| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Committee on Immunology, and Departments of
Ophthalmology and Visual Sciences and
Pathology, University of Chicago, Chicago, IL 60637
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
m) across the inner membrane of
the mitochondria is one early indicator of apoptosis that can occur in
many death pathways, including the FasL/Fas and ROS pathways
(7, 8, 9, 10). Caspases, a family of cysteine proteases
(11), are downstream effectors of both the FasL/Fas and
ROS pathways that promote DNA fragmentation. In the Fas/FasL death
pathway, caspases are activated either directly or indirectly following
a loss of mitochondrial integrity. The proteolytic activity of caspases
is tightly regulated. Effector caspases exist as inactive procaspases
that require specific cleavage by upstream initiator caspases to
initiate processing that will result in catalytic activity
(12). The effectiveness of the CD28/B7 pathway in
costimulation has been attributed, in part, to the induction of the
anti-apoptotic Bcl-xL protein, which is
involved in maintaining m (13, 14).
Biologically active IL-12 (p35/p40) has a critical role in promoting
the differentiation of naive Th cells into Th1 cells (15).
In addition, IL-12 is known to have a strong adjuvant effect in
promoting T cell-mediated immune responses (16). This
adjuvant effect is likely due, in part, to the capacity of IL-12 to
enhance IL-2 secretion and responsiveness (17, 18), which
can augment proliferation, and/or to induce IFN-
secretion, which
can up-regulate expression of APC costimulatory molecules, such as B7
family members (19) and ICAM-1 (20). Using an
in vitro model of Th cell costimulation by accessory cells, we have
shown that IL-12 can promote ICAM-1-mediated, CD28-independent
costimulation of human naive Th cells by preventing cell death and
enhancing cell expansion (E. Palmer and G. van Seventer, unpublished
observations, and Ref. 21). Similarly, a regulatory role
for IL-12 in preventing apoptosis has also been reported by other
investigators studying different models, including HIV-induced Th cell
apoptosis (22), Fas-mediated apoptosis in murine and human
Th cells (22, 23), and -irradiation-induced apoptosis
in monocytes (24). However, the mechanism by which IL-12
can prevent apoptosis in all these models remains unknown.
The goal of this study was to characterize the mechanism(s) by which
IL-12 prevents AICD in human naive Th cells costimulated by ICAM-1 in
the absence of CD28-mediated costimulation. To do this, we used an in
vitro model of costimulation by ICAM-1 in the absence of CD28-dependent
costimulation to both measure the effects of IL-12 alone in preventing
AICD and to directly compare the effects of IL-12- and CD28-mediated
costimulation on regulating AICD. Using this model system, we found
that the AICD following ICAM-1 costimulation of naive Th cells
correlated with a loss of m, cleavage and
processing of procaspase-9 and -3, and generation of caspase-3
catalytic activity. This cell death was partially inhibited by
agonistic anti-Fas mAb. Addition of IL-12 strongly promoted the
expansion and viability of ICAM-1-costimulated Th cells through effects
that predominantly, if not exclusively, inhibited Fas signaling without
restoring mitochondrial integrity. Indeed, the most striking effects of
IL-12 on apoptotic pathways were the induction of altered processing of
procaspase-9 and -3 and inhibition of caspase-3 catalytic activity.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The following Abs and purified ligands were used to stimulate or block T cells: anti-CD3 mAb OKT3 (IgG2a), obtained from American Type Culture Collection (Manassas, VA); anti-CD28 mAb 9.3, provided by Dr. C. June (University of Pennsylvania, Philadelphia, PA); and recombinant human ICAM-1 Ig (extracellular domain of ICAM-1 fused to Fc portion of IgG), a gift of P. Hoffman (ICOS, Bothwell, WA). Stimulatory Abs and ligands were indirectly immobilized with goat anti-human IgG-coated and sheep anti-mouse IgG-coated wells (ICN Pharmaceuticals, Aurora, OH). Fas Abs CH-11 (agonist, cross-linking) and ZB4 (antagonist, blocking) were purchased from Kamiya Biomedical (Seattle, WA). Isotype control Abs MOPC21 (IgG1), MOPC195 (IgG2b), and TEPC183 (IgM) were purchased from ICN Pharmaceuticals. The following Abs to human proteins were used for Western blotting: caspase-3 (rabbit), caspase-9 (rabbit), Bcl-2 (mouse), Bax (mouse), and Bak (mouse), all purchased from BD PharMingen (San Diego, CA); Bcl-xL (rabbit) Ab was purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Secondary reagents anti-mouse-HRP, anti-rabbit-HRP, and HRP-conjugated NeutrAvidin were purchased from Pierce (Rockford, IL). Biotinylated mouse mAbs to the following molecules were used for flow cytometry: CD25 (clone B-B10; BioSource International, Camarillo, CA) and Fas and FasL (BD PharMingen); streptavidin-PE (BD Biosciences, San Jose, CA) was used to recognize the biotinylated Abs.
Recombinant human IL-12 (p35/p40) was generously provided by Hoffman-LaRoche (Nutley, NJ). CFSE was purchased from Molecular Probes (Eugene, OR). The following reagents were used to assess cell death: mitochondrial dye 3,3'-dihexyloxacarbocyanine iodide (DiOC6(3); Molecular Probes), annexin V-PE (R&D Systems, Minneapolis, MN), propidium iodide (PI; Sigma, St. Louis, MO), PhiPhiLux-G1D2 caspase-3-like caspase fluorescent substrate (OncoImmunin, Gaithersburg, MD).
