reaching wild-type levels of MHC class I expression. A similar effect is seen when TAPl-L2o-Fl/2 is transfected into the TAP1- deficient cells (Figure 3.3.3 A), where there is also a shift in fluorescence with two peaks forming—the second peak resembles that seen in the STF1-169/TAP1.2 cells. These data suggests that the attachment of fragment Fl/2 or fragment F3 to the TAPs does not abolish their ability to function in antigen presentation on MHC class I molecules. It should be noted, however, that this assay investigated TAP1- and TAP2-DHFR fusion proteins individually, where only one protein of the TAP1-TAP2 heterodimer is a TAP-DHFR fragment chimera. There is a possibility that a TAP 1-TAP2 heterodimer where both proteins contain DHFR fragment fusions might function differently. Fluorescence Figure 3.3.3: FACS of MHC class I expression. Dotted histogram marked TAPr" is BRE- 169 (TAP 1-deficient) cells; dotted histogram marked TAP2"A is STF1-169 (TAP2-deficient) cells. Dashed histograms marked wt are STF1-169/TAP1.2 (TAP2-complemented STF1- 169) cells which served as positive controls. The filled peak in A is BRE-169 cells transfected with TAPl-L20-Fl/2. The filled peak in B is STF1-169 cells transiently transfected with TAP2-L2o-F3. MHC class I expression was determined by labeling with R- phycoerythrin-conjugated W6/32 antibody. Peaks were generated by FACS analysis with 488 nm excitation and 575 nm emission. Zero point for fluorescence is set arbitrarily with each experiment; thus, absolute values in A cannot be compared with values in B. 3.4 - Discussion In my screens, false positives were observed at a rate comparable to those observed for screening of protein-protein interactions by PCA (0.0002%)(5). However, the rate of false positives was, in all cases, significantly less than that observed for true positives. Thus, although there is considerable variation between samples, positive samples are unambiguously positive. Western blot analysis and RT-PCR indicated that these colonies express full-length endogenous hamster DHFR. Because full-length transcripts for hamster DHFR were present in non-transfected CHO DUKX-B11 cells, I can conclude that the DHFR-deficient phenotype of the cells is due to a defect in the translation of the DHFR transcript. How the CHO DUKX-B11 cells were able to re-express a full-length DHFR protein is not known, but may involve mutations in cis-acting elements of the mRNA. These mutations could be in sequences required for the proper initiation of translation such as the ribosome binding site in the 5' UTR, or elongation initiation factors in the 3'UTR. Fusing DHFR fragments Fl/2 or F3 to the C-termini of TAP genes does not seem to abrogate the function of TAP 1 and TAP2. The MHC class I expression assay did not show a complete restoration of a wild-type phenotype when TAP-DHFR fusions were introduced; this is probably due to having less than 100 % transfection efficiency and the possibility that TAP-DHFR is less effective than TAP in transporting peptides. Nonetheless, the introduction of TAPI-L20-FI/2 and TAP2-L20-F3 did produce a substantial restoration of MHC class I molecules on the surface of TAP 1- and TAP2-deficient cells, suggesting that TAP1- and TAP2-DHFR fusions are both able to transport peptides. These observations are consistent with data indicating that TAP1 with green fluorescent protein at its carboxy terminus is functional (4). Thus the DHFR PCA assay has the potential for more detailed functional investigations such as determining if proteins affecting TAP function work by disrupting TAP dimerization. It is also likely that this technology will be fully applicable to the study of other half ABC transporters. Bibliography 1. Abele, R., and R. Tampe. 2004. The ABCs of immunology: structure and function of TAP, the transporter associated with antigen processing. Physiology 19:216-24. 2. Heintke, S., M. Chen, U. Ritz, B. Lankat-Buttgereit, J. Koch, R. Abele, B. Seliger, and R. Tampe. 2003. Functional cysteine-less subunits of the transporter associated with antigen processing (TAP1 and TAP2) by de novo gene assembly. FEBS Letters. 533:42-6. 3. Pelletier, J. N., F. X. CampbeU-Valois, and S. W. Michnick. 1998. Oligomerization domain-directed reassembly of active dihydrofolate reductase from rationally designed fragments. Proceedings of the National Academy of Sciences of the United States of America 95:12141-6. 4. Reits, E. A., J. C. Vos, M. Gromme, and J. Neefjes. 2000. The major substrates for TAP in vivo are derived from newly synthesized proteins.[comment]. Nature. 404:774-8. 5. Remy, I., and S. W. Michnick. 2001. Visualization of biochemical networks in living cells. Proceedings of the National Academy of Sciences of the United States of America 98:7678-83. 6. Urlaub, G., and L. A. Chasin. 1980. Isolation of Chinese hamster cell mutants deficient in dihydrofolate reductase activity. Proceedings of the National Academy of Sciences of the United States of America 77:4216-20. IV Peptide transport by TAPL 4.1 - Introduction With the discovery that TAPL forms homodimers (Chapter 2), my next objective was to determine the function of TAPL. More specifically, I wanted to test my prediction that the TAPL homodimer transports peptides like the TAP1-TAP2 heterodimer does. To do this I applied an in vitro transport assay with microsomes from TAPL-expressing insect cells. 4.1.1 — Peptide transport by TAP1/2 A requirement for a peptide transporter into the ER lumen was first elucidated in studies of cell lines with severely reduced MHC class I levels on the surface of the cells (20, 30). The cell lines that had normal expression levels of MHC class I and 132-microglubulin molecules could present exogenous peptides and express peptides targeted to the ER by signal sequences, but they could not present intracellular antigens. However, by transfecting tap! and/or tapl into these cells, the defective phenotypes could be restored (26, 29). Subsequent experiments with semi-permeabilized cells and isolated microsomes demonstrated that this peptide transport was ATP- and TAP-dependent (4, 23,28). Further studies with TAP expressed in insect cells and yeast found that TAP can function without other factors of the adaptive immune system (21, 32). The peptide binding site of the TAP1-TAP2 transporter was shown to consist of regions from both proteins when photo-reactive peptide analogues labeled both TAP1 and TAP2 (5, 6, 25). To determine more specifically what the peptide binding regions are, crosslinked subunits were digested and immunoprecipitated with various antibodies generated to different epitopes of TAP 1 and TAP2 (24,25). Analysis of the cross-linked TAP subunits indicated that TAP1 and TAP2 have similar peptide binding regions. The peptide binding region consists of the cytoplasmic loops between putative transmembrane helix (TM) 4 and TM5 of the core domain and the 15 amino acids following TM6 (Figure 4.1.1). These regions identified correspond to the polymorphic amino acids 374 and 380 of TAP2, which have been shown to affect substrate specificity in rat TAP2 (7,22). TMD TMD < <- N Core Core N domain domain domain domain < >|< > < >< > ER lumen Cytosol N NBD1 Figure 4.1.1: Schematic model of TAP. Based on sequence alignments with members of the ABC-B subfamily, hydrophobicity analyses, and truncation studies (18), the