| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Program in Molecular Cardiobiology, Boyer Center for Molecular Medicine, Yale University School of Medicine, New Haven, CT 06510
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
secretion, and
perforin expression. The input cell number affected specificity and
potency of the resulting CTL. At low input (<105
cells/well), anti-EC CTL were rarely detected. At high input
(>106 cells/well), anti-EC CTL developed that
displayed unrestricted, low-titer killing and an unstable phenotype. At
intermediate input (1.0–2.5 x 105 cells/well),
classical class I MHC-restricted, CD8+, and
perforin-positive anti-EC CTL developed with reproducible
frequencies. However, under all conditions EC were less efficient
stimulators than BLC from the same donor. Anti-EC CTL did not kill BLC,
whereas anti-BLC CTL killed BLC and EC from the same donor with
comparable efficiency. When CD8+ T lymphocytes were grown
in the presence of EC and BLC together, the differentiation of
anti-BLC CTL was completely suppressed, while the anti-EC
response was intact. The inhibition of the allogeneic anti-BLC CTL
response was independent of T cell-EC contact, and proliferation of
CD8+ T cells was inhibited by EC-conditioned medium. We
conclude that EC are competent but less efficient activators of CTL
differentiation than are BLC and that EC actively regulate
differentiation and/or expansion of allospecific
CTL. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
It has been proposed that human endothelial cells (EC) may substitute for "passenger leukocytes" as initiators of the host anti-graft response (14, 15). Human EC are competent in culture to present allogeneic class I MHC molecules directly to resting CD8+ memory T cells, leading to cytokine secretion and proliferation (16, 17). Recently, CTL have been recovered from rejecting allografts that appear to recognize EC specifically, i.e., they can lyse donor EC but not donor leukocytes (18, 19). Because each mature CTL must recognize the same Ag as its preCTL progenitor, this finding implies that donor EC stimulated the differentiation of such CTL from precursors. However, these data do not rule out the possibility that professional APC may also have played a role (e.g., as "trans-costimulators" (20)).
The simplest way to test the sufficiency of human EC to promote CTL differentiation is to coculture EC with preCTL. To date, such experiments have given conflicting results. Clayberger et al. (21), using human foreskin microvascular EC as stimulators, found that PBMC differentiated into nonspecific (i.e., nonallorestricted) CD8+ killers, whereas BLC under the same conditions stimulated the development of class I MHC-restricted CTL. In contrast, Pardi and Bender (22) observed that human foreskin microvascular EC were able to stimulate allospecific CTL differentiation from purified CD8+ T cell precursors, but no comparison to BLC or other professional APC were made. The present study was undertaken to assess the relative capacity of cultured EC to promote allogeneic CD8+ T cell differentiation in comparison to BLC derived from the same donor as the EC. In addition, we wished to test the hypothesis that stimulation of CD8+ T cells by allogeneic EC would lead to the emergence of CTL reactive with EC but not BLC targets.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Recombinant human IL-2, IFN-, and IL-12 were purchased from
R&D Systems (Minneapolis, MN). Mitomycin C, indomethacin,
N
-nitro-L-arginine methyl
ester (L-NAME) and
N-nitro-D-arginine methyl ester
(D-NAME) were obtained from Sigma (St. Louis, MO).
Neutralizing Abs reactive with TGF-ß (used at a final concentration
of 5 µg/ml) were obtained from R&D Systems. OKT8 hybridoma cells were
obtained from American Type Culture Collection (Manassas, VA), and mAbs
(IgG2a) were affinity purified from crude ascites using HiTrap Protein
G according to the manufacturer’s instructions (Pharmacia Biotech,
Piscataway, NJ). K1616, a nonbinding control Ig (IgG1), was purified in
the same way.
Cell isolation
Human EC were isolated from individual umbilical veins as previously described (16) and cultured at 37°C and 5% CO2-humidified air in Medium 199 supplemented with 2 mM L-glutamine, 100 U/ml penicillin, 100 µg/ml streptomycin (all from Life Technologies, Grand Island, NY), 50 µg/ml endothelial cell growth supplement (Collaborative Research/Becton Dickinson, Bedford, MA), and 100 µg/ml heparin (Sigma). In the experiments performed in this study, EC were used at passage levels 3 to 9. Such EC cultures have been found to be devoid of CD45+-contaminating leukocytes, yet retain uniform von Willebrand factor positivity.
