| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
R. W. Johnson Pharmaceutical Research Institute, San Diego, CA 92121; and
Department of Immunology, The Scripps Research Institute, La Jolla, CA 92037
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The consequences of subjecting naive T cells to signal 1 alone are controversial. On the basis of exposing T cells to cross-linked anti-CD3 mAb or to Ag presented by fixed APC, it has been argued that signal 1 alone leads to an incomplete form of activation where the cells synthesize cytokine receptors, e.g., IL-2R, but not cytokines (IL-2) (5, 6). In the absence of exogenous cytokines, the T cells enter an anergic state where the cells are refractory to secondary stimulation. Despite these findings, some workers have found that the failure of purified naive phenotype T cells to respond to cross-linked anti-CD3 mAb in the absence of APC cannot be overcome by addition of exogenous IL-2 (7), implying that signal 1 alone may be incapable of synthesizing IL-2R. Conversely, at the other extreme, there are several reports that full activation of naive T cells can be induced by exposure to anti-CD3 mAb cross-linked on plastic or Sepharose beads (8). Similar evidence that signal 1 alone can be directly immunogenic has come from studies with the MHC class I (H-2 Ld)-restricted 2C TCR transgenic mouse (9). Here, culturing naive 2C T cells with purified peptide/Ld complexes on plastic was found to cause T cell proliferative responses in the absence of added APC.
We recently examined the effects of Signal 1 on 2C T cells with the aid of transfected Drosophila cells as APC (10). In this model, culturing naive 2C CD8+ cells with specific peptides presented by Ld-transfected Drosophila cells was essentially nonimmunogenic unless these APC also expressed costimulatory molecules, either B7–1 or ICAM-1. In the absence of costimulation, exposure to peptide/Ld complexes on Drosophila cells caused marked TCR down-regulation and a slight increase in IL-2R (CD25) expression on 2C CD8+ T cells, but failed to induce proliferation even with high concentrations of a high affinity peptide (11).
One explanation for the conflicting reports on the effects of signal 1 alone is that TCR signaling is much more intense when the ligand concerned (peptide/MHC complexes or anti-CD3/TCR mAb) is presented on an immobile matrix rather than on the cell surface. We have now examined this possibility by exposing 2C CD8+ T cells to peptide/Ld complexes coupled to either the surface of mouse erythrocytes or to cell-sized latex beads. Confirming our studies with Drosophila cells as APC (10), we show here that peptide/Ld complexes presented by mouse erythrocytes (mRBC)4 are essentially nonimmunogenic. However, in marked contrast to mRBC, peptide/Ld complexes presented on beads or bound to microtiter plates lead to overt T cell activation. Significantly, similar activation of T cells occurred when mRBC were fixed with glutaraldehyde before Ld coupling. The data thus suggest that signal 1 is only immunogenic when the ligand recognized by the TCR is immobilized.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
2C TCR transgenic mice (12) were maintained on a C57BL/6 background and kept under specific pathogen-free conditions at the rodent breeding colony at the R. W. Johnson Pharmaceutical Research Institute (San Diego, CA). 1B2 mAb specific for the 2C TCR (13) was used for screening the progeny.
Purification of CD8+ T cells
As described previously (14), cell suspensions prepared from pooled lymph nodes (inguinal, axillary, cervical, iliac, and mesenteric lymph nodes) of young adult 2C mice (8–12 wk) were first treated with a mixture of mAbs (anti-CD4, anti-HSA, and anti-I-Ab) plus C (complement) for 45 min at 37°C. The surviving 2C cells were further separated into CD8+ cells by panning at 4°C for 60–90 min on petri dishes coated with anti-CD8 mAb.
Cell lines, cytokines, and mAbs
The mAbs 3.168 (anti-CD8), RL172 (anti-CD4), J11D (anti-heat-stable Ag), and 28-16-8s (anti-I-Ab) were used as described previously (15). The Abs used for blocking experiments (anti-ICAM-1 and anti-IL-2) and those used for FACS analysis (FITC-conjugated anti-B7-1, anti-ICAM-1, anti-CD25, anti-CD69, and phycoerythrin-conjugated anti-CD8) were purchased from PharMingen (San Diego, CA). The cell line producing CTLA4Ig fusion protein was a gift from Peter Lane (Basel Institute for Immunology, Basel, Switzerland). The Ld-expressing RMA-S cell line (RMA-S.Ld) and the hybridoma producing the anti-clonotypic 1B2 mAb were provided by H. Eisen (Massachusetts Institute of Technology, Boston, MA). Recombinant human IL-2 was purchased from Genzyme (Boston, MA).