membrane topology of TAP is predicted. Both subunits contain a core transmembrane domain (TMD) with 6 helices and an NF^-terminal extension of 4 and 3 helices for TAP1 and TAP2, respectively. The NDBs for TAP1 and TAP2 are indicated. The peptide binding region of TAP consists of the last cytosolic loop and the amino acid extension of the last transmembrane helix of TAP1 and TAP2 (24). Walker A and B (A, B) sequences and the C loop sequence (C) of the NBDs are involved in binding and/or hydrolyzing ATP. Site I and site II are defined as the ATP binding pocket composed of residues of the Walker A and B motifs of NBD 1 and the C loop of NBD2 and vice versa. The L loop is speculated to interact (double arrow) with the Q loop. Figure reproduced from Abele et al. (1). The length selectivity of TAP has been determined via randomized peptide libraries in conjunction with competition binding assays (33) and from transport assays with peptides with a N-glycosylation motif at one terminus and a radio-iodinated tyrosine at the other end (19). From these studies, 8-16 amino acid peptides are good substrates for TAP, with 9-12 amino acids being the optimal length. However, peptides up to 40 amino acids in length were also transported, albeit less efficiently. To determine the recognition principle of TAP, a combinatorial approach was used which measured the average affinity of a random mixture of peptides with one residue in common (31). By studying a complex peptide library, the authors were able to determine the effect each peptide residue has on the affinity of TAP. As indicated in Figure 4.1.2, this work found that the N-terminal three positions and the C-terminal residue are critical for efficient binding, whereas the peptides between these residues had negligible effects. An interesting observation about the TAP recognition principle is that T cell receptors make contact mainly between residues 5-8 of MHC class I associated peptides (9,10); therefore TAP shows the lowest specificity (or highest variability) in the area where T cell recognition takes place. > CD CD CO K •1 N o R R •i position 1 2 6 8 9 T3 D P D 2 • E N 1 S CO B promiscuity in sequence and length contact sites (TAP1/2) contact sites T-ceil N-terminal recognition HLA frimming restricted Figure 4.1.2: Substrate recognition motif and substrate binding pocket of human TAP. TAP selectivity is illustrated for positions of the peptide. Favored (white) and disfavored residues (black) of TAP are given at the individual positions as extracted using combinatorial peptide libraries (31). A model of the substrate binding pocket is shown in the lower panel. Figure reproduced from Abele et al. (2). The Km of the transport of the peptide TYNRTRALI by TAP was calculated at 661.3 ± 86.8 nM via a microsomal transport assay (35). As a Lineweaver-Burk transformation of the transport data yielded a correlation coefficient of 0.994, the authors suggest that TAP follows Michaelis-Menten kinetics. Another analysis of TAP kinetics was achieved by studying the rate of peptide-stimulated ATP hydrolysis by TAP (11). By comparing the amount of inorganic phosphate released in the presence or absence of the peptide RRYQKSTEL, the TAP-specific ATPase activity had a Km of 0.3 ± 0.06 mM. The authors also calculated the TAP ATP hydrolysis rate at 5 molecules of ATP per TAP complex per second. The authors concluded that the turnover rate of TAP was sufficient to account for a role for TAP in peptide loading of MHC class I molecules, and that rate-limiting step of antigen processing was not the intracellular transport by TAP. This latter approach of determining the Km for TAP is particularly useful because they used purified and reconstituted TAP and were able to show that ATP hydrolysis is peptide-stimulated. 4.1.2 — Why TAPL may be a peptide transporter The first indication that TAPL may be a peptide transporter is its amino acid sequence identity with TAP1 and TAP2. Overall, TAPL shares 38 and 40 % amino acid sequence identity with TAP1 and TAP2, respectively. When one looks at just the regions of TAP 1 and TAP2 involved in peptide binding, the shared amino acid identity with TAPL to these regions is very high as well (Figure 4.1.3). While the peptide-binding regions of TAP 1 and TAP2 share 27 % amino acid identity, 41 % of TAPL matches the binding sites of either TAP1 or TAP2. Thus, the predicted peptide-binding domains of TAPL appear to be a hybrid of the TAP1 and TAP2 peptide binding domains. This argues for TAPL being a peptide transporter and perhaps transporting a similar range of substrates as the TAP1-TAP2 transporter. 2 0 JQ LUE VCJHR E S|SOE TAPlgLLFLLPKKV ^^ffiB n EVHLR MMVYK m TAPLQIIMMVSNIE FRHQ BVL RE REA TAP2HFTIAAEKV MffiEMwcRHKBAffEWCROlW TAP1 10 TAPL QMQFTQAVgVLtglE TAP2 EFVLHDCM@SVGE B8VH3YVQTLVYI Figure 4.1.3: Amino acid similarity between peptide binding regions of TAP1, TAP2, and TAPL. Peptide regions for TAP1 and TAP2 were determined by Nijenhuis et al. (24, 25). Alignments were created using sequences from the National Center for Biotechnology Information NCBI (accession codes: TAP1, NP_000584; TAP2, NP_000535). Black boxes indicate matches between TAPL (middle) and TAP1 (top) or TAP2 (bottom). The second line of evidence that TAPL may be a peptide transporter lies in a route of antigen presentation known as cross-priming or cross-presentation. Recently, it was shown that cross-presentation can occur in the ER-phagosome (3,12, 14). In macrophage cells, Houde et al. (14) have found that phagosomes in macrophages have all the components necessary for cross-presentation. Figure 4.1.4 shows their working model for cross presentation in macrophages. This process involves hydrolases (H) acquired sequentially during maturation and processing of exogenous peptides in the phagosome lumen. These pathogen-derived peptides are then retrotranslocated from phagosomes into the cytoplasm where they can access the ubiquitin/proteasome complex on the cytoplasmic side of phagosomes. The proteasome then processes the proteins into peptides suitable for MHC class I molecules, and these peptides are transported back by TAP into the phagosome lumen where they bind to MHC class I molecules. The membrane recycling machinery then delivers these MHC class I molecules to the cell surface. Figure 4.1.4: A working model for phagosomal cross-presentation. Shown is a ER- phagosome of an antigen presenting cell. Description of Figure is in the text. Reproduced from Houde et al. (14) Ackerman et al. (3) have shown that the same process of cross-presentation is likely to occur in dendritic cells as well. Their work also indicated all the necessary components for cross-presentation to be present in ER-phagosomes. Furthermore, they demonstrated that peptide transport occurred in phagosomes isolated from these cells. As the phagosomes contained TAP1, and the peptide transport could be inhibited by the TAP inhibitor ICP47, the authors concluded that TAP was responsible for the transport of peptide into the phagosome lumen during cross-presentation. More recently, Johnson et al. (15,16) have demonstrated that dendritic cells express TAPL after they have been matured with TNFa, and that the cross-presentation of antigens by dendritic cells is impaired in TAPL -/- mice. These data, as well as the localization of TAPL to endosomes and lysosomes (36) suggests a possible role for TAPL in transporting peptides into phagosomes, perhaps in concert with TAP. It may also be that ICP47 is able to inhibit the transport of both TAPL and other TAPs. This would necessitate a re-evaluation of conclusions of Ackerman et al. ((3) described above) who believe that TAP is responsible for the cross-presentation in dendritic cells. 