BLC were prepared from cord blood harvested from the same individual as the EC. Cord blood mononuclear cells were isolated by density gradient centrifugation using lymphocyte separation medium (LSM, Organon Teknika, Durham, NC). A portion of these cells were used for serologic tissue typing (courtesy of Dr. M. I. Lorber, Yale Tissue Typing Laboratory, New Haven, CT), and the remainder were used for the generation of BLC by EBV transformation with the EBV secreting cell line 95.8 (a generous gift from Dr. G. Miller, Yale Medical School, New Haven, CT) as described (23). BLC lines were cultured at 37°C and 5% CO2-humidified air in RPMI 1640 (Life Technologies) in 10% FCS with 2 mM L-glutamine, 100 U/ml penicillin, and 100 µg/ml streptomycin.
CD8+ T cells were purified from PBMC isolated from healthy volunteer donors by leukapheresis and density gradient centrifugation. A portion of the cells were tissue typed serologicly for HLA class I molecule expression (courtesy of the Yale Tissue Typing Laboratory) and the remainder, about 2 to 3 x 109 PBMC per donor, were suspended in ice cold FCS/10% DMSO, aliquoted, and stored in liquid nitrogen until use. Thawed PBMC were washed in RPMI 1640/5% FCS, and CD8+ lymphocytes were positively selected using anti-CD8 Ab-coated magnetic beads (Dynabeads M450-CD8, Dynal, Lake Success, NY) according to the manufacturer’s instructions. Briefly, PBMC were suspended at 2 x 107 cells/ml in the same medium and incubated with 1.2 x 107/ml Dynabeads for 30 min at 4°C. Beads with attached CD8+ cells were washed five to six times until the supernatant was clear. The Detachabead solution (Dynal) was added to the suspension, which was then incubated for 45 to 60 min at room temperature. The recovery was 70 to 75% of the total input CD8+ T lymphocytes with >96% CD8+ expression determined by FACS analysis (see below) and >99% viability assessed by trypan blue exclusion. No discrete contaminating populations of CD4+, CD19+, HLA-DR+, or CD16+ cells were detected. In pilot experiments, CD8+ T cells purified by negative selection (for HLA-DR, CD16, and CD4) were never >85 to 90% CD8+. However, no significant differences were noted between CD8+ CTL generated from positively or negatively selected cells. Because of the higher degree of purity, all of the experiments reported in this paper used positively selected populations. For FACS analysis, cells were washed in PBS/1% BSA and incubated with directly conjugated mAbs (isotype matched, FITC- or phycoerythrin-conjugated control mAbs, or mouse anti-human HLA-DR, HLA class I (W6/32), CD45, CD25, CD16, CD19, CD14, CD4, CD8, and CD3 Abs, all from Immunotech, Westbrook, ME) according to the manufacturer’s instructions. Stained cells were washed three times with PBS/1% BSA, and then fixed in 2% paraformaldehyde in PBS and analyzed within 24 h using a FACSort (Becton Dickinson, San Jose, CA). All staining, washing, and incubation steps were conducted on ice.
CTL differentiation
Both EC and BLC were used as stimulator cells for CTL differentiation from CD8+ preCTL. EC were suspended from confluent cultures using trypsin-EDTA and were seeded at confluent density into 96-well round-bottom microtiter plates (for limiting dilution analysis and microcultures), 24-well plates (for bulk cultures), or 6-well plates for Transwell experiments (all plasticware from Falcon, Becton Dickinson, Bedford, MA) 1 to 2 days before the cocultures were started. Where indicated, EC were fixed with 1% paraformaldehyde for 10 min at room temperature, extensively washed, and rested for 10 min before cocultures were initiated. BLC were washed in HBSS, and mitomycin C was added at 50 µg/ml final concentration. Cells were incubated at 37°C for 20 min and washed three times with RPMI 1640/5% FCS. After resting for 1 h at 37°C, cells were again washed three times and then added to wells at numbers comparable to confluent EC.
