| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Division of Immunology and Cancer Research Laboratory, Department of Molecular and Cell Biology, University of California, Berkeley, CA 94720
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and 
genes on mouse chromosome 14. We have
investigated the role of Dad1 during T cell development and activation.
We observe that endogenous Dad1 levels are modulated during T cell
development to reach maximal expression in mature thymocytes.
Transgenic mice that overexpress Dad1 in both the thymus and peripheral
immune system have been generated. Apoptosis of thymocytes from such
mice is largely unaffected, but peripheral T cells display
hyperproliferation in response to stimuli. Therefore, the linkage
between the TCR and Dad1 genes may have important consequences for T
cell function. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and Dad1 are linked in the chicken, mouse, and human
genomes (1, 2). Although TCR is expressed exclusively
in T cells of the immune system (3, 4), Dad1 is expressed
at varying levels in all tissues (1). On mouse chromosome
14, these two genes are separated by 12 kb that include regulatory
elements important for TCR expression, including a locus control
region (LCR)4
(5). This LCR is unusual in that it is comprised of DNase
I hypersensitive sites (HS) which form in distinct patterns in all
tissues tested (1). Because the presence of HS in
chromatin of a particular cell type is normally associated with
transcriptional activity in this cell type, we have suggested that the
region of genomic DNA between TCR and Dad1 is involved in regulation
of Dad1 expression. The nature of Dad1 expression in T cells is of
particular interest because this is a cell type in which both genes are
active. If Dad1 shares regulatory sequences with TCR, it is possible
that they are expressed similarly in T cells. Given the data that Dad1 can act to prevent apoptosis, its role in apoptosis-prone T cells becomes more intriguing. Experiments with yeast and the hamster cell line BHK have shown that Dad1 encodes a gene required for viability of these cells (6, 7), with the loss of Dad1 in BHK cells leading to death via apoptosis (6). Furthermore, both yeast and BHK cells have incomplete N-linked glycosylation due to the lack of Dad1 (7, 8). As a subunit of the oligosaccharyltransferase enzyme (OST) complex (7, 9), Dad1 is critical for establishing normal levels of glycosylation and has been postulated to play a regulatory role in the OST. Although Caenorhabditis elegans overexpressing Dad1 have been shown to contain cells rescued from apoptosis (10), Dad1 has yet to be overexpressed in mammalian systems. In particular, the importance of Dad1 in T cell function has not been previously addressed in vivo. We have asked what the effect of increased Dad1 levels is on T cells by overexpressing Dad1 under the T cell-specific lck proximal promoter. Mice transgenic for lck-Dad1 exhibit normal T cell development but have a substantially heightened response to mitogens.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
RNA was made from adult thymocytes sorted for CD4 and CD8 expression using Trizol reagent (Tel-Test, Friendswood, TX). Poly(A)+ RNA was selected on oligo(dT) minicolumns (Qiagen, Chatsworth, CA), and cDNA was then made using Life Technologies/BRL (Gaithersburg, MD) Superscript II reverse-transcriptase. Dilutions of cDNA were used as templates in PCRs containing [32P]dCTP and primers against either GAPDH (11) or Dad1 (see below). Reactions were run onto polyacrylamide gels, dried, and exposed to PhosphorImager (Molecular Dynamics, Sunnyvale, CA) cassettes for quantitating and normalization.
Antibodies
We generated Abs against Dad1 as has been described previously (6), with an additional purification over a peptide-affinity column. Abs used in flow cytometry were anti-CD4, anti-CD8, anti-CD3, anti-TCRß, anti-IgM, and anti-B220 (Caltag, South San Francisco, CA). Flow cytometry was performed on a Coulter (Palo Alto, CA) EPICS XL-MCL.
Mice
The transgenic construct consists of the Dad1 cDNA library clone
insert introduced into the BamHI site of the lck proximal
promoter vector, p1017 (12). Transgenic founders were
produced via standard oocyte microinjection and were identified by
Southern blotting of tail DNA. The Dad1 probe (Fig. 1
) is an
EcoRI-PstI fragment from exon 1 of Dad1. It
recognizes a 1.1-kb EcoRI/BamHI genomic fragment
as well as the 0.5-kb transgenic cDNA. Subsequent litters
were routinely typed by PCR with the following primers: ORFDad1F,
CTGAAGTTGCTGGACGCCTATC; and ORFDad1R, GACGACAAGGTGCAGGATCG. PCR was
conducted using 1.0 mM MgCl2 at a 58°C
annealing temperature. Northern blotting was performed on
RNA isolated with Trizol (Boehringer Mannheim, Indianapolis, IN) using
the same Dad1 exon 1 cDNA probe as above. F5 mice transgenic for
anti-influenza TCR have been described (13). These
mice were genotyped with the following primers: F5F,
GCAGAACCAACAAATGCTGGTGTC; and F5R, GCCAAGCACACGAGGGTAGC. PCR was
conducted as above, except using 1.5 mM MgCl2.
Fifty nanomoles of influenza nucleoprotein (NP) peptide was injected
i.p. for 4 days before analysis.
|
A total of 106 cells were added to each well of triplicate samples on a 96-well plate together with the indicated stimuli in a total of 200 µl of complete RPMI 1640 medium and incubated 24 to 48 h. One microliter of [3H]thymidine (1 µCi, DuPont/NEN, Boston, MA) was added to each well 8 h before analysis with a 96-well plate harvester (Inotech, Rockville, MD) and plate reader (Packard (Meriden, CT) Matrix 9600). To purify lymph node T cells, B cells were depleted with anti-B220 Dynabeads (Dynal, Great Neck, NY) and confirmed by flow cytometry to be >97% pure. Ascites were produced at the University of California, Berkeley (anti-CD3 clone 500A2 and anti-CD28 clone 37.51). PMA and A23187 were from Sigma (St. Louis, MO).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Dad1 and TCR are separated by a set of DNase I hypersensitive
sites that can either drive ubiquitous expression or T cell-specific
expression of a linked transgene depending on their context within the
transgene (5, 14). Therefore, it is possible that Dad1 and
TCR are regulated by a common set of elements and may in turn share
a similar expression pattern in T cells. We have used semiquantitative
RT-PCR and immunoblotting to compare Dad1 levels with the previously
characterized expression pattern of TCR. Thymocytes from 4- to
6-wk-old mice were stained for the developmental markers CD4 and CD8
and sorted according to their expression profiles. In early T cell
precursors that lack expression of CD4 and CD8 (double negatives or
DN), TCR is not expressed at detectable levels. In contrast, Dad1
RNA (Fig. 1
A) and protein
(Fig. 1B) is detected in this population. TCR expression
becomes detectable in thymocytes doubly positive for CD4 and CD8 (DP),
whereas Dad1 levels decrease relative to DN levels. Finally, expression
in the most mature
CD4+CD8- or
CD4-CD8+ (single positives
or SP) T cells is highest for both TCR and Dad1. During the
transition from DP to SP, Dad1 levels parallel those of TCR,
supporting the idea that Dad1 and TCR may share regulatory elements
during this last stage of T cell development.
T cells of lck-Dad1 transgenic mice
To ask whether increasing levels of Dad1 in T cells would
influence their development and/or function, we constructed a transgene
using the lck proximal promoter and the Dad1 cDNA (Fig. 2A). We generated six
transgenic founders of varying copy numbers, