| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Department of Immunology and Rheumatology, Merck Research Laboratories, Rahway, NJ 07065; and
Howard Hughes Medical Institute and Department of Immunology, University of Washington, Seattle, WA 98195
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-chain genes to
generate an enormous repertoire of diverse TCR specificities, under
circumstances in which each developing cell in general expresses only a
single receptor species. Subsequent selection eliminates autoreactive
and nonfunctional cells from this repertoire (reviewed in Ref.
1). Programmed cell death (apoptosis) appears to be the
default fate of thymocytes if they do not receive signals to
proliferate and/or differentiate, thus eliminating nonfunctional cells
(death by neglect). This is reflected at the earliest developmental
stage when the expression of cytokine receptors and the pre-TCR
(composed of a newly rearranged TCR ß-chain and a surrogate TCR
(pre-T) -chain) is required for expansion and differentiation. In
the absence of survival signals delivered via these receptors, the
CD4-8- (double-negative,
DN3) cells undergo
apoptosis (reviewed in Refs. 2 and 3).
Thymocytes that advance beyond this regulatory checkpoint express low
levels of the TCR as well as both CD4 and CD8 coreceptors
(double-positive cells, DP). Such cells are subject to positive and
negative selection within the thymus. DP cells that express TCRs of
intermediate affinity for peptide ligands in the context of MHC
molecules are positively selected to differentiate into CD4 or CD8
single-positive cells. Consequently, these cells up-regulate their TCR
expression levels and eventually populate the periphery. In contrast,
both very low as well as very high affinity interactions of the TCR
with self-MHC/peptide ligand result in apoptosis. It is believed that
most thymocytes express TCRs which cannot mediate signals that promote
survival and differentiation. Hence, these cells die by neglect. On the
other end of the spectrum, very high affinity TCR interactions with
self MHC cause negative selection, i.e., the clonal deletion of these
potentially autoreactive cells (reviewed in Refs. 4 and
5). To investigate the mechanisms responsible for negative selection, we sought to inhibit caspase-mediated apoptosis in vivo. Caspases, a family of at least 14 cysteine proteases, have been found to play a critical role in mediating apoptosis in species ranging from nematodes to mammals (reviewed in Refs. 6 and 7). In general, these proteases are constitutively expressed as inactive proenzymes, which become proteolytically activated during apoptosis. Based on sequence homologies and cellular function, caspases can be divided into two major subfamilies with close resemblance to either caspase-1 or to caspase-3. According to their respective function during apoptosis, they have also been categorized as initiator or effector caspases (7). Caspases are abundantly expressed in lymphoid cells, and their involvement in apoptosis suggested that they might play a role in negative selection. Indeed, studies performed using fetal thymic organ culture implicate caspase-3-related enzyme activation in thymocyte apoptosis (8). These authors reported that the generic pharmacological caspase inhibitor zVAD-FMK (benzyloxycarbonyl-Val-Ala-Asp-fluoro-methylketone) inhibited deletion of thymocytes induced by either anti-CD3 Abs, the glucocorticoid dexamethasone, or antigenic peptide in vitro. Moreover, caspase-3 activation was detected specifically during apoptosis induced by TCR stimulation, but not during spontaneous cell death (9).
These studies leave open the mechanism whereby TCR stimulation might induce caspase activation and the importance of this process in vivo. It remains unclear whether caspase activation is actually required for negative selection to occur or whether it occurs as a secondary effect that accompanies negative selection-induced thymocyte death.
Gene disruptions of individual caspases yield phenotypes associated with the inflammatory response (caspase-1 and -11) or defects in cell- and/or stimulus-specific apoptosis (caspase-2, -3, -8, and -9). However, none of these mice manifest abnormalities in thymocyte maturation (10, 11, 12, 13, 14, 15, 16, 17, 18), suggesting that several caspases may play redundant roles in thymocytes. We therefore pursued a strategy designed to block the function of most caspases expressed in the thymus.
Our studies make use of the baculovirus protein p35, which inhibits all caspase subfamilies through a stable interaction with their catalytic sites (19, 20, 21, 22, 23). In addition to its ability to protect baculovirus-infected insect cells from apoptosis, the ectopic expression of p35 blocks apoptosis involving diverse death signals, e.g., developmental death, growth factor withdrawal, or DNA damage, in many different cell types and species (reviewed in Ref. 24). These observations suggested that p35 might provide a potent tool to investigate mechanisms of apoptosis in thymocytes. In this study, we present results obtained from transgenic mice expressing the p35 protein under the control of the thymocyte-specific lck proximal promoter.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The p35 cDNA (25) was amplified by PCR on linearized baculovirus DNA from Autographa californica nucleopolyhedrovirus (BaculoGold; Phar-Mingen, San Diego, CA) using the following primers (BamHI restriction sites are underlined): p35–5' primer, 5'-GCG GGA TCC CCA TAG CAA AAT GTG TGT AAT TTT TCC GGT-3'; p35–3' primer, 5'-GCG GGA TCC TTA TTT AAT TGT GTT TAA TAT TAC ATT TTT GTT GAG-3'.
