Download Modulation of the Antigenic Peptide Transporter TAP by

doi:10.1016/j.jmb.2007.02.102
J. Mol. Biol. (2007) 369, 95–107
Modulation of the Antigenic Peptide Transporter TAP by
Recombinant Antibodies Binding to the Last Five
Residues of TAP1
Gabriele Plewnia 1 , Katrin Schulze 1 , Carola Hunte 2 , Robert Tampé 1
and Joachim Koch 1 ⁎
1
Institute of Biochemistry,
Biocenter, Johann Wolfgang
Goethe-University,
Max-von-Laue-Strasse 9,
D-69438 Frankfurt a. M.,
Germany
2
Max-Planck-Institute
of Biophysics,
Max-von-Laue-Strasse 3,
D-60438 Frankfurt a. M.,
Germany
The transporter associated with antigen processing (TAP) plays a pivotal role
in the major histocompatibility complex (MHC) class I mediated immune
response against infected or malignantly transformed cells. It belongs to the
ATP-binding cassette (ABC) superfamily and consists of TAP1 (ABCB2) and
TAP2 (ABCB3), each of which possesses a transmembrane and a nucleotidebinding domain (NBD). Here we describe the generation of recombinant Fv
and Fab antibody fragments to human TAP from a hybridoma cell line
expressing the TAP1-specific monoclonal antibody mAb148.3. The epitope
of the antibody was mapped to the very last five C-terminal amino acid
residues of TAP1 on solid-supported peptide arrays. The recombinant
antibody fragments were heterologously expressed in Escherichia coli and
purified to homogeneity from periplasmic extracts by affinity chromatography. The monoclonal and recombinant antibodies bind with nanomolar
affinity to the last five C-terminal amino acid residues of TAP1 as
demonstrated by ELISA and surface plasmon resonance. Strikingly, the
recombinant antibody fragments confer thermal stability to the heterodimeric TAP complex. At the same time TAP is arrested in a peptide
transport incompetent conformation, although ATP and peptide binding to
TAP are not affected. Based on our results we suggest that the C terminus of
TAP1 modulates TAP function presumably as part of the dimer interface of
the NBDs.
© 2007 Elsevier Ltd. All rights reserved.
*Corresponding author
Keywords: ABC transporter; adaptive immune system; antigen processing;
epitope mapping; SPOT
Introduction
Major histocompatibility complex (MHC) class I
molecules present endogenous peptides on the
surface of nucleated cells to cytotoxic T-lymphocytes
(CD8+), which scan these peptide-MHC I complexes
to eliminate infected or malignantly transformed
Abbreviations used: ABC, ATP-binding cassette; CDR,
complementarity determining region; ER, endoplasmic
reticulum; ICP, infected cell protein; mAb, monoclonal
antibody; MHC, major histocompatibility complex; NBD,
nucleotide-binding domain; PLC, peptide-loading
complex; TAP, transporter associated with antigen
processing; TBS-T, Tris-buffered saline with Tween 20;
TMD, transmembrane domain; US, unique short.
E-mail address of the corresponding author:
[email protected]
cells.1,2 The majority of antigenic peptides are
derived from proteolytic processing of poly-ubiquitylated proteins by the proteasome complex in the
cytoplasm. These peptides are translocated into the
endoplasmic reticulum (ER) via the transporter
associated with antigen processing (TAP) and are
subsequently loaded onto MHC I molecules.3,4 This
process requires a sophisticated macromolecular
machinery, the peptide-loading complex (PLC).5–9
The TAP-dependent peptide supply is crucial in the
assembly and quality control of peptide-MHC I
complexes, since only peptide-loaded MHC I
molecules can leave the ER for transport to the cell
surface.10–12
DNA viruses have evolved manifold strategies to
escape the host's immune response leading to
persistence and repeated reactivation. The PLC
represents a major target for viral effector proteins.13,14 Two viral inhibitors of human TAP have
0022-2836/$ - see front matter © 2007 Elsevier Ltd. All rights reserved.
96
been characterized in detail. ICP47 of herpes simplex virus 1 (HSV-1) acts as a high-affinity inhibitor
in the cytosol by blocking peptide binding to
TAP.15,16 By contrast, US6 of human cytomegalovirus (HCMV) binds to TAP in the ER and inhibits
peptide translocation by blocking ATP but not
peptide binding.17–21 Both proteins proved useful
to investigate TAP function in vivo and in vitro.
Beside these fascinating strategies of viral evasion,
synthetic or protein engineered inhibitors have not
been described so far. Antibodies, which recognize
discrete sites within TAP are promising tools to shed
light on the functional and structural dynamics of
the transport complex. One of the candidates, the
human TAP1-specific monoclonal antibody (mAb)
148.3, has been used for indirect immunofluorescence, immunoblotting, co-immunoprecipitation
and purification of the heterodimeric TAP-complex.
Moreover, the mAb148.3 allowed for the identification and isolation of the macromolecular peptideloading complex composed of TAP1, TAP2, tapasin,
MHC class I heavy chain, β2-microglobulin, calreticulin, and ERp57.5,7,9
Since impaired TAP expression is directly linked to
down-regulation of MHC class I antigen presentation and immunevasion of many tumors the TAP
heterodimer represents an important diagnostic target in tumorigenesis.14,22,23 However, to date, the
production of TAP-specific recombinant antibodies
such as Fv and Fab fragments has not been reported
although recombinant antibodies are superior to fulllength antibodies in a variety of applications (for
recent review see Schaedel & Reiter24).
Here, we generated for the first time recombinant
antibody fragments (Fv and Fab) to TAP. The
recombinant antibody fragments were produced in
Escherichia coli and purified from periplasmic
extracts by affinity chromatography. The epitope
of the antibodies was mapped to the last five
C-terminal amino acid residues of TAP1 and their
affinity binding constants were determined. Strikingly, the recombinant antibodies inhibit the TAPspecific peptide transport in semi-permeabilized
cells and at the same time confer thermal stability
to the TAP complex. These antibodies are therefore
important tools to investigate the functional role of
the C terminus of TAP1 and provide the possibility
to modulate the activity of the TAP complex.
Furthermore, the variable heavy (VH) and variable
light (VL) chains are now accessible for genetic
engineering, thus allowing for functionalization
and modulation of their binding properties, desirable features for therapy and diagnostic imaging.
Results
The mAb148.3 binds to the very last five
C-terminal amino acid residues of human TAP1
So far, only three monoclonal antibodies to human
TAP have been generated.25,26 The mAb148.3, which
is specific for human TAP1, is an important diag-
Antibodies Modulating TAP Function
nostic tool to investigate the antigen processing
machinery in tumor development and viral escape
mechanisms.21,23 Moreover, the mAb148.3 has led to
the identification and subunit mapping of the
macromolecular peptide-loading complex in the
ER membrane;7,9 however, up to date its biochemical properties have not been addressed.