Peptides
Peptides used in this study were synthesized on an Applied Biosystem model 431A synthesizer (Foster City, CA) by a standard solid phase peptide synthesis method (tBoc chemistry). All peptides were purified with C18 reverse phase HPLC. The concentrations of peptides were determined by quantitative amino acid analysis. The sequences of peptides used in this study were: p2Ca, LSPFPFDL (16); QL9, QLSPFPFDL (17); and P1A, LPYLGWLVF (18).
Media
RPMI 1640 medium was supplemented with 10% FCS (Irvine Scientific, Santa Ana, CA), 5% NCTC 109 (Life Technologies, Gaithersburg, MD), 2 mM glutamine, 5 x 10-5 M 2-ME, and antibiotics.
Proliferation assay
Purified populations of 2C CD8+ cells (5 x 104/well) were cultured with Ld beads, Ld mRBC, or Ld-transfected Drosophila cells (2 x 105/well) in 200-µl wells in the presence of 10 µM of the indicated peptides. Cultures were pulsed with 1 µCi of [3H]thymidine (DuPont, Wilmington, DE) for 8 h before harvest (1 Ci = 37 GBq). All data shown refer to the mean of triplicate cultures; SD were generally within 5–15% of the mean.
IL-2 production
The biologic activity of IL-2 produced by 2C CD8+ T cells was measured using an IL-2-dependent cell line, CTLL-2. At the time indicated, 50 µl of supernatants were collected from each culture well and added to 5000 CTLL-2 cells for 24 h; 1 µCi of [3H]thymidine (DuPont) was added, and the cultures were harvested 16 h later.
CTL
2C CD8+ T cells were cultured with Ld beads or Ld mRBC in a volume of 2 ml in a 24-well culture plate in the presence of 10 µM peptides with or without IL-2. After 3 or 4 days, the cells were pooled and adjusted to the required number. To prepare targets, RMA-S-Ld cells were labeled with 51Cr (100 µCi/1–2 x 106 cells; DuPont) at 37°C for 90 min in the presence of peptides. After labeling, the cells were thoroughly washed and resuspended in medium with peptides. The CTL and labeled targets were cultured for 4 h, and the specific 51Cr release was calculated as previously described (14).
Flow cytometric analysis
For analysis of surface expression of Ld on latex beads and mRBC, 1 x 106 cells or beads were incubated with fluorescein-labeled anti-Ld mAb (30-5-7) for 30 min on ice and then washed. FITC-conjugated anti-B7-1 and anti-ICAM-1 mAbs (PharMingen, San Diego, CA) were used to analyze the expression of B7-1 and ICAM-1 on 2C CD8+ T cells. For analysis of surface expression of CD25 and CD69, 2C CD8+ T cells were stained with FITC-conjugated anti-CD25 or anti-CD69 mAb and phycoerythrin-conjugated anti-CD8 mAb. Propidium iodide was included in the last step of staining at 0.5 µg/ml. Live cells (propidium iodide negative) were acquired and analyzed on a FACScan (Becton Dickinson, Mountain View, CA).
Preparation of Ld-coated beads and RBC
Soluble MHC class I Ld molecules expressed in Drosophila melanogaster cells transfected with Ld and ß2-microglobulin (19) were purified as previously described (20) and biotinylated using biotin-BMCC (Pierce, Rockford, IL). Soluble class I molecules were shown to be correctly folded heterodimers, since they bind conformation-sensitive anti-Ld Ab 30-5-7 (21). In this article, Ld-ß2-microglobulin heterodimers are referred to as Ld.
Latex sulfate beads with diameters of 6.7 or 3 µm (Interfacial Dynamics, Portland, OR) were coated with neutravidin (Pierce) as follows. One hundred microliters of beads (3 x 108 beads/ml suspension) were incubated with 4 µg of neutravidin on ice for 2 h, then washed four times and incubated for 2 h on ice with a different concentration of biotinylated Ld. For simplicity, the beads 6.7 µm in diameter will be referred to as 6 µm. Ld immobilization on 5-µm magnetic beads was performed as previously described (22). Mouse erythrocytes were biotinylated and coated with neutravidin as previously described (23). Briefly, after centrifugation the buffy coat was removed, and mRBC were washed four times in PBS. A 10% mRBC suspension in PBS was then incubated in the presence of NHS-LC-biotin (Pierce) at a final concentration of 167 µg/ml for 30 min at room temperature. After four washes in PBS containing 3% dialyzed FCS, cells were resuspended in an equal volume of 1 mg/ml neutravidin and kept at room temperature for 30 min. They were then washed four times and incubated with biotinylated Ld for 2 h on ice. In some cases, biotinylated anti-CD28 mAb (PharMingen) was immobilized together with biotinylated Ld to mRBC or latex beads. The molar ratio of Ld:anti-CD28 mAb used for coupling was 24:1.