The obtained 917-bp cDNA fragment was cloned into the BamHI site of the transgenic p1017 vector (26), placing its expression under the control of the lck proximal promoter. Transgenic mice were generated and propagated as described previously (27). For studies of negative selection or early thymocyte development, lck-p35 mice were also crossed with OT1 (28) or HY TCR transgenic mice (29) or rag1-/- mice (purchased from The Jackson Laboratory, Bar Harbor, ME) (30), respectively. The presence of the respective transgenes was monitored by PCR strategies. All mice were housed under specific pathogen-free conditions.
Thymocyte stimulation assays in vitro
For CD3 and Fas (CD95) stimulation, 24-well plates were coated
with 10 µg/ml anti-CD3
(145-2C11) or anti-Fas Abs (Jo2;
PharMingen). A total of 3 x 106 thymocytes
were seeded into Ab-coated or into control wells, and cultured in 1 ml
medium (RPMI 1640 supplemented with 10% FCS, 2 mM
L-glutamine, 50 µM 2-ME, and antibiotics) at 37°C for
indicated periods. Thymocytes were cultured similarly for the OVA
peptide deletion assay, except for the replacement of Ab stimuli with 1
nM OVA peptide (SIINFEKL; 28) presented by 3 x
105 irradiated EL4 cells. For OVA peptide
titration experiments, 5 x 105 thymocytes
were plated into 96-well round-bottom plates in the presence of 5
x 104 EL4 cells and indicated peptide
concentrations. Where indicated, the caspase inhibitor zVAD-FMK (Enzyme
Systems Products, Livermore, CA) was added at a final concentration of
100 µM.
Flow-cytometric analysis
A total of 5 x 105 to 1 x
106 cells were stained with saturating
concentrations of Abs at 4°C for 30 min, using combinations of the
following mouse-specific Abs: PE-conjugated anti-CD4 and
FITC-conjugated anti-CD8 (Caltag Laboratories, San Francisco, CA),
biotinylated (BIO) anti-CD3, BIO anti-CD69, FITC
anti-CD25, PE anti-CD44, and BIO annexin V (PharMingen).
Transgenic HY or OT1 TCR -chains were detected with BIO T3.70 (gift
from Dr. H. S. Teh, University of British Columbia, Vancouver,
Canada) or BIO anti-V2 (PharMingen), respectively. BIO Abs were
visualized by tricolor-conjugated streptavidin (SA-TRI; Caltag). Data
were collected on a FACScan flow cytometer (Becton Dickinson, San Jose,
CA). Changes in mitochondrial transmembrane potential were visualized
by staining with 3,3'-dipropylthiadicarbocyanine iodide
(DiSC3(5); Molecular Probes, Eugene, OR). In
brief, cells were resuspended in 1 ml of staining medium (Hanks’
buffered saline, 10 mM HEPES, 1% BSA), and 1 µl
DiSC3(5) in DMSO was added for a final
concentration of 40 nM. Cells were incubated for 20 min at room
temperature in the dark, then washed and resuspended in staining
medium, followed by data collection on a
FACStarPlus flow cytometer (Becton Dickinson)
within the next hour. All analyses were performed using ReproMac 2.3
software (TrueFacts Software, Seattle, WA).
Protein extracts and Western blot analysis
Single cell suspensions from thymus or cultured thymocytes were washed in PBS twice and resuspended in hypotonic lysis buffer (10 mM HEPES, pH 7, 50 mM NaCl, 2 mM MgCl2, 40 mM ß-glycerophosphate, 5 mM EGTA) at 5 x 108 cells/ml. Cells were disrupted by four alternating freeze-thaw steps, and the lysates were cleared by centrifugation at 20,000 x g for 10 min at 4°C. Protein concentrations were determined by BCA protein assay (Pierce, Rockford, IL).
For Western blot analysis, protein extracts were separated on 12% SDS-PAGE gels, transferred onto polyvinylidene difluoride membranes (Amersham Life Science, Arlington Heights, IL), and probed with polyclonal chicken anti-p35 Abs (1 µg/ml; provided by Dr. A. Niles, Promega, Madison, WI). Incubation with HRP- or alkaline phosphatase-conjugated anti-chicken IgY (IgY-HRP, 1 µg/ml; IgY-AP, 200 ng/ml; Promega) was followed by enhanced chemiluminescence or chemifluorescence (Amersham Life Science) detection methods, respectively.
