| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
,
,
Departments of
* Biochemistry and Biophysics and
Microbiology and Immunology, and
Lineberger Comprehensive Cancer Center, University of North Carolina, Chapel Hill, NC 27599; and
Laboratory of Cell Biology, National Cancer Institute, National Institutes of Health, Bethesda, MD 20892
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The exact mechanism by which the TCR/pMHC interaction results in
transmission of signals inside the T cell is unclear. Crystal
structures of TCR with and without class I pMHC indicate a
conformational change does not occur upon binding (4, 5).
The 
- and 
-chains of the TCR are short and contain no
recognizable signaling motifs, however, the associated CD3
complex
contains several repeats of the immune tyrosine-base activation motif
(ITAM) (6). The ITAMs are phosphorylated after TCR
engagement of pMHC Ag, via the Src family kinase
p56lck (7).
p56lck has been shown to associate with the
coreceptor on the surface of CD4+ or CD8+ T
cells (8). p56lck interacts with
the cysteine residues in the CD8 -chain (9, 10), and
the palmitoylation of both p56lck and CD8
result in the presence of both proteins in lipid rafts at the cell
membrane (11). CD8 makes specific interactions with
conserved regions in the 2 and 3 domains of MHC independently of
the identity of the bound peptide (12, 13). These data
have led to a model for T cell signaling where pMHC Ag interacts with
both TCR and CD8, and this interaction initiates the events that result
in T cell activation (14, 15). But, as plausible as this
mechanism for T cell activation seems, it does not account for the
existence of functional pMHC ligands that are CD8-independent
(16, 17, 18, 19, 20, 21, 22).
Most studies examining CD8 interactions with class I MHC have used the
CD8 homodimer, but the CD8 homodimer is found primarily on
intraepithelial T cells in the gut, NK cells, and 
T cells
(23, 24). The CD8 isoform found almost exclusively on
CD8+ T cells is the CD8 heterodimer.
Structures of both the human and murine CD8 homodimer
cocrystallized with pMHC complexes have been determined (12, 13). The structures confirm conclusions from earlier mutagenesis
experiments showing that CD8 binds to class I MHC mostly on the 3
domain and partially on the 2 domain (22, 25, 26).
Apparently, the CD8 -chain does not significantly change the
affinity of CD8 for class I MHC, as the affinity constant for the
binding of CD8 to pMHC as determined by surface plasmon resonance
(SPR) is 
65 µM (27), in the range of
KD values reported for CD8 homodimer
binding to MHC (11–220 µM) (24).
The function of CD8 binding to class I MHC in T cell activation is still disputed. The predominant function attributed to CD8 is that of a coreceptor, which increases the avidity of the fairly weak TCR:pMHC interaction (23, 24, 28). However, CD8 has also been implicated in cell-cell adhesion (29, 30, 31), and the H-2Dk tetramer has been shown to bind CD8 on naive CD8+ T cells independently of TCR:pMHC binding (32). Experimental evidence even suggests that CD8 is not necessary for functional T cell activation, as some pMHC:TCR interactions have been identified as "CD8-independent" based on the ability of CTL to lyse target cells in the presence of an anti-CD8 Ab (22, 33). Recent data indicate that T cells lacking CD8 are still able to kill and pMHC tetramers still bind (17, 20). Additionally, it has been shown that the presence of CD8 is absolutely necessary to trigger T cell calcium responses to pMHC (15, 17). Together, these data indicate that some, but not all, T cell functions may occur independently of CD8.
The murine AHIII 12.2 T cell clone recognizes murine class I molecule H-2Db in complex with a synthetic peptide p1058 (sequence FAPGFFPYL) (34). This murine CD8+ T cell was originally generated by injection of a human B cell into a C57BL/6 mouse (35). The clone was shown to be xenoreactive, and restricted to (human) class I MHC HLA-A2.1 in complex with peptide p1049 (sequence ALWGFFPVL) (35). Our earlier comparison of the x-ray crystal structures of p1049/A2 and p1027/Db (a p1058 variant also recognized by AHIII 12.2) demonstrated that the molecular surfaces recognized by this TCR are not similar (36), thus molecular mimicry does not play a significant role in the recognition of dissimilar pMHC by the AHIII 12.2 T cell clone. Exactly how this TCR recognizes two very different molecular surfaces is still not known.
In this study, we show affinity and kinetics for binding of the recombinant AHIII 12.2 TCR to syngeneic pMHC (p1058/Db) and xenogeneic pMHC (p1049/A2) ligands by SPR. These experiments demonstrate that the affinity of p1049/A2 for the AHIII 12.2 TCR is significantly greater than that of p1058/Db. In addition, functional assays with AHIII 12.2 T cells and anti-CD8 Abs show that xenogeneic p1049/A2 is recognized by the TCR in a CD8-independent fashion, while syngeneic p1058/Db activity is diminished in the presence of anti-CD8 Abs. Additional experiments prove that the early signaling pathways thought to require CD8 coengagement are intact when the T cells are presented with either p1049/A2 or p1058/Db. Confocal fluorescence microscopy and fluorescence resonance energy transfer (FRET) flow cytometry with pMHC tetramers and CD8 Abs reveal TCR and CD8 colocalize on the T cell membrane even in the absence of CD8 binding to class I MHC. Thus, binding of pMHC by CD8 is not necessary for TCR and CD8 to be brought together and for CD8-associated signals to be sent.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
All peptides were synthesized at the National Cancer Institute (Bethesda, MD). Peptides (p1049: ALSGFFPVL, p1058: FAPGFFPYL, MLL: MLLSVPLLL, or FLU: ASNENMETM) were purified to >95% purity by reverse-phase HPLC and their identity was confirmed by matrix-assisted laser desorption ionization-time-of-flight spectroscopy. Peptides were dissolved in DMSO at 20 mg/ml by weight. Final peptide concentrations were determined by amino acid analysis (Protein Chemistry Laboratory, Department of Chemistry, University of North Carolina, Chapel Hill, NC).