Here we have determined by immunological subtyping that the mAb148.3 belongs to the immunoglobulin G1 (IgG1) κ-chain subtype (data not
shown). To map the epitope of the mAb148.3 we
have used peptide arrays of overlapping oligopeptides (15mers off-set by three amino acid residues)
covering the entire sequence of human TAP1 and
TAP2 (Figure 1). In accordance with the fact that a
C-terminal peptide of TAP1 was used for immunization of the mice to establish a hybridoma, the
epitope of the mAb148.3 is located within the last 15
C-terminal amino acid residues of human TAP1
(Figure 1). However, most importantly no other
peptide derived of TAP1 or TAP2 is recognized. The
same results were obtained in mapping experiments
on peptide arrays with one amino acid off-set (data
not shown). To determine the precise epitope and
critical residues for epitope recognition, systematically permutated variants and arrays of N and
C-terminal truncations of the identified peptide were
probed with hybridoma supernatant (Figure 2). A
linear epitope was identified corresponding to the
very last five C-terminal amino acid residues of TAP1
( 744 ADAPE 748 ) (Figure 2(b) and (c)). Here, the
binding efficiency was fully preserved when compared to the 15mer peptide. Further mutational
analyses demonstrate that the residues D745 and
E748 are mandatory for recognition, as they cannot
be replaced by alanine, whereas A744 has less impact
(Figure 2(a)). By constrast, exchange of A746 and
P747 did not affect epitope recognition (Figure 2(a)).
Surprisingly, binding of the mAb148.3 is not dependent on a free carboxyl group at the C terminus, since
the peptides were linked via a peptide bond to the
PEG-spacer on the cellulose membrane. Notably,
three independently synthesized peptides corresponding to the last 15 amino acid residues of TAP1
(Figure 2, always first spot on the peptide arrays)
show about the same signal intensity when incubated
with equal amounts of mAb148.3, demonstrating
quantitative reproducibility of the assay. In summary, the epitope of the mAb148.3 recognizes
specifically the very last five C-terminal amino acid
residues ADAPE (essential amino acids underlined)
of human TAP1.
Cloning, expression, and purification of
recombinant antibody fragments of
the mAb148.3
To convert the TAP1-specific mAb148.3 into Fv
and Fab formats, cDNA was prepared from hybridoma cells and the VH and VL chain genes were
PCR-amplified using a set of degenerate primers.
After sequencing, the complementarity determining
regions (CDR) and framework regions (FR) of the
Antibodies Modulating TAP Function
97
Figure 1. Epitope mapping of the monoclonal antibody mAb148.3. (a) Schematic representation of the TAP
heterodimer. The transmembrane domains (TMDs) of TAP1 and TAP2 and the nucleotide-binding domains (NBDs) are
shown in dark and light grey, respectively. The TMDs are subdivided into a core-domain, which was recently shown to
comprise the peptide translocation pore and N-terminal domains, which recruit the adapter protein tapasin.6,53 (b) Arrays
of overlapping oligopeptides (15mers, off-set three amino acids) of human TAP1 and TAP2 were generated by Fmoc
chemistry on PEG-functionalized cellulose membranes and probed with hybridoma supernatant (mAb148.3). Bound
antibodies were detected with a goat anti-mouse HRP-conjugated antibody and visualized by chemiluminescence
imaging. The areas on the membrane that correspond to the TMDs and the NBDs of TAP are shown in grey and as open
boxes, respectively.
antibody chains were assigned as described by
Kabat et al.27 (Figure 3). The genes of the identified
VH and VL chains were subsequently cloned into the
expression vectors pASK68 and pASK85 for periplasmic production of Fv and Fab fragments,
respectively. Recombinant Fv and Fab fragments
were produced at different levels in E. coli (Figure 4).
Both antibody fragments were detected after 1 h of
induction as shown by SDS-PAGE and immunoblotting with tag-specific antibodies (Figure 4). The
recombinant antibody fragments were initially
purified via the C-terminal Strep-tag of the Fv-VH
chain and the C-terminal His6-tag of the Fab-VHCH1
chain. However, the Fv fragment displayed a
tendency to dimerize, leading to reduced affinity
to its epitope (data not shown). Therefore, the
recombinant antibody fragments were purified by
affinity chromatography employing the peptide
CYWAMVQAPADAPE (epitope underlined). Here,
the Fv fragment was purified to homogeneity in a
single affinity chromatography step as shown by
SDS-PAGE (silver-stained) and immunoblotting
using a Streptactin-HRP conjugate (Figure 5(a)).
By contrast, the purity as well as the total amount
of purified Fab fragment was considerably lower
than for the Fv fragment (Figure 5(b)). The Fab
fragment was however clearly detectable by
immunoblotting using an Fab-specific antibody
(Figure 5(b)). Heterodimer assembly of the purified
Fab fragment was confirmed by SDS-PAGE under
non-reducing conditions. Here, the Fab fragment
has an apparent molecular mass of 50 kDa due to
the intermolecular disulfide bridge at the C termini
of the constant regions (between CH1 and CL) (data
not shown). The yields of purified Fv and Fab
fragments were 100 μg and 20 μg per liter of E. coli
culture, respectively. Notably, comparable amounts
of the Fab fragment were purified by IMAC.
Correct folding and monodispersity of the purified
Fv fragment was demonstrated by analytical size
exclusion chromatography (Figure 5(c)). The Fv
fragment appears as a single peak with a molecular
mass of 28 kDa. A proteolytic Fab fragment
(50 kDa), which was generated by papain digestion
of mAb148.3, eluted shortly before the Fv fragment
and was used as an internal standard (Figure 5(c)).
Notably, the proteolytic Fab fragment appears as a
monomeric peak with an additional sub-fraction
98
Antibodies Modulating TAP Function
Fab fragments by papain cleavage of the mAb148.3.
The resulting Fab fragments were purified on
Protein G Sepharose and analyzed by SDS-PAGE
under non-reducing conditions and immunoblotting using Fab and Fc-specific antibodies (Figure
7(a) and (b)). In addition to the monovalent Fab
fragment a small amount of the bivalent (Fab)2
fragment was detected with the Fab-specific antibody. The proteolytic Fab fragment displays a KD
value of 14.8( ± 0.8) nM (Figure 7(c)), demonstrating
that the KD values of the isolated paratopes of the
bivalent mAb148.3 for binding to the TAP1-NBD are
comparable with those of the recombinant Fv and
Fab fragments.
The association and dissociation kinetics of the Fv
fragment were analyzed by surface plasmon resonance using the epitope peptide CYWAMVQAPADAPE immobilized on a CM5 biosensor chip (Figure
6(c)). Rate constants and equilibrium binding constants were determined based on a standard model
for a 1:1 interaction. The association rate constant
derived from the various concentrations of Fv
fragments was determined to ka = (8.57 ± 0.31) × 104
M−1s−1 (4 °C). The dissociation rate constant was
kd = (1.43 ± 0.06) × 10−3 s−1 (4 °C). Diffusion limitation
and multivalency effects (crowding at the sensor
interface) could be excluded. The KD derived from
these rate constants was 16.80( ± 0.69) nM. The KD
value calculated from the resonance units at
equilibrium binding of the Fv fragment to the
surface (Req) is 32.1( ± 1.10) nM. Notably, the Req
value deduced from binding experiments at 25 °C is
35.4( ± 2.20) nM (data not shown). These results
prove the self-consistency of the data and confirm
the nanomolar affinity of the recombinant antibody
fragments for TAP1 determined by ELISA.