Isolation of T cells
Human PBMC from anonymous donors (buffy coats received from LifeSource Blood Services, Glenview, IL, and United Blood Service, Chicago, IL) were isolated by lymphocyte separation medium (Ficoll) density-gradient centrifugation. Resting CD4+CD45RA+CD45RO- naive T cells were obtained by negative selection with mouse mAbs to MHC class II, CD19, CD16, CD14, glycophorin, CD8, and CD45RO and magnetic beads (Polysciences, Warrington, PA) as previously described (25). Selected CD45RA+ cells were consistently 99% pure by FACScan analysis.
Accessory cell-independent T cell stimulation
The 48-well plates were coated with 0.25 ml of 5 µg/ml each sheep anti-mouse Ig and goat anti-human Ig in PBS overnight at 4°C, washed three times with PBS, and coated with 0.25 ml of 0.3 µg/ml OKT3 and with 0.25 ml of either 0.3 µg/ml of ICAM-1 Ig or anti-CD28 mAb 9.3 in PBS for 1 h at room temperature. After being washed three times with PBS, 0.25 x 106 Th cells were added in culture medium (RPMI 1640 supplemented with 10% FBS, 20 mM L-glutamine, 100 IU/ml penicillin, and 100 µg/ml streptomycin; all from BioWhittaker, Walkersville, MD), in a final volume of 0.5 ml, to coated wells with or without 1 ng/ml rIL-12. On day 4 after initiation of culture, cells and supernatants were transferred to 12-well plates for expansion with 1 ml fresh culture medium. T cell counts and viability were determined by no-wash flow cytometry on days 4–7 (see below). To determine cell division, naive Th cells were labeled with 1.7 µM CFSE for 10 min at room temperature, quenched with FCS, and washed two times with culture medium before addition to the assay. Cell division analysis was performed using Modfit software (Verity Software House, Topsham, ME).
Cell surface flow cytometry
Flow cytometric analysis was performed on a FACScan (BD Biosciences) as previously described (25) and was analyzed with Lysys II and CellQuest (both from BD Biosciences) software.
No-wash flow cytometry
To determine relative cell number, Th cells were activated for 4 days and then transferred to larger wells with supernatant as described above. No-wash flow cytometry was performed by collecting a portion of the expansion culture on days 4–7. Events were collected on FACScan for a constant amount of time so as to allow for calculation of cell number. Assessment of live cells was achieved either by gating on cells based on forward scatter and side scatter characteristics or PI exclusion (see below).
Staining cells for cell death analysis
To determine plasma membrane breakdown, 
10 µl of 50 µg/ml
PI was added to cell samples immediately before analysis in FL3
parameter on FACScan. For m, 0.1-ml
cultures of T cells were collected, and 0.4 ml of 40 nM
DiOC6(3) diluted in culture medium was added to
the cells, which were then incubated at 37oC for
30 min. Samples were immediately analyzed in FL1 parameter on FACScan.
To determine phosphatidylserine (PS) exposure, T cells were collected,
washed twice in bead separation buffer, resuspended in annexin V
binding buffer (10 mM HEPES, 10 mM NaOH, 140 mM NaCl, 2.5 mM
CaCl2, pH 7.4) at 1 x
106/ml, and incubated with directly labeled
annexin V for 15 min at room temperature, in the dark, according to the
manufacturer’s specifications. Samples were then diluted in annexin V
binding buffer and analyzed on FACScan within 1 h of staining. To
determine caspase activity in living cells, 0.5–1 x
106 T cells were collected in 1.5-ml Eppendorf
tubes and centrifuged to remove all supernatant. A total of 50 µl of
10 µM PhiPhiLux-G1D2
substrate and 5 µl of FCS were added to each sample, and cells were
resuspended by gently tapping the tube with the fingers. Open tubes
were incubated in 5% CO2, 37°C for 60 min.
Subsequently, cells were washed once with 1 ml ice-cold flow cytometry
diluting buffer (provided by manufacturer) and gently resuspended in
0.5–1 ml flow cytometry diluting buffer. Samples were analyzed on
FACScan within 60–90 min of the final wash.
ELISA
The mAb pairs obtained from BD PharMingen were used at 1–4 µg/ml in a sandwich ELISA to measure IL-2 (sensitivity, 0.2 U/ml) in supernatants. Avidin-peroxidase and ABTS substrate were purchased from Sigma and used as described in the BD PharMingen protocol.
Cell lysates
Stimulated T cell were collected at various time points, washed one time in PBS, and snap-frozen in LN2 and stored at -80°C until detergent lysis. Cell pellets were thawed on ice and lysed in either buffer containing 1% Triton X-100, 50 mM Tris, 300 mM NaCl, 10 µg/ml aprotinin, 5 µg/ml leupeptin, 1 mM pefabloc, 1 mM sodium orthovanadate, 2 mM EDTA, and 10 mM NaF, or in modified Laemmli buffer containing 60 mM Tris (pH 6.8), 10% glycerol, and 2% SDS, a buffer reported to prevent procaspase processing as result of cell lysis (26). Cell membranes were pelleted by microcentrifugation, and the protein content of each lysate-containing supernatant was determined by bicinchoninic acid protein assay (Pierce). Next, 5x loading dye (125 mM Tris (pH 6.8), 25% glycerol, 4% SDS, 10% 2-ME, and 0.5% bromophenol blue) was added to samples before boiling and loading, or boiling, snap-freezing, and storage at -80°C.
SDS-PAGE and Western blotting
Equal microgram amounts (30–50 µg) of cell lysate protein were separated by SDS-PAGE through a 15% polyacrylamide gel and were electroblotted on to polyvinylidene difluoride membranes. Membranes were blocked with 5% milk in TBST for 1 h at room temperature and blotted overnight at 4°C with primary Abs diluted, as recommended by the manufacturer, in 1% BSA in TBST. After washing, membranes were incubated with appropriate HRP-conjugated secondary reagent, and chemiluminescence detection was performed according to the manufacturer’s instructions (Amersham Pharmacia Biotech, Piscataway, NJ). Densitometry analysis was done with ImageQuant (Molecular Dynamics, Sunnyvale, CA).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To study the regulation of T cells by LFA-1-mediated costimulatory signals, we stimulated peripheral blood-derived human naive (CD45RA+CD45RO-) CD4+ Th cells with coimmobilized anti-CD3 mAb and ICAM-1 Ig fusion protein in the absence or presence of recombinant human IL-12, added on day 0. Th cells stimulated with coimmobilized anti-CD3 and anti-CD28 mAbs with or without IL-12 were used as a positive control for cell expansion. The concentrations of anti-CD3 mAb, ICAM-1 Ig (ICAM-1), and anti-CD28 mAb used were chosen to provide consistent and comparable activation of naive Th cells (between both costimulatory conditions and different donors), as determined by [3H]thymidine incorporation at day 4 (data not shown). Th cells were stimulated for 4 days in the presence of Ab and fusion protein (activation phase of the assay) and then were transferred to a larger well to expand for an additional 3 days (expansion phase of the assay) in the absence of Ab and fusion protein.
The relative number of live naive Th cells obtained under the various
costimulatory conditions was determined by PI exclusion in a no-wash
flow cytometric analysis on days 4–7 of culture. The results are
summarized in Fig. 1
A. These results indicated that up to day 5 of culture, there
was equivalent cell expansion among all four costimulatory conditions
(ICAM-1, ICAM-1 + IL-12, anti-CD28, and anti-CD28 + IL-12).
However, after day 5, the relative number of live ICAM-1-costimulated
Th cells began to plateau, and by day 7, these numbers actually
declined to levels significantly lower than that of ICAM-1 + IL-12-,
anti-CD28-, or anti-CD28 + IL-12-costimulated Th cells. On day
7 of culture, there was no significant difference in the relative
number of live Th cells occurring under ICAM-1 + IL-12 and
anti-CD28 costimulatory conditions, and both these costimulatory
conditions generated approximately four times more live cells than
ICAM-1 costimulation, whereas anti-CD28 + IL-12 costimulation
generally exceeded the viable cell recoveries of ICAM-1 + IL-12 or
anti-CD28. It should be noted that viable cell recovery following
stimulation with anti-CD3 alone, IL-12 alone, or anti-CD3 plus
IL-12 did not differ significantly from that of unstimulated Th cells
(data not shown).
|
B shows that, under all three
costimulatory conditions, most naive Th cells entered the cell cycle
and underwent a similar pattern of cell division while in contact with
stimulating Ab and ligand (day 4). However, ICAM-1-costimulation
maintained proliferation only until day 5 or 6, whereas both ICAM-1 +
IL-12- and anti-CD28-costimulated Th cells continued to proliferate
until day 7 (Fig. 1B).
The Th cell proliferation data suggested that ICAM-1 + IL-12 and
anti-CD28 costimulation may induce Th cells to produce levels of
autocrine growth factors that support late proliferation. Consequently,
we determined the amount of the T cell growth factor, IL-2, detectable
in the culture supernatant at day 4 and the expression levels of the
inducible IL-2R 
-chain (CD25) induced by each costimulatory
condition on days 4–6. We found that the differences in late expansion
of the various costimulatory populations could not be explained by
differences in the amounts of IL-2 secreted or CD25 expressed. In all
populations, similar levels of IL-2 were detected in supernatants on
day 4, and comparable levels of CD25 were expressed on days 4, 5, and 6
(data not shown). Together, these results show that ICAM-1, ICAM-1 +
IL-12, and anti-CD28 costimulation induced similar levels of early
naive Th cell expansion and IL-2 production, and that IL-12 enhanced
late proliferation of Th cells costimulated with ICAM-1.
It should be noted that Th cells costimulated with ICAM-1 + IL-12 increased expression levels of CD28 ligands (CD80 and CD86) compared with ICAM-1-costimulated Th cells (data not shown). However, blocking studies with a combination of anti-CD80 and anti-CD86 mAbs demonstrated that CD28-mediated costimulation did not contribute to the IL-12-mediated effects on Th cell viability and late proliferation (data not shown).
IL-12 decreases the portion of ICAM-1-costimulated T cells that die
In other model systems, ICAM-1 costimulation has been shown to
induce death in T cells (4, 28). Thus, we were interested
in determining whether ICAM-1 costimulation also induced Th cell death
in our model and whether IL-12 may affect the viability of
ICAM-1-costimulated T cells. To address these issues, the death of
CFSE-labeled Th cells was assessed by PI exclusion after stimulation
with ICAM-1, ICAM-1 + IL-12, and anti-CD28. The results of no-wash
flow cytometry showed that, after the cells have divided multiple
times, cell death occurs in all populations, although to varying
extents (Fig. 2). In particular, the percentage of PI+ dead cells
(Fig. 2, upper sections of dotplots) increased in
ICAM-1-costimulated cultures throughout expansion, resulting in the
death of the majority of cells by day 7. In contrast, although ICAM-1 +
IL-12 and anti-CD28 costimulation cultures also accumulated dead
cells after a few rounds of cell division, these costimulatory signals
induced cell death to a much lesser extent than ICAM-1-costimulatary
signals. Taken together with our previous results, these data indicate
that although ICAM-1 can costimulate the early activation and expansion
of Th cells, it cannot sustain late proliferation or Th cell survival
but, rather, results in Th cell death. Furthermore, they show that the
addition of IL-12 to ICAM-1-costimulated Th cells can rescue late
proliferation and promote cell survival, resulting in a comparable
viable Th cell recovery at day 7 as with anti-CD28 costimulation.