Cocultures were initiated by addition of CD8+ T cells from donors allogeneic to the EC or BLC. Purified CD8+ T cells were suspended in RPMI 1640, 10% human type AB serum (Irvine Scientific, Santa Ana, CA), 2 mM L-glutamine, 100 U/ml penicillin, and 100 µg/ml streptomycin and added to confluent EC monolayers or BLC at the indicated input cell number. For limiting dilution analysis, CD8+ input cell numbers were titered over a broad range (250,000 to 2,500 cells/well) to observe the different response patterns against EC and BLC, respectively. Microcultures were typically initiated with 36 replicates (2 times 18 replicates for testing against stimulator vs third party cell line or EC vs BLC) per test group. When the allospecificity of CTL was tested, 200,000 CD8+ T cells/well were used with EC stimulators and 50,000 CD8+ T cells/well were used with BLC stimulators. When EC and BLC were compared for the efficiency with which they stimulate CTL differentiation, cocultures were initiated with the same CD8+ T cell input number (100,000 cells/well) for both cell types. Bulk cultures were initiated in replicate cultures in 24-well plates at input cell numbers of 1 to 4 x 106 CD8+ T cells/well. Transwell cultures were performed using 6-well culture inserts (PET, pore size 0.4 µm, Falcon). CD8+ T cells (2 x 106) were added together with BLC to the upper compartment of the culture insert and confluent living or fixed EC were grown in the bottom of the 6-well plate. Where indicated, pharmacologic inhibitors of cyclooxygenase (5 µM indomethacin) or nitric oxide synthase (5 mM L-NAME) were included in the cultures.
In all cases, cocultures were fed with fresh medium containing IL-2 at a final concentration of 5 ng/ml after 3 days. After 7 days, 150 µl of supernatant per well was removed from the microcultures, pooled, and frozen at -70°C for cytokine measurements. Fresh medium containing IL-2 (100 µl/well) was added, and the CD8+ T cells were transferred to fresh EC or BLC. The bulk cultures were suspended and collected into tubes, spun down, resuspended in fresh medium containing IL-2 and transferred to fresh EC or BLC. The cultures were fed again after 10 days in medium supplemented with IL-2. Where indicated, fresh medium was replaced by a 1:1 mixture of fresh and conditioned medium. After 2 wk, supernatant was harvested, pooled, and frozen for cytokine measurements. The medium was replaced by Medium 199, 5% FCS, 5 mM HEPES, 2 mM L-glutamine, 100 U/ml penicillin, and 100 µg/ml streptomycin, and cells were used for cytotoxicity assays.
In selected experiments, proliferation of anti-BLC CTL was measured as previously described (16). In brief, 105 CD8+ T cells/well were incubated with mitomycin C-treated BLC in quadruplicate. After 78 h, 1 µCi/well [3H]thymidine (NEN Life Science Products, Boston, MA) in complete medium was added. Cells were harvested after an additional 18 h using a 96-well harvester (Tomtec, Orange, CT) and counted on a Microbeta scintillation counter (Wallac, Gaithersburg, MD). The specific cpm were corrected for background (mitomycin C-treated BLC), typically <200 cpm.
Cytotoxicity assay
Cytotoxicity was measured by a calcein fluorescence release assay performed according to published methods (24). In brief, target EC were grown in flat-bottom 96-well plates to confluence (2 x 104 cells/well). Cells were incubated with 20 µM calcein-AM (Molecular Probes, Eugene, OR) in Medium 199 and 5 mM HEPES for 30 min at 37°C. The medium was replaced by complete EC growth medium, and cells were rested overnight. After 8 to 12 h, cells were washed twice with Medium 199, 5% FCS, 5 mM HEPES, 2 mM L-glutamine, 100 U/ml penicillin, and 100 µg/ml streptomycin. Alternatively, target BLC were harvested from suspension culture, washed once with Medium 199 and 5 mM HEPES, and incubated with 20 µM calcein-AM for 30 min at 37°C. The cells were washed three times, counted, and added to the microculture wells at numbers comparable to confluent EC.
Effector cells from microcultures were suspended and 150 µl/well transferred onto calcein-loaded target cells. The effector cell number (mean of cell counts from three randomly chosen replicate wells) was counted with a hemocytometer and the average E:T ratio calculated. Effector cells from bulk cultures were washed once, counted, and added at titered E:T ratios in duplicates. Twelve replicate wells/plate were incubated with lysis buffer (50 mM sodium borate, 0.1% Triton X-100, pH 9.0, maximal release). After 4-h incubation at 37°C, supernatant (75 µl/well) was carefully removed and transferred into a flat-bottom 96-well plate. Released calcein was measured using a fluorescence multiwell plate reader (Cytofluor 2, Perseptive Biosystems, Framingham, MA; excitation wavelength 485 nm, emission wavelength 530 nm). Percent specific killing was calculated as [(release sample - spontaneous release)/(maximal release - spontaneous release)] x 100%. Spontaneous release was generally <25%.