Figure 2. Determination of the minimal epitope of the
mAb148.3. The immunoreactive 15mer peptide corresponding to the C terminus of TAP1 was further dissected
by alanine scanning mutagenesis (a) and C (b) or N-terminal
truncation (c).
corresponding to an (Fab)2 fragment as described
for other antibodies.28
Recombinant antibody fragments bind with
nanomolar affinity to the TAP1 epitope
The binding affinity of the recombinant antibody
fragments was analyzed by ELISA using the NBD of
TAP1 as antigen.29 The equilibrium binding constants (KD) for the monovalent recombinant Fv
(31.8( ± 2.7) nM) and the Fab fragment (39.5( ± 2.7)
nM) are very similar (Figure 6(a) and (b)), however,
they differed drastically from the KD value of the
mAb148.3 (0.42( ± 0.04) nM) by two orders of
magnitude (Figure 7(d)). To investigate whether
this difference is based on avidity effects of the
bivalent IgG molecules, we generated proteolytic
Fv binding to the C terminus of TAP1 arrests
the TAP complex in a peptide transport
incompetent state
It was recently shown that the NBDs of TAP1 and
TAP2 have non-equivalent C-terminal tails, which
may account for functional differences in ATP
binding and hydrolysis at both NBDs.30 Moreover,
the C terminus of the dimeric HlyB-NBD, the closest
homologue of a dimer NBD structure, is involved in
stabilization of the dimer interface.31 Based on this
information we speculated that the Fv fragment
modulates peptide translocation into the ER lumen
by binding to the C terminus of TAP1. We therefore
performed transport inhibition studies in semipermeabilized cells after pre-incubation with graded
amounts of recombinant antibodies. Strikingly, the
Fv fragment inhibits the TAP-dependent peptide
transport in a concentration dependent manner
(Figure 8). The inhibitory effect is fully reversible in
the presence of a 1000-fold molar excess of the 5mer
epitope (Ac-ADAPE-OH), demonstrating that the
translocation arrest is specifically linked to the C
terminus of TAP1. Note that the 5mer peptide cannot
be recognized and transported by TAP, since it is too
short and blocked at its N terminus.32 In the presence
99
Antibodies Modulating TAP Function
Figure 3. Amino acid sequence
of the variable heavy and variable
light chains of the mAb148.3. The
complementarity determining regions (CDRs), which are involved
in epitope recognition, are annotated as described by Kabat et al.27
(yellow boxes). Interspersed βsheets, responsible for the stability
of the scaffold of the antigen-binding site, are indicated by arrows.
The variable region of the heavy
(VH) and light (VL) chain consists of
130 and 124 amino acid residues,
respectively. The intramolecular
disulfide bonds between Cys22 and
Cys98, and between Cys23 and
Cys93, respectively, are indicated
with an asterisk.
of 10 μM mAb148.3, peptide transport was blocked
to a similar extent when compared to the Fv
fragment. By contrast, the same amount of the
TAP2-specific mAb435.3 did not inhibit peptide
transport. Notably, the epitope of the mAb435.3 is
localized within a sequence stretch C-terminal of the
TMD of TAP2 (469KFQDVSFAYPNR480) as determined by epitope mapping (data not shown). These
findings argue against simple steric hinderance by
the Fv fragment or the mAb148.3. In the absence of
either ATP or in the presence of the TAP-specific
inhibitor ICP47 (10 μM with ATP), peptide transport
was reduced to background level.
The TAP complex is stabilized by antibody
fragments
Since the C-terminal tails of TAP1 and TAP2
control nucleotide interaction30 we investigated
whether binding of the Fv fragment to the C
terminus of the TAP1-NBD influences the ATPbinding properties of TAP. However, ATP-agarose
binding experiments (for experimental details see
Koch et al.6) in the absence and presence of 10 μM of
Fv fragment revealed that ATP binding to the
heterodimeric TAP complex as well as TAP1 alone
was not altered by the recombinant antibodies (see
Supplementary Data Figure 1).
Peptides and ATP bind independently from each
other to TAP.26,33,34 Moreover, the heterodimeric TAP
complex is highly unstable in the absence of ATP or
ADP.35 Based on these data we speculated that the Fv
fragment could compensate for the absence of
nucleotides. TAP-containing microsomes were preincubated in the absence of ATP and in the presence
of either ATP (3 mM) or Fv fragment (5 μM) for up to
120 min at 27 °C and subjected to peptide binding
experiments (at 4 °C) with the radiolabeled peptide
RR( 125 I)YQKSTEL at saturation concentration
(1 μM). Interestingly, the Fv fragment could stabilize
TAP to a similar extent as ATP (Figure 9). Even after
2 h 80% of the peptide-binding capacity of TAP was
maintained. By contrast, in the absence of ATP and
stabilizing Fv fragment the amount of peptidereceptive TAP decreased in a time-dependent manner with only 10% of bound peptide after 2 h. The
decrease of peptide binding directly reflects the loss
of peptide transport function as shown previously.35
In summary, these observations led to two
important conclusions: (i) the recombinant antibody
fragments do not affect ATP or peptide binding to
TAP and (ii) strongly stabilize the heterodimeric
TAP complex similar to ATP, possibly by mimicking
a nucleotide-bound state within the transport cycle.
Discussion
The major and rather unexpected finding of this
systematic study is that the peptide transport of
100
Antibodies Modulating TAP Function
Figure 4. Expression of recombinant antibodies. E. coli were transformed with expression vectors pASK68 (Fv) and
pASK85 (Fab). Expression was induced with IPTG (pASK68) or AHT (pASK85) for 3 h. Aliquots of the bacterial cell
lysates (0 h, 1 h, 2 h, 3 h), the supernatant after harvesting (sn), the periplasmic fraction (pp, 1/25 aliquot), and the
cytoplasmic fraction (cp) were analyzed by reducing Tricine-SDS-PAGE ((a), 16%, expression of Fv; (d), 12%, expression of
Fab). The VH (b) and VL (c) chains of the Fv fragment were detected with tag-specific antibodies at 14.4 kDa and 13.8 kDa,
respectively. (e) Heavy chain (VH-CH1) of the Fab fragment at 25 kDa. Molecular mass standard in kDa.
human TAP can be inhibited by antibodies that
specifically recognize a discrete linear epitope
comprised of the very last five C-terminal amino
acid residues of TAP1 (744ADAPE748). In addition,
these antibodies strongly stabilize the heterodimeric
TAP complex against rapid thermal inactivation in
the absence of ATP. Since no other region within TAP
was recognized by the antibodies both effects are
exclusively linked to the C terminus of TAP1.
Here, we generated recombinant Fv and Fab
fragments by genetic engineering from hybridoma
cells expressing the mAb148.3. Recombinant antibody fragments were heterologously expressed in
E. coli and secreted into the periplasm. Both, the Fv
and Fab fragments were purified to homogeneity
and monodispersity as demonstrated by SDS-PAGE,
immunoblotting and size exclusion chromatography. Based on kinetic and thermodynamic analyses
the epitope–paratope interaction was characterized
in detail and compared with the parental antibody
demonstrating nanomolar affinity of the monovalent antibody fragments for TAP1. The equilibrium
binding constants of the recombinant antibodies as
well as of the mAb148.3 were determined by ELISA
and surface plasmon resonance using the isolated
NBD and the C-terminal epitope peptide CYWAMVQAPADAPE of TAP1, respectively, as antigen. The
KD values for the monovalent recombinant Fv and
the Fab fragments are very similar. Moreover, the KD
of the monovalent proteolytic Fab fragment of the
mAb148.3 for the NBD of TAP1 is in the same range
as for the recombinant antibody fragments.