Caspase enzyme assay
A total of 3 or 15 µg of protein extracts from fresh or cultured thymocytes were incubated in 96-well plates in 100 µl assay buffer (100 mM HEPES, pH 7.5, 10% sucrose, 0.1% CHAPS, 10 mM DTT, 0.1 mg/ml OVA), containing the substrate Ac-DEVD-AMC (1 µM; Peptides International, Louisville, KY) at 37°C for 30 or 60 min. The release of AMC was detected on a SpectraFluor Plus instrument (excitation 360 nm; emission 460 nm) with XFluor software (Tecan US, Durham, NC). Standard dilutions of AMC were included in each assay to determine absolute concentrations of released AMC in samples and convert measured fluorescent units into a rate of substrate cleavage per mg protein (pmol cleaved DEVD-AMC x min-1 x mg-1). Values were corrected for spontaneous release of AMC in the absence of extract.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To direct the expression of p35 in immature thymocytes, we placed
the p35 cDNA from baculovirus (25) under the control of
the lck proximal promoter (26). Eight
transgenic mouse lines were generated that exhibited a wide range of
p35 expression levels, as determined by Northern and Western blot
analysis (Fig. 1
A, and
unpublished data). We used these lines to establish a dose-response
relationship, correlating p35 expression levels with caspase
inhibition. Thymocytes were isolated from mice representing different
transgenic lines or corresponding normal littermate controls (LMC) and
stimulated with cross-linking anti-Fas (CD95) Abs in vitro for
4 h (Fig. 1B). Protein extracts from freshly isolated
or cultured cells were analyzed for caspase activity using a synthetic
tetrapeptide substrate (Ac-DEVD-AMC; Ref. 31) that is
preferentially cleaved by caspase-3-related proteases, the main targets
for inhibition by p35. The release of the fluorogenic compound AMC
provides a measure of caspase activity. In agreement with the proposed
stochiometric inhibition of caspases by p35 (19), we
observed a linear inhibition of caspase activity as a function of p35
expression, reaching maximal enzyme inhibition at 
1–2 ng p35 per
1 x 107 cells. The residual background of
substrate degradation, which is seen even at clearly saturating p35
expression levels, may be caused by proteasome-mediated cleavage (Nancy
Thornberry, personal communication).
|
High p35 expression in thymocytes inhibits caspase-3-related enzyme activity and reduces apoptosis in vitro
Immature DP T cells, which represent the vast majority of thymocytes, undergo spontaneous cell death when cultured in single cell suspensions. Moreover, they are exquisitely sensitive to TCR stimulation with anti-CD3 Abs (a process that is thought to mimic negative selection in vitro) and to Ab-mediated cross-linking of Fas. Therefore, we tested the lck-p35 transgene for its ability to inhibit caspase activity and to rescue thymocytes under these apoptosis-inducing conditions in vitro.
Thymocytes from lck-p35 transgenic and LMC mice were
isolated and cultured in the absence or presence of anti-CD3 or
anti-Fas (CD95) Abs for 2, 4, and 24 h (Fig. 2). At each time point, flow-cytometric
samples and protein extracts were prepared to determine the percentage
of live DP thymocytes and the enzymatic activity of caspases,
respectively. The expression of p35 protein improved the survival of DP
cells slightly (7–10% more live cells) at all time points when
cultured in medium only. In the presence of anti-CD3 Abs, a similar
improvement of survival of DP thymocytes by p35 was observed at the 2-
and 4-h time points; however, this protective effect was lost after
24 h. In contrast, p35 profoundly reduced apoptosis in
anti-Fas-treated cells.
|
, lower panels). A dose-response relationship
obtained with additional transgenic lines expressing low to
intermediate levels of p35 underscored the causal link between the
observed caspase inhibition and improved thymocyte survival, i.e.,
caspase inhibition by intermediate or low levels of p35 expression
resulted in intermediate or weak protection from apoptosis,
respectively (data not shown). Additional experiments focused on lines
expressing very high levels of p35 (Fig. 1). Expression of transgenic TCRs early in thymocyte development reduces p35 expression levels
In a wild-type mouse with a normal mixed TCR repertoire, the number of negatively selected thymocytes is almost certainly very low (33). To facilitate the investigation of negative selection, we introduced the well-characterized HY and OT1 TCR transgenes onto a lck-p35 mouse background (28, 29). The HY TCR recognizes the male-specific HY Ag in the context of H-2Db MHC class I, causing positive selection of CD8+ HY T cells in females and strong negative selection of DP thymocytes in male mice (29). The OT1 TCR specifically recognizes the OVA peptide 257–264 in the context of H-2Kb MHC class I (28). In the absence of peptide Ag, thymocytes are positively selected to the CD8+ T cell compartment. However, the addition of the specific OVA peptide to these thymocytes in vivo or in vitro induces a concentration-dependent deletion of the DP compartment, thereby providing another model for negative selection (28).