IFN- ELISA
IFN- was measured in tissue culture supernatants with a
sandwich ELISA using a commercially available Ab pair (monoclonal mouse
anti-human IFN-, MAB285, and biotinylated polyclonal goat
anti-human IFN-, BAF285, both from R&D Systems) according to the
manufacturer’s instructions. IFN- was measured either in the pooled
and frozen coculture supernatants after 7 or 14 days, or in the
individual microculture supernatants from cytotoxicity assays. In the
latter case, the medium was replaced after harvesting the supernatant
from the CTL assay to measure calcein release, and the cultures further
incubated overnight at 37°C. Supernatant (100 µl) from six
replicate microculture wells were then transferred directly to the
precoated ELISA plate.
Perforin staining
Fresh CD8+ T cells or CTL from cocultures were fixed with freshly prepared, ice cold 4% paraformaldehyde for 10 min and spun onto gelatin-coated glass slides (Cytospin 2, Shandon, Pittsburgh, PA). After washing the slides in PBS, cells were permeabilized and blocked with 0.1% saponin (Sigma) in PBS supplemented with 1% BSA, 5% normal human serum, and 5% normal goat serum for 10 min at room temperature. Monoclonal mouse anti-human perforin (1/250, T-Cell Diagnostics, Cambridge, MA) and rat anti-human CD8 Abs (1/250, MCA351, Serotec, Washington, DC) were added, and cells were incubated for 60 min at room temperature. An isotype-matched (IgG2b) nonbinding mouse mAb (Cappel, Organon Teknika) was used as control. After the slides were washed for 10 min with PBS/0.1% saponin, polyclonal Cy2-conjugated goat anti-mouse and Texas red-conjugated goat anti-rat Abs (both from Jackson ImmunoResearch Laboratories Inc, West Grove, PA) were added for 30 min. After the slides were washed for 10 min, they were air dried and embedded in aqueous mounting medium. Cells were examined and photographed with a fluorescence microscope (Microphot FXA, Nikon, Tokyo, Japan). For optimal staining of perforin granules, it appeared critical to stain cells immediately after fixation.
Statistical analysis
Data analysis of limiting dilution experiments was performed according to likelihood maximization using a computer program kindly provided by Dr. C. Orosz (Ohio State University, Columbus, OH). Results from different groups were compared with ANOVA or with the Wilcoxon test.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We compared the capacity of cultured EC and BLC, derived from the same donor, to stimulate the differentiation of allogeneic peripheral blood CD8+ T cells into CTL. Preliminary experiments confirmed the observation of Pardi and Bender (22) that the addition of small amounts of exogenous IL-2 was necessary to develop CTL activity in vitro. Subsequently, IL-2 (5 ng/ml) was introduced to all cocultures at day 3 and replenished at restimulation (day 7) and at subsequent refeeding (day 10).
At limiting dilutions of purified CD8+ T cells/well (Fig. 1
A), the frequency of a
positive response against allogeneic BLC stimulators conformed to
"single hit" and suggested an allogeneic precursor CTL frequency
ranging from 1/8,000 to 1/80,000 in different experiments. In contrast,
the frequency of a positive response against allogeneic EC stimulators
was very low and generally did not fulfill the criteria for "single
hit" interactions. In the single instance when they did, limiting
dilution analysis suggested very low precursor frequencies of
<1/500,000.
|
B). However, at
intermediate input cell numbers (e.g., at 2 x 105
CD8+ T cells/well) both anti-BLC and anti-EC CTL
could be reproducibly generated. Such anti-EC CTL displayed
allospecificity, i.e., anti-EC CTL grown and tested under these
conditions killed the stimulator with much higher frequency than they
killed an unrelated third party EC line (Fig. 2A). This degree of
allospecificity in such anti-EC cultures was comparable to that
measured with anti-BLC CTL (Fig. 2B).
|
A).
Among different experiments, the same CD8+ T cell
responder/EC stimulator combination led to results that were highly
variable and unpredictable when assayed at 2 wk. Longer-term bulk
cultures (i.e., greater than 2 wk) of CD8+ T cells
stimulated by allogeneic EC frequently stopped growing or lost
cytotoxic activity, or both. Furthermore, the cytotoxicity of
anti-EC bulk cultures did not appropriately titer with varying
effector number and sometimes actually decreased at increasing E/T
ratios. The inclusion of mAbs to CD8, which strongly inhibits killing
by long-term anti-BLC CTL lines (data not shown), actually enhanced
killing by anti-EC CTL and revealed titerable effector cells (Fig. 4A). BLC, on the other hand,
routinely stimulated titerable allospecific CTL responses under the
identical bulk culture conditions (Fig. 3B). Such
CD8+ CTL lines could be stably propagated for several
weeks, often increasing in potency (data not shown).