Cell lines, Abs, and MHC tetramer reagents
The CD8+ CTL clone AHIII 12.2, derived from a C57BL/6 mouse (37), was restricted to both HLA-A*0201 (38) and H-2 Db (34). AHIII 12.2 cells were maintained by weekly stimulation with irradiated HSB stimulator cells in the presence of IL-2. IL-2 was removed from culture media for at least 36 h before performing assays for T cell activity. EL-4 cells were obtained from R. Tisch (Department of Microbiology and Immunology, University of North Carolina). Chinese hamster ovary (CHO)-A2 cells were a gift from Dr. P. Parham (Department of Microbiology and Immunology, Stanford University, Palo Alto, CA)
The -CD8 (53-6.7), -CD8 (53-5.8) Abs (conjugated and
unconjugated), and the hamster IgG isotype Ab control were purchased
from BD PharMingen (San Diego, CA). The H57–597 anti-C TCR Ab
was purified from hybridoma supernatant (no. HB-218; American Type
Culture Collection, Manassas, VA). The 3A5 -
p56lck Ab (agarose conjugate) was purchased from
Santa Cruz Biotechnology (Santa Cruz, CA).
MHC tetramers were made as described previously (39).
Briefly, solubilized class I MHC H chain containing a C-terminal
biotinylation sequence, 2m, and peptide were
folded in vitro in a buffer; the crude mixture was then concentrated in
an Amicon ultrafiltration cell (Millipore, Bedford, MA), and purified
by gel filtration (40). Folded MHC was biotinylated using
BirA enzyme and biotinylation buffer (Avidity, Denver, CO) according to
manufacturer’s specifications. Biotinylated MHC monomers were
complexed with avidin, avidin-HRP, or avidin-PE (Leinco, St. Louis, MO)
to form tetramers, filtered through 0.22 µm, and used in experiments.
The extent of biotinylation was assessed by SDS-PAGE gel as described
(41). Octamers were formed by adding -PE
(Sigma-Aldrich, St. Louis, MO) Ab to pMHC tetramers as previously
described (41). Nonbiotinylated MHC protein used in SPR
experiments was produced as described above, except MHC H chain does
not contain a biotinylation target sequence as described previously
(40).
Construction of TCR expression plasmids, protein expression, and inclusion body preparation
cDNAs for the - and -chains of the AHIII 12.2 TCR (D.
Loftus, unpublished data) were used to construct expression vectors.
Residues 1–202 and 1–237 of the mature and AHIII 12.2 TCR
chains, respectively, were cloned into the pLM1 vector (a gift of G.
Verdine, Harvard University, Cambridge, MA) for expression in
Escherichia coli. The expression of TCR ectodomains in
E. coli was as previously described (42). After
induction, E. coli containing either AHIII 12.2 TCR - or
-chain were pelleted by centrifugation, and resuspended in buffer
containing lysozyme. The cells were lysed by French press, and DNA was
degraded by the addition of DNase. Cells were washed once with
detergent buffer and once with Tris buffer, then solubilized in
guanidium-HCl. Following ultracentrifugation, protein purity was
assessed by SDS-PAGE, and protein concentration was determined by
Bradford assay (Bio-Rad, Hercules, CA). Aliquots were stored at
-80°C until use.
Protein folding and purification
TCR was folded as described (42), with a few
modifications. Briefly, 70 mg of TCR -chain and 40 mg of TCR
-chain were combined and brought to a total volume of 30 ml with 10
mM Tris, 6 M guanidine-HCl, 0.2 mM DTT, pH 8.0. Ten milliliters of this
solution were injected into 600 ml of a buffer composed of 1 M
L-arginine, 100 mM TrisCl, pH 8.5, 2 mM EDTA, 0.5 mM
reduced glutathione, 0.05 mM oxidized glutathione, 0.2 mM PMSF, and 120
mg sodium azide. Two additional 10-ml injections of protein were added
successively in 3–12 h intervals to give a total protein concentration
of 4 µM. After the final protein injection, the buffer was kept at
10°C for 24 h. The buffer was dialyzed extensively against 10 L
of 100 mM urea, and finally 10 L of 100 mM urea, 10 mM Tris, pH 7.5.
After centrifugation and filtration, the solution was purified and
concentrated by DE-52 ion exchange (Whatman, Clifton, NJ) and injected
onto a 26/60 S300 gel filtration column (Amersham-Pharmacia Biotech,
Piscataway, NJ) at 0.5 ml/min in TBS, pH 8.0. The correctly folded TCR
eluted at a Mr of 45,000 kDa. Fractions
(5 ml) containing correctly folded TCR were concentrated to 200 µM,
and stored at -80°C. A typical yield was 7 mg protein/L of folding
buffer (4%).
ELISA for refolded AHIII 12.2 TCR
An ELISA with soluble TCR and pMHC tetramer has been previously
described (43). Briefly, the anti-C TCR Ab H57-597
was used to coat the wells of a 96-well ELISA plate at 2 µg/ml.
Subsequently, soluble AHIII 12.2 TCR was added to the plate at 2
µg/ml, followed by different concentrations of pMHC tetramers. For
this experiment, the avidin used to make pMHC tetramers was covalently
linked to HRP (Sigma-Aldrich) for detection. Following the addition of
3,3',5,5'-tetramethylbenzidine (an HRP substrate), the amount of
pMHC-tetramer binding was observed at OD405 using
a SPECTRAmax 190 plate reader (Molecular Devices, Sunnyvale, CA) and
quantitated using SOFTmax Pro 3.1 software.
Soluble TCR binding to cell surface
This experiment was performed as in Ref. 44 , with a few modifications. Briefly, T2 cells, which express peptide-deficient HLA-A2 molecules, were pulsed with 1 µM p1049 peptide or a control peptide (MLL), then incubated with various concentrations of soluble AHIII 12.2 TCR, followed by FITC labeled-H57-597 (Southern Biotechnology Associates, Birmingham, AL). After washing, median fluorescence was measured using a FACScan flow cytometer.