4.2 - Materials and Methods 4.2.1 — Creation of a TAPL expression vector for insect cells To create High Five insect cells (from ovarian cells of the cabbage looper, Trichoplusia ni. Invitrogen) which stably express TAPL, an expression plasmid named pIB-TAPL-His, was constructed. First, using TAPL-Li0-Fl/2 as a template, TAPL cDNA was amplified using the PCR primers TL5KozBam (5'-GCTAGGATCCGCCACCATGCGGCTGTGGAA GGCG-3') and TL3NOT (5'-GCTAGCGGCCGCCGGCCTTGTGACTGCCGTTGGC-3') at an annealing temperature of 45 °C for 5 cycles followed by 25 cycles at 58 °C. TL5KozBam contains a BamHl RE site followed by a Kozak consensus sequence and a sequence complementary to the 5' end of TAPL. TL3NOT is a reverse primer containing a Notl RE site followed by a sequence complementary to the 3' end of TAPL. The resulting PCR product and pIB/V5-His Vector (blasticidin resistance gene, OplEl promoter, Invitrogen) were digested with BamHl and Notl, the pIB/V5-His vector was dephosphorylated, the PCR product and vector were gel-purified, and the two were ligated together. The resulting plasmid had the TAPL gene with a His-tag at its C-terminus so that the TAPL protein could be purified later, if required. DNA sequence analysis confirmed that no mutations were introduced in the cloning process. 4.2.2 — Maintenance, transfection, and selection of insect cells High Five Cells were cultured, transfected, and selected for stable expression of TAPL as detailed in the InsectSelect BSD system with pIB/V5-His manual (Version F, Invitrogen). Briefly, 60 mm plates were seeded with 1 x 106 cells (50 - 60 % confluence) in Grace's Insect Cell Culture Medium with L-glutamine (Invitrogen), rocked slowly for 2-3 minutes to distribute the cells, and then left to settle and attach to the bottom of the plate for 1 hour. Media was then removed and 1 mL of Grace's medium was added containing 20 jaL of Cellfectin (Invitrogen) and 2.3 ug of pIB-TAPL-His DNA. Cells were then rocked slowly (2 side-to-side motions / minute) for 4 hours. Each plate then received 2 mL of medium and cells were grown for 48 hours. Cells were then split 1:5 to give 20 % confluence and left overnight. Medium was then replaced with medium containing 50 fig / mL blasticidin (Invitrogen) to select for stable integration of the pIB-TAPL-His-16 plasmid. Medium was replaced with fresh media containing blasticidin every 3 days for 7 days. Colonies were then isolated and sub-cultured in medium containing 25 jag / mL blasticidin. 4.2.3 — Western blots Crude membranes were prepared by suspending approximately 8 x 105 cells in 500 uL PBS containing 10 uL of Protease Inhibitor Cocktail (Pharmingen) and sonicating for 30 seconds on ice. Nuclei and whole cells were removed by centrifuging at 500 x g for 5 minutes at 4 °C. The supernatant was transferred to a new tube and membranes were pelleted by centrifuging at 48,000 x g for 3 hours at 4 °C. Pellets were solubilized with 100 uL of 2% SDS in PBS without calcium and magnesium (Invitrogen). Protein concentration was determined via the Bicinchoninic Acid Assay (Pierce Biotechnology, Rockford, IL, U.S.A.). Each sample (8 jig total protein) was loaded onto a No vex 4-12 % Bis-Tris gel (Invitrogen) and run according to the manufacturer's instructions. Anti-His antibody was from Invitrogen. Anti-TAPL polyclonal antibody IB3B was produced in our laboratory. All blots were visualized with either anti-mouse or anti-rabbit HRP- conjugated antibodies (Jackson ImmunoResearch Laboratories, Inc.) and ECL+ (Amersham Biosciences). 4.2.4 — Microsome preparation Microsomes were prepared from both untranfected High Five Insect cells and High Five cells which have stable expression of TAPL. First, approximately 6.3 x 107 cells were re- suspended in 800 uL cavitation buffer (50 mM Tris at pH 7.4, 250 mM sucrose, 25 mM potassium acetate, 5 mM magnesium acetate, and 0.5 mM calcium acetate) containing complete protease inhibitor (Roche, Laval, Qc, Canada) and lysed by drawing 12 times through a 25-gauge needle. Then the lysate was centrifuged at 500 x g for 5 minutes to pellet any nuclei and unlysed cells. The supernatant was transferred to a new tube and diluted to 5 mL with 2.5 M sucrose solution (all sucrose solutions in cavitation buffer). The mixture was then vortexed briefly, transferred to a new tube, and overlaid with 2 mL of 2.0 M sucrose solution followed by 3.6 mL of 1.3 M sucrose solution and finally 1 mL of cavitation buffer. After the gradient was centrifuged for 17 hours at 4 °C at 200,000 x g in a Beckman SW 41 Ti rotor (Beckman Coulter), microsome fractions were collected at the interfaces below and above the 1.3 M sucrose layer, pooled, and washed in 10 mL PBS and centrifuged for 1.5 hours in a Type 70 Ti rotor at 73,014 x g. Washed microsomes were then re-suspended in 1 mL of transport assay buffer (PBS at pH 7.4 with 1 mM dithiothreitol, 2 mM magnesium chloride) alone, or with transport assay buffer containing 5 mM ATP or 5 mM AMP. 4.2.5 — Peptide transport assay For the translocation assays, peptides of sequences SIINFEKL, FAPGNYPAL, RRYQNSTEL, RYWANATRST, with the specific activities of 1,257 Ci / mmol, 871 Ci / mmol, 931 Ci / mmol, and 857 Ci / mmol, respectively, were synthesized, labeled with I125, and purified by high performance liquid chromatography by Peninsula Laboratories Inc. (San Carlos, CA, U.S.A.). As it does not contain a tyrosine group, SIINFEKL, was labeled by using the Bolton Hunter reagent; the other three peptides were labeled by using the Chloramine-T method. Microsomes were thawed on ice and passed through a 25-gage needle 5 times before using. In 150 uL of total reaction volume, an appropriate concentration of radiolabeled peptide was added to transport assay buffer and the sample was pre-warmed to 37 °C for 2 to 5 minutes. At time zero, 80 |ig of total microsomes was added to the reaction buffer. To quench the reaction, 350 uL of ice-cold PBS was added to the reaction tube, which was then immediately placed on ice. Microsomes were then pelleted by centrifugation at 16,000 x g, washed once with 500 uL cold PBS and centrifuged again at the same speed. The pellet was then resuspended in 1 mL of lysis buffer (25 mM Tris pH 7.4, 150 mM NaCl, 0.5 % Nonidet P-40 (Roche)) and the radioactivity was determined using a Cobra II Auto-gamma y-counter (Perkin-Elmer, Woodbridge, Ontario, Canada). Samples for each trial were all in triplicate. Graphs and histograms of transport data were created with Prism software package (GraphPad Software, Inc., San Diego, CA, U.S.A.). To determine if the direction of peptide transport was away from the ATP cassette, as is the case with TAP1/2 (reviewed in (1)), or if it is an atypical half ABC transporter that transport peptides towards its ATP-binding cassette, microsomes were prepared that contain a mixture of right-side-out and inside-out vesicles (Figure 4.2.1). As the chosen method of detecting transport relies on the accumulation of radioactive peptides inside microsomes, only transport into