We crossed the two mouse lines with highest p35 expression with HY and
OT1 mice and analyzed transcript and protein expression levels in males
and females from lck-p35 x HY and
lck-p35 x OT1 litters. Both Northern and Western blot
analysis revealed that mice with the single lck-p35
transgene maintained high p35 mRNA and protein expression. However,
their lck-p35/HY and lck-p35/OT1 double
transgenic littermates had 5-fold reduced p35 mRNA and protein
expression levels (Fig. 3, and
unpublished data). The same relative reduction was observed for both
high p35-expressing lines and for additional lck-p35 x
HY and lck-p35 x OT1 litters when low or intermediate
p35-expressing mouse lines had been used for breedings (data not
shown). These observations suggest that artificially early and high
expression of a TCR reduces the activity of the transgenic
lck proximal promoter at a much earlier step in thymocyte
development than in normal thymi, in which high TCR expression is only
achieved upon positive selection to single-positive T cells.
Nevertheless, the previously established dose-response relationship
(Fig. 1B) indicated that even these reduced p35 levels in
the doubly transgenic mice would be sufficient to block caspase
activity.
|
To study the deletion of DP thymocytes by endogenous Ag in the HY
TCR transgenic mice, we compared HY mice with lck-p35/HY
male littermates. Fig. 4 depicts the
CD4/CD8 flow-cytometric profile of an HY female control, showing the
typical skewing of CD4-8+
cells due to positive selection of HY TCR-expressing cells (upper
left panel). As previously reported, this effect becomes more
obvious if only cells that actually express the HY TCR are examined (as
determined by staining with T3.70, an Ab specific for the V-chain of
the HY TCR; upper right panel). In contrast, HY
TCR-expressing thymocytes are autoreactive in male mice.
Consequently, the majority of DP thymocytes in male animals are
eliminated and total thymocyte numbers are dramatically reduced
(10% of control). As some thymocytes can escape negative selection
by replacing the transgenic TCR -chain with a rearranged endogenous
TCR -chain (providing a different specificity), we again
electronically gated on cells that are recognized by the T3.70 Ab to
restrict the analysis to authentic HY TCR transgenic cells (Fig. 4, lower panels). Clearly, negative selection was not affected
by the lck-p35 transgene because neither the total cell
numbers nor the relative representation of thymic subpopulations were
changed. Positive selection in female HY TCR transgenic mice is
likewise unchanged in the presence of p35 (data not shown).
|
To examine the importance of caspases for thymocyte survival in
more detail, we used peptide-mediated deletion of T cells in vitro as a
surrogate for negative selection in vivo (28). Thymocytes
from animals expressing the OT1 TCR, p35, or both were exposed to
increasing amounts of OVA peptide presented by EL4 cells in vitro.
After 18–20 h, cells were analyzed by flow cytometry to determine the
number of surviving DP thymocytes (Fig. 5A). As expected,
lck-p35 transgenic and normal control thymocytes were not
responsive to OVA peptide, but both OT1 and lck-p35/OT1
thymocytes underwent apoptosis in an OVA peptide dose-dependent manner.
Interestingly, the shape of the deletion curve, which indicates the
relative sensitivity of the thymocytes to the OVA peptide, was
identical for OT1 and lck-p35/OT1 cells. However, the total
number of surviving DP cells was slightly (5%) but consistently
increased by the presence of p35. A more pronounced increase in basal
survival (12%) is found for lck-p35 cells compared with
normal controls, which is consistent with the higher p35 expression in
the absence of the TCR transgene.
|
A). Fig. 5B shows that caspase activity in both OT1 and
lck-p35/OT1 thymocytes increased in culture even in the
absence of peptide stimulation (medium control). This correlates with
the spontaneous death of a large subpopulation of cells with low CD4
and CD8 expression that is typically found in these mice (see Fig. 7).
Nonetheless, p35 expression clearly reduced the magnitude of this
spontaneous enzyme activation. OVA peptide stimulation increased
caspase activity about 2-fold in both OT1 and lck-p35/OT1
thymocytes but, as in the medium control, p35 expression suppressed the
enzyme activity to 50% when compared with OT1 thymocytes. The
persistently high level of p35 expression in the
lck-p35 control restricted caspase activity to background
levels as before (Figs. 2 and 5B).
|
Because the reduced levels of p35 in lck-p35/OT1
transgenic thymocytes permitted some increase in caspase activity (Fig. 5B), it remained conceivable that negative selection and
peptide-mediated deletion in vitro both require a threshold level of
caspase activity. To test this possibility, we took advantage of the
potent generic caspase inhibitor zVAD-FMK. In a pilot experiment, we
determined that a concentration of 100 µM zVAD-FMK was sufficient to
fully block caspase activity and prevent the loss of membrane polarity
characteristic of apoptosis. Subsequent assays confirmed that this
concentration also completely inhibits caspase-mediated
oligonucleosomal DNA degradation following anti-Fas-,
anti-CD3-, or dexamethasone-induced apoptosis of normal thymocytes
(data not shown). zVAD-FMK does, however, not generally interfere with
signal transduction pathways, as flow-cytometric analysis confirmed
that all thymic subpopulations were capable of responding normally to
peptide stimulation with respect to the induction of the T cell
activation marker CD69 (unpublished data).