SPR experiments
Five-thousand resonance units (RUs) of H57-597 (capturing
molecule, anti-TCR C Ab) were covalently bound to a Biacore CM5
sensor chip (Uppsala, Sweden) using standard amine coupling. Soluble
AHIII 12.2 TCR (ligand) was then added to the Ab at a concentration of
67 nM to generate 300–400 RU of bound TCR. Soluble class I MHC
(analyte) was injected onto the surface at a flow rate of 100 µl/min
in a 30-s pulse. TCR and MHC were removed from the surface with 0.1 M
Glycine, 0.5 M NaCl, pH 2.5, and the procedure was repeated until at
least three curves were obtained for the different concentrations of
analyte. Curves obtained at each concentration were subtracted from a
reference surface that contained Ab alone without TCR. After background
subtraction, curves were imported into BIAevaluation 3.0, normalized
along the x- and y-axes, and fit globally to determine kinetic
constants. The suitability of the fit was measured based on
2 values and the appearance of residuals. In all cases,
2 was below 1, residuals were small and random, and the
experimental curves visually matched the predicted curves. Curves and
statistics were calculated from the average of at least three curves
for each concentration.
Multimer staining of AHIII 12.2 cells
For T cell binding assays, the tetramer or octamer complexes
were added at the indicated concentrations to 2 x
105 AHIII 12.2 T cells. After a 45-min incubation
at 4°C, cells were washed three times, and analyzed for median
fluorescence by flow cytometry (FACScan; BD Biosciences, Franklin
Lakes, NJ). As a negative control in all tetramer or octamer binding
experiments, multimers composed of HLA-A2.1 complexed to an irrelevant
peptide (MLL) were used. The median fluorescence of the MLL/A2 octamers
or tetramers at each concentration or time point was subtracted in each
experiment. For two-color staining with anti-CD8 and
anti-CD8 Abs, tetramer or octamer were added at saturating
concentrations after staining with the appropriate CD8 Ab for 30 min at
4°C.
Proliferation assay
AHIII 12.2 T cells (3 x 105) were
added to each of three wells in a 96-well flat-bottom plate after being
IL-2 starved for at least 36 h. Various concentrations, from 5 to
0.0005 µg/ml, of pMHC tetramers were then added to triplicate wells,
with or without 5 µg/ml -CD8 (53-6.7) or an isotype control Ab.
Three additional wells receive no tetramer. The plates were then
incubated at 37°C, 5% CO2, for 2 days, or
until cells clumped and formed T cell blasts.
[3H]Thymidine was then added at 1 µCi/well.
Cells were again incubated at 37°C, 5% CO2,
for 6–18 h, then either frozen at -20°C for future harvesting, or
harvested immediately using a multiple sample harvester (Otto
Hiller, Madison, WI). Incorporation of
[3H]thymidine was measured by scintillation
counting using a Beckman LS5000 counter (Palo Alto, CA).
Cytotoxicity assay
Target cells (2 x 106; CHO,
CHO-A2.1, or EL-4) were centrifuged and resuspended in 50 µl of FBS,
0.1 mCi (100 µl) 51Cr, with incubation at
37°C, 5% CO2 for 1 h. After several
washes with PBS, target cells were added at 5 x
103/well to a 96-well round bottom plate
containing the indicated amount of peptide. Recombinant human
2m (100 nM) was also added to CHO and CHO-A2.1
cells. 51Cr-labeled target cells and peptide were
incubated at 37°C, 5% CO2 for 1 h. AHIII
12.2 T cells with or without CD8 blocking Ab (53-6.7) were then
added at 5 x 104 cells/well to target
cells. T cells and target cells were incubated at 37°C, 5%
CO2. After 4 h, supernatants were harvested
and 51Cr release was measured with a Packard
Cobra gamma counter (Downers Grove, IL). All assays were performed in
triplicate, and the percent specific lysis was calculated as follows:
((experimental release - spontaneous release)/(maximum
release - spontaneous release)) x 100. Spontaneous release
is defined as counts per minute released from target cells in the
absence of T cells, and maximum release is defined as counts per minute
released from target cells lysed with 2.5% Triton X-100.
Calcium mobilization assay
AHIII 12.2 T cells were loaded with the calcium-specific dye
Indo-1-AM as previously described (45). After Indo-1-AM
labeling, some cells were additionally incubated with anti-CD8
or anti-CD8 FITC-conjugated Abs. Dye-loaded cells were examined
on a MoFlo flow cytometer (Cytomation, Fort Collins, CO) for 30 s to
generate a baseline, 10 µg of the indicated tetramer complexes were
added, and fluorescence was measured in real time for 9 min. Data was
analyzed using the FloJo software package (Treestar, San Carlos,
CA).
In vitro kinase assay
Cells (4 x 106 per sample) were
resuspended in 100 µl of RPMI 1640 at 37°C. The appropriate
tetramers and/or Abs were added for 3 min at 37°C, and the cells were
lysed for 10 min on ice with 1% Brij-96, 50 mM Tris, pH 7.5, 150 mM
NaCl, 1 mM Na3VO4, 10 mM
NaF, and protease inhibitors. After centrifugation, supernatant was
precleared and incubated with agarose-conjugated
anti-p56lck for 2 h. After washing,
agarose beads were incubated with 150 µl of kinase buffer (20 µCi
[-32P]ATP, 25 mM HEPES, pH 7.5, 7.5 mM MgCl,
1.5 mM MnCl, 1 mM Na3VO4,
and 20 µg of a peptide corresponding to the C-terminal ITAM of
CD3 (ITAMc) (46), for 15 min at 30°C
(47). To assess p56lck
phosphorylation, beads were pelleted, and equal amounts of the kinase
reaction were spotted onto triplicate circles of P81 phosphocellulose
paper (Whatman). After drying, the circles were washed four times with
1% phosphoric acid, and radioactivity was measured using the Beckman
LS5000 scintillation counter.