Surprisingly, the recombinant Fv fragment inhibits TAP-dependent peptide translocation into the
ER similar to viral factors such as ICP47 of herpes
simplex virus or US6 of human cytomegalovirus.
Transport inhibition of the monovalent Fv fragment
and the intact mAb148.3 was equally efficient. This
observation demonstrates that inhibition results
from a direct interaction of the paratopes with the
C terminus of TAP1 and not from steric effects
caused by the Fc regions of the intact antibodies or
cross-linking of adjacent TAP heterodimers because
of bivalent interactions of the mAb148.3.
Antibodies Modulating TAP Function
101
Figure 5. Affinity purification of recombinant antibody fragments. Fv and Fab fragments were affinity-purified via the
epitope peptide CYWAMVQAPADAPE of TAP1. Aliquots of the periplasmic fraction (pp), the flow through (ft), the
washing step (w), and the elution fractions (1–5) were analyzed. (a) Purification profile of the Fv fragment analyzed by
Tricine-SDS-PAGE (16%, silver-stained). The VH chain was detected with a Streptactin-HRP conjugate. (b) Purification of
the Fab fragment was monitored by Tricine-SDS-PAGE (12%, silver-stained). The Fab fragment was detected with an Fabspecific antibody. Molecular mass standard in kDa. (c) The affinity-purified Fv and proteolytic Fab fragments were
separated by size exclusion chromatography (Superdex 200 PC3.2/30). Proteins (0.3 mg total protein) were monitored at
280 nm. The void volume (V0 = 0.85 ml) and the total volume (Vt = 2.4 ml) are indicated by arrows in the graph.
To elucidate the inhibition mechanism we considered interference with ATP or peptide binding to
the TAP complex. However, we could not detect an
adverse effect on ATP or peptide binding in the
presence of the Fv fragment, illustrating that the
overall integrity of the heterodimeric TAP complex
is maintained. Strikingly, the recombinant antibody
fragments stabilize the TAP complex similar to ATP.
The peptide-binding capacity of TAP was maintained over 2 h, whereas only 10% of remaining TAP
activity was observed in the absence of Fv fragment.
These results are reminiscent of observations made
during characterization of the recently identified
TAP-specific viral inhibitor UL49.5 from bovine
herpesvirus 1 (BHV-1). This protein employs two
mechanisms to inhibit peptide translocation: (i)
arrest of the transporter in a functionally incompetent conformation and (ii) induction of proteasomal
degradation of components of the peptide-loading
complex.36 Interestingly, the TAP complex was
resistant to proteasomal degradation when green
fluorescent protein was fused to the C terminus of
TAP1. However, this effect could be explained by
either protection from proteolytic degradation or
stabilization of the TAP complex.
By binding to the C terminus of TAP1, the
antibodies presented in our study arrest TAP in a
transport-incompetent state. Several lines of argumentation have to be considered: The X-ray crystal
structure of the isolated NBD of human TAP1 bound
to ADP-Mg2+ was solved at a resolution of 2.5 Å.37
However, the structure of the last six C-terminal
amino acid residues, which contain the epitope of the
mAb148.3, was not resolved, indicating high flexibility of this region in the ADP-bound monomeric
state of the NBD. Based on truncation studies and
chimeras of rat TAP1 and TAP2, Knittler and
colleagues could show that the C-terminal tails of
TAP1 and TAP2 are non-equivalent.30 It became also
evident that the control of nucleotide interaction in
the NBD of TAP1 is more complex than in the NBD of
TAP2. Although nucleotide binding to the NBDs can
be modulated by both C-terminal tails, the exchange
of ADP to ATP within the NBD of TAP1 is restricted
to the intrinsic C-terminal tail. Interestingly, the Cterminal tail of the NBD of TAP1 can force the NBD
102
Figure 6. Affinity binding constants of the recombinant antibodies. ELISA plates were coated with purified
TAP1-NBD (10 μg/ml) and probed with various concentrations of the affinity-purified Fv (a) and Fab fragments
(b) (for details see Materials and Methods). The experimental data were fitted to a 1:1 Langmuir binding model.
(c) The binding kinetics of the Fv fragment to the epitope
peptide CYWAMVQAPADAPE were analyzed by surface
plasmon resonance at 4 °C. Rate constants and equilibrium binding constants were determined based on a
standard model for a 1:1 interaction.
of TAP2 to adapt the same ATP binding properties as
the NBD of TAP1, and the transporter containing two
C-terminal tails of TAP1 has a remaining peptide
transport activity of only 30% when compared to
wild-type TAP.30 More recently, it was suggested
that the so-called α6/β10-loops close to the C termini
of TAP1 and TAP2 determine the non-synonymous
nucleotide binding of the two NBDs.38 However,
there is no indication from structural or biochemical
studies that the α6/β10-loop region makes direct
Antibodies Modulating TAP Function
contact with bound ATP or ADP, suggesting that the
α6/β10-loops contribute to the regulation of nucleotide binding in a conformational rather than a sequence-specific manner.
The recombinant antibody fragments generated in
our study did not affect ATP binding to TAP.
However, peptide translocation was inhibited by
60%. Considering the potential importance of the
α6/β10-loop, the distance between this region and
the epitope of the antibodies at the C terminus of
TAP1 might be too large to fully disturb the
intramolecular cross-talk within the NBD of TAP1.
Inhibition of peptide transport was highly specific,
since the mAb435.3 directed against TAP2 showed
absolutely no effect on peptide transport even at a
concentration of 10 μM. If the affinity of mAb435.3
were to be 1000-fold lower than that of mAb148.3
we would still have expected to see an effect on
peptide transport. Furthermore, the fact that
mAb435.3 is widely used in co-immunoprecipitation studies of TAP and the peptide-loading complex, demonstrates that the native conformation of
TAP2 is recognized with a reasonable affinity.
Notably, the TAP-specific viral inhibitor ICP47,
which displays a similar equilibrium binding constant (50 nM39) to the Fv fragment (32 nM), inhibits
peptide translocation to a similar extent (75%) when
used at a concentration of 10 μM.
It is commonly accepted that ATP binding to the
NBDs of ABC-transporters induces the formation of
a composite dimer, which in turn is required for
productive ATP hydrolysis. 40 Several sequence
motifs of both NBD monomers contribute to dimer
formation, most importantly the Walker A motif and
the C-loop.31,41,42 The crystal structure of the HlyBNBD dimer bound to ATP-Mg2+provides evidence
that also the C terminus of the NBDs contributes to
the dimer interface.31 This suggests one possible
explanation of our data, namely that the antibodymediated inhibition of the TAP-dependent peptide
transport is due to interference with the formation of
a stable dimer interface between the NBDs of TAP1
and TAP2. Notably, alignment of HlyB and TAP1
shows little conservation of the C-terminal 20
residues. However, of the few dimeric NBD structures of ABC transporters available, HlyB is the
closest relative to TAP. At this stage, therefore, we
can only suggest a direct contribution of the C
terminus of TAP1 to the dimer interface. In order to
provide strong experimental proof we have to rely
on high-resolution 3D structures of a TAP heterodimer, which are not available yet.