We found that these effects of IL-12 on ICAM-1-costimulated Th cells
are dose dependent and optimal at concentrations of 1 ng/ml (data not
shown). In addition, the effects are obtained if IL-12 is added to the
culture as late as day 3 of stimulation, but they do not occur with the
addition of IL-12 on day 4 of culture or later (data not shown).
|
Our observations thus far corroborated the reports of others
(4, 28) indicating that costimulation with ICAM-1 results
in Th cells undergoing cell death. However, it was not known whether
the Fas/FasL death pathway, which is involved in some forms of Th cell
AICD (6), was also involved in the Th cell death that
follows activation by ICAM-1 costimulation. Consequently, we performed
experiments to examine the contribution of Fas signaling to the death
of ICAM-1-costimulated Th cells. Our initial studies focused on
determining the expression of Fas and FasL on the surface of Th cells
costimulated with ICAM-1, ICAM-1 + IL-12, or anti-CD28. We found
that expression of Fas on the vast majority of cells in each
costimulatory condition persisted throughout expansion (days 4–7)
(Fig. 3A). Furthermore, the percentage of cells expressing low
levels of FasL was comparable on day 4, after which it decreased
similarly for all populations (Fig. 3B). Together, these
results indicated that there was the potential for Fas-mediated death
in all culture conditions. In addition, coculture with IL-12 did not
appear to either lower the percentage of cells expressing Fas or FasL
(Fig. 3, A and B) or the intensity of Fas and
FasL staining (data not shown). Interestingly, the expression of FasL
rapidly decreases in all conditions after day 5, a point at which the
percentage of nonviable cells is still increasing in the ICAM-1
costimulatory condition.
|
C confirm that
ICAM-1-costimulated Th cells undergo apoptosis, as demonstrated by
binding of annexin V. ZB4 mAb inhibited the percentage of live Th cells
positive for PS by 35% in ICAM-1-costimulated cultures compared
with cultures containing isotype control Ab, and the mAb had no
inhibitory effect on the PS expression of live Th cells costimulated
with ICAM-1 + IL-12 (Fig. 3C). Thus, although Fas-blocking
experiments indicate that Fas signaling contributes to the death of
ICAM-1-costimulated Th cells, these combined results (Fig. 3, B and C) also suggest that other death mechanisms
must be involved in the induction of ICAM-1-dependent AICD.
Furthermore, the data indicate that the major, if not only, mechanism
by which IL-12 inhibits the death of ICAM-1-costimulated Th cells is by
inhibition of Fas-mediated death. Expression of Bcl-2 family members does not correlate with survival of ICAM-1 + IL-12-costimulated Th cells
Antiapoptotic Bcl-2 family members are effective in inhibiting
Fas-mediated death in certain types of T cells (10). Bcl-2
family members regulate both the loss of m
and release of cytochrome c from mitochondria in cells
undergoing apoptosis (14, 30, 31). Consequently, we
investigated whether IL-12 may inhibit ICAM-1-costimulated T cell
death by up-regulating expression of the anti-apoptotic Bcl-2
family members, Bcl-xL and Bcl-2, and/or by
down-regulating expression of the proapoptotic members, Bax and Bak. As
shown in Fig. 4A, Bcl-xL was detected at low levels in
ICAM-1-costimulated Th cells and at high levels in
anti-CD28-costimulated Th cells at 48 h of stimulation.
However, the results also show that addition of IL-12 to
ICAM-1-costimulated T cells does not alter Bcl-xL
expression. Thus, the survival advantage of costimulation with ICAM-1 +
IL-12 did not correlate with early enhancement of
Bcl-xL expression. Surprisingly, IL-12 was found
to inhibit expression of Bcl-xL in Th cells
costimulated with either ICAM-1 or anti-CD28 for 96 h (Fig. 4B). Furthermore, this negative influence of IL-12 on
Bcl-xL expression in ICAM-1-costimulated cells
was maintained through days 4–7 of culture, during which time a
similar IL-12-inhibitory effect was found for Bcl-2 expression (Fig. 4C).
|
C).
However, whereas coculture with IL-12 was found to actually up-regulate
Bak expression in all experiments, the expression of Bax was decreased
in the majority of experiments (4 of 7; Fig. 4C). However,
the IL-12-induced relative decrease in expression of the antiapoptotic
Bcl-2 family members, Bcl-xL and Bcl-2, was
observed to be greater than the IL-12-induced relative decrease in Bax
expression (Fig. 4C). Thus, together, these data strongly
suggest that IL-12 does not inhibit AICD in ICAM-1-costimulated human
naive Th cells through effects on the expression of Bcl-2 family
members.