Confocal microscopy
AHIII 12.2 T cells (5 x 105) were
washed once with PBS and settled for 30 min on adherent-coated
Colorfrost Plus glass slides (Fisher, Pittsburgh, PA). After blocking
for 10–30 min with 10% FBS in PBS, cells were washed and stained with
tetramers (10 µg of p1049/A2, 20 µg of
p1058/Db, and MLL/A2) conjugated to Alexa 568
(Molecular Probes, Eugene, OR) or Abs (3 µg of anti-CD8 FITC)
for 5 min on ice. After washing, stained cells were fixed with 3%
sucrose/0.02% sodium azide/3% paraformaldehyde in PBS. Slides were
mounted with ProLong antifade (Molecular Probes), and examined using a
digital deconvolution fluorescence microscope (3I, Denver, CO). Data
was collected at the same fluorescence intensity for each sample, and
intensity was renormalized to approximately the same magnitude in the
final images.
FRET flow cytometry
FRET experiments were performed as described previously
(48). Briefly, 2 x 105 AHIII
12.2 T cells were stained on ice with 5 µg/ml
anti-CD8-allophycocyanin for 30 min. Cells were washed and the
indicated pMHC octamers conjugated to PE were added at 200 µg/ml, and
incubated on ice for an additional 30 min. After further washing, cells
were run on a FACSCalibur flow cytometer (BD Biosciences). The
allophycocyanin was excited between 615 and 655 nm, but was not
excited with the 488-nm laser as was PE. However, the excitation
spectrum for PE overlaps the absorption spectrum of allophycocyanin
allowing for the production of FRET. After stimulation of PE, FRET
between PE and allophycocyanin was visualized as an increase in signal
in the FL-3 channel, relative to negative control or single-color
staining. The FL3 channel was used because the little emission from PE
at wavelengths >670 nm was easily compensated. Compensation values
were almost identical to those published previously (48).
To compare the FRET produced from p1058/Db and
p1049/A2, the FRET found from p1058/Db was
adjusted to the level that would have been obtained if
p1058/Db octamer bound as well as did p1049/A2.
The p1058/Db octamers bound at a consistently
smaller fraction (59% of mean channel fluorescence) than p1049/A2
octamers at the concentrations used.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Previous studies with AHIII 12.2 T cells demonstrate that this T
cell clone is xenogeneic and recognizes both human HLA-A2 in complex
with the p1049 (ALWGFFPYL) peptide, and murine H-2
Db in complex with the p1058 (FAPGFFPVL) peptide
(34, 35). To determine the mechanism AHIII 12.2 uses to
detect these two disparate ligands, biophysical experiments were
performed with rTCR produced in vitro in E. coli. To confirm
that the rAHIII 12.2 TCR was folded properly, its ability to bind
p1049/A2 was measured by ELISA. The TCR was captured on an ELISA plate
using an anti-TCR C Ab, H57-597. Binding was assessed by
addition of class I MHC tetramers complexed to avidin-HRP. The p1049/A2
tetramers bound well to the captured TCR and this response was
dose-dependent (Fig. 1
A).
Binding of the rTCR was measured on live APCs as a second assay to
confirm proper folding of the TCR. T2 cells expressing HLA-A2.1 were
incubated with p1049 or control peptide, followed by different
concentrations of soluble AHIII 12.2 TCR. TCR binding to surface MHC
was detected using the anti-TCR C Ab, H57-597, conjugated to
FITC. T2 cells pulsed with p1049 exhibited a dose-dependent increase in
fluorescence due to TCR binding, while those cells pulsed with control
peptide did not (Fig. 1B). These data indicate that rAHIII
12.2 TCR expressed and refolded in vitro is correctly folded and
biologically active.
|
Much of literature on T cell activation implicates either the
affinity or half-life of the pMHC/TCR interaction as the trigger for
events that result in differentiation into effector T cells. SPR was
used to measure the binding of rAHIII 12.2 TCR to p1049/A2 or
p1058/Db. TCR was immobilized by binding to a
sensor chip previously coated with an Ab against TCR C (H57-597).
Subsequently, either p1049/A2 or p1058/Db pMHC
complexes were injected at various concentrations over the sensor chip.
Recombinant p1049/A2 bound well and p1058/Db
bound relatively poorly. Low concentrations of p1049/A2 generated the
same response units as high concentrations of
p1058/Db (Fig. 2).
The calculated affinity of the AHIII 12.2 TCR for p1049/A2 was 11.3
µM, 8-fold greater than for p1058/Db at 84
µM (Table I). This result is unexpected
as CTL assays using AHIII 12.2 T cells demonstrated equal lysis of
target cells containing A2 or Db (prepulsed with
p1049 and p1058, respectively) (34). Even though a
disparity exists between the affinities of these two ligands for
the AHIII 12.2 TCR, it is interesting to note that the off-rates for
the interaction of either p1049/A2 or p1058/Db
with the AHIII 12.2 TCR are similar (Table I).
|
|
). Previous studies in
our laboratory had indicated that some agonist ligands bind T cells too
weakly to be measured by tetramer staining (41).
Therefore, we increased the valency of the soluble pMHC ligands using a
mAb to the fluorophore covalently linked to avidin on the tetramer,
making octamers. Upon this modification, we saw binding of both
p1058/Db and p1049/A2 to the live AHIII 12.2 T
cells (Fig. 3). As seen in the SPR experiments,
p1058/Db octamers bound more poorly than did
p1049/A2 octamers. Therefore, the avidity contributed by the presence
of CD8 does not greatly enhance binding to
p1058/Db-presenting cells and does not compensate
for the differences in affinity between pMHC and TCR observed by
SPR.
|
Although binding of p1058/Db tetramers on
the T cell surface could not be detected using flow cytometry, AHIII
12.2 T cells lyse either p1049/A2 or
p1058/Db-presenting cells with equal
efficiency (34). We have previously shown that
class I MHC tetramers may stimulate CTL without apparent
requirements for costimulation (49). Tetramers composed of
p1049/A2 stimulated proliferation of AHIII 12.2 T cells (Fig. 4). Surprisingly,
p1058/Db tetramers stimulated proliferation of
AHIII 12.2 T cells equally well as did p1049/A2 (Fig. 4). These results
indicate that even tetramers that bind so poorly as to be undetected by
flow cytometry may produce an agonist response in functional assays. In
addition, as seen with peptide-pulsed target cells, A2 and
Db cause similar responses to AHIII 12.2 when
late effector responses are measured.