Fig. 6 documents the effect of zVAD-FMK
on OVA peptide-induced thymocyte deletion in vitro. As in Fig. 5, the
data represent the percentage of surviving DP thymocytes relative to
that observed in freshly isolated cells. zVAD-FMK improves the overall
survival of DP cells for both LMC and OT1 transgenic thymocytes by
20–30% (Fig. 6, and unpublished data). However, the specific
responsiveness of the OT1 TCR-expressing cells to OVA peptide remains
exactly the same. The shift of the peptide response curve occurred by
the same magnitude at every OVA peptide concentration analyzed,
suggesting that this difference is exclusively accounted for by the
loss of spontaneous cell death.
|
m), which occur at
a very early stage of apoptotic cell death and can be visualized with
m-sensitive, lipophilic dyes such as
DiSC3(5) or 3,3'-dihexyloxacarbocyanine iodide
(DiOC6) (3, 36). Fig. 7 shows representative results for this
analysis, using experimental conditions as described for Figs. 5 and 6.
For clarity, the raw data are restricted to OT1 transgenic cells that
were freshly isolated or incubated with 1 nM OVA peptide in the absence
or presence of 100 µM zVAD-FMK. During apoptosis, thymocytes show a
characteristic shift to lower FSC and higher SSC properties, which we
exploited to determine the representation of live cells in all
experiments described in
Figs. 2–6. However, in cultures treated with
zVAD-FMK, a significant proportion of cells displayed aberrant FSC and
SSC properties (Fig. 7A, top panels). Analysis of
mitochondrial conditions in cells cultured without zVAD-FMK confirmed
that the FSC/SSC gate included almost exclusively cells with high
m, a characteristic of living cells
(DiSC3(5)-bright, Fig. 7A). In
contrast, in the presence of zVAD-FMK, one half of the population
within the FSC/SSC gate manifested a low m,
indicating that more cells were committed to apoptosis than revealed by
the FSC/SSC properties. Therefore, we compared the numbers of live DP
and CD4low8low cells under
all different conditions defined by either FSC/SSC or by high
m (Fig. 7B). Both methods of
enumerating live cells performed equivalently in cells not exposed to
zVAD-FMK, thus confirming the results for previous experiments with OT1
and lck-p35/OT1 transgenic cells. However, the comparison
for zVAD-FMK-treated cells yielded much lower absolute numbers for
total live cells and consequently for live DP and
CD4low8low cells when based
on m. Moreover, by gating on
DiSC3(5)-bright cells, the relative
representation of
CD4low8low cells was
reduced. Additional characterization showed that, irrespective of
culture conditions, all cells with high m
were negative for annexin V staining (which assesses membrane
phospholipid polarity) (37) and uptake of
7-amino-actinomycin D (7AAD, a DNA-intercalating agent that stains
apoptotic or dead cells upon loss of plasma membrane integrity late in
apoptosis) (38), i.e., these cells appear healthy by all
criteria. Electronic gating of cells with low
m revealed that in the absence of zVAD-FMK,
90% of this population was annexin V positive and 24% was 7AAD
positive, while in the presence of zVAD-FMK, only 33% of these
cells were annexin V positive, but still 22% were 7AAD positive,
confirming cell death within this population (data not shown). The
restriction of the analysis to cells with high
m therefore provides a more
accurate assessment of live cells in zVAD-FMK-treated
cultures than the FSC/SSC gate (or annexin V). This becomes especially
important in the case of
CD4low8low cells, which
might otherwise be overestimated and mistaken for intermediate cells
during positive selection. Also for DP cells, the mitochondrial
analysis indicates that the rescue by zVAD-FMK is less effective than
was suggested by the analysis of FSC/SSC. These results are
nevertheless consistent with those presented in Fig. 6, in that caspase
inhibition by zVAD-FMK did not block peptide-mediated deletion of OT1
thymocytes in vitro. In summary, the data obtained with zVAD-FMK confirm the observations with lck-p35 transgenic thymocytes and indicate that inhibition of caspase activity can reduce apoptosis due to lack of survival signals, but does not change the sensitivity of the OT1 TCR-expressing cells to negative selection.
p35 does not rescue immature thymoctes in the absence of pre-TCR signaling
The rescue of DP thymocytes from spontaneous cell death by p35 and
zVAD-FMK in vitro suggested that p35 might also prevent apoptosis
caused by the lack of survival signals during thymocyte development.
For example, immature thymocytes die if they are unable to express a
pre-TCR complex on the surface (reviewed in Ref. 3). In
rag1-/- mice, thymocyte differentiation
is blocked at the DN stage and total cell numbers are reduced to 1%
of those in normal thymi because of failure to rearrange TCR genes
(30). We therefore assessed whether p35 expression could
rescue rag1-/- thymocytes in
vivo.
We crossed the high p35-expressing transgenic lines with
rag1-/- mice and analyzed
rag1-/- lck-p35 and control mice
(Fig. 8). In heterozygous
rag1-deficient mice, p35 expression did not affect thymocyte
development. Similarly, p35 expression proved incapable of ameliorating
the defects imposed by homozygous deletion of rag1 genes.