microsomes can be detected. To detect transport away from the TAPL ABC, microsomes prepared without ATP were placed in buffer containing ATP (Fig. 4.2.1, left), so that only TAPL molecules oriented with the ABC on the outside of vesicles will have access to ATP. To detect transport towards the TAPL ABC, microsomes were prepared in the presence of ATP and placed in buffer containing AMP, so that only TAPL molecules oriented with the ABC on the inside of vesicles will have access to ATP. Legend: m Transmembrane domain of TAPL *P Radiolabeled peptide m ATP-binding cassette (ABC) of TAPL \ S Microsome Direction of Transport TAPL is orientated with ABC TAPL is orientated with ABC ^ outside of the microsomes ^ inside the microsome.s V /microsomal ATP \ *P *P • D (not detectable) ^ B (not detectable) ^ *P Figure 4.2.1: Determining direction of peptide transport. Microsomes prepared will be a mixture of right-side-out (left) and inside-out (right) vesicles. In each peptide transport assay, transport is indicated by an accumulation of radioactivity within microsomes. Thus, transport into vesicles can be monitored (A and C) while transport out of vesicles cannot (B and D). If TAPL is oriented with the ABC on the exterior of the microsomes and ATP is present in the reaction buffer (left), increased radioactivity in microsomes will indicate transport away for the ABC. If TAPL is oriented with the ABC on the interior of the microsomes (inside-out vesicles) which contain ATP and AMP is in the reaction buffer (right), increased radioactivity in microsomes will indicate transport towards the ABC. 4.3 - Results 4.3.1 — Construction of a TAPL-expressing cell line and preparation of microsomes High Five cells with stable expression of His-tagged TAPL were achieved by the transfection of pIB-TAPL-His plasmid and selection with blasticidin. After drug selection, a clone which had high expression of TAPL was selected for peptide transport studies. L15 L16 L19 L25 L26 L32 L34 L37 L39 L40 L41 anti-His Tag anti-TAPL Figure 4.3.1: Selection of TAPL-His expressing High Five cells. Shown are two western blots of High Five cells transfected with and selected for expression of TAPL-His. Lane labels indicate 11 different colonies assayed. After probing with anti-His tag monoclonal antibodies (top), the western blot was stripped and re-probed with anti-TAPL polyclonal antibodies. Clone L25 was chosen for use in peptide transport assays. Microsomes were then prepared from the selected insect cell lines. Additional western blots indicated that TAPL was equally abundant in the membrane layers above and below the 1.3 M sucrose layer (data not shown), so these layers were collected and pooled. 4.3.2 — Peptide transport assay To determine if TAPL could function as a peptide transporter, 4 putative substrates were assayed. The first peptide, SIINFEKL, was assayed for peptide transport at 0,1, 5, and 10 minutes. At each time point, triplicate samples were analyzed as follows: TAPL-containing microsomes with vesicular ATP and AMP in the buffer, TAPL-containing microsomes with AMP in the buffer, and TAPL-containing microsomes with ATP in the buffer— corresponding samples for control, non-transfected High Five cells were assayed as well. At all time points, there were not significant differences between all 6 samples, indicating that SIINFEKL is not a substrate of TAPL. The second peptide assayed as a TAPL substrate was RRYQNSTEL. The same conditions were used for this peptide except that the time points were 0, 2, 5, and 15 minutes. As can be seen in Figure 4.3.2, peptide transport was indicated only when TAPL-containing microsomes were supplied with ATP in the buffer. This indicates that RRYQNSTEL is a substrate of TAPL and that TAPL transports this peptide away from the ATP-binding cassette (Figure 4.2.1, A). 1000' TAPL microsomes, AMP in buffer 900' TAPL microsomes, ATP in buffer TAPL microsomes containing ATP 800-I High Five microsomes, AMP in buffer B 700- 3 High Five microsomes, ATP in buffer c E 600- High Five microsomes containing . i_ 0) °- 500- c 3 400- O O 200- 300- 100- 0 -T- T" l r- -1 1 1 1 1 1 i 9 4 6 7 8 10 11 12 13 14 15 16 Minutes Figure 4.3.2: ATP-dependent transport of RRYQNSTEL by TAPL. Microsomes prepared from either untransfected control or TAPL-containing High Five insect cells were assayed for the ability to transport peptides in an ATP-dependent manner. At time zero, microsomes were added to the reaction buffer containing I125 labeled RRYQNSTEL peptide. After the reaction was quenched, microsomes were collected by centrifugation, washed, and counted on a ••counter. Transport is indicated when radioactive peptide accumulates inside microsomes. Bars represent the mean of 3 trials with the standard deviation indicated. Next, two other peptides (RYWANATRST and FAPGNYPAL) were tested as substrates for TAPL. Since it had already been determined that TAPL transports away from its ABC, only transport in this direction was assayed. Figure 4.3.3 indicates that after 15 minutes, TAP-containing microsomes accumulate RYWANATRST peptide in an ATP-dependent manner. The peptide FAPGNYPAL, however, was not transported by TAPL. 1500- CH] High Five microsomes, AMP in buffer 1250- E=3 High Five microsomes, ATP in buffer TAPL microsomes, AMP in buffer 2 1000' EZDTAPL microsomes, ATP in buffer °- 7S0- JS c 3 O O 500' B 111 M I lili:: 250- FAPGNYPAL RYWANATRST Figure 4.3.3: Peptide transport assay with peptides FAPGNYPAL and RYWANATRST. Microsomes prepared from either untransfected control or TAPL- containing High Five insect cells were assayed for the ability to transport peptides in an ATP-dependent manner. At time zero, microsomes were added to the reaction buffer containing I125 labeled FAPGNYPAL or RYWANATRST peptide. After 15 minutes, the reaction was quenched and microsomes were collected by centrifugation, washed, and counted on a ••counter. Transport is indicated when radioactive peptide accumulates inside microsomes. Bars represent the mean of 3 trials with the standard deviation indicated. To determine an apparent Km for the transport of peptide RYWANATRST by TAPL, time course experiments were performed to determine the initial velocity at various concentrations of peptides. A representative time course experiment is shown in Figure 4.3.4. From these results, it is apparent that accumulation of peptide occurs in a time-dependent manner and that the curve is approaching an asymptote; this indicates that TAPL has a saturable binding site. Initial velocities were also calculated from time-course experiments with 0.20, 0.80,100, and 500 nM (Figure 4.3.5). A double-reciprocal, or Lineweaver-Burk, transformation of the data yielded an r2 value of 0.96, suggesting that the transport of peptide by TAPL follows Michaelis-Menton kinetics. From this plot the apparent Km of TAPL is 3.2 ±0.4nM. 10000-1 Minutes Figure 4.3.4: Transport of radioactive 0.39 nM RYWANATRST as a function of time. At time zero, microsomes prepared from TAPL-containing High Five insect cells were added to reaction buffer containing I125 labeled RYWANATRST peptide and ATP. At various time points, the reaction was quenched and microsomes were pelleted, washed, and counted on a ••counter. Transport is indicated when radioactive peptide accumulates inside microsomes. Each trial was performed in triplicate, and the standard deviation