Within the current study a detailed analysis of the
inhibition mechanism of TAP-dependent peptide
transport is provided (peptide binding, ATP binding). The only open issue is whether TAP is inhibited
by steric hindrance of the recombinant antibody
fragments or by neutralizing a possible direct
functional role of the C terminus of TAP1. In this
respect, truncation of the C-terminal residues of
TAP1 will not provide experimental proof to
distinguish between these two inhibition modes.
This distinction might, if at all, be possible on the
103
Antibodies Modulating TAP Function
Figure 7. Affinity binding constant of the proteolytic Fab fragment. After papain cleavage, the proteolytic Fab
fragment was affinity-purified on Protein G Sepharose and analyzed by non-reducing SDS-PAGE (10%). The Fab (a) and
Fc fragments (b) were detected with specific antibodies. The binding constants of the proteolytic Fab fragment (c) and the
mAb148.3 (d) for the NBD of TAP1 were determined by ELISA (for details see Materials and Methods). The experimental
data were fitted to a 1:1 Langmuir binding model.
level of TAP NBDs dimerization, which cannot be
analyzed within the context of a full-length transporter since its transmembrane domains confer
heterodimerization of TAP1 and TAP2. Therefore,
studies with the isolated NBDs would be required.
However, nobody in the field has so far managed to
form a non-artificial heterodimer of the NBDs of
TAP1 and TAP2 in solution. Finally, even if available,
it remains questionable whether the isolated NBD
dimer represents a valid model system for the TAP
heterodimer within the ER membrane.
In conclusion, the recombinant antibody fragments generated here proved useful to analyze the
functional importance of the very last amino acid
residues at the C terminus of TAP1. Moreover, the
antibodies represent valuable tools to modulate TAP
function and might prove useful in the structural
analysis and 3D-crystallization of the TAP complex.
Material and Methods
Subtyping of monoclonal antibodies
The Fc subtype of the human TAP1-specific mouse
monoclonal antibody mAb148.325 was determined using
the mouse hybridoma subtyping kit (Roche) according to
the manufacturer's instructions.
Epitope mapping
Oligopeptides covering the entire sequence of human
TAP1A (accession no. L21204) and TAP2E (accession
no. Z22936) were synthesized on activated cellulose
membranes employing the automated SPOT-robot
ASP222 (Intavis) as described.43 15mer oligopeptide
arrays of TAP1 and TAP2 (off-set by one or three
amino acids), mutated peptides with sequential exchange of all amino acids by alanine or glycine, and N
or C-terminally truncated peptides were generated.
After saturation of non-specific binding sites with 2%
(w/v) skimmed milk powder in TBS (20 mM Tris-HCl
(pH 8.0), 250 mM NaCl), the arrays were probed with
hybridoma supernatant (mAb148.3, 1:10 dilution in
blocking buffer) for 2 h at 20 °C. The cellulose
membranes were washed three times for 5 min with
TBS supplemented with 0.2% Tween 20 (TBS-T(0.2%))
and three times with TBS. Bound antibodies were
detected with a goat anti-mouse HRP-conjugated antibody (Sigma) and quantified by chemiluminescence
imaging (Lumi-Imager F1™, Roche). Analogous experiments were performed for the mAb435.3, a mouse
monoclonal antibody recognizing the NBD of human
TAP2.26
104
Antibodies Modulating TAP Function
side (IPTG, 1 mM, pASK68) or anhydrotetracyclin (AHT,
0.2 mg/l, pASK85). Growth was continued for 3 h at
22.5 °C and 200 rpm. Cells were harvested (5000g, 15 min,
4 °C), resuspended in 20 ml of periplasm extraction buffer
(100 mM Tris-HCl (pH 8.0), 500 mM sucrose, 1 mM EDTA)
and incubated on ice for 30 min. Spheroblasts were
removed (5000g, 45 min, 4 °C) and the periplasmic fraction
was stored at −20 °C.
Figure 8. Inhibition of intracellular peptide transport
in semi-permeabilized cells. Transport assays were
performed in semi-permeabilized cells with fluoresceinlabeled peptide (500 nM of RRYQNSTC(F)L) for 3 min at
32 °C in the presence of ATP (10 mM). For inhibition
studies the samples were pre-incubated with either ICP47
(10 μM), Fv fragment (10 μM) together with the
competitor peptide Ac-ADAPE-OH (10 mM), different
concentrations of Fv fragment (0.01–10 μM), mAb148.3
(10 μM) or mAb435.3 (10 μM). After cell lysis, N-core
glycosylated (transported) peptides were bound to Concanavalin A-Sepharose and quantified after specific
elution. Background transport activity was determined
by replacing ATP with apyrase (one unit). Data represent
the mean of triplicate measurements after subtraction of
background transport activity.
cDNA synthesis, PCR amplification, and cloning of
antibody genes
Total RNA of hybridoma cells expressing the mAb148.3
was extracted with the RNAeasy Mini Kit (Qiagen)
followed by poly(A)-mRNA isolation with the Oligotex
Direct mRNA Mini Kit (Qiagen). The VL and VH chain
genes were PCR-amplified using a set of degenerate
primers,44 cloned into the pDrive vector (Invitrogen) and
sequenced. To determine genetic heterogeneity, sequences
were compared with antibody genes in the Kabat
database.27 The identified VH and VL chain genes were
PCR-amplified and subsequently cloned into the vectors
pASK6845 and pASK8546 for periplasmic expression of Fv
and Fab fragments in E. coli. The following PCR primer
pairs were used: VH,for 5′-AGGTGAAGCTGCAGGAGACTG-3′ and VH,back 5′-GGAGACGGTGACCGAGG TTCCTTGAC-3′, VL,for 5′-TTGAGCTCAC TCAGGCTGCACCCTCTG-3′ and VL,back 5′-TCAGCTCGAGCTTGGTCCCAGCACCGAACGTGAG-3′ (introduced restriction sites are
underlined). The identity of the VH and VL chain genes was
confirmed by DNA sequencing.
Preparation of periplasmic extracts
Antibody fragments were expressed in E. coli (strain
JM83) based on published procedures.47,48 Briefly, an
overnight culture (grown at 30 °C in 2YT medium
supplemented with ampicillin (100 μg/ml) and streptomycin (30 μg/ml)) was diluted 40-fold in 2 l of the same
medium without streptomycin and grown at 22.5 °C till
mid exponential phase (A550 nm = 0.5). Protein expression
was induced by adding isopropyl-β-D-thiogalactopyrano-
Figure 9. Stabilization of the TAP complex. TAPcontaining microsomes (15 μg protein) were pre-incubated with 3 mM of ATP or 5 μM of Fv fragment for the
indicated periods at 27 °C. Peptide binding was
performed with 1 μM of radiolabeled peptide for
15 min at 4 °C. TAP-associated peptides were quantified
by γ-counting. Measurements without ATP and Fv
fragment (a), in the presence of ATP (b), and in the
presence of Fv fragment but without ATP (c). Data
represent the mean of triplicate measurements and were
normalized to the peptide-binding data of TAP without
additives (0 min).