|
|
B). In contrast, anti-BLC
CTL from pooled microcultures continued to kill significantly,
consistent with the behavior of bulk cultures of anti-BLC CTL. This
result suggests that EC but not BLC may stimulate a CD8+
"suppressor" population, although we have not shown that these
regulatory T cells are alloantigen or Id specific. EC stimulate conventional, class I MHC-restricted CTL differentiation
In the experiments described above, we identified a set of
conditions involving an intermediate number of input preCTL, which led
to the differentiation of allospecific anti-EC CTL responses. We
proceeded to characterize these effector cells more fully. To determine
whether these allospecific cultures displayed class I MHC restriction,
we compared killing of third party EC lines sharing
HLA-A or -B alleles with the
stimulator cell line with third party cells sharing no
HLA-A or -B alleles. As may be seen in
Figure 5, cross-reactivity was detectable
only in cultures that share class I alleles, indicative of MHC
restriction.
|
, B and D and
7C). Perforin appeared to be
relevant for CTL functions because cytotoxicity against the stimulator
cell line was never observed in cultures that lacked perforin
expression.
|
secretion
Although EC could stimulate allospecific CTL differentiation, they
appeared to be quantitatively less efficient stimulator cells than BLC
from the same individual donor (Fig. 1B). We further
analyzed differences between the capacities of EC and BLC by assessing
several other parameters of the CD8+ T cell response (Fig. 7). CD8+ T cell growth was consistently less when
stimulated with allogeneic EC than with BLC (Fig. 7A).
Growth in both cases was absolutely dependent upon exogenous IL-2.
However, IL-2 alone in the absence of BLC or EC was unable to support
CD8+ T cell proliferation or even survival, because the T
cell number fell precipitously by 2 wk to 4 ± 2% of input cells.
Furthermore, growth differences between cultures stimulated by BLC or
by EC did not correlate with the fraction of IL-2R
(CD25) expressing
cells, which were 21 ± 6% and 19 ± 8%, respectively.
|
secretion was assayed. In the 7 day coculture supernatant, IFN- was
not detectable in cocultures with EC but reproducibly reached over 100
pg/ml when BLC were used as stimulators (Fig. 7B). At this
point, cell numbers in the microcultures grown against EC and against
BLC were still comparable, and IFN- levels were not adjusted to CTL
number.
We attempted to improve the efficiency with which EC stimulate CTL
differentiation and IFN- production by various manipulations. First,
BLC are known to produce IL-12, a 70-kDa heterodimeric glycoprotein
originally identified as differentiation factor for CTL and NK cells
that increases IFN- secretion (25, 26). Therefore, we added rIL-12
to the culture medium to determine whether this manipulation would
promote anti-EC CTL differentiation. As shown in Figure 8, IL-12 actually reduced the frequency
of CTL development, although IL-12, as expected, increased IFN-
secretion in the same cultures (data not shown). Variations in the
timing or concentration of IL-12 addition did not improve the efficacy
of this cytokine in our system. Second, BLC are also known to express
high levels of class I MHC molecules, which might also influence the
outcome of the cocultures. Therefore, we pretreated the EC for 24
h with IFN- (27) before the CD8+ T cells were added, a
manipulation that increased class I MHC expression six- to sevenfold
(data not shown). However, this approach also failed to increase the
frequency of CTL differentiation by EC (Fig. 8).
|
The observation that EC are less efficient than BLC as stimulators
of allospecific CTL responses could either arise from a deficiency of
activating signals or from provision of inhibitory signals or both. To
address this question, we performed a mixing experiment by culturing
CD8+ T cells with EC + BLC and compared the results to
cultures stimulated with EC or BLC alone. Cell growth, IFN-
secretion during the culture period, and percent perforin-positive
cells was similar for CD8+ T cells cocultured with EC or EC
+ BLC, and all of these parameters were dramatically less than in
cultures of CD8+ T cells cocultured with BLC (Table I). Consistent with our previous results,
EC were less efficient than BLC as stimulators of CTL in these
experiments (compare Fig. 9, A
(dark bars) and B (open bars), respectively). The frequency
of microcultures with CTL activity stimulated by EC + BLC was similar
to that using EC stimulators alone. While anti-BLC CTL generally
lysed both BLC and EC, anti-EC CTL appeared specific for EC and
often did not kill BLC from the same donor. CTL from the mixed
cocultures, grown in the presence of EC and BLC, were also able to lyse
EC and failed to kill BLC (Fig. 9C). In other words, in the
mixed cocultures EC appear to inhibit the anti-BLC CTL response
while simultaneously stimulating an anti-EC-specific response.