is indicated. Dotted line indicates the initial velocity as the curve approaches zero. 3.0x101S Initial velocity [Substrate] (moles / minute) 19 2.0x101* 0.20 nM 4.01 x lO" 0.39 nM 6.09 x 1019 0.80 nM 9.43 x 10"19 -17 I.OxlO1^ 100 nM 9.96 x 10 500 nM 4.32 x lO-16 Km = 3.2 ± 0.4 nM 6 Figure 4.3.5: Lineweaver-Burk plot of initial velocity as a function of substrate concentration for the transport of RYWANATRST by TAPL. Initial velocities of transport were calculated as shown in Figure 4.3.4. The Km was then calculated as an absolute value of the inverse of the x-intercept. Note: Points for l/[Substrate] for 100 nM and 500 nM overlap and appear very close to the origin of the plot. 4.4 - Discussion This Chapter presents the first direct evidence that TAPL can transport peptides. As could be predicted from the sequence identity of TAPL with the peptide-binding regions of TAP1 and TAP2, two classical TAP substrates (RRYQNSTEL (4, 13, 17) and RYWANATRST (27, 33)) were transported by TAPL. These substrates are within the optimal size length for TAP and are high-affinity substrates of TAP (Figure 4.1.2). The transport of peptides by TAPL occurred in the direction away from the ABC, as it is with TAP. As the peptide RYWANATRST seemed to be transported slightly more efficiently than the peptide RRYQNSTEL, RYWANATRST was chosen for further kinetic characterization. Time course experiments with RYWANATRST demonstrated a time-dependent accumulation of peptide with a curve that approached an asymptote, indicating TAPL to have a saturable binding site. A double-reciprocal, or Lineweaver-Burk, transformation of the data suggested that transport of peptide by TAPL follows Michaelis-Menton kinetics, and indicated an apparent Km of 3.2 ± 0.4 nM. This Km is approximately 10 to 20 times lower than the TAP Km values of 661.3 ± 86.8 nM for peptide TYNRTRALI (35) and 300 ± 60 nM for RRYQKSTEL peptide-stimulated ATPase activity (11). This suggests that TAPL is able to transport peptides effectively and at a lower cytosolic concentration than TAP. It should be noted, however, that Km values can vary significantly depending on factors such as microsome composition and preparation, buffer composition and temperature, substrate used, and detection method. Surprisingly, the two exogenous peptides assayed, the MHC class I molecule-restricted Sendai virus antigen (324-332) peptide FAPGNYPAL and the MHC class I molecule- restricted Chicken egg ovalbumin peptide (Ova) (257-264) SIINFEKL did not appear to be substrates of TAPL. As Ova peptide is a commonly used substrate for cross-presentation assays (3,14), it would be a likely candidate for TAPL transport into ER-phagosomes. However, I cannot rule out Ova as a substrate because it was radiolabeled with the Bolton- Hunter reagent. The N-terminal modification the Bolton-Hunter reagent creates could interfere with the ability of TAPL to use this peptide as a substrate and may have produced misleading results. Alternative transport assays which do not require labeling with the Bolton-Hunter reagent labeling or which have modified the Ova peptide to include a tyrosine, may find that Ova peptide can be a substrate for TAPL, but this is speculative. However, if TAPL has a different range of substrate specificity, having both TAP and TAPL in ER- phagosome membranes could greatly increase the diversity of antigens which could be cross- presented on MHC class I. Another function of TAPL could be contributing to the immune privilege of the brain and testis (8, 34). The high expression of TAPL in testis and brain (36) could be explained if TAPL is acting as a "peptide sink" transporting peptides from the cytoplasm into lysosomes so that they cannot be transported by TAP and expressed on MHC class I. This would decrease the recognition of tissues in the brain and testis by the immune system by lowering the amount of antigen presented. Thus, I propose a dual-role for TAPL as facilitator of both immune privilege in brain and testis and cross-presentation in macrophages and dendritic cells. Bibliography 1. Abele, R., and R. Tampe. 2004. The ABCs of immunology: structure and function of TAP, the transporter associated with antigen processing. Physiology 19:216-24. 2. Abele, R., and R. Tampe. 1999. Function of the transport complex TAP in cellular immune recognition. Biochimica et Biophysica Acta 1461:405-19. 3. Ackerman, A. L., C. Kyritsis, R. Tampe, and P. Cresswell. 2003. Early phagosomes in dendritic cells form a cellular compartment sufficient for cross presentation of exogenous antigens. Proceedings of the National Academy of Sciences of the United States of America 100:12889-94. 4. Androlewicz, M. J., K. S. Anderson, and P. Cresswell. 1993. Evidence that transporters associated with antigen processing translocate a major histocompatibility complex class I-binding peptide into the endoplasmic reticulum in an ATP-dependent manner. Proceedings of the National Academy of Sciences of the United States of America. 90:9130-4. 5. Androlewicz, M. J., and P. Cresswell. 1994. Human transporters associated with antigen processing possess a promiscuous peptide-binding site. Immunity. 1:7-14. 6. Androlewicz, M. J., B. Ortmann, P. M. van Endert, T. Spies, and P. Cresswell. 1994. Characteristics of peptide and major histocompatibility complex class I/beta 2- microglobulin binding to the transporters associated with antigen processing (TAP1 and TAP2). Proceedings of the National Academy of Sciences of the United States of America 91:12716-20. 7. Deverson, E. V., L. Leong, A. Seelig, W. J. Coadwell, E. M. Tredgett, G. W. Butcher, and J. C. Howard. 1998. Functional analysis by site-directed mutagenesis of the complex polymorphism in rat transporter associated with antigen processing. Journal of Immunology. 160:2767-79. 8. Filippini, A., A. Riccioli, F. Padula, P. Lauretti, A. D'Alessio, P. De Cesaris, L. Gandini, A. Lenzi, and E. Ziparo. 2001. Control and impairment of immune privilege in the testis and in semen. Human Reproduction Update 7:444-9. 9. Garboczi, D. N., P. Ghosh, U. Utz, Q. R. Fan, W. E. Biddison, and D. C. Wiley. 1996. Structure of the complex between human T-cell receptor, viral peptide and HLA-A2. [comment]. Nature 384:134-141. 10. Garcia, K. C, M. Degano, R. L. Stanfield, A. Brunmark, M. R. Jackson, P. A. Peterson, L. Teyton, and I. A. Wilson. 1996. An alphabeta T cell receptor structure at 2.5 A and its orientation in the TCR-MHC complex.fsee comment]. Science 274:209-219. 11. Gorbulev, S., R. Abele, and R. Tampe. 2001. Allosteric crosstalk between peptide- binding, transport, and ATP hydrolysis of the ABC transporter TAP. Proceedings of the National Academy of Sciences of the United States of America. 98:3732-7. 12. Guermonprez, P., L. Saveanu, M. Kleijmeer, J. Davoust, P. Van Endert, and S. Amigorena. 2003. ER-phagosome fusion defines an MHC class I cross-presentation compartment in dendritic cells. Nature 425:397-402. 13. Heintke, S., M. Chen, U. Ritz, B. Lankat-Buttgereit, J. Koch, R. Abele, B. Sehger, and R. Tampe. 2003. Functional cysteine-less subunits of the transporter associated with antigen processing (TAP1 and TAP2) by de novo gene assembly. FEBS Letters. 533:42-6. 14. Houde, M., S. Bertholet, E. Gagnon, S. Brunet, G. Goyette, A. Laplante, M. F. Princiotta, P. Thibault, D. Sacks, and M. Desjardins. 2003. Phagosomes are competent organelles for antigen cross-presentation.