Ld immobilization was monitored by FACS analysis using
anti-Ld mAb (30-5-7). Immobilization of Ld
on mRBC increased linearly when amounts of Ld in the
incubation medium ranged from 2 to 66 ng/106 mRBC (MFI
increase from 
80 to 1500) and reached a plateau beyond 200
ng/106 mRBC (MFI = 2500). Conditions with latex
beads were optimized so as to obtain a similar range of MFI values on
beads and mRBC. Ld immobilization on latex beads increased
linearly when amounts of Ld ranging from 1.8–150
ng/106 beads used in the incubation (MFI increase from
30 to 2300). Amounts of Ld immobilized under these
conditions were quantitatively assessed by Scatchard analysis using
125I-labeled Fab of anti-Ld mAb 30-5-7. We
found 12,200–50,700 Ld molecules immobilized per mRBC
(average of three independent measurements). Comparable amounts
(8,100–49,000 molecules of Ld/bead; average of four
independent measurements) were immobilized on latex beads. Unless
indicated otherwise, the highest amounts of Ld were used in
experiments (Ld2+, i.e., 200 ng/106 mRBC or 150
ng/106 beads). Where indicated, suboptimal amounts of
Ld (Ld+), corresponding to about 2.5 times
fewer immobilized molecules, or differing amounts of Ld in
the range indicated above were used.
Glutaraldehyde-fixed Ld-coated mRBC were prepared as follows. The mRBC were biotinylated as described above and then incubated in 1% glutaraldehyde in PBS for 30 min. Glycine was added (0.1 M final concentration), and cells were washed four times and coupled to neutravidin as described above. After four washes, fixed mRBC were incubated with biotinylated Ld. We determined by Scatchard analysis that fixed and nonfixed mRBC were coated with comparable amounts of Ld.
Drosophila APC
Drosophila cells transfected with Ld molecules alone (Ld.Fly) or Ld plus B7-1 (Ld.Fly.B7–1) molecules were prepared and used as described previously (10).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The mRBC and latex beads of two different sizes (6 and 3 µm)
were coupled with soluble recombinant Ld class I molecules
either with or without anti-CD28 mAb as described in
Materials and Methods. In brief, beads coated with
neutravidin and mRBC subjected to surface biotinylation followed by
exposure to neutravidin were incubated with various concentrations of
biotinylated Ld molecules in the presence or the absence of
biotinylated anti-CD28 mAb and then washed thoroughly. FACS
analysis of Ld density on beads and mRBC is shown in Fig. 1
A. For the functional studies
discussed below, CD8+ T cells were cultured with
6-µm-diameter latex beads or mRBC coated with about 5 x
104 Ld molecules/bead or cell unless stated
otherwise.
|
-ketoglutarate dehydrogenase, a Krebs cycle enzyme (16, 24). In terms of immunogenicity and binding to Ld and 2C
TCR molecules, QL9 is a substantially higher affinity peptide than p2Ca
(17). An unrelated peptide, P1A (18), was used as a negative control;
this peptide is not recognized by 2C cells but has very high affinity
for Ld molecules. For functional studies, purified naive 2C
CD8+ cells (CD44low, 1B2+) were
prepared from lymph node suspensions of young 2C mice; for
proliferative responses, doses of 5 x 104 2C
CD8+ cells and 2 x 105 beads or mRBC/well
were used unless stated otherwise. It should be noted that the
proportion of memory phenotype CD44high, 1B2+
CD8+ cells was extremely low in the 2C mice used, typically
2% in an 8-wk-old mouse (14). The main findings reported below were
confirmed with FACS-sorted cells depleted of contaminating
CD44high 1B2+ cells. T cell activation and proliferation
As mentioned in the Introduction, previous studies
showed that overt activation of 2C CD8+ cells induced by
QL9 peptide presented by Ld-transfected
Drosophila cells (Ld.Fly) as APC required that
these APC coexpressed either B7 (B7-1 or B7-2) or ICAM-1
costimulatory/adhesion molecules (10). With Drosophila APC
expressing Ld alone, addition of a low concentration of QL9
peptide was sufficient to cause marked TCR down-regulation (11).
However, even a high concentration of QL9 (10 µM) induced only
minimal up-regulation of the activation markers CD69 and CD25
(IL-2R); proliferative responses were undetectable (10).