Induction of loss of m is similar in ICAM-1- and
ICAM-1 + IL-12-costimulated T cells
Loss of m across the inner membrane of
the mitochondria is an early indicator of apoptosis generated by many
death stimuli (8, 9), including some types of
AICD-inducing stimuli (10, 31). We investigated whether
the death of ICAM-1-costimulated T cells might correlate with a loss of
m during expansion, despite the enhanced
expression of Bcl-2 and Bcl-xL during this
period. Loss of m was determined by
staining with the lipophilic cationic dye
DiOC6(3). High DiOC6(3)
staining is detected in cells in which m is
intact, whereas low staining is detected in cells in which
m has dissipated. Live cells were analyzed
to determine whether the ICAM-1-costimulatory signal induced a loss of
m before the ability to exclude PI was
lost. We found that 20–30% of live ICAM-1- and ICAM-1 +
IL-12-costimulated Th cells stained low for
DiOC6(3) on days 4, 5, and 6, whereas negligible
staining was observed in anti-CD28-costimulated live Th cells (Fig. 5). These results indicate that ICAM-1 costimulation initiated a loss of
m regardless of the addition of IL-12,
whereas CD28 costimulation did not. Based on these results, dead cells
would be predicted to accumulate in both ICAM-1- and ICAM-1 +
IL-12-costimulated Th cell cultures. However, the results depicted in
Fig. 2 show that only ICAM-1-costimulated Th cells accumulate a
majority of dead cells. These combined results suggest that IL-12 can
inhibit death pathways downstream of mitochondria in
ICAM-1-costimulated Th cells.
|
To determine the potential downstream effectors of Fas signaling
that may be affected by mitochondrial dysregulation in
ICAM-1-costimulated Th cells, we investigated the cleavage of initiator
caspase-9 and effector caspase-3 in the costimulated Th cell
populations. Caspases are synthesized as inactive proenzymes that
require cleavage and subsequent heterodimerization of large and small
subunits for activity (11). Procaspase-9 processing
results from the interaction of Apaf-1 with mitochondria-derived
cytochrome c (32). Western blot analysis of
cellular lysates collected on day 6 demonstrated that although a
portion of caspase-9 is processed in all populations, the resulting
fragments of caspase-9 are unique depending on the stimulus (Fig. 6A). Two aspartic acid residues near the end of the large
subunit of caspase-9 result in cleavage fragments sized 35 kDa (p35)
and 37 kDa (p37), both of which include the prodomain and large subunit
(33). In Th cells costimulated with ICAM-1, only p35 was
generated (Fig. 6A, lane 1). In contrast, ICAM-1
+ IL-12 costimulation resulted in the generation of predominantly p37
(Fig. 6A, lane 2). p35 was generated in
anti-CD28-costimulated Th cells, although substantial levels of
uncleaved procaspase-9 remained (Fig. 6A, lane
3). We also investigated the effect of IL-12 on caspase-9
processing in anti-CD28-costimulated T cells and found that,
similar to its effects on ICAM-1-costimulated Th cells, IL-12
promoted the processing of procaspase-9 to primarily the p37
fragment (Fig. 6A, lane 4).
|
B, lane 1).
Addition of IL-12 to ICAM-1-costimulated Th cells did not inhibit the
processing of procaspase-3 (p32) but, rather, it inhibited the
generation of the p17 and p12 subunits in favor of the p21 subunit
(Fig. 6B, lane 2). CD28-derived signals generated
primarily the p21 fragment, with little or no detectable p17 and p12
active subunits, while retaining a substantial portion of procaspase-3
(Fig. 6B, lane 3). It should be noted that,
despite equal loading in each lane (as determined by micrograms of
protein in the cell lysate), the anti-CD28 lane consistently showed
a stronger total caspase (procaspase plus caspase fragments) signal
than with the other costimulatory conditions. Nevertheless, the
anti-CD28 lane always showed a very weak signal for the p17
fragment and an undetectable signal for the p12 fragment, consistent
with the lack of significant cell death found with this
costimulation condition. Interestingly, the addition of IL-12 to
anti-CD28-costimulated Th cells resulted in an alteration of
procaspase-3 processing such that nearly all procaspase-3 was processed
to the p19 and p21 caspase-3 fragments (Fig. 6B, lane
4). In summary, these results show that IL-12 can alter processing of procaspase-9 and -3 in both ICAM-1- and anti-CD28-costimulated Th cells. Specifically, IL-12 can induce the processing of procaspase-9 to the p37 fragment and inhibit the generation of the active p12 and p17 subunits of procaspase-3 while increasing the generation of a p21 fragment. Taken together, the results suggest the possibility that IL-12 may inhibit the occurrence of death in ICAM-1-costimulated Th cells by inducing the generation of a caspase-9 p37 fragment and inhibiting the generation of the catalytically active p12 and p17 subunits of caspase-3.
IL-12 prevents the generation of catalytic active caspase-3 in ICAM-1-costimulated Th cells
The results shown in Fig. 6B suggested that ICAM-1
costimulation may lead to the generation of catalytically active
caspase-3, whereas addition of IL-12 may promote incomplete processing
of procaspase-3 and the generation of an enzymatically inactive
caspase-3 form. To directly determine whether this might be the case,
ICAM-1- and ICAM-1 + IL-12-costimulated Th cells were loaded with
PhiPhiLux-G1D2, a substrate
for caspase-3-like enzymes that contains the DEVD sequence. The
fluorescence of the PhiPhiLux substrate increases when it is cleaved by
caspase-3. The results in Fig. 7
show that on day 7, 74% of ICAM-1-costimulated Th cells
demonstrated caspase-3-like enzyme activity, whereas only 21% of
ICAM-1 + IL-12-costimulated Th cells exhibited increased fluorescence
associated with the PhiPhiLux substrate (Fig. 7, right
quadrants). Simultaneous PI staining revealed that virtually all
cells positive for caspase activity were also in the late stages of
death, as demonstrated by their inability to exclude PI (Fig. 7, upper right quadrants). Similar PI exclusion results were
observed on days 5 and 6 (data not shown). These results indicate that
caspase-3-like enzyme activation is associated with the late stages of
death in primary human Th cells costimulated with ICAM-1. Furthermore,
they indicate that IL-12 can inhibit caspase-3-like enzyme activity and
cell death in ICAM-1-costimulated Th cells.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We found that coengagement of the TCR and LFA-1 with anti-CD3 mAb
and ICAM-1 Ig is capable of stimulating naive Th cells to secrete high
levels of IL-2 and of inducing Th cell division (data not shown and
Fig. 1). This was a result inferred from prior studies using
[3H]thymidine incorporation to measure
proliferation in both murine and human Th cells costimulated by ICAM-1
(1, 2, 3, 34, 35). Consistent with reports from Damle et al.