[see comment]. Nature 425:402- 6. 15. Johnson, L., F. Zhang, A. Chen, W. Renxue, V. Ling, and W. Jefferies. 2004. An ABC-transporter gene that plays a crucial role in cross-priming by dendritic cells. Submitted for publication. 16. Johnson, L. A. 2004. The ABCs of Antigen Presentation: Defining Antigen Processing Pathways. Ph.D. The University of British Columbia, Vancouver. 17. Karttunen, J. T., P. J. Lehner, S. S. Gupta, E. W. Hewitt, and P. Cresswell. 2001. Distinct functions and cooperative interaction of the subunits of the transporter associated with antigen processing (TAP). Proceedings of the National Academy of Sciences of the United States of America 98:7431-6. 18. Koch, J., R. Guntrum, S. Heintke, C. Kyritsis, and R. Tampe. 2004. Functional dissection of the transmembrane domains of the transporter associated with antigen processing (TAP). Journal of Biological Chemistry 279:10142-7. 19. Koopmann, J. O., M. Post, J. J. Neefjes, G. J. Hammerling, and F. Momburg. 1996. Translocation of long peptides by transporters associated with antigen processing (TAP). European Journal of Immunology 26:1720-8. 20. Ljunggren, H. G., N. J. Stam, C. Ohlen, J. J. Neefjes, P. Hoglund, M. T. Heemels, J. Bastin, T. N. Schumacher, A. Townsend, K. Karre, L. Hidde, and H. L. Ploegh. 1990. Empty MHC class I molecules come out in the cold. Nature 346:476-80. 21. Meyer, T. H., P. M. van Endert, S. UebeL B. Ehring, and R. Tampe. 1994. Functional expression and purification of the ABC transporter complex associated with antigen processing (TAP) in insect cells. FEBS Letters. 351:443-7. 22. Momburg, F., E. A. Armandola, M. Post, and G. J. Hammerling. 1996. Residues in TAP2 peptide transporters controlling substrate specificity. Journal of Immunology. 156:1756-63. 23. Neefjes, J. J., F. Momburg, and G. J. Hammerling. 1993. Selective and ATP- dependent translocation of peptides by the MHC-encoded transporter, [erratum appears in Science 1994 Apr 1;264(5155):16]. Science. 261:769-71. 24. Nijenhuis, M., and G. J. Hammerling. 1996. Multiple regions of the transporter associated with antigen processing (TAP) contribute to its peptide binding site. Journal of Immunology 157:5467-77. 25. Nijenhuis, M., S. Schmitt, E. A. Armandola, R. Obst, J. Brunner, and G. J. Hammerling. 1996. Identification of a contact region for peptide on the TAP1 chain of the transporter associated with antigen processing. Journal of Immunology. 156:2186-95. 26. Powis, S. J., A. R. Townsend, E. V. Deverson, J. Bastin, G. W. Butcher, and J. C. Howard. 1991. Restoration of antigen presentation to the mutant cell line RMA-S by an MHC-linked transporter. Nature. 354:528-31. 27. Saveanu, L., S. Daniel, and 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. Journal of Biological Chemistry. 276:22107- 13. 28. Shepherd, J. C, T. N. Schumacher, P. G. Ashton-Rickardt, S. Imaeda, H. L. Ploegh, C. A. Janeway, Jr., and S. Tonegawa. 1993. TAP 1-dependent peptide translocation in vitro is ATP dependent and peptide selective, [erratum appears in Cell 1993 Nov 19;75(4):613]. Cell. 74:577-84. 29. Spies, T., and R. DeMars. 1991. Restored expression of major histocompatibility class I molecules by gene transfer of a putative peptide transporter, [see comment]. Nature 351:323-4. 30. Townsend, A., C. Ohlen, L. Foster, J. Bastin, H. G. Ljunggren, and K. Karre. 1989. A mutant cell in which association of class I heavy and light chains is induced by viral peptides. Cold Spring Harbor Symposia on Quantitative Biology 54 Pt 1:299-308. 31. Uebel, S., W. Kraas, S. Kienle, K. H. Wiesmuller, G. Jung, and R. Tampe. 1997. Recognition principle of the TAP transporter disclosed by combinatorial peptide libraries. Proceedings of the National Academy of Sciences of the United States of America 94:8976-81. 32. Urlinger, S., K. Kuchler, T. H. Meyer, S. Uebel, and R. Tampe. 1997. Intracellular location, complex formation, and function of the transporter associated with antigen processing in yeast. European Journal of Biochemistry. 245:266-272. 33. van Endert, P. M., R. Tampe, T. H. Meyer, R. Tisch, J. F. Bach, and H. O. McDevitt. 1994. A sequential model for peptide binding and transport by the transporters associated with antigen processing. Immunity. 1:491-500. 34. Wiendl, H., M. Mitsdoerffer, and M. Weller. 2003. Hide-and-seek in the brain: a role for HLA-G mediating immune privilege for glioma cells. Seminars in Cancer Biology 13:343-51. 35. Yang, B., and T. J. Braciale. 1995. Characteristics of ATP-dependent peptide transport in isolated microsomes. Journal of Immunology 155:3889-96. 36. Zhang, F., W. Zhang, L. Liu, C. L. Fisher, D. Hui, S. Childs, K. Dorovini-Zis, and V. Ling. 2000. Characterization of ABCB9, an ATP binding cassette protein associated with lysosomes. Journal of Biological Chemistry. 275:23287-94. V Overall Discussion and Future Prospects 5.1 - Discussion The work described in this thesis was successful in achieving all of its three primary objectives: to determine the applicability of the DHFR PCA to human half ABC transporters, to find novel interactions, and to determine the function of the interactions. In Chapter 2,1 performed the first comprehensive assay to look for all the in vivo interactions which could occur between TAP1, TAP2, and TAPL. By applying the DHFR PCA to this family, I was able to confirm the known TAP1-TAP2 heterodimer and provide evidence that these proteins do not form homodimers or interact with TAPL. This is an important finding as TAP1 and TAP2 had both been speculated to form homodimers (2, 8, 14). The most significant finding in this Chapter was that TAPL forms homodimers, as determining its dimerization state is a prerequisite to studying its function. The finding that TAP1, TAP2, and TAPL all have only one dimerization state raises questions about why half ABC transporters exist as half ABC transporters. At the beginning of this work, I speculated that the reason these transporters exist as interacting halves rather than full transporters was so that they would be able to interact with multiple partners in order to have different functions or activities. Furthermore, other investigations had found indirect evidence that both TAP1 and TAP2 may form homodimers (2, 8,14). Thus my finding that all three of these transporters only have one dimerization state forces us to re• evaluate the raison d'etre of half ABC transporters. While the White proteins in drosophila exist as half transporters so they can form multiple dimerization states to transport different substrates (6,16-18), there must be other advantages to existing as half transporters. I speculate that, at least in the human TAP family, having half transporters could: (1) facilitate the regulation of gene expression; (2) regulate transport by association and dissociation of the halves; or (3) generate more diversity in a population by having two separate genes encoding for one transporter. TAPL not interacting with TAP1 or TAP2 agrees with the finding that it localized to lysosomes (25) where no other mammalian ABC transporters have been found. This, as well as the fact that TAPL is expressed in testis, brain, spinal cord, and thyroid (25), suggests that TAPL may have a unique role required by those tissues. What role TAPL may play was examined in Chapter 4 (discussed below). In Chapter 3,1 investigated in detail