|
|
secretion during the effector phase, i.e., in
the supernatants of the CTL assay, using CTL generated by EC
stimulators, by BLC stimulators, or by EC + BLC stimulators. Anti-BLC
CTL produced significant amounts of IFN- when cultured overnight
with BLC (Fig. 9E, open bar). CTL grown in the presence of
EC or EC + BLC produced less IFN- when cultured overnight with BLC
(Fig. 9, D and F, open bars), but this difference
may reflect the reduced number of effector cells generated in
cultures containing EC. The most consistent observation was that
CTL produced less IFN- when cultured overnight with EC compared
with culture with BLC, irrespective of the stimulator cell type (EC,
BLC, or EC + BLC) used to generate the CTL (Fig. 9, D–F,
dark bars). To determine whether the failure to detect anti-BLC CTL in mixed cocultures resulted from the generation of suppressor cells by EC + BLC stimulators, we mixed CD8+ CTL stimulated by EC + BLC with those stimulated by BLC alone in the effector phase. In this case, anti-EC CTL did not inhibit the cytotoxic activity of anti-BLC CTL when they were pooled and tested on BLC targets (data not shown). This experiment suggests that EC-mediated inhibition of the anti-BLC response appears to occur during the differentiation of CTL from preCTL and not during the CTL assay.
Analysis of the inhibitory effect of EC upon generation of anti-BLC CTLs
In a final series of experiments, we analyzed the basis of the
inhibitory effects mediated by EC on anti-BLC CTL differentiation.
To determine whether cell contact was required for EC to inhibit the
development of anti-BLC CTL we used a Transwell system. When
anti-BLC CTL were grown in the upper compartment of a Transwell,
the presence of EC in the lower compartment strongly reduced CTL growth
(Fig. 10A). The resulting
anti-BLC CTL expressed little perforin (Fig. 6F), were
essentially unable to kill BLC targets (Fig. 10B), and
produced lesser amounts of IFN- during the effector phase (Fig. 10C) compared with control anti-BLC CTL generated in the
absence of EC. Fixing the EC in the lower compartment with 1%
paraformaldehyde before the initiation of the Transwell coculture
markedly reduced the extent of growth inhibition, and the resulting
anti-BLC CTL secreted more IFN- during the effector phase and
were as effective killers as the control CTL grown in the absence of EC
(data not shown). The addition of fixed EC into mixed cocultures with
CD8+ T cells and BLC also failed to inhibit CTL
differentiation (data not shown). In other words, metabolically active
EC are required to inhibit anti-BLC CTL differentiation and growth,
and this inhibition is independent of EC-T cell contact.
|
A) as well as
function, we attempted to demonstrate the presence of a soluble
inhibitor in a simpler, short-term assay. We found that
CD8+ T cell proliferation to BLC was detectable on day 3
and peaked at day 4 of culture. Three-day EC-conditioned medium was
tested for its effects on CD8+ T cell proliferation to BLC.
As shown in Figure 11B, such EC-conditioned medium inhibited
allostimulated CD8+ T cell proliferation by 30% when added
to CD8+ T cell/BLC cocultures. In contrast, conditioned
medium from BLC slightly enhanced T cell proliferation under the same
conditions. This degree of growth inhibition, if sustained, could
explain the profound inhibitory effects on growth found in the
Transwell experiments (Fig. 11A), where a 10-fold difference
in cell numbers was observed after 2 wk. The inhibitory activity
produced by EC does not appear to be TGF-ß-mediated because
neutralizing Abs reacting with this molecule did not prevent
EC-mediated inhibition in coculture experiments with mixed stimulators
(EC + BLC), nor did it boost the anti-EC CTL response (data not
shown). We also ruled out NO and prostaglandin as potential candidates
for EC inhibitory activity because neither L-NAME (5 mM)
nor indomethacin (5 µM) were able to reduce the level of inhibition
of anti-BLC differentiation produced by EC in a Transwell system.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The second major conclusion from our study is that EC are inefficient
stimulators of CTL compared with BLC. This conclusion was suggested by
the limiting dilution experiments and from the quantitative comparison
of positive microcultures at low input cell numbers. In general, EC are
less efficient than BLC at promoting CD8+ T cell growth,
IFN- secretion, and acquisition of cytolytic functions, but equally
effective at inducing IL-2R (CD25) expression (16). The failure to
observe "single hit" responses at low input cell number could imply
either that additional stimulator signals were needed (e.g., because of
lack of costimulators) or that responder cells needed to interact
(e.g., by provision of stimulatory cytokines) (30). Four manipulations,
namely addition of exogenous IL-12, up-regulation of class I MHC
molecule expression, neutralization of TGF-ß, or addition of BLC, did
not appear to help. However, in a pilot experiment, inclusion of
irradiated PBMC feeder cells from the responder did allow input
CD8+ T cell differentiation to conform to a "single
hit" response, supporting the notion of missing or inadequate
cytokines.