(28), we also showed that ICAM-1 costimulation leads to
the death of human Th cells. Results obtained with a murine ICAM-1
costimulation model differed from those presented here in that, in
the absence of exogenous IL-2, there was little IL-2 production or cell
division in murine naive TCR-transgenic Th cells stimulated with
antigenic peptide and ICAM-1-positive (B7-1- and B7-2-negative) APC,
although there was extensive cell death (4, 36). The
nature and strength of the TCR- and LFA-1-derived signals may account
for at least some of the differences between our model and the
TCR-transgenic model. Nevertheless, despite differences in IL-2
secretion and early proliferation, it is evident that ligation of LFA-1
by ICAM-1 leads to death in both human and murine naive Th cells.
The activation-dependent cell death seen upon naive Th cell
costimulation with ICAM-1 occurs as a consequence of a primary
stimulation. The mechanism(s) involved in this type of
activation-dependent cell death may therefore differ from that of the
AICD that takes place after restimulation of previously activated Th
cells. Our data are consistent with a model of activation-dependent Th
cell death upon ICAM-1 costimulation in which downstream events
resulting from activation of the FasL/Fas pathway and some other
apoptotic pathway contribute to Th cell death. Specifically,
Fas-blocking studies indicated that 35% of the death seen in naive
Th cells costimulated with ICAM-1 can be attributed to FasL/Fas
interaction.
Addition of IL-12 to ICAM-1-costimulated Th cells results in an
increase in the number of cell divisions in cycling Th cells, a lower
percentage of dead cells, and ultimately, four times more live Th cells
than with ICAM-1 costimulation alone (Figs. 1 and 2). These results
suggest that IL-12 acts in a manner similar to CD28 in sustaining
proliferation and enhancing cell survival (5). Does IL-12
promote proliferation, survival, or both in ICAM-1-costimulated Th
cells? Our results show that IL-12 enhanced the proliferative capacity
of ICAM-1-costimulated Th cells just as it is reported to enhance the
proliferation of anti-CD28-costimulated T cells (37, 38). It is difficult to separate proliferation and survival in
our system, because they are always linked. Thus, it is possible that
IL-12 may induce the outgrowth of Th cells that survive ICAM-1
costimulation, because the percentages of PI+
cells differed greatly between ICAM-1- and ICAM-1 + IL-12-costimulated
Th cells, although the number of PI+ cells did
not (data not shown). However, our analysis of death indicators
supports the notion that IL-12 actively inhibits death induced by
ICAM-1 costimulation while promoting proliferation.
Our investigation of the expression of antiapoptotic and proapoptotic
Bcl-2 family members during Th cell activation and expansion in our
costimulatory models revealed that changes in the relative expression
levels of these proteins are unlikely to explain the increased survival
of ICAM-1 + IL-12-costimulated Th cells. We found that the relative
expression of the antiapoptotic Bcl-2 family members, Bcl-2 and
Bcl-xL, in Th cells costimulated with ICAM-1 +
IL-12 is decreased compared with Th cell costimulated with ICAM-1 only
(Fig. 4C). Furthermore, the expression of the proapoptotic
Bcl-2 family member Bak is actually up-regulated following IL-12
coculture (Fig. 4C). The expression of the proapoptotic
Bcl-2 family member Bax is often (although not always) down-regulated
by IL-12 coculture and is not likely to influence the overall outcome,
because the relative decrease in expression of Bax is much less than
the IL-12-induced decreases in Bcl-2 and Bcl-xL
expression (Fig. 4C). Therefore, these observations, in
combination with results showing that a loss of
m was not inhibited by IL-12 (Fig. 5),
suggest that IL-12 induces a survival factor acting downstream of the
mitochondria.
The strongest indications that IL-12 actively inhibited the death of
ICAM-1-costimulated Th cells are evident in the ability of IL-12 to
inhibit Fas-mediated death and to prevent death events downstream of
the loss of m. Although one-third of
ICAM-1-costimulated Th cells were susceptible to Fas-mediated signals,
ICAM-1-costimulated Th cells exposed to IL-12 showed no evidence of
Fas-mediated death (Fig. 3C). Furthermore, we found that
IL-12 altered ICAM-1-mediated processing of procaspase-9 and -3, (Fig. 6, A and B) and prevented both the activation of
procaspase-3 and the accumulation of dead cells (Fig. 7). Loss of
m is an early indicator of apoptosis
(39). Cytochrome c release from the
mitochondria is an event that is known to promote caspase activation
and, thus, lead to cell death (32), and to take place
independently of but coordinated with loss of
m (31, 40). Although we do not
know the status of cytochrome c in ICAM-1-costimulated Th
cells, it is possible that IL-12 did not prevent a loss of
m but did prevent the release of cytochrome
c from the mitochondria of ICAM-1-costimulated Th cells in a
Bcl-2/Bcl-xL-independent fashion. Alternatively,
we believe it is more likely that cytochrome c was released
from the mitochondria in all cells costimulated with ICAM-1, and that
IL-12 altered the procaspase-9 processing promoted by this cytochrome
c and possibly altered procaspase-3 processing, as well, in
a manner that prevented caspase-3 activation and cell death.