the utility of the DHFR PC A as a tool for studying half ABC transporters. An unfortunate aspect of this assay was the occurrence of false positive colonies in the survival assay. While these colonies occurred at a low frequency (0 .0002%), they appear as a significant background because the colony counts of my positive samples were also relatively low (Figure 2.3.1). Nevertheless, the rate of false positives was, in all cases, significantly less than that observed for true positives. Fortunately, by studying the protein and mRNA expression of the colonies, I was able to determine that these colonies arose by a reversion of the cell line to a wild type phenotype. Thus, false positives in this assay are not a serious concern as they can be identified and distinguished from "true" positive results. A great benefit of this assay, however, is the fact that I found the TAP 1- and TAP2-DHFR chimeric proteins were able to function at detectable levels. This gives the DHFR-PCA assay potential for more detailed functional investigations such as determining if proteins affecting TAP function work by disrupting TAP dimerization. It is also likely that this technology will be fully applicable to the study of other half ABC transporters. In Chapter 4,1 found evidence that TAPL can transport peptides. This is the first time that anyone has been able to attribute a function to this transporter. I found that two substrates of TAP, the peptides RYWANATRST and RRYQNSTEL, could be transported by TAPL. Kinetic characterization of the transport of RYWANATRST found that TAPL appears to follow Michaelis-Menton kinetics, and has an apparent Km of 3.2 ± 0.4 nM. This raises the exciting possibility that TAPL could have an important immunological role. As TAPL resides in endosomes / lysosomes and is found primarily in the brain and testis, I propose two possible biological roles for TAPL (Figure 5.1.1). One possibility is that TAPL might contribute to the cross-presentation of antigens by transporting peptides into the lumen of ER-phagosomes of dendritic cells and macrophages. In this case, we could revise the model of cross-presentation for macrophages proposed by by Houde et al. (Figure 4.1.4, (10)) which also appears to occur in dendritic cells. (1, 9). In this model, fragments of pathogenic peptides are retrotranslocated from phagosomes into the cytoplasm by Sec61, processed by ER-phagosome-associated proteasomes, and then transported back into the lumen of the ER-phagosome by TAP to be bound by MHC class I molecules. I propose that TAPL could also responsible for transporting the peptides into the ER-phagosome and contributing to cross-presentation. It is also possible that TAPL has a different specificity than TAP, which would allow it to expand the diversity of transported peptides in this route of antigen presentation. We could also speculate that, due to the elevated expression of TAPL in the brain and testis, that TAPL could facilitate the immune privilege of these organs. Immune privilege is where these tissues are less immunological active—a process which, among other factors, can involve reduced MHC class I expression (7, 22). In these tissues, TAPL could act as a "peptide sink" transporting peptides from the cytoplasm into lysosomes so that they cannot be transported by TAP and expressed on MHC class I. The result would be a decrease in the recognition of the brain and testis by the immune system. This dual role for TAPL integrates the subcellular localization of TAPL, tissue distribution, and function as a peptide transporter. Figure 5.1.1: Proposed functions of TAPL. A: TAPL is contributing to cross-presentation by transporting peptides into ER-phagosomes. B: TAPL contributes to immune privilege by taking peptides away from the TAP transporter, which will decrease the amount of peptide bound by MHC class I molecules. For more details, see text above. 5.2 - Future prospects 5.2.1 — Peptide specificity of TAPL Following my successful demonstration that isolated microsomal membranes of insect cells transfected with the TAPL are able to transport peptides in an ATP-dependent manner, the next step would be an investigation of the range of peptides TAPL can transport. Initially one should determine if other peptides recognized by TAP 1/2 dimers compete with the radiolabeled TAPL substrate. Using a wide array of competitor peptides (19), the lengths and sequences of peptides transported most efficiently by TAPL can be determined. If TAPL transports the same range of substrates as TAP 1/2, it would be consistent with an immune function for TAPL, as TAP 1/2 substrates are presented to the immune system by MHC class I. On the other hand, if TAPL has a different range of substrates, it could be important in regulating intracellular levels of other peptides, such as peptide hormones. It is possible that TAPL could serve both functions. Which function predominates may depend on the tissue in which it is expressed. A comprehensive investigation of the TAPL substrate specificity would, at a minimum, provide a more in-depth understanding of the role of TAPL in immune function. It may also reveal another role for TAPL, such as the transport of peptide hormones. Peptide hormones stimulate the proliferation of human pancreatic cancer cell lines of ductal origin (5) and contribute to the pathogenesis of human breast cancer (15). Natural peptide hormones have also been shown to decrease the proliferation of human pancreatic adenocarcinoma cells (20). These data raise could raise exciting possibilities for the involvement of TAPL in endocrine regulation, and possibly cancer proliferation. Studying which residues in TAPL are necessary for this peptide and substrate specificity would also improve our understanding of the TAP family and transporters in general. As shown in Figure 4.1.3, 41 % of TAPL matches the binding sites of either TAP1 or TAP2. Thus, determining which of these residues are important for TAPL's specificity could help us understand specificity, or lack thereof, of the TAP1-TAP2 heterodimer. Approaches to study the key residues in the binding site could involve site directed mutagenesis and cross- linking transported peptides to TAPL and identifying which regions are crucial for binding. 5.2.2 — Localization of TAPL homodimers. The first step in doing comprehensive localization studies of TAPL would be to generate an array of monoclonal antibodies as the currently available polyclonal antibody for TAPL is not sensitive enough to readily detect TAPL expression in various tissues (25). Concordance of staining using different TAPL monoclonal antibodies would confirm the specific expression of TAPL in, for example, antigen presenting cells such as macrophages and dendritic cells before and after they have been stimulated (i.e. with INF-y). This will enable a better understanding of what roles TAPL plays in these tissues and cell types. Moreover, highly specific monoclonal antibodies would allow for critical sub-cellular localization studies using electron microscopy or confocal microscopy. By using the TAPL-DHFR dimers from the DHFR PCA, it should be possible to also generate monoclonal antibodies which recognize TAPL homodimers. This line of research would allow the evaluation of the amount and location of the biologically active dimers of TAPL. It would also allow us to look more deeply at where and when dimerization of these transporters occurs and what other proteins or conditions are necessary for this dimerization. An approach to determine the normal tissue expression of TAPL could be to "knock-in" a marker, such as enhanced green fluorescent protein (EGFP), at the 3' end of the coding region. In this way, the distribution of TAPL under the physiological promoter could be investigated. 