Total thymocyte numbers in rag1-/-lck-p35 transgenic mice were as low as in
rag1-/- mice (1.55 x
106 vs 1.65 x 106
cells), and no differentiation beyond the DN stage was detectable. The
DN population can be further characterized by assessing the expression
profile of CD25 and CD44 (39); cells mature from
CD25-44+ to
CD25+44+ and subsequently
extinguish CD44 expression. CD25 expression is normally lost upon
successful signaling from the pre-TCR, leading to the appearance of
CD25-44- cells that
rapidly gain CD4 and CD8 expression (39). Fig. 8 clearly
shows that in both rag1-/- and
rag1-/- lck-p35 transgenic animals,
the majority of thymocytes are
CD25+44- (69% and 74%,
respectively), and the distribution of cells in earlier stages is
likewise indistinguishable. Consistent with the complete absence of DP
cells in these mice, no
CD25-44- thymocytes were
detectable (Fig. 8). Importantly, mRNA analysis confirmed high
expression of p35 transcripts in DN thymocytes from
rag1-/- lck-p35 transgenic mice. In
conclusion, p35 expression cannot rescue maturation of
rag1-/- thymocytes.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Several viruses encode proteins that inhibit caspase function (reviewed in Ref. 24). Most notably, the cowpox virus protein crmA and the baculovirus protein p35 function as pseudosubstrates, which become cleaved by caspases, but do not subsequently dissociate from the enzymes. CrmA, which inhibits caspases-1 and -8 most efficiently, was shown to inhibit Fas-induced cell death, but not negative selection when overexpressed in transgenic thymocytes (40, 41). We decided to probe thymocyte apoptosis with p35, which inhibits caspases-1, -3, -4, -6, -7, -8, and -10, and can probably be considered a general caspase inhibitor (19, 20, 21, 22). To complement and confirm results obtained for lck-p35 transgenic thymocytes, we repeated critical in vitro experiments with the pharmacological caspase inhibitor zVAD-FMK, which can be titrated for maximal effect.
Our results in vitro clearly show that the high p35 expression levels obtained in lck-p35 transgenic thymocytes effectively block the activity of caspase-3-related enzymes following apoptotic stimuli such as anti-Fas or anti-CD3 Abs, or the process initiated by prolonged cultivation of single cell thymocyte suspensions. However, this caspase inhibition only partially improved thymocyte survival. Consistent with previous studies in cell lines (42, 43), p35 efficiently suppressed anti-Fas-induced cell death, in which caspase-8 is known to play a primary, death-initiating role (reviewed in Ref. 44). In contrast, protection from anti-CD3-induced apoptosis and from spontaneous cell death was modest, implying the involvement of caspases that are not inhibitable by p35 and/or an additional, caspase-independent cell death mechanism under these conditions. In vivo, the number of thymocytes was unaffected by p35, indicating that death by neglect, which is thought to account for most of the observed cell death in the thymus (33), occurred normally.
p35 expression does not rescue the early developmental block of rag1-deficient thymocytes
To investigate the function of caspases during early thymic development, we analyzed the effect of p35 expression in rag-deficient mice. Rag-/- thymocytes lack a functional pre-TCR, which normally provides survival, expansion and differentiation signals at the DN stage. Signals that mimic pre-TCR stimulation fully restore thymocyte development from the DN to the DP stage (reviewed in Ref. 3). Surprisingly, the inactivation of the transcription factor p53, which plays a critical role in inducing apoptosis especially upon DNA damage, or the expression of a transgene encoding the anti-apoptotic molecule Bcl-2 enabled rag-/- thymocytes to extinguish CD25 expression and differentiate into intermediate CD4+ or DP cells (45, 46). These experiments suggested that prolonged survival of DN thymocytes might permit them to continue an intrinsic differentiation program in the absence of pre-TCR signaling. However, the results for the Bcl-2 transgene were controversial (47).
Our results for rag1-/- lck-p35 transgenic mice indicate that the developmental block in rag1-/- thymocytes cannot be rescued by inhibiting caspase activity, and they underscore that death by neglect is not inhibited by p35 expression. These findings and the studies of Maraskovsky et al. (47) suggest that rag1-/- thymocytes lack an essential differentiation signal from the pre-TCR that is also a prerequisite for survival and proliferation. It remains possible, however, that the results by Linette et al. (46), studying Bcl-2 overexpression, or by Jiang et al. (45) in p53-deficient thymocytes may reflect the importance of other, caspase-independent pathways in the cell death process.