|
The CD8 coreceptor, expressed on CD8+ T
cells, has been shown to increase the avidity of the TCR/pMHC
interaction by binding to MHC (24, 31). The AHIII 12.2
CD8+ T cell line is derived from a mouse. Murine
CD8 binding (from AHIII 12.2 T cells) to human A2 should be greatly
reduced compared with murine CD8 binding to murine class I MHC
(H-2Db) as the residues in murine
Db that interact with CD8 are different from
those in A2. It has been shown that CD8 preferentially interacts with
MHC from the same species (19, 21). Additionally, murine
CD8 does not bind to human HLA-A2 when examined by SPR
(50), and when tested in cell-cell adhesion assays
(51). To confirm that murine CD8 is not involved in AHIII
12.2 T cell association with human pMHC, binding of human and murine
MHC octamers to AHIII 12.2 T cells was measured in the presence and
absence of anti-CD8 Abs. Two anti-CD8 Abs, one anti-CD8
and one anti-CD8, were used in these experiments. The
anti-CD8 Ab has been shown to block association of CD8 and class
I MHC (53-6.8, blocking) (17). Conversely, the
anti-CD8 Ab does not block association of CD8 and class I MHC
(53-6.7, nonblocking). It does increase binding of multimeric pMHC to T
cells presumably due to an increase in the local concentration of CD8
by virtue of the bivalent nature of the Ab (17). We have
previously shown that the nonblocking anti-CD8 Ab interferes
with proliferation and lysis even though it does not interfere with MHC
binding to CD8 (49). As predicted, the nonblocking
anti-CD8 Ab increases binding for p1058/Db
octamers to AHIII 12.2 T cells (Fig. 5A). Nonblocking
anti-CD8 Ab does not affect p1049/A2 octamer binding. The
blocking anti-CD8 Ab, 53-5.8, has been shown to impede binding
of multimeric class I MHC to T cells (17). If AHIII 12.2 T
cells are treated with blocking anti-CD8 Ab, binding of
p1058/Db octamers is completely abolished.
However, p1049/A2 octamer binding is only slightly decreased in the
presence of blocking anti-CD8 Ab. Similar results were observed
using either p1049/A2 tetramer or octamer at 25 or 37°C (data not
shown). These results imply xenoreactive p1049/A2 multimers do not
require an association with CD8 to bind AHIII 12.2 T cells.
|
Ab, does not block binding of
MHC to T cells (Fig. 5
A), it has been previously shown to
inhibit CTL activity (49). The presence of the nonblocking
anti-CD8 Ab in a CTL assay with p1049-pulsed CHO-A2 target cells
has a very small effect on lysis, while use of the Ab in an assay with
EL-4 pulsed p1058 cells results in a dramatic reduction in lysis (Fig. 5B). These data confirm the CD8 independence of xenoreactive
p1049/A2 and CD8 dependence of p1058/Db seen by
multimer binding.
T cell proliferation was tested in the presence of the nonblocking
anti-CD8 Ab, to determine whether the xenoreactive complex
responded differently from the syngeneic complex. AHIII 12.2 T cells
proliferate equally well when presented with p1049/A2 or
p1058/Db tetramers (Fig. 4). However, when nonblocking
anti-CD8 is added to AHIII 12.2 T cell culture in addition to
pMHC tetramers, proliferation decreases for both p1049/A2 or
p1058/Db tetramers (Fig. 5C). The
pattern of decrease is different for the two ligands. For p1049/A2, the
effect is a "dampening" of the saturating signal. Nonblocking
anti-CD8 has two effects on proliferation in the presence of
p1058/Db tetramers; it decreases the saturating
signal and right-shifts the dose-response curve. The right-shift is a
classic demonstration of CD8 dependence of
p1058/Db. The decrease in the saturating signal
in the presence of nonblocking anti-CD8 in both cases is not due
to tetramer-induced apoptosis, as Abs against the early apoptosis
marker protein, annexin V, do not bind to proliferating T cells. In
addition, similar numbers of T cells divide in either p1049/A2- or
p1058/Db-stimulated T cell cultures as measured
by the decrease in CFSE fluorescence over time (data not shown).
Therefore, the decrease in saturating signal in the presence of the
nonblocking anti-CD8 Ab is likely due to a decrease in the rate
of T cell divisions. These data show that the presence of nonblocking
anti-CD8 Abs do not cause a shift in the dose-response curve for
p1049/A2. Hence, by this assay the xenoreactive complex is classically
defined as CD8-independent.
In summary, anti-CD8 Abs reduce binding, proliferation, and lysis of AHIII 12.2 T cells when presented with syngeneic p1058/Db. These Abs do not impair the response of AHIII 12.2 T cells when presented with the xenoreactive p1049/A2. Interestingly, close examination of the effect of the anti-CD8 Abs shows that there is a small but reproducible reduction in binding, lysis, and rate of proliferation for the CD8-independent p1049/A2. These data suggest that even though there is no direct binding of the CD8 to the class I MHC, CD8 is nearby and involved in signaling in the cell for the xenoreactive ligand.
CD8-independent, xenoreactive responses generate the same early signals as CD8-dependent responses
The above experiments examine proliferation and cytolysis which
are late events in T cell activation. The data show that late T cell
events generated by p1049/A2 interactions are similar to those
generated by p1058/Db multimers even though
p1049/A2 does not bind CD8. These data imply that the early responses
do not require CD8 binding to class I MHC either. Previous experiments
have demonstrated that T cells lacking CD8 expression cannot produce a
calcium flux, even if the CD8- T cells can still
bind pMHC multimers (17). To determine the effects of
anti-CD8 Abs on calcium flux by CD8-dependent and CD8-independent
pMHC multimers, AHIII 12.2 T cells were loaded with the
calcium-specific fluorophore, Indo-1AM. Fluorescence from
Ca2+ binding to Indo-1AM was measured in real
time, immediately after addition of p1049/A2,
p1058/Db, or irrelevant tetramer. Both p1049/A2
and p1058/Db tetramers produced a sustained
calcium flux, while the irrelevant tetramer produced no flux (Fig. 6). Calcium mobilization induced by the
p1058/Db tetramer was delayed and decayed faster
than did the response to p1049/A2. This difference is likely due to the
poor binding of p1058/Db tetramer (Fig. 3), as a
greater proportion of AHIII 12.2 T cells responded, and the kinetics of
the calcium flux was faster and slightly more sustained when higher
concentrations of p1058/Db tetramer were used
(data not shown). Also, a transient calcium response to weak pMHC
ligands, such as seen in this study for p1058/Db,
has been demonstrated for some class II MHC ligands (52).