Certain caspase processing patterns are correlated with death, such as
the generation of the p35 fragment of caspase-9 and the p17 and p12
fragments of caspase-3. We detected all of these caspase fragments in
ICAM-1-costimulated T cells (Fig. 6, A and B),
thus indicating that cell death in our model is caspase-mediated.
However, in ICAM-1-costimulated cells exposed to IL-12 during
activation, caspase-9 was processed to primarily a p37 fragment and
caspase-3 to primarily a p21 fragment, while the cells continued to
proliferate. The significance of these alternative cleavage patterns is
not known. These cleavage products may have no function as incompletely
processed enzymes. Alternatively, if they are capable of pairing with
fully processed subunits, they may act as decoys that limit the amount
of active enzyme generated. It is also possible that the alternative
cleavage products may have a function in promoting cellular
proliferation, as has been shown for certain caspases in early Th cell
proliferation (41, 42, 43).
In addition to inhibiting apoptosis in ICAM-1-costimulated Th cells,
IL-12 has been shown to inhibit Fas-mediated death in T cells from
HIV+ donors (22), both
irradiation-induced and Fas-mediated death in macrophages
(24), and death of a human Th1 clone resulting from either
IL-2 withdrawal or treatment with HIV gp120 (44). In all
of these scenarios, death is executed by effector caspases. The ability
of IL-12 to block many different death stimuli suggests a model in
which IL-12 is capable of modifying apoptotic signals at a distal stage
in multiple death pathways. Our results demonstrating an inhibitory
effect of IL-12 on caspase-3 activity are consistent with this
model.
One way in which IL-12 may influence caspase processing at a distal stage of multiple death pathways is by inducing expression of an inhibitor of apoptosis (IAP), a family of proteins that can inhibit apoptosis through direct interaction with caspases (45). In support of such a mechanism of IL-12 function, IAP family members have been shown to interact with caspase-3, -7, and -9 and to be expressed in activated T cells (46, 47). Furthermore, similar to the effects of IL-12, certain IAPs, such as survivin, are involved in promoting cell division (48).
In conclusion, we have characterized a mechanism by which IL-12 prevents a form of AICD in ICAM-1-costimulated Th cells. We believe that the significance of this novel regulatory role of IL-12 is clearly beyond its role in our in vitro ICAM-1 costimulation in the absence of CD28. We speculate that the regulatory role of IL-12 in preventing cell death and promoting cell expansion is, for example, a crucial component of the previously recognized adjuvant effect of IL-12 seen in vivo (16). Consistent with this notion, Marth et al. (23) have recently shown that IL-12 can promote the clonal expansion phase of a physiological immune response. Moreover, they demonstrated that this positive effect of IL-12 was lost in mice deficient for the FasL/Fas pathway. Our results would predict that the expansion effect of IL-12 observed by Marth and colleagues is due to the capacity of IL-12 to prevent FasL/Fas-mediated cell death through inhibition of caspase-3 activity. In addition to its beneficial effect on immune responses, the novel regulatory role of IL-12 in promoting cell expansion can also be envisioned to have detrimental effects. The expression of IL-12 in gut-associated tissues of patients with inflammatory bowel disease has, thus far, been correlated mainly with the presence of Th1-type cytokines (49). Our findings suggest another way in which IL-12 may influence normal gut homeostasis; specifically, by inhibiting cell death and, thus, promoting the local expansion of inflammatory cytokine-secreting Th1 cells in the gut, IL-12 may contribute significantly to the pathogenesis of inflammatory bowel disease.
|
Footnotes |
|---|
2 Current address: Department of Environmental Health, Boston University School of Public Health, Boston, MA 02118.
3 Address correspondence and reprint requests to Dr. Gijs A. van Seventer at the current address: Department of Environmental Health, Boston University School of Public Health, 715 Albany Street, Talbot 2E, Boston, MA 02118. E-mail address:gvsevent{at}bu.edu
4 Abbreviations used in this paper: AICD, activation-induced cell death; FasL, Fas ligand; m, mitochondrial membrane potential; ROS, reactive oxygen species; DiOC6(3), 3,3'-dihexyloxacarbocyanine iodide; PI, propidium iodide; PS, phosphatidylserine; IAP, inhibitor of apoptosis.