105
Antibodies Modulating TAP Function
SDS-PAGE and immunoblotting
ELISA assay
Proteins were separated by Tricine-SDS-PAGE49 and
electro-transferred onto nitrocellulose membranes. Nonspecific binding sites were saturated with 2% (w/v)
skimmed milk powder in TBS-T(0.05%) (blots with Fab
fragments) or 3% (w/v) BSA in TBS-T(0.05%) supplemented with 0.5% (w/v) avidin (blots with Fv fragments). All washing steps were performed with TBS-T
(0.05%). The Strep-tag of the Fv-VH chain was detected
with a Streptactin-HRP conjugate (Sigma), the myc-tag of
the Fv-VL chain with the myc-specific monoclonal antibody 9E10, and the histidine-tag of the Fab-VHCH1 chain
with a His6-specific monoclonal antibody (Novagen).
Bound antibodies were detected with a goat-anti mouse
HRP-conjugated antibody (Sigma) and visualized by
chemiluminescence imaging.
96 well plates (MaxiSorb, Nunc) were coated with 1 μg of
soluble NBD of TAP129 in coating buffer (1 M NaHCO3 (pH
9.0); 100 μl/well) for 16 h at 4 °C. After saturation of nonspecific binding sites with 3% (w/v) BSA in TBS-T(0.05%),
the mAb148.3 or the antibody fragments were bound for
1 h at 25 °C and detected with a goat anti-mouse HRPconjugated antibody (Sigma). All washing steps were
performed with TBS-T(0.05%). Bound antibodies were
visualized with 3,3′,5,5′-tetramethylbenzidine (TMB, Progen) and quantified in an ELISA reader at a wavelength of
450 nm. All experiments were performed in triplicate.
Affinity purification of monoclonal and recombinant
antibodies
The peptide CYWAMVQAPADAPE corresponding to
the C terminus of human TAP1 was coupled to Ultralink
Iodoacetyl Gel (Pierce) via the N-terminal cysteine
according to the manufacturer's instructions. After
equilibration of the column with equilibration buffer
(50 mM sodium phosphate (pH 7.0), 150 mM NaCl), 50 ml
of hybridoma supernatant (mAb148.3, adjusted to pH 7.0)
or periplasmic fractions of the recombinant antibodies (Fv
and Fab fragments, dialyzed against equilibration buffer)
were loaded. After washing, bound antibodies were
eluted with 100 mM glycine-HCl (pH 2.7). The antibodies
were immediately buffered to neutral pH (1 M Tris-HCl,
pH 9.0), concentrated by ultrafiltration (Amicon, Millipore) and quantified by Micro BCA Protein Assay
(Pierce).
Generation of proteolytic Fab fragments
Papain cleavage of the mAb148.3 was carried out as
described50 with some modification. Digestion was
performed in 50 μl aliquots of reaction buffer (10 μM
mAb in 100 mM NaOAc, 1 mM EDTA, pH 5.5). To activate
the protease, 500 ng papain (Sigma) were pre-incubated in
reaction buffer supplemented with 100 mM cysteine
(Sigma) for 15 min at 20 °C. The reaction was started by
adding the mAb and continued for 16 h at 35 °C. The
digest was terminated with 70 mM iodoacetamide for
30 min at 20 °C. After neutralization (1 M Tris-HCl, pH
9.0) the mixture was loaded onto Protein G Sepharose
(Amersham Biosciences) pre-equilibrated with ten column
volumes of equilibration buffer (50 mM sodium phosphate, pH 7.0) and subsequently washed with ten column
volumes of the same buffer. The Fab fragments were
eluted with 100 mM glycine-HCl (pH 2.7), immediately
buffered to neutral pH (1 M Tris-HCl, pH 9.0), and
concentrated by ultrafiltration (Amicon, Millipore).
Size exclusion chromatography
Size exclusion chromatography was performed via a
Superdex 200 PC3.2/30 column (GE Healthcare) on a
SMART system (GE Healthcare) in running buffer (50 mM
sodium phosphate (pH 7.0), 150 mM NaCl) with a flow
rate of 40 μl/min at 4 °C. To determine the molecular mass
of the purified proteins carbonic anhydrase (29 kDa) and
albumin (66 kDa) were used as standards.
Surface plasmon resonance measurements
Binding kinetics were determined by SPR on a CM5 chip
(Biacore) using a Biacore® T100 (Biacore) surface plasmon
spectrometer. The peptide epitope CYWAMVQAPADAPE
was coupled to the dextran matrix in citric acid buffer
(0.1 M, pH 2.5) using a thiol coupling kit (Biacore). All
experiments were carried out at a flow rate of 30 μl/min
using HBS buffer (20 mM Hepes (pH 7.5), 150 mM NaCl,
0.05% surfactant P20). Different concentrations of recombinant antibody were injected onto the chip and dissociation was followed in a constant flow of HBS buffer without
antibody. The surface was regenerated with 0.1 M glycineHCl (pH 2.7). The analysis of the binding data was carried
out using the BIAevaluation software (Biacore) and Origin
(OriginalLab Corporation, Northampton, U. S. A.) based
on a standard model for a 1:1 interaction.
Cell culture
Insect cells (Spodoptera frugiperda (Sf9)) were grown in
Sf900II medium (Invitrogen) following standard procedures. Infection with recombinant baculovirus, encoding
human TAP1 and TAP2, was performed as described.25 For
infections, a multiplicity of infection (MOI) of 3 was used.
Peptide transport assay
2.5 × 106 Sf 9 cells were semi-permeabilized in 50 μl of
AP-buffer (5 mM MgCl2 in PBS, pH 7.0) supplemented
with saponin (0.05% (w/v)) for 1 min at 20 °C. After
washing with AP-buffer, the cells were pre-incubated with
the mAb148.3 (10 μM), the mAb435.3 (10 μM) or different
concentrations of the Fv fragment for 20 min at 4 °C and
then mixed with fluorescein-labeled peptide (500 nM,
RRYQNSTC(F)L, N-core glycosylation site underlined;52)
and ATP (10 mM) in 100 μl of AP-buffer. The transport
assay was performed for 3 min at 32 °C and terminated
with 1 ml of ice-cold PBS supplemented with 10 mM
EDTA. After centrifugation, the cells were solubilized in
1 ml of lysis buffer (50 mM Tris-HCl (pH 7.5), 150 mM
NaCl, 5 mM KCl, 1 mM CaCl2, 1 mM MnCl2, 1% (v/v) NP40) for 15 min at 20 °C. Insoluble debris was removed by
centrifugation. N-core glycosylated and therefore transported peptides were bound to 60 μl of Concanavalin A
(ConA) Sepharose (50% (v/v) in lysis buffer) for 16 h at
4 °C. After three times washing of the beads with 1 ml of
lysis buffer, peptides were eluted with 200 mM methyl αD-mannopyranoside for 30 min at 4 °C and quantified with
a fluorescence plate reader (λex/em = 485/520 nm). Background transport activity was determined by replacing
ATP with 1 unit of apyrase (Sigma). Specific inhibition of
the Fv fragment was determined in the presence of a 1000-
106
fold molar excess of 5mer epitope peptide (Ac-ADAPEOH). ICP47 was used as a control at identical conditions as
described for the Fv fragment. All experiments were
performed in triplicate and corrected for background
transport activity.