The third conclusion is that EC regulate CTL responses. This result comes both from the bulk culture experiments and from the mixed stimulator, the Transwell, and the conditioned medium experiments. The bulk culture experiments suggest that EC induce regulatory T cells that inhibit allospecific killing and/or promote allo-unrestricted killing. This observation may explain the previous results of Clayberger et al. (21). Further evidence for such regulatory T cells is provided by the observation that anti-CD8 mAbs increase the potency of anti-EC bulk cultures and that mixing of anti-EC CTL microcultures does not result in high titer allospecific killing. Such regulatory T cells may be "suppressor" cells that act in the effector phase, but we have no evidence that they are alloantigen or Id specific. The observation that anti-CD8 mAbs increase killing also implies that it does not block killing by anti-EC CTL.
The mixed stimulator, conditioned media, and Transwell experiments all point to the fact that EC inhibit the ability of BLC to stimulate differentiation of preCTL into CTL by elaboration of a stable secretory activity present in EC-conditioned medium. We do not know if this activity is a single factor or even if it is a protein, but Ab neutralization suggests that it is not TGF-ß1, a well-characterized immunosuppressant made by cultured EC (31). EC-conditioned medium is much less efficient than EC in Transwell cultures, suggesting that the inhibitory activity has a relatively short half-life. We ruled out both NO and prostaglandins as candidate molecules for this inhibitor. Anti-EC CTL emerge in settings where EC completely suppress the anti-BLC response, e.g., in the mixed-stimulator experiments. This observation further indicates that anti-EC CTL may differ from anti-BLC CTL by more than simply Ag specificity, i.e., EC-specific CTL may not be inhibited by the EC-secreted activity. Alternatively, it may suggest that EC provide costimulation that overcomes the inhibitory effect of this suppressor molecule but are unable to do so in "trans", i.e., when BLC are stimulating the TCRs of the responding CD8+ T cells.
Our findings have two potentially important implications for acute allograft rejection. First, they suggest that the generation of anti-EC CTL may differ from the generation of CTL reactive with passenger leukocytes. If such anti-EC CTL mediate acute vascular rejection in vivo, a lesion that often portends a therapy-resistant form of graft rejection (29) remains to be shown. If the development of anti-EC CTL is, in fact, the harbinger of therapy-resistant rejection, then inhibition of this response may require alternative therapies, potentially testable in our in vitro model. Second, the capacity of EC to suppress conventional CTL differentiation may point to a natural EC-derived immunosuppressant that can be added to the armamentarium of antirejection therapy.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Jordan S. Pober, Boyer Center for Molecular Medicine, 295 Congress Ave., New Haven, CT 06510. E-mail address:
3 Abbreviations used in this paper: preCTL, precursor CTL(s); EC, endothelial cell(s); BLC, B lymphoblastoid cell(s); L-NAME, N-nitro-L-arginine methyl ester.