Essentially identical findings occurred when 2C CD8+
cells were cultured with QL9 peptide (10 µM) presented by
Ld coupled to the surface of mRBC (Ld.mRBC).
Thus, despite marked TCR down-regulation (Fig. 1
B), levels
of CD69 (Fig. 1B) and CD25 expression (data not shown) on 2C
CD8+ cells after 12-h culture with Ld.mRBC and
QL9 were only slightly elevated, and proliferative responses (measured
on day 2) were undetectable (Fig. 1, B and D). By
contrast, coupling of both Ld and anti-CD28 mAb on the
surface of mRBC (Ld.mRBC.CD28) led not only to TCR
down-regulation but also to conspicuous up-regulation of CD69 and CD25
and a strong proliferative response in the absence of added cytokines
(Fig. 1, B and D). These changes applied to the
highly immunogenic QL9 peptide. When Ld.mRBC.CD28 plus
the nonimmunogenic P1A peptide were used as a control, the surface
phenotype (including TCR expression) of 2C cells was essentially
identical with that of 2C cells cultured in medium alone (Fig. 1B).
The above data indicate that, as for
Ld.Drosophila cells (10), presentation of the
immunogenic QL9 peptide by Ld.mRBC to 2C cells caused
marked TCR down-regulation but was incapable of inducing T cell
activation. The results with Ld-coupled latex beads
(Ld.beads) as APC were quite different. Thus, unlike
Ld.mRBC, Ld.beads plus QL9 peptide induced
marked up-regulation of CD69 (Fig. 1B) and CD25 (Fig. 1C) and led to a moderately strong proliferative response
(Fig. 1, B–D); these data applied in the absence of added
cytokines and with concentrations of QL9 peptide ranging from 10 µM
(Fig. 1, B–D) down to 0.1 µM (data not shown). Coupling
of anti-CD28 mAb on Ld.beads
(Ld.beads.CD28) caused further elevation of CD25 and
CD69 expression and increased the proliferative response by about
three- to fourfold (Fig. 1, B–D).
In contrast to the strong QL9 peptide, presentation of the weaker p2Ca
peptide by Ld.beads caused little or no change in CD69 and
CD25 expression and failed to cause a proliferative response even in
the presence of added IL-2 (Fig. 1C and data not shown).
However, p2Ca peptide did cause CD25 up-regulation with
Ld.beads.CD28 (Fig. 1C). In this situation,
addition of IL-2 led to a moderately strong proliferative response
(data not shown). As for Ld.mRBC.CD28, addition of the
control P1A peptide to Ld.beads.CD28 failed to cause 2C
CD8+ proliferation or a change in CD25 and CD69 expression
(Fig. 1C and data not shown).
The above data indicated that Ld beads acted as efficient
APC for 2C CD8+ cells, although only with the high affinity
QL9 peptide. These findings with Ld.beads contrast sharply
with the complete inability of either Ld.mRBC or
Ld.Fly APC to stimulate 2C CD8+ cells unless
costimulation (anti-CD28 mAb for mRBC and B7-1 for
Drosophila cells) was provided (Fig. 1D).
The conspicuous APC function found with Ld.beads plus QL9
peptide raised the question of whether certain physical features of the
latex beads provided an unusual form of costimulation. To examine this
possibility, we tested whether adding uncoupled (no Ld)
beads to Ld.mRBC could provide "bystander"
costimulation, which is known to stimulate 2C CD8+ cells in
another system (25). As shown in Fig. 1D, however, addition
of either uncoupled beads or beads coupled with neutravidin
alone was totally unable to overcome the failure of 2C
CD8+ cells to respond to Ld.mRBC plus QL9
peptide. In addition, overt activation of 2C CD8+ cells
could be induced by QL9 peptide presented by another type of beads,
i.e., Ld-coupled magnetic beads (data not shown). In view
of these findings, the possibility that latex beads possessed intrinsic
costimulatory activity seems unlikely.
Ld density
For the above experiments, 6-µm beads expressing about 5 x
104 Ld molecules/bead were used. The effect of
Ld density on activation of T cells was tested using latex
beads coupled with different concentrations of Ld. FACS
analysis of the coupled beads showed that the number of Ld
molecules on the beads correlated directly with the concentration of
Ld molecules used during the coupling procedure (data not
shown). The number of Ld molecules on the Ld
beads was determined by Scatchard plot analysis (Fig. 2A).
|
C). For TCR down-regulation, a slightly higher density of
Ld molecules (
10,000 Ld/bead) was required
(Fig. 2B). For proliferation (Fig. 2D), responses
were near background levels with low concentrations of Ld
molecules (<10,000 Ld/bead), but increased sharply as the
Ld density was increased up to 50,000 Ld/bead.