Peptide binding assay
Peptide binding to TAP was measured in filter assays as
described.51 TAP-containing microsomes (15 μg total
protein) were pre-incubated in the absence of ATP, with
ATP (3 mM) or Fv fragment (5 μM) for different periods at
27 °C. Subsequently, the microsomes were incubated with
the radiolabeled peptide RR(125I)YQKSTEL (1 μM) in 50 μl
of AP-buffer for 15 min at 4 °C. Microsomes were
transferred to the filter plate and washed three times
with 100 μl of ice-cold AP-buffer. Peptides bound to the
filters were quantified by γ-counting. The data were
corrected for background binding determined in the
presence of a 100-fold molar excess of unlabeled peptide.
All experiments were performed in triplicate.
Antibodies Modulating TAP Function
7.
8.
9.
10.
11.
12.
Acknowledgements
We thank Dr Arne Skerra for providing the
pASK68 and pASK85 vectors, Eckhard Linker for
technical assistance in cell culture, Dr David Parcej
for stimulating discussions, and Klaus Hoffmeier for
help in preparing Figure 3. This work was supported by the Deutsche Forschungsgemeinschaft
(SFB 628) and the Center for Membrane Proteomics
(Frankfurt/M.).
13.
14.
15.
16.
Supplementary Data
Supplementary data associated with this article
can be found, in the online version, at doi:10.1016/
j.jmb.2007.02.102
References
1. Lehner, P. J. & Cresswell, P. (2004). Recent developments in MHC-class-I-mediated antigen presentation.
Curr. Opin. Immunol. 16, 82–89.
2. Yewdell, J. W. & Haeryfar, S. M. (2005). Understanding
presentation of viral antigens to CD8+ T cells in vivo:
the key to rational vaccine design. Annu. Rev. Immunol.
23, 651–682.
3. Rock, K. L., Gramm, C., Rothstein, L., Clark, K., Stein,
R., Dick, L. et al. (1994). Inhibitors of the proteasome
block the degradation of most cell proteins and the
generation of peptides presented on MHC class I
molecules. Cell, 78, 761–771.
4. Abele, R. & Tampé, R. (2004). The ABCs of immunology:
structureandfunctionofTAP,thetransporterassociated
with antigen processing. Physiology (Bethesda), 19,
216–224.
5. Koch, J. & Tampé, R. (2006). The macromolecular
peptide-loading complex in MHC class I-dependent
antigen presentation. Cell. Mol. Life Sci. 63, 653–662.
6. Koch, J., Guntrum, R., Heintke, S., Kyritsis, C. &
Tampé, R. (2004). Functional dissection of the
17.
18.
19.
20.
21.
22.
23.
transmembrane domains of the transporter associated
with antigen processing (TAP). J. Biol. Chem. 279,
10142–10147.
Bangia, N. & Cresswell, P. (2005). Stoichiometric
tapasin interactions in the catalysis of major histocompatibility complex class I molecule assembly.
Immunology, 114, 346–353.
Antoniou, A. N., Ford, S., Pilley, E. S., Blake, N. &
Powis, S. J. (2002). Interactions formed by individually
expressed TAP1 and TAP2 polypeptide subunits.
Immunology, 106, 182–189.
Ortmann, B., Copeman, J., Lehner, P. J., Sadasivan, B.,
Herberg, J. A., Grandea, A. G. et al. (1997). A critical role
for tapasin in the assembly and function of multimeric
MHC class I-TAP complexes. Science, 277, 1306–1309.
Townsend, A., Ohlen, C., Bastin, J., Ljunggren, H. G.,
Foster, L. & Karre, K. (1989). Association of class I
major histocompatibility heavy and light chains
induced by viral peptides. Nature, 340, 443–448.
York, I. A., Roop, C., Andrews, D. W., Riddell, S. R.,
Graham, F. L. & Johnson, D. C. (1994). A cytosolic
herpes simplex virus protein inhibits antigen presentation to CD8+ T lymphocytes. Cell, 77, 525–535.
Hughes, E. A., Hammond, C. & Cresswell, P. (1997).
Misfolded major histocompatibility complex class I
heavy chains are translocated into the cytoplasm and
degraded by the proteasome. Proc. Natl Acad. Sci.
USA, 94, 1896–1901.
Loch, S. & Tampé, R. (2005). Viral evasion of the MHC
class I antigen-processing machinery. Pflugers. Arch.
451, 409–417.
McCluskey, J., Rossjohn, J. & Purcell, A. W. (2004). TAP
genes and immunity. Curr. Opin. Immunol. 16, 651–659.
Tomazin, R., Hill, A. B., Jugovic, P., York, I., van
Endert, P., Ploegh, H. L. et al. (1996). Stable binding of
the herpes simplex virus ICP47 protein to the peptide
binding site of TAP. EMBO J. 15, 3256–3266.
Ahn, K., Meyer, T. H., Uebel, S., Sempe, P., Djaballah,
H., Yang, Y. et al. (1996). Molecular mechanism and
species specificity of TAP inhibition by herpes simplex
virus ICP47. EMBO J. 15, 3247–3255.
Ahn, K., Gruhler, A., Galocha, B., Jones, T. R., Wiertz,
E. J., Ploegh, H. L. et al. (1997). The ER-luminal domain
of the HCMV glycoprotein US6 inhibits peptide
translocation by TAP. Immunity, 6, 613–621.
Hengel, H., Koopmann, J. O., Flohr, T., Muranyi, W.,
Goulmy, E., Hämmerling, G. J. et al. (1997). A viral
ER-resident glycoprotein inactivates the MHC-encoded peptide transporter. Immunity, 6, 623–632.
Lehner, P. J., Karttunen, J. T., Wilkinson, G. W. &
Cresswell, P. (1997). The human cytomegalovirus US6
glycoprotein inhibits transporter associated with
antigen processing-dependent peptide translocation.
Proc. Natl Acad. Sci. USA, 94, 6904–6909.
Hewitt, E. W., Gupta, S. S. & Lehner, P. J. (2001). The
human cytomegalovirus gene product US6 inhibits
ATP binding by TAP. EMBO J. 20, 387–396.
Kyritsis, C., Gorbulev, S., Hutschenreiter, S., Pawlitschko, K., Abele, R. & Tampé, R. (2001). Molecular
mechanism and structural aspects of transporter
associated with antigen processing inhibition by the
cytomegalovirus protein US6. J. Biol. Chem. 276,
48031–48039.
Spies, T. & DeMars, R. (1991). Restored expression of
major histocompatibility class I molecules by gene
transfer of a putative peptide transporter. Nature, 351,
323–324.