Received for publication January 29, 2001. Accepted for publication May 9, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
stimulated monocytes: a potential mechanism for amplification of T cell activation through the CD28 pathway. Cell. Immunol. 137:429.[Medline]
: tissue distribution, biochemistry, and function of a natural adherence molecule (ICAM-1). J. Immunol. 137:245.[Abstract]
converting enzyme-like homolog Mch6 and the lamin-cleaving enzyme Mch2 are substrates for the apoptotic mediator CPP32. J. Biol. Chem. 271:27099. production by mouse T helper clones that are unresponsive to B7 costimulation. J. Exp. Med. 180:223.This article has been cited by other articles:
|
|
 |
|
  A. J M Watson In vivo single-photon emission computed tomography imaging of apoptosis in Crohn's disease and anti-tumour necrosis factor therapy Gut, April 1, 2007; 56(4): 461 - 463. [Full Text] [PDF]
|
|
|
|
|
|
J Mudter and M F Neurath Apoptosis of T cells and the control of inflammatory bowel disease: therapeutic implications Gut, February 1, 2007; 56(2): 293 - 303. [Full Text] [PDF]
|
|
|
|
 |
|
 H. Riemann, K. Loser, S. Beissert, M. Fujita, A. Schwarz, T. Schwarz, and S. Grabbe IL-12 Breaks Dinitrothiocyanobenzene (DNTB)-Mediated Tolerance and Converts the Tolerogen DNTB into an Immunogen J. Immunol., November 1, 2005; 175(9): 5866 - 5874. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Wuthrich, T. Warner, and B. S. Klein IL-12 Is Required for Induction but Not Maintenance of Protective, Memory Responses to Blastomyces dermatitidis: Implications for Vaccine Development in Immune-Deficient Hosts J. Immunol., October 15, 2005; 175(8): 5288 - 5297. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. G. Feng, D. Jankovic, M. Kullberg, A. Cheever, C. A. Scanga, S. Hieny, P. Caspar, G. S. Yap, and A. Sher Maintenance of Pulmonary Th1 Effector Function in Chronic Tuberculosis Requires Persistent IL-12 Production J. Immunol., April 1, 2005; 174(7): 4185 - 4192. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Kandula and C. Abraham LFA-1 on CD4+ T Cells Is Required for Optimal Antigen-Dependent Activation In Vivo J. Immunol., October 1, 2004; 173(7): 4443 - 4451. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Nolt and J. L. Flynn Interleukin-12 Therapy Reduces the Number of Immune Cells and Pathology in Lungs of Mice Infected with Mycobacterium tuberculosis Infect. Immun., May 1, 2004; 72(5): 2976 - 2988. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Chang, J.-H. Cho, S.-W. Lee, S.-Y. Choi, S.-J. Ha, and Y.-C. Sung IL-12 Priming during In Vitro Antigenic Stimulation Changes Properties of CD8 T Cells and Increases Generation of Effector and Memory Cells J. Immunol., March 1, 2004; 172(5): 2818 - 2826. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. S. Seaman, F. W. Peyerl, S. S. Jackson, M. A. Lifton, D. A. Gorgone, J. E. Schmitz, and N. L. Letvin Subsets of Memory Cytotoxic T Lymphocytes Elicited by Vaccination Influence the Efficiency of Secondary Expansion In Vivo J. Virol., January 1, 2004; 78(1): 206 - 215. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S.-J. Ha, D.-J. Kim, K.-H. Baek, Y.-D. Yun, and Y.-C. Sung IL-23 Induces Stronger Sustained CTL and Th1 Immune Responses Than IL-12 in Hepatitis C Virus Envelope Protein 2 DNA Immunization J. Immunol., January 1, 2004; 172(1): 525 - 531. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. W. Lee, Y. Park, J. K. Yoo, S. Y. Choi, and Y. C. Sung Inhibition of TCR-Induced CD8 T Cell Death by IL-12: Regulation of Fas Ligand and Cellular FLIP Expression and Caspase Activation by IL-12 J. Immunol., March 1, 2003; 170(5): 2456 - 2460. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. M. Cleary, W. Tu, A. Enright, T. Giffon, R. Dewaal-Malefyt, K. Gutierrez, and D. B. Lewis Impaired Accumulation and Function of Memory CD4 T Cells in Human IL-12 Receptor {beta}1 Deficiency J. Immunol., January 1, 2003; 170(1): 597 - 603. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Takahashi, T. Tashiro, M. Miyazaki, M. Kobayashi, R. B. Pollard, and F. Suzuki An essential role of macrophage inflammatory protein 1{alpha}/CCL3 on the expression of host's innate immunities against infectious complications J. Leukoc. Biol., December 1, 2002; 72(6): 1190 - 1197. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Kobayashi, H. Takahashi, A. P. Sanford, D. N. Herndon, R. B. Pollard, and F. Suzuki An Increase in the Susceptibility of Burned Patients to Infectious Complications Due to Impaired Production of Macrophage Inflammatory Protein 1{alpha} J. Immunol., October 15, 2002; 169(8): 4460 - 4466. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Kwang Yoo, J. Ho Cho, S. Woo Lee, and Y. Chul Sung IL-12 Provides Proliferation and Survival Signals to Murine CD4+ T Cells Through Phosphatidylinositol 3-Kinase/Akt Signaling Pathway J. Immunol., October 1, 2002; 169(7): 3637 - 3643. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Y. Park, B. Hondowicz, M. Kopf, and P. Scott The Role of IL-12 in Maintaining Resistance to Leishmania major J. Immunol., June 1, 2002; 168(11): 5771 - 5777. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
), ICAM-1- (•), ICAM-1 + IL-12- (
),
anti-CD28- (
), and anti-CD28 + IL-12- (
) costimulated Th
cells. Data represent the mean and SEM of results derived from five
donors. B, Naive Th cells were labeled with CFSE and
stimulated with coimmobilized anti-CD3 mAb and ICAM-1 Ig with (
). Th cells were transferred to new wells with
supernatant and fresh culture medium on day 4, and Th cells were
collected for no-wash flow cytometry on days 4–7. Analysis of cell
division was performed on live cells based on forward and side scatter
characteristics. The number of live events collected in each generation
was determined by no-wash flow cytometric
analysis over 1 min, followed by further analysis using Modfit
software. Undivided cells are represented in the parent (P) generation.
Results depict the mean and SE of duplicate cultures derived from one
representative of six tested donors different from those donors in
A.