Prior incubation of AHIII 12.2 T cells with nonblocking anti-CD8
Ab increased the calcium response for both p1049/A2 and
p1058/Db (data not shown) most likely due to the
bivalence of the Ab as described earlier. As shown for multimer pMHC
binding, incubation with blocking anti-CD8 Ab does not eliminate
calcium flux after addition of p1049/A2 tetramer, but the
anti-CD8 Ab abolished Ca2+ mobilization by
p1058/Db.
|
). The
activity of p56lck is greater in p1049/A2
tetramer-stimulated T cells than in
p1058/Db-stimulated T cells.
p56lck activity in
p1058/Db-stimulated cells is not significantly
greater than the MLL/A2-tetramer background (p
= 0.304). The inability of p1058/Db tetramers to
significantly activate p56lck may reflect poor
TCR:pMHC binding, as weak agonists require longer pMHC:TCR contact
times, and higher concentrations for stimulation of signaling activity
(54). Because p1058/Db tetramers
cause a calcium flux and proliferation of these T cells, these data
imply that either the in vitro kinase activity assay is not sensitive
enough to show functional differences between p1058/Db tetramer and
irrelevant tetramer treatment or there is some unknown integrating
signal that allows for downstream events that is separate from
p56lck kinase activity.
|
in
combination with tetramers, a significant increase in
p56lck kinase activity is seen for both p1049/A2
(p = 0.003) and p1058/Db
vs irrelevant tetramer (p = 0.005). However,
when AHIII 12.2 T cells are treated with blocking anti-CD8 in
combination with tetramers, lck kinase activity of
p1058/Db-stimulated AHIII 12.2 T cells decreases
compared with stimulation with 1058/Db tetramer
alone or in combination with nonblocking anti-CD8. Once again,
blocking anti-CD8 has little effect on AHIII 12.2 T cell
stimulation by p1049/A2 tetramer compared with tetramer alone. These
data once again demonstrate the CD8 independence of the xenoreactive
pMHC, and indicate that binding of MHC with CD8 is not necessary for
signal transduction during T cell activation by p1049/A2. Yet, as seen
previously, CD8 engagement by syngeneic p1058/Db
is critical for efficient T cell activation. TCR and CD8 colocalize in the absence of direct pMHC:CD8 binding
AHIII 12.2 T cells proliferate, kill, and signal after
presentation with a CD8-independent ligand, p1049/A2. The kinase that
initiates the signal, p56lck, is activated in
p1049/A2-tetramer stimulated T cells. These data suggest that CD8 is
recruited to the immunological synapse in the absence of binding to
class I MHC. Confocal microscopy shows that both p1049/A2 and
p1058/Db tetramers colocalize in discrete patches
with the anti-CD8 Ab, but not in the irrelevant tetramer control
(Fig. 8, A–C). Furthermore,
p1049/A2 tetramers, but not p1058/Db tetramers,
colocalized with the blocking anti-CD8 Ab (data not shown). In
addition, no colocalization was apparent when staining with tetramer
and anti-CD90 (Fig. 8D). Colocalization is confirmed
with FRET between fluorophores conjugated to octameric class I MHC and
anti-CD8 Ab (Fig. 9A).
Although this particular use of FRET cannot be used to calculate exact
distances between the fluorophores reliably, after corrections for the
amount of octamer binding to the T cells, the fluorescence due to
p1058/Db octamer is slightly larger than for
p1049/A2. This suggests that the distance between the fluorophores is
smaller for p1058/Db than p1049/A2 as expected if
CD8 is binding to Db but not to A2 on AHIII 12.2
T cells (Fig. 9B).
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Using recombinant proteins and SPR, we show that the binding affinity
of p1049/A2 is about eight times greater than
p1058/Db for AHIII 12.2 TCR. The difference in
affinity is primarily due to a much slower on-rate of the murine
complex (Table I). Biologically, the apparent difference in affinity
may be illusory because p1058/Db binds to CD8 in
addition to the TCR when the T cell binds to the target cell. If the
binding of CD8 compensated for the lack of affinity of the
p1058/Db complex for the AHIII 12.2 TCR compared
with p1049/A2, tetramers composed of the two complexes should bind
equally well to the T cells, but p1058/Db
tetramers do not bind by flow cytometry and yet p1049/A2 tetramers bind
well. Surprisingly, even though the affinities are significantly
different, p1058/Db and p1049/A2 have similar
dissociation rates from AHIII 12.2 TCR. Because the T cell recognizes
cells presenting p1058/Db as well as they
recognize cells presenting p1049/A2, these data appear to support the
idea that the dissociation rate of the complex controls T cell
activation.
Most of the allo- and xenogeneic TCR examined have been classified as
CD8-independent (16, 17, 18, 19, 20, 21) and all CD8-independent TCR
studied bind their ligands with abnormally high affinity (20, 55). The 2C TCR allo-response to p2Ca/Ld
is CD8-independent (17, 20) with a
KD around 3 µM. However, 2C has an
10-fold lower affinity and is CD8-dependent when presented with
syngeneic SIYR/Kb (56). BM3.3 is
CD8-independent during recognition of pBM1/Kb
(57) with an affinity of 2.5 µM (55).