Seliger, B., Ritz, U., Abele, R., Bock, M., Tampé, R.,
Sutter, G. et al. (2001). Immune escape of melanoma:
107
Antibodies Modulating TAP Function
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
first evidence of structural alterations in two distinct
components of the MHC class I antigen processing
pathway. Cancer Res. 61, 8647–8650.
Schaedel, O. & Reiter, Y. (2006). Antibodies and their
fragments as anti-cancer agents. Curr. Pharm. Des. 12,
363–378.
Meyer, T. H., van Endert, P. M., Uebel, S., Ehring, B. &
Tampé, R. (1994). Functional expression and purification of the ABC transporter complex associated with
antigen processing (TAP) in insect cells. FEBS Letters,
351, 443–447.
van Endert, P. M., Tampé, R., Meyer, T. H., Tisch, R.,
Bach, J. F. & McDevitt, H. O. (1994). A sequential
model for peptide binding and transport by the
transporters associated with antigen processing.
Immunity, 1, 491–500.
Kabat, E. A., Wu, T. T., Reid-Miller, M., Perry, H. M. &
Gottesmann, K. (1987). Sequences of Proteins of Immunological Interest. Public Health Science, National
Institutes of Health, Bethesda, Maryland, USA.
Better, M., Chang, C. P., Robinson, R. R. & Horwitz, A.
H. (1988). Escherichia coli secretion of an active
chimeric antibody fragment. Science, 240, 1041–1043.
Müller, K. M., Ebensperger, C. & Tampé, R. (1994).
Nucleotide binding to the hydrophilic C-terminal
domain of the transporter associated with antigen
processing (TAP). J. Biol. Chem. 269, 14032–14037.
Bouabe, H. & Knittler, M. R. (2003). The distinct
nucleotide binding states of the transporter associated
with antigen processing (TAP) are regulated by the
nonhomologous C-terminal tails of TAP1 and TAP2.
Eur. J. Biochem. 270, 4531–4546.
Zaitseva, J., Jenewein, S., Jumpertz, T., Holland, I. B. &
Schmitt, L. (2005). H662 is the linchpin of ATP
hydrolysis in the nucleotide-binding domain of the
ABC transporter HlyB. EMBO J. 24, 1901–1910.
Uebel, S., Kraas, W., Kienle, S., Wiesmüller, K. H., Jung,
G. & Tampé, R. (1997). Recognition principle of the
TAP transporter disclosed by combinatorial peptide
libraries. Proc. Natl Acad. Sci. USA, 94, 8976–8981.
Androlewicz, M. J., Anderson, K. S. & Cresswell, P.
(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.
Proc. Natl Acad. Sci. USA, 90, 9130–9134.
Uebel, S., Meyer, T. H., Kraas, W., Kienle, S., Jung, G.,
Wiesmüller, K. H. & Tampé, R. (1995). Requirements
for peptide binding to the human transporter associated with antigen processing revealed by peptide
scans and complex peptide libraries. J. Biol. Chem. 270,
18512–18516.
van Endert, P. M. (1999). Role of nucleotides and
peptide substrate for stability and functional state of
the human ABC family transporters associated with
antigen processing. J. Biol. Chem. 274, 14632–14638.
Koppers-Lalic, D., Reits, E. A., Ressing, M. E., Lipinska,
A. D., Abele, R., Koch, J. et al. (2005). Varicelloviruses
avoid T cell recognition by UL49.5-mediated inactivation of the transporter associated with antigen processing. Proc. Natl Acad. Sci. USA, 102, 5144–5149.
Gaudet, R. & Wiley, D. C. (2001). Structure of the ABC
ATPase domain of human TAP1, the transporter
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
associated with antigen processing. EMBO J. 20,
4964–4972.
Ehses, S., Leonhardt, R. M., Hansen, G. & Knittler,
M. R. (2005). Functional role of C-terminal sequence
elements in the transporter associated with antigen
processing. J. Immunol. 174, 328–339.
Beinert, D., Neumann, L., Uebel, S. & Tampé, R. (1997).
Structure of the viral TAP-inhibitor ICP47 induced by
membrane association. Biochemistry, 36, 4694–4700.
Jones, P. M. & George, A. M. (2004). The ABC
transporter structure and mechanism: perspectives
on recent research. Cell. Mol. Life Sci. 61, 682–699.
Smith, P. C., Karpowich, N., Millen, L., Moody, J. E.,
Rosen, J., Thomas, P. J. & Hunt, J. F. (2002). ATP
binding to the motor domain from an ABC transporter
drives formation of a nucleotide sandwich dimer. Mol.
Cell, 10, 139–149.
Chen, J., Lu, G., Lin, J., Davidson, A. L. & Quiocho,
F. A. (2003). A tweezers-like motion of the ATPbinding cassette dimer in an ABC transport cycle. Mol.
Cell, 12, 651–661.
Koch, J. & Mahler, M. (2002). Peptide Arrays on
Membrane Supports. Springer, Heidelberg, Germany.
Hunte, C. & Müncke, C. (2004). Application of
antibody-fragments as crystallization enhancers. In
Molecular Biology in Medicinal Chemistry (Dingermann,
T., Steinhilber, D. & Folkers, C., eds), vol. 21, pp.
Wiley-VCH, Weinheim, Germany.
Kleymann, G., Ostermeier, C., Ludwig, B., Skerra, A.
& Michel, H. (1995). Engineered Fv fragments as a tool
for the one-step purification of integral multisubunit
membrane protein complexes. Biotechnology (N.Y.), 13,
155–160.
Skerra, A. (1994). Use of the tetracycline promoter for
the tightly regulated production of a murine antibody
fragment in Escherichia coli. Gene, 151, 131–135.
Skerra, A. & Plückthun, A. (1988). Assembly of a
functional immunoglobulin Fv fragment in Escherichia
coli. Science, 240, 1038–1041.
Fiedler, M. & Skerra, A. (2001). proBA complementation of an auxotrophic E. coli strain improves plasmid
stability and expression yield during fermenter production of a recombinant antibody fragment. Gene,
274, 111–118.
Schägger, H. (2003). SDS electrophoresis techniques. In
Membrane Purification and Crystallization (Hunte, C.,
von Jargow, G. & Schägger, H. eds), 2nd edit., Elsevier
Academic Press, Burlington, Maryland, USA.
Parham, P. (1986). Preparation and purification of
active fragments from mouse monoclonal antibodies.
In Cellular Immunology (Weir, D. M., ed.), 4th edit.
Blackwell Scientific Publications, California, USA.
Chen, M., Abele, R. & Tampé, R. (2003). Peptides induce
ATP hydrolysis at both subunits of the transporter
associated with antigen processing. J. Biol. Chem. 278,
29686–29692.
Neumann, L. & Tampé, R. (1999). Kinetic analysis of
peptide binding to the TAP transport complex:
evidence for structural rearrangements induced by
substrate binding. J. Mol. Biol. 294, 1203–1213.
Koch, J., Guntrum, R. & Tampé, R. (2005). Exploring the
minimal functional unit of the transporter associated
with antigen processing. FEBS Letters, 579, 4413–4416.
Edited by I. Wilson
(Received 14 September 2006; received in revised form 19 February 2007; accepted 23 February 2007)
Available online 15 March 2007