For most of the experiments discussed below, we used beads displaying a
high density of Ld (50,000 Ld/beads;
Ld2+.beads); in some experiments beads containing a two- to
threefold lower Ld density (Ld+.beads) were
used. Bead size vs Ld density
The above data refer to cell-sized 6-µm beads used at 2 x
105/well with 5 x 104 responder cells.
The effects of culturing 2C CD8+ cells with QL9 peptide (10
µM) presented by 6- vs 3-µm beads containing either high
(Ld2+.beads) or low density (Ld+.beads) of
Ld are shown in Fig. 3;
except in Fig. 3B, the beads were all used at the same
concentration (2 x 105/well).
|
, Aa and Ab). Cocoupling of anti-CD28 mAb on
the beads did not affect TCR down-regulation but amplified CD69
expression (Fig. 3, Ac and Ad; note the different
scales in Ab and Ad).
The results were quite different for proliferative responses (Fig. 3B). Here, both Ld density and bead size were
highly important. With a low Ld density (Fig. 3Ba), proliferative responses (measured without added
cytokines) were very limited, although the responses were clearly
higher with 6- than 3-µm beads (see inset). With a high
Ld density (Fig. 3Bb), even small numbers of
6-µm beads (2 x 104/well) generated high responses;
by contrast, large numbers of 3-µm beads led to low (although clearly
significant) responses. These differences were less marked when the
beads coupling both Ld and anti-CD28 mAb (Fig. 3, Bc and Bd; note the difference in scales for
Ba and Bb vs Bc and Bd); in
this situation, even 3-µm beads with low Ld density
became highly immunogenic. Thus, bead size and Ld density
became much less important when costimulation was provided.
Kinetics
The above findings indicate that in the absence of costimulation,
production of strong 2C proliferative responses by Ld.beads
plus QL9 peptide required both a high density of Ld on the
beads and the use of cell-sized beads (6 µm rather than 3 µm).
For these experiments and the experiments in Figs. 1 and 2,
proliferative responses were measured early in culture, i.e., on day 2.
Interestingly, the proliferative responses with Ld.beads
dropped sharply on day 3 and declined to background levels on day 4
(Fig. 3, Ca and Cd). These abbreviated kinetics
applied to each type of bead tested, including the immunogenic 6-µm
Ld2+ beads and were largely unrelated to the dose of
responder cells used (Fig. 3Cd). When the immunogenic 6-µm
Ld2+ beads were also coupled with anti-CD28 mAb, the
proliferative response was augmented and reached a peak on day 3 rather
than on day 2 (Fig. 3Cb, note the difference in scales
between Ca and Cb). With less immunogenic beads,
e.g., 3-µm Ld+ beads, cocoupling of anti-CD28 mAb
caused a marked elevation of the response on day 2, but the response
declined thereafter (Fig. 3Cb).
The above data make two points. First, the proliferative responses elicited by Ld.beads were of brief duration and fell markedly after day 2. Second, beads containing anti-CD28 mAb prolonged the proliferative response, but only with the more immunogenic beads (3- or 6-µm Ld2+.beads).
IL-2 production and CTL activity
The simplest explanation for the abbreviated proliferative
responses elicited by Ld.beads is that IL-2 production in
the cultures was too low to stimulate more than transient
proliferation. In support of this possibility, IL-2 production elicited
by Ld.beads was high when the beads contained anti-CD28
mAb (Fig. 4Aa) but very low
when the beads lacked this Ab (Fig. 4Ab, note the different
scales between Aa and Ab). Although very limited,
IL-2 production elicited by Ld.beads seemed to be real
because cytokine production increased appreciably with high doses of
responder cells (Fig. 4Ac), was higher at 24 h than at
48 h (Fig. 4Ac), and was virtually undetectable with
beads coated with suboptimal amounts of Ld (Fig. 4Ab); in addition, adding anti-IL-2 mAb to the cultures
caused a partial decrease in the proliferative response (see below).