Affinity constants of syngeneic pMHC:TCR interactions are usually in
the range of 20–100 µM (56, 57, 58, 59). The HLA-A2-restricted
TCR, Jm22z, has a KD of 20 µM for its
syngeneic pMHC ligand and could not function independently of CD8
(50). Taken together with our results, these data imply
that above a certain threshold affinity, a direct interaction between
CD8:pMHC is no longer necessary. As most CD8-independent, high affinity
pMHC ligands are allo- or xenoreactive, it is likely that these
cross-reactive TCR would have been eliminated by negative selection if
they had undergone T cell development upon their allo- or xenoreactive
MHC. Thus, there must be selective pressure during T cell development
for low affinity, CD8-dependent syngeneic ligands (60) to
reduce the potential for autoreactive T cells. Abs also contact their
ligands with high affinity, and stimulation of T cells with TCR- or
CD3-specific Abs initiates T cell activation (61, 62). Thus, our findings agree well with fluorescence
microscopy studies that show cocapping of CD8 and TCR with
anti-CD8 and anti-TCR Abs (63), and
immunoprecipitation experiments that demonstrate an association between
CD8 and TCR/CD3 after anti-CD3 stimulation (53).
Two Abs were used to explore the CD8 dependence of murine AHIII 12.2
when presented with p1058/Db and p1049/A2.
Multimer binding, cytolysis, proliferation, calcium mobilization, and
lck activity were examined in the presence and absence of
these Abs. The anti-CD8 Ab (53-5.8) has been shown to block
binding of CD8 to MHC (17). This Ab slightly reduced
binding of p1049/A2 multimers to AHIII 12.2 T cells and decreased
calcium flux similarly. However, the anti-CD8 Ab completely
abrogates binding, abolished calcium flux, and greatly reduced
lck activity when stimulated with
p1058/Db. Conversely, the anti-CD8 Ab
(53-5.7) does not block binding of CD8 to MHC, but it has been shown to
block proliferation and lysis (49). Presumably, this
blockage has to do with disruption of an optimal complex of proteins at
the immunological synapse. Treatment with the anti-CD8 Ab
improves binding of p1058/Db, presumably because
of an aggregation of CD8 receptors by the bivalent Ab. Because murine
CD8 does not bind to A2, we would not expect an increase in binding of
p1049/A2 and that is indeed the case. In the presence of
anti-CD8, we see decreased AHIII 12.2, proliferation, but
increased calcium flux (data not shown) and lck activity
when presented with p1058/Db and p1049/A2. The
only difference we see is with respect to cytolysis. The
anti-CD8 Ab had no effect on CTL recognition of p1049/A2, but
significantly decreased recognition of p1058/Db.
These data confirm that AHIII 12.2 is CD8-independent by the classical
definition when presented with p1049/A2 and CD8-dependent when
presented with p1058/Db.
CD8 mobilizes to the area of TCR capping during T cell activation
(62). These areas may be called foci when treating with
tetramers, but on cells presented with competent APCs they would be
fully functional immunological synapses. The mechanism that CD8 uses to
migrate to the synapse is not clear. CD8 contains a palmitylation
target sequence such that CD8 is likely partitioned into lipid
microdomains (11, 63). During recognition of CD8-dependent
ligands, CD8 could be recruited by either actively binding to class I
MHC or CD8 may be recruited by mass action, literally by being
associated with the other proteins because of its location in the lipid
microdomain. Because CD8 is recruited to the foci without binding to
MHC, our data suggest that CD8 is initially recruited nonspecifically
by virtue of its location in a lipid microdomain. In the CD8-dependent
case such as p1058/Db, CD8 is recruited to and
remains in the synapse by virtue of association with pMHC, and also
probably because of continued recruitment signals. This would suggest
that CD8 recruitment is dependent on the strength of the signals that
stimulate aggregations of lipid microdomains. As CD8 recruitment also
seems dependent on the strength of the binding between pMHC and TCR
(data presented in this study and S. E. Kerry and J. A. Frelinger,
unpublished observations), this would suggest that the lipid domain
aggregation mechanisms are dependent on the strength of the
interaction. We are testing this hypothesis using a set of
CD8-independent pMHC:TCR partners of varying affinities.
Coligation of CD8 and TCR to pMHC leads to
p56lck activation (14, 15) as one
of the initial signaling events in T cell activation. Calcium flux does
not occur when CD8 is absent from the T cell surface (15, 17). This requirement may be seen using anti-CD3 Abs. T cell
activation upon anti-CD3 stimulation is dependent on
CD8/p56lck association, as mutation of the
p56lck-binding site in the CD8 -chain results
in a loss of calcium flux and protein tyrosine phosphorylation
(64). In this study, we show that xenogeneic pMHC elicits
p56lck kinase activity typical of T cell
activation, in the absence of binding between pMHC and CD8. In
addition, we show that there is not necessarily a correspondence
between lck activity and calcium mobilization. Treatment of
AHIII 12.2 T cells with p1058/Db tetramer results
in a well-defined calcium mobilization even if it is slower and weaker
than the response to p1049/A2. However, lck activity
generated by p1058/Db is undistinguishable
from the activity engendered by irrelevant tetramer. Because the time
frames of the experiments are comparable, there appear to be two
possible explanations. Either the reproducibility of the lck
kinase assay is such that we cannot distinguish between an active
signal and background or there is another signal that connects TCR:MHC
complex and calcium mobilization.
Our visual, biophysical and functional data all suggest that the CD8 coreceptor may be recruited in the absence of binding to MHC. The question becomes: what controls CD8 recruitment to the synapse? We suggest that it may be aggregation of lipid microdomains, because the instance in which we observe recruitment without binding is with a high affinity interaction between TCR and pMHC. In the instance where we observe low affinity, the signal to aggregate microdomains is not maintained and CD8 binding to MHC is required to keep CD8 in the foci. Therefore, we propose that high affinity TCR:pMHC binding obviates the requirement for CD8:pMHC binding to keep CD8 in the foci. This idea is supported by work with mutant TCR that have a higher affinity for pMHC (65). The rationale for the existence of the two mechanisms, CD8-dependent and CD8-independent, is not yet clear. Additionally, as the data presented in this study suggest that high affinity TCR:pMHC engagement occurs predominantly in the case of allo- or xeno-reactive ligands, it is possible that this high-affinity, CD8-independent activation mechanism presents different signals that may be used as a means to identify MHC ligands destined for removal by negative selection during T cell development.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Edward J. Collins, Department of Microbiology and Immunology, University of North Carolina, CB Number 7290, 804 M. E. Jones Building, Chapel Hill, NC 27599. E-mail address: edward_collins{at}med.unc.edu
3 Abbreviations used in this paper: pMHC, peptide MHC; ITAM, immune tyrosine-base activation motif; SPR, surface plasmon resonance; FRET, fluorescence resonance energy transfer; CHO, Chinese hamster ovary; RU, resonance unit.