Caspase inhibition represses spontaneous cell death in vitro, but not peptide-specific negative selection in vitro and in vivo
Evidence for caspase activation during anti-CD3 Ab or
peptide-induced thymocyte deletion implied that caspases might be
required for negative selection (8, 9). However, our
results show that caspase inhibition by p35 or zVAD-FMK reduced
spontaneous cell death of OT1 thymocytes in medium without peptide, but
the relative dose response to deleting OVA peptide was
indistinguishable from controls for either caspase inhibitor. Moreover,
p35 expression did not prevent the deletion of DP thymocytes in the HY
TCR model for negative selection by an endogenous Ag. Although
thymocytes can be detected in male HY mice that have replaced the
transgenic HY TCR -chain by a rearranged endogenous TCR -chain of
a different specificity and thereby escaped negative selection, the
percentage of these cells is not increased by p35 (data not shown). In
summary, these data indicate that negative selection is not prevented
by the inhibition of caspase activity.
During the preparation of this manuscript, Izquierdo and colleagues (48) reported their studies of p35-expressing mouse lines that were generated using a similar transgenic construct. Their p35 protein expression levels appear comparable with our high-expressing mouse lines, as judged from signals on Western blots using p35 Abs from the same source. Consistent with our results, Izquierdo et al. (48) demonstrated that the p35 transgene did not perturb T lymphocyte development and homeostasis, but conferred resistance to anti-Fas-induced apoptosis. They also find less significant protection from CD3 cross-linking, dexamethasone, UV, and several additional inducers of apoptosis. A moderate degree of protection from spontaneous cell death by p35 in vitro could be dramatically improved by the addition of zVAD-FMK, indicating residual caspase activity in the p35 transgenic cells. However, their effort to investigate the effect of p35 on negative selection led them to a different conclusion. Apoptosis of DP cells was reduced in the F5 TCR transgenic model and upon exogenous superantigen stimulation in the presence of p35 when low to moderate Ag concentrations were applied. This result suggested to the authors that p35 could block negative selection. However, they report that p35 expression did not inhibit thymocyte deletion induced by high Ag concentrations and/or by chronic Ag treatment, nor was negative selection by endogenous superantigen blocked (48).
In view of these and our own results, we believe that the inhibition of caspase activity by p35 or by zVAD-FMK can reduce the spontaneous apoptosis of thymocytes, providing an advantage for survival even in the presence of deleting Ags. Nonetheless, the relative response to negative selection remains intact. In lck-p35 transgenic mice, persistent TCR activation by endogenous or chronic Ag stimulation probably causes the down-regulation of the lck proximal promoter and consequently reduced p35 expression levels, but even a persistently high concentration of zVAD-FMK only affects spontaneous cell death, not negative selection.
The relative improvement of thymocyte survival by zVAD-FMK, which we observe for every given peptide concentration, may be consistent (although less pronounced) with the results by Clayton et al. (8), in which zVAD-FMK protected DP cells in FTOC from apoptosis induced by a single dose of anti-CD3 Ab, dexamethasone, or specific Ag peptide. However, in the absence of a titration of the apoptotic stimuli in FTOC, it is difficult to judge whether these results reflect an intrinsic difference between FTOC and adult thymocytes.
Implications of the results for caspase inhibition for the mechanism of TCR-induced apoptosis
Caspase inhibitors block Fas-induced cell death and suppress
morphological hallmarks of apoptosis. However, evidence is accumulating
that a variety of apoptotic stimuli induces alterations in
mitochondrial function that suffice to cause cell death in the absence
of caspase activation (reviewed in Ref. 49). For example,
the recently cloned mitochondrial apoptosis-inducing factor (AIF)
induced the loss of plasma membrane polarity, nuclear chromatin
condensation, and DNA degradation to large (50-kb) fragments despite
caspase inhibition (50). Moreover, the mitochondrial
release of cytochrome c, an important electron carrier in
the respiratory chain, can be observed upon UV or staurosporin-induced
apoptosis even if caspase activity is blocked (51). The
resulting disruption of the mitochondrial respiratory chain may, per
se, be sufficient to cause cell death. Very recent studies indicate
that activation-induced cell death of peripheral T cells cannot be
prevented by caspase inhibitors, while protection of these cells from
reactive oxygen species inhibited apoptotic changes, e.g., the loss of
the mitochondrial transmembrane potential (52).
In summary, we have shown that the caspase inhibitors p35 and zVAD-FMK may delay, but do not block the TCR-induced loss of the mitochondrial transmembrane potential and subsequent apoptosis. Although we cannot formally exclude residual caspase activity even under conditions when no enzyme activity was measurable, the caspase-dependent morphologic characteristics of apoptosis, such as plasma membrane depolarization and oligonucleosomal DNA fragmentation, were efficiently suppressed. Although caspase activation may normally contribute to apoptosis in negative selection and death by neglect, our results clearly indicate that other caspase-independent pathways are sufficient to cause thymocyte death upon TCR stimulation and to eliminate nonfunctional or autoreactive T cells.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Petra Doerfler, Merck Research Laboratories, P.O. Box 2000, Rahway, NJ 07065.