With Ld.mRBC, IL-2 production was undetectable (Fig. 4Ab) unless these cells were also coated with anti-CD28
mAb (Fig. 4Aa).
|
Ba) unless the cultures were supplemented with exogenous
IL-2 (Fig. 4Bc). By contrast, when the beads contained
anti-CD28 mAb, strong CTL activity was generated in the absence of
added IL-2 (Fig. 4Bb). Similar findings applied with
Ld.mRBC (Fig. 4Bd). Surface markers
Cell surface staining revealed that culturing 2C CD8+
cells with Ld.beads (6-µm Ld2+.beads) plus
QL9 peptide caused a slight increase in B7 expression and a modest
increase in ICAM-1 expression (Fig. 5A). This finding raised the
possibility that 2C CD8+ cells interacting with
Ld.beads plus peptide received bystander stimulation
through recognition of B7 or ICAM-1 on neighboring T cells. However,
adding the B7 antagonist CTLA4Ig or anti-ICAM-1 mAb to cultures of
2C CD8+ cells plus Ld.beads and QL9 peptide
failed to reduce CD69 up-regulation on 2C CD8+ cells and
caused only a modest reduction of the proliferative response (Fig. 5, B and C); adding a mixture of these reagents was
no more effective than anti-ICAM-1 mAb alone (Fig. 5B).
As a control for these experiments, adding anti-IL-2 mAb to the
cultures reduced the proliferative response by about 60% (Fig. 5B), indicating that the minimal production of IL-2 elicited
by Ld.beads plus QL9 (see above) played a significant role
in the proliferative response.
|
A). Thus, as with
Ld.mRBC (Fig. 6B) and Ld-expressing
Drosophila cells (Fig. 6C), no obvious signs of
cell aggregation were apparent when 2C CD8+ cells were
cultured with Ld.beads plus QL9 peptide (Fig. 6A). By contrast, coupling of anti-CD28 mAb caused
conspicuous formation of large conjugates for both Ld.mRBC
and Ld.beads (Fig. 6, D and E); no
conjugates occurred in the presence of the control P1A peptide (Fig. 6F).
|
In previous studies up-regulation of the activation markers, CD69
and CD25, on 2C CD8+ cells in response to QL9 peptide
presented by Ld-expressing Drosophila cells
coexpressing ICAM-1 was markedly impaired by addition of CCD, an
inhibitor of actin filament polymerization; by contrast, CCD failed to
impair up-regulation of CD69 and CD25 in response to
Ld-expressing Drosophila APC coexpressing
B7-1 (11). In view of these findings, the question arises of whether
CCD would impair the response of 2C CD8+ cells to
Ld.beads. As shown in Fig. 5, adding CCD to cultures of 2C
CD8+ cells plus Ld.beads and QL9 peptide was
highly effective in blocking both initial signaling of 2C
CD8+ cells, i.e., CD25 and CD69 up-regulation, and T cell
proliferation (Fig. 5, B–D). By contrast, CCD had
minimal effects with Ld.beads containing anti-CD28 mAb
(Fig. 5D).
Effects of APC fixation
The key finding in the above experiments is that 2C CD8+ cells responded strongly to Ld/QL9 complexes presented on latex beads (and also on magnetic beads) but were almost totally unresponsive to these complexes on mRBC or Drosophila cells. The possibility that certain cell surface components of mRBC and Drosophila cells are inhibitory for T cell activation is unlikely because coexpression of costimulatory molecules converted the cells to strong APC. Moreover, adding Ld.mRBC did not inhibit the response of 2C CD8+ cells to QL9 peptide presented by Ld.beads (data not shown).
In considering other explanations, it may be relevant that
Ld molecules coupled to beads are immobilized, whereas
Ld molecules on mRBC or Drosophila cells are
presumably subject to lateral diffusion in the cell membrane. Hence,
TCR signaling may be less intense when the ligand recognized is capable
of lateral movement rather than immobilized. If so, preventing such
movement, e.g., by cell fixation, would be expected to improve APC
function. To test this possibility, we examined the effects of
fixing mRBC with glutaraldehyde (1%) before Ld
coupling (Ld.F.mRBC); other workers have reported
that 1% glutaraldehyde is sufficient to impair lateral mobility of
proteins in the cell membrane (26, 27). As shown in Fig. 7, both Ld.mRBC and
Ld.F.mRBC plus QL9 peptide had similar levels of
Ld molecules (Fig. 7A) and induced a comparable
level of TCR down-regulation on 2C CD8+ cells (Fig. 7Ba). However, in marked contrast to Ld.mRBC,
Ld.F.mRBC plus QL9 peptide induced conspicuous CD69
expression (Fig. 7Bb) and proliferation (Fig. 7Ca). In fact, in terms of APC function,
Ld.F.mRBC closely resembled Ld.beads. Thus,
both types of APC induced a strong, but abbreviated, proliferative
response (Fig. 7, Ca and Cb) and very limited
IL-2 production (Fig. 7Cc).