Received for publication May 10, 2002. Accepted for publication October 17, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
T cell receptor structure at 2.5 A and its orientation in the TCR-MHC complex. Science 274:209. and HLA-A2. Nature 387:630.[Medline]
ectodomain fragment in complex with H-2Kb. Immunity 9:519.[Medline]
3 domain into HLA B27. Nature 348:642.[Medline]
3 domain of class I influence the magnitude of the xenogeneic response. J. Exp. Med. 170:1091.3 domain of HLA-A molecules affects binding to CD8. Nature 338:345.[Medline]
3 domain of HLA-A2. Nature 345:41.[Medline]
and CD8 dimers. J. Biol. Chem. 274:27237.2 and 3 domains of MHC class I. J. Exp. Med. 182:1275. domains in CD8 coreceptor function: importance for MHC I binding, signaling, and positive selection of CD8+ T cells in the thymus. Immunity 12:409.[Medline]
with an antigenic Tax peptide from human T lymphotropic virus type 1 and the class I MHC molecule HLA-A2. J. Immunol. 157:5403.[Abstract]
/CD8 interaction on the surface of T cells. J. Immunol. 167:821. chain. J. Biol. Chem. 276:32786. complex. Eur. J. Immunol. 26:2093.[Medline]
T cell receptor interactions with syngeneic and allogeneic ligands: affinity measurements and crystallization. Proc. Natl. Acad. Sci. USA 94:13838. endows CD8 with efficient coreceptor function by coupling T cell receptor/CD3 to raft-associated CD8/p56lck complexes. J. Exp. Med. 194:1485. polypeptides and intracellular associations of CD8 with p56lck. Eur. J. Immunol. 28:745.[Medline]
This article has been cited by other articles:
|
|
 |
|
  D. Gibbings and A. D. Befus CD4 and CD8: an inside-out coreceptor model for innate immune cells J. Leukoc. Biol., August 1, 2009; 86(2): 251 - 259. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Mareeva, E. Martinez-Hackert, and Y. Sykulev How a T Cell Receptor-like Antibody Recognizes Major Histocompatibility Complex-bound Peptide J. Biol. Chem., October 24, 2008; 283(43): 29053 - 29059. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Tian, R. Maile, E. J. Collins, and J. A. Frelinger CD8+ T Cell Activation Is Governed by TCR-Peptide/MHC Affinity, Not Dissociation Rate J. Immunol., September 1, 2007; 179(5): 2952 - 2960. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Laugel, H. A. van den Berg, E. Gostick, D. K. Cole, L. Wooldridge, J. Boulter, A. Milicic, D. A. Price, and A. K. Sewell Different T Cell Receptor Affinity Thresholds and CD8 Coreceptor Dependence Govern Cytotoxic T Lymphocyte Activation and Tetramer Binding Properties J. Biol. Chem., August 17, 2007; 282(33): 23799 - 23810. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Martinez-Hackert, N. Anikeeva, S. A. Kalams, B. D. Walker, W. A. Hendrickson, and Y. Sykulev Structural Basis for Degenerate Recognition of Natural HIV Peptide Variants by Cytotoxic Lymphocytes J. Biol. Chem., July 21, 2006; 281(29): 20205 - 20212. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Wooldridge, H. A. van den Berg, M. Glick, E. Gostick, B. Laugel, S. L. Hutchinson, A. Milicic, J. M. Brenchley, D. C. Douek, D. A. Price, et al. Interaction between the CD8 Coreceptor and Major Histocompatibility Complex Class I Stabilizes T Cell Receptor-Antigen Complexes at the Cell Surface J. Biol. Chem., July 29, 2005; 280(30): 27491 - 27501. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J.-L. Chen, G. Stewart-Jones, G. Bossi, N. M. Lissin, L. Wooldridge, E. M. L. Choi, G. Held, P. R. Dunbar, R. M. Esnouf, M. Sami, et al. Structural and kinetic basis for heightened immunogenicity of T cell vaccines J. Exp. Med., April 18, 2005; 201(8): 1243 - 1255. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. O. Holman, E. R. Walsh, and S. C. Jameson Characterizing the Impact of CD8 Antibodies on Class I MHC Multimer Binding J. Immunol., April 1, 2005; 174(7): 3986 - 3991. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. E. Kerry, J. Buslepp, L. A. Cramer, R. Maile, L. L. Hensley, A. I. Nielsen, P. Kavathas, B. J. Vilen, E. J. Collins, and J. A. Frelinger Interplay between TCR Affinity and Necessity of Coreceptor Ligation: High-Affinity Peptide-MHC/TCR Interaction Overcomes Lack of CD8 Engagement J. Immunol., November 1, 2003; 171(9): 4493 - 4503. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) or MLL/A2 (•)
pMHC tetramers containing streptavidin-HRP. After addition of
3,3',5,5'-tetramethylbenzidine substrate, OD405 was
measured. B, A2-expressing T2 cells were pulsed with 1
µM p1049 (
),
p1058/Db tetramers (
), or p1058/Db octamers
(
), or FLU/Db (*) were
incubated at the indicated concentrations with 3 x
105 IL-2-starved AHIII 12.2 T cells. After 24–48 h
[3H]thymidine was added and the cultures were reincubated
for an additional 6–12 h. [3H]Thymidine incorporation
was determined by scintillation counting. Data shown is from one
experiment performed in triplicate and is representative of three
separate experiments.