3 Abbreviations used in this paper: DN, double negative; Ac-DEVD, Acetyl-Asp-Glu-Val-Asp; AMC, amino-methyl-coumarin; BIO, biotinylated; DiSC3(5), 3,3'-dipropylthiadicarbocyanine iodide; DP, double positive; FSC, forward scatter; FTOC, fetal thymic organ culture; LMC, littermate control; RAG, recombination-activating gene; SA-TRI, tricolor-conjugated streptavidin; SSC, side scatter; zVAD-FMK, benzyloxycarbonyl-Val-Ala-Asp-fluoromethylketone; m, mitochondrial transmembrane potential; 7AAD, 7-amino-actinomycin D.
Received for publication December 6, 1999. Accepted for publication February 7, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. D. Liu, C. C. Whiting, T. Tomassian, M. Pang, S. J. Bissel, L. G. Baum, V. V. Mossine, F. Poirier, M. E. Huflejt, and M. C. Miceli Endogenous galectin-1 enforces class I-restricted TCR functional fate decisions in thymocytes Blood, July 1, 2008; 112(1): 120 - 130. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Diaz-Guerra, R. Vernal, M. J. del Prete, A. Silva, and J. A. Garcia-Sanz CCL2 Inhibits the Apoptosis Program Induced by Growth Factor Deprivation, Rescuing Functional T Cells J. Immunol., December 1, 2007; 179(11): 7352 - 7357. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Bras, V. J. Yuste, G. Roue, S. Barbier, P. Sancho, C. Virely, M. Rubio, S. Baudet, J. E. Esquerda, H. Merle-Beral, et al. Drp1 Mediates Caspase-Independent Type III Cell Death in Normal and Leukemic Cells Mol. Cell. Biol., October 15, 2007; 27(20): 7073 - 7088. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Woelfel, J. Bixby, M. A. Brehm, and F. K.-M. Chan Transgenic Expression of the Viral FLIP MC159 Causes lpr/gld-Like Lymphoproliferation and Autoimmunity J. Immunol., September 15, 2006; 177(6): 3814 - 3820. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Chen, Y.-H. Wang, Y. Wang, L. Huang, H. Sandoval, Y.-J. Liu, and J. Wang Dendritic cell apoptosis in the maintenance of immune tolerance. Science, February 24, 2006; 311(5764): 1160 - 1164. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. K. Nussbaum and J. L. Whitton The Contraction Phase of Virus-Specific CD8+ T Cells Is Unaffected by a Pan-Caspase Inhibitor J. Immunol., December 1, 2004; 173(11): 6611 - 6618. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S.-C. Chen, C.-C. Huang, C.-L. Chien, C.-J. Jeng, H.-T. Su, E. Chiang, M.-R. Liu, C. H. H. Wu, C.-N. Chang, and R.-H. Lin Cross-linking of P-selectin glycoprotein ligand-1 induces death of activated T cells Blood, November 15, 2004; 104(10): 3233 - 3242. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y.-W. He Orphan nuclear receptors in T lymphocyte development J. Leukoc. Biol., September 1, 2002; 72(3): 440 - 446. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Ohara, K. Koyama, Y. Kusunoki, T. Hayashi, N. Tsuyama, Y. Kubo, and S. Kyoizumi Memory Functions and Death Proneness in Three CD4+CD45RO+ Human T Cell Subsets J. Immunol., July 1, 2002; 169(1): 39 - 48. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Hara, A. Takeda, M. Takeuchi, A. C. Wakeham, A. Itie, M. Sasaki, T. W. Mak, A. Yoshimura, K. Nomoto, and H. Yoshida The Apoptotic Protease-Activating Factor 1-Mediated Pathway of Apoptosis Is Dispensable for Negative Selection of Thymocytes J. Immunol., March 1, 2002; 168(5): 2288 - 2295. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J.-H. Lai, L.-J. Ho, K.-C. Lu, D.-M. Chang, M.-F. Shaio, and S.-H. Han Western and Chinese Antirheumatic Drug-Induced T Cell Apoptotic DNA Damage Uses Different Caspase Cascades and Is Independent of Fas/Fas Ligand Interaction J. Immunol., June 1, 2001; 166(11): 6914 - 6924. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Foghsgaard, D. Wissing, D. Mauch, U. Lademann, L. Bastholm, M. Boes, F. Elling, M. Leist, and M. Jaattela Cathepsin B Acts as a Dominant Execution Protease in Tumor Cell Apoptosis Induced by Tumor Necrosis Factor J. Cell Biol., May 21, 2001; 153(5): 999 - 1010. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. C. Y. Wang, A. Thern, A. Denzel, J. Kitson, S. N. Farrow, and M. J. Owen DR3 Regulates Negative Selection during Thymocyte Development Mol. Cell. Biol., May 15, 2001; 21(10): 3451 - 3461. [Abstract] [Full Text]
|
|
|
|
|
|
R. D. Pettersen, G. Bernard, M. K. Olafsen, M. Pourtein, and S. O. Lie CD99 Signals Caspase-Independent T Cell Death J. Immunol., April 15, 2001; 166(8): 4931 - 4942. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