|
Cd). As for Ld.beads and Ld.F.mRBC,
the proliferative response to Ld/QL9 complexes on plastic
peaked on day 2 and then fell abruptly. |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Since it is well established that chemical fixation of lymphoid APCs strongly inhibits APC function (5), the current finding that fixation of mRBC before Ld coupling markedly improved APC function might seem surprising. However, the present data have a precedent in the observations of Wade et al. on APC expressing class II molecules with a truncated cytoplasmic tail (28). The key finding with these APC was that the immunogenicity of class II molecules was inversely correlated with the lateral mobility of these molecules in the cell membrane. Thus, on normal APC, the truncated class II molecules were shown to have increased lateral mobility (relative to normal class II molecules) but were poorly immunogenic for T cells. After fixation, however, lateral mobility of class II molecules was decreased, whereas immunogenicity was increased. These data are thus consistent with the present finding that QL9/Ld complexes were far more immunogenic on fixed mRBC than on normal mRBC.
In speculating on why lateral mobility of peptide/class I complexes reduces immunogenicity, it should be stressed that presentation of QL9/Ld complexes by normal unfixed mRBC was highly effective at causing TCR down-regulation. Thus, lateral mobility of these complexes in the cell membrane seemed to be crucial only for TCR signaling and not for TCR endocytosis. In considering this paradox, the effects of CCD are of interest. With Ld.beads as APC, addition of CCD to culture had a minor effect on 2C proliferative responses to QL9 peptide when the beads coexpressed anti-CD28 mAb. In the absence of costimulation, however, CCD virtually abolished T cell proliferation but had little effect on TCR down-regulation. Since CCD inhibits actin filament polymerization (29), the implication is that with immobilized QL9/Ld complexes, TCR association with the cytoskeleton (30) is crucial for T cell activation but is not essential for initial TCR recognition, i.e., for the events leading to TCR endocytosis.
Based on these findings, contact with immobilized peptide/class I complexes may considerably augment TCR association with the cytoskeleton; this association may then serve to potentiate TCR-dependent signaling and thereby lead to overt T cell activation in the absence of costimulation. This model could explain our prior finding that the failure of 2C CD8+ cells to respond to QL9/Ld complexes on Drosophila APC could be overcome by cotransfecting these APC with ICAM-1 (10). As manifested by CD69 and CD25 up-regulation, T cell stimulation in this situation was blocked by CCD (11), implying that T cell interaction with ICAM-1 (presumably via LFA-1) enhanced TCR association with the cytoskeleton. By contrast, CCD failed to inhibit T cell activation when the APC expressed B7-1, i.e., a classic costimulatory molecule.
The capacity of immobilized peptide/MHC complexes to activate T cells in the absence of costimulation is of course a highly artificial situation. Could this finding have physiologic significance? On this point it is of interest that marked immobilization of class I molecules on human APC can be induced by coligating class I and LFA-3 molecules by mAbs (31). Since LFA-3 is the ligand for CD2, an adhesion molecule on T cells, a prime function of CD2/LFA-3 (and LFA-1/ICAM-1) interactions may be to promote immobilization of MHC/peptide complexes on APC, thereby intensifying signal 1. Hence immobilizing MHC/peptide complexes experimentally may simply bypass the natural function of adhesion molecules on APC. Our suggestion therefore is that the strong TCR signaling induced by artificially immobilized MHC/peptide complexes mimics the augmentation of TCR signaling induced by adhesion molecules on normal APC.
It should be emphasized that in the absence of costimulation, activation of T cells by immobilized peptide/MHC complexes was only seen under defined conditions. Thus, proliferative responses of 2C CD8+ cells to QL9/Ld complexes on beads were strongly dependent on both the size of the beads and the relative density of Ld molecules. Confirming prior studies of Mescher with primed T cells (3), larger (6 µm) beads were much more effective at stimulating naive 2C CD8+ cells than smaller (3 µm) beads; likewise, even with the larger beads, reducing the density of Ld molecules by threefold (from 50,000 molecules/bead) markedly reduced the proliferative response. The influence of bead size is surprising because Ld density on 3-µm beads was fivefold higher than that on 6-µm beads (1769 vs 355 Ld/µm2). The implication therefore is that even a slight reduction in the contact area on the APC substantially reduces the T proliferative response (32). It is worth noting that bead size and Ld density were much less critical when the beads expressed costimulatory molecules. Thus, even 3-µm beads with low Ld density became strongly immunogenic when the beads expressed costimulatory molecules. This finding exemplifies the crucial importance of costimulation (signal 2) when signaling via the TCR (signal 1) is limiting.
At face value, the strong costimulation-independent stimulation of 2C CD8+ cells induced by immobilized QL9/Ld complexes might suggest that signal 2 becomes largely redundant when signal 1 is intense. Here, two points should be made. First, T proliferative responses in the absence of costimulation required stringent conditions (see above) and were only seen with the high affinity QL9 peptide; responses with the weaker p2Ca peptide were barely detectable. Second, even with high concentrations of peptide, the T proliferative response to immobilized QL9/Ld complexes were of only brief duration; peak responses were seen on day 2 of culture. In addition, cytokine (IL-2) production was very limited, and CTL activity was undetectable unless the cultures were supplemented with exogenous cytokines. By contrast, providing the cells with costimulation led to a prolonged proliferative response, high IL-2 production, and strong CTL activity. These findings indicate that for generation of effector function, even intense TCR signaling cannot compensate for a lack of costimulation.
In conclusion, the data in this paper document two main points. First, peptide/class I complexes are essentially nonimmunogenic when expressed on a normal cell membrane but are capable of inducing an intense, but brief, proliferative response when lateral mobility of these complexes in the membrane is inhibited. Second, despite inducing proliferation, strong signaling via the TCR fails to induce cytokine synthesis or T effector function unless accompanied by costimulation. Thus, even under extreme conditions, productive activation of T cells requires a combination of signal 1 and signal 2.
Note added in proof. In agreement with our data, Goldstein et al. (1998. Purified MHC class I and peptide complexes activate naive CD8+ cells independently of the CD28/B7 and LFA-1/ICAM-1 costimulatory interactions. J. Immunol. 160:3180) have recently reported that 2C cells give high proliferative responses to immobilized Ld/peptide complexes. In contrast to our findings, however, Goldstein et al. observed high concentrations of IL-2 in the cultures and prolonged proliferative responses. This difference may reflect that the synthetic peptide studies (QY5) has even higher affinity for the 2C TCR than QL9.
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. J. Sprent, Department of Immunology, The Scripps Research Institute, 10550 N. Torrey Pine Rd., La Jolla, CA 92037. E-mail address:
3 Address correspondence and reprint requests to Dr. Z. Cai, The R. W. Johnson Pharmaceutical Research Institute, 3535 General Atomic Ct., #100, San Diego, CA 92121. E-mail address:
4 Abbreviations used in this paper: mRBC, mouse erythrocytes; MFI, mean fluorescence intensity; CCD, cytochalasin D.
Received for publication April 15, 1998. Accepted for publication July 10, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1 and 2 domains during de novo synthesis. J. Immunol. 140:3522.[Abstract]
chain of MHC class II molecules results in inefficient antigen presentation to antigen-specific T cells. Int. Immunol. 6:1457.
chain associates with the actin cytoskeleton upon activation of mature T lymphocytes. Immunity 3:623.[Medline]
This article has been cited by other articles:
|
|
 |
|
  S. Walter, L. Herrgen, O. Schoor, G. Jung, D. Wernet, H.-J. Buhring, H.-G. Rammensee, and S. Stevanovic Cutting Edge: Predetermined Avidity of Human CD8 T Cells Expanded on Calibrated MHC/Anti-CD28-Coated Microspheres J. Immunol., November 15, 2003; 171(10): 4974 - 4978. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. H. Schafer, A. K. Gandhi, M. A. Loveland, R. S. Chen, H.-W. Man, P. P. M. Schnetkamp, G. Wolbring, S. Govinda, L. G. Corral, F. Payvandi, et al. Enhancement of Cytokine Production and AP-1 Transcriptional Activity in T Cells by Thalidomide-Related Immunomodulatory Drugs J. Pharmacol. Exp. Ther., June 1, 2003; 305(3): 1222 - 1232. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. Hwang, X. Shen, and J. Sprent Direct stimulation of naive T cells by membrane vesicles from antigen-presenting cells: Distinct roles for CD54 and B7 molecules PNAS, May 27, 2003; 100(11): 6670 - 6675. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Zhai, L. Meng, F. Gao, R. W. Busuttil, and J. W. Kupiec-Weglinski Allograft Rejection by Primed/Memory CD8+ T Cells Is CD154 Blockade Resistant: Therapeutic Implications for Sensitized Transplant Recipients J. Immunol., October 15, 2002; 169(8): 4667 - 4673. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. D. Brown and D. Lillicrap Dangerous liaisons: the role of "danger" signals in the immune response to gene therapy Blood, July 30, 2002; 100(4): 1133 - 1140. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H.-W. Mittrucker, M. Kursar, A. Kohler, R. Hurwitz, and S. H. E. Kaufmann Role of CD28 for the Generation and Expansion of Antigen-Specific CD8+ T Lymphocytes During Infection with Listeria monocytogenes J. Immunol., November 15, 2001; 167(10): 5620 - 5627. [Abstract] [Full Text] [PDF]
|
|