Received for publication March 10, 1998. Accepted for publication June 30, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. A. Hunt and F. Haddad The Changing Face of Heart Transplantation J. Am. Coll. Cardiol., August 19, 2008; 52(8): 587 - 598. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. N. Mitchell and P. Libby Vascular Remodeling in Transplant Vasculopathy Circ. Res., April 13, 2007; 100(7): 967 - 978. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Doebis, S. Schu, J. Ladhoff, A. Busch, F. Beyer, J. Reiser, R. F. Nicosia, S. Broesel, H.-D. Volk, and M. Seifert An anti-major histocompatibility complex class I intrabody protects endothelial cells from an attack by immune mediators Cardiovasc Res, November 1, 2006; 72(2): 331 - 338. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. L. Shiao, J. M. McNiff, and J. S. Pober Memory T Cells and Their Costimulators in Human Allograft Injury J. Immunol., October 15, 2005; 175(8): 4886 - 4896. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Kummer, A. Lev, Y. Reiter, and B. C. Biedermann Vascular Endothelial Cells Have Impaired Capacity to Present Immunodominant, Antigenic Peptides: A Mechanism of Cell Type-Specific Immune Escape J. Immunol., February 15, 2005; 174(4): 1947 - 1953. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Choi, J. Walker, S. Boichuk, N. Kirkiles-Smith, N. Torpey, J. S. Pober, and L. Alexander Human Endothelial Cells Enhance Human Immunodeficiency Virus Type 1 Replication in CD4+ T Cells in a Nef-Dependent Manner In Vitro and In Vivo J. Virol., January 1, 2005; 79(1): 264 - 276. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. F. Halloran Immunosuppressive Drugs for Kidney Transplantation N. Engl. J. Med., December 23, 2004; 351(26): 2715 - 2729. [Full Text] [PDF]
|
|
|
|
|
|
L. Zheng, T. F. Gibson, J. S. Schechner, J. S. Pober, and A. L. M. Bothwell Bcl-2 Transduction Protects Human Endothelial Cell Synthetic Microvessel Grafts from Allogeneic T Cells In Vivo J. Immunol., September 1, 2004; 173(5): 3020 - 3026. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Kreisel, A. S. Krupnick, K. R. Balsara, M. Riha, A. E. Gelman, S. H. Popma, W. Y. Szeto, L. A. Turka, and B. R. Rosengard Mouse Vascular Endothelium Activates CD8+ T Lymphocytes in a B7-Dependent Fashion J. Immunol., December 1, 2002; 169(11): 6154 - 6161. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. C. Biedermann Vascular Endothelium: Checkpoint for Inflammation and Immunity Physiology, April 1, 2001; 16(2): 84 - 88. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. J. Dengler, D. R. Johnson, and J. S. Pober Human Vascular Endothelial Cells Stimulate a Lower Frequency of Alloreactive CD8+ Pre-CTL and Induce Less Clonal Expansion than Matching B Lymphoblastoid Cells: Development of a Novel Limiting Dilution Analysis Method Based on CFSE Labeling of Lymphocytes J. Immunol., March 15, 2001; 166(6): 3846 - 3854. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Andriambeloson, M. Bigaud, E. O. Schraa, T. Kobel, V. Lobstein, C. Pally, and Hans-Gunter Zerwes Endothelial Dysfunction and Denudation in Rat Aortic Allografts Arterioscler Thromb Vasc Biol, January 1, 2001; 21(1): 67 - 73. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. M. Marelli-Berg, D. Scott, I. Bartok, E. Peek, J. Dyson, and R. I. Lechler Activated Murine Endothelial Cells Have Reduced Immunogenicity for CD8+ T Cells: A Mechanism of Immunoregulation? J. Immunol., October 15, 2000; 165(8): 4182 - 4189. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. J. Dengler and J. S. Pober Human Vascular Endothelial Cells Stimulate Memory But Not Naive CD8+ T Cells to Differentiate into CTL Retaining an Early Activation Phenotype J. Immunol., May 15, 2000; 164(10): 5146 - 5155. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Zheng, T. J. Dengler, M. S. Kluger, L. A. Madge, J. S. Schechner, S. E. Maher, J. S. Pober, and A. L. M. Bothwell Cytoprotection of Human Umbilical Vein Endothelial Cells Against Apoptosis and CTL-Mediated Lysis Provided by Caspase-Resistant Bcl-2 Without Alterations in Growth or Activation Responses J. Immunol., May 1, 2000; 164(9): 4665 - 4671. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Gatti, M. A. Velleca, B. C. Biedermann, W. Ma, J. Unternaehrer, M. W. Ebersold, R. Medzhitov, J. S. Pober, and I. Mellman Large-Scale Culture and Selective Maturation of Human Langerhans Cells from Granulocyte Colony-Stimulating Factor-Mobilized CD34+ Progenitors J. Immunol., April 1, 2000; 164(7): 3600 - 3607. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Mahboubi, B. C. Biedermann, J. M. Carroll, and J. S. Pober IL-11 Activates Human Endothelial Cells to Resist Immune-Mediated Injury J. Immunol., April 1, 2000; 164(7): 3837 - 3846. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. C. Biedermann and J. S. Pober Human Vascular Endothelial Cells Favor Clonal Expansion of Unusual Alloreactive CTL J. Immunol., June 15, 1999; 162(12): 7022 - 7030. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