5.2.3 — Function in immune tolerance. To determine if TAPL expression in brain and testis contributes to the immunological privilege of these sites, there are avenues of experimentation that could be followed. The first step would be to determine if TAPL -/- mice have increased MHC class I expression or if they can generate a stronger immune response than wild-type tissues from these organs. Another approach would be to take cells which normally express MHC class I and can elicit an immune response and see if introducing TAPL would alter their immunological phenotype. These types of studies could help to solve the complex phenomenon of immune privilege. 5.2.4 — Detection of interactions within the White family of half ABC transporters. As I have established that the DHFR PCA is a valuable tool for detecting in vivo interactions between ABC half transporters, one could now apply this method to other half ABCs, such as the human ABCG, or White family. This is of particular interest as one member of the ABCG family, ABCG2, has been implicated in resistance to a large number of chemotherapy drugs (4,12, 23). Other members of the ABCG family seem to have roles in sterol regulation and transport (3,4,11,13,21, 24) but it is possible that ABCG1, ABCG4, ABCG5, and ABCG8 could interact with ABCG2. Thus it is important to do a comprehensive assay of all the interactions which could occur within this family in vivo. These data will give new insights into the dynamics of the ABCG family and could elucidate novel drug transporters. Bibliography 1. Ackerman, A. L., C. Kyritsis, R. Tampe, and P. Cresswell. 2003. Early phagosomes in dendritic cells form a cellular compartment sufficient for cross presentation of exogenous antigens. Proceedings of the National Academy of Sciences of the United States of America 100:12889-94. 2. Antoniou, A. N., S. Ford, E. S. Pilley, N. Blake, and S. J. Powis. 2002. Interactions formed by individually expressed TAP1 and TAP2 polypeptide subunits. Immunology. 106:182-9. 3. Berge, K. E., H. Tian, G. A. Graf, L. Yu, N. V. Grishin, J. Schultz, P. Kwiterovich, B. Shan, R. Barnes, and H. H. Hobbs. 2000. Accumulation of dietary cholesterol in sitosterolemia caused by mutations in adjacent ABC transporters.[comment]. Science. 290:1771-5. 4. Cserepes, J., Z. Szentpetery, L. Seres, C. Ozvegy-Laczka, T. Langmann, G. Schmitz, H. Glavinas, I. Klein, L. Homolya, A. Varadi, B. Sarkadi, and N. B. Elkind. 2004. Functional expression and characterization of the human ABCG1 and ABCG4 proteins: indications for heterodimerization. Biochemical & Biophysical Research Communications 320:860-7. 5. Douziech, N., A. Lajas, Z. Coulombe, E. Calvo, J. Laine, and J. Morisset. 1998. Growth effects of regulatory peptides and intracellular signaling routes in human pancreatic cancer cell lines. Endocrine 9:171-83. 6. Dreesen, T. D., D. H. Johnson, and S. Henikoff. 1988. The brown protein of Drosophila melanogaster is similar to the white protein and to components of active transport complexes. Molecular & Cellular Biology. 8:5206-15. 7. Filippini, A., A. Riccioli, F. Padula, P. Lauretti, A. D'AIessio, P. De Cesaris, L. Gandini, A. Lenzi, and E. Ziparo. 2001. Control and impairment of immune privilege in the testis and in semen. Human Reproduction Update 7:444-9. 8. Gabathuler, R., G. Reid, G. Kolaitis, J. Driscoll, and W. A. Jefferies. 1994. Comparison of cell lines deficient in antigen presentation reveals a functional role for TAP-1 alone in antigen processing. Journal of Experimental Medicine. 180:1415-25. 9. Guermonprez, P., L. Saveanu, M. Kleijmeer, J. Davoust, P. Van Endert, and S. Amigorena. 2003. ER-phagosome fusion defines an MHC class I cross-presentation compartment in dendritic cells. Nature 425:397-402. 10. Houde, M., S. Bertholet, E. Gagnon, S. Brunet, G. Goyette, A. Laplante, M. F. Princiotta, P. Thibault, D. Sacks, and M. Desjardins. 2003. Phagosomes are competent organelles for antigen cross-presentation.[see comment]. Nature 425:402- 6. 11. Ito, T. 2003. Physiological function of ABCG1. Drug News & Perspectives 16:490-2. 12. Johnson, L., F. Zhang, A. Chen, W. Renxue, V. Ling, and W. Jefferies. 2004. An ABC-transporter gene that plays a crucial role in cross-priming by dendritic cells. Submitted for publication. 13. Lee, M. H., K. Lu, S. Hazard, H. Yu, S. Shulenin, H. Hidaka, H. Kojima, R. Allikmets, N. Sakuma, R. Pegoraro, A. K. Srivastava, G. Salen, M. Dean, and S. B. Patel. 2001. Identification of a gene, ABCG5, important in the regulation of dietary cholesterol absorption. Nature Genetics. 27:79-83. 14. Marusina, K., G. Reid, R. Gabathuler, W. Jefferies, and J. J. Monaco. 1997. Novel peptide-binding proteins and peptide transport in normal and TAP-deficient microsomes. Biochemistry. 36:856-63. 15. Normanno, N., F. Ciardiello, R. Brandt, and D. S. Salomon. 1994. Epidermal growth factor-related peptides in the pathogenesis of human breast cancer. Breast Cancer Research & Treatment 29:11-27. 16. O'Hare, K., C. Murphy, R. Levis, and G. M. Rubin. 1984. DNA sequence of the white locus of Drosophila melanogaster. Journal of Molecular Biology. 180:437-55. 17. Pepling, M., and S. M. Mount. 1990. Sequence of a cDNA from the Drosophila melanogaster white gene. Nucleic Acids Research. 18:1633. 18. Tearle, R. G., J. M. Belote, M. McKeown, B. S. Baker, and A. J. HoweUs. 1989. Cloning and characterization of the scarlet gene of Drosophila melanogaster. Genetics. 122:595-606. 19. Uebel, S., W. Kraas, S. Kienle, K. H. Wiesmuller, G. Jung, and R. Tampe. 1997. Recognition principle of the TAP transporter disclosed by combinatorial peptide libraries. Proceedings of the National Academy of Sciences of the United States of America 94:8976-81. 20. Vesely, B. A., Q. McAfee, W. R. Gower, Jr., and D. L. Vesely. 2003. Four peptides decrease the number of human pancreatic adenocarcinoma cells. European Journal of Clinical Investigation 33:998-1005. 21. Wang, N., D. Lan, W. Chen, F. Matsuura, and A. R. Tall. 2004. ATP-binding cassette transporters Gl and G4 mediate cellular cholesterol efflux to high-density lipoproteins. Proceedings of the National Academy of Sciences of the United States of America 101:9774-9. 22. Wiendl, H., M. Mitsdoerffer, and M. Weller. 2003. Hide-and-seek in the brain: a role for HLA-G mediating immune privilege for glioma cells. Seminars in Cancer Biology 13:343-51. 23. Yoshikawa, M., A. Ito, T. Ishikawa, and Y. Ikegami. 2004. [Drug resistance mediated by ABC transporters]. Gan to Kagaku Ryoho [Japanese Journal of Cancer & Chemotherapy] 31:1-6. 24. Yu, L., K. von Bergmann, D. Lutjohann, H. H. Hobbs, and J. C. Cohen. 2004. Selective sterol accumulation in ABCG5/ABCG8-deficient mice. Journal of Lipid Research 45:301-7. 25. Zhang, F., W. Zhang, L. Liu, C. L. Fisher, D. Hui, S. Childs, K. Dorovini-Zis, and V. Ling. 2000. Characterization of ABCB9, an ATP binding cassette protein associated with lysosomes. Journal of Biological Chemistry. 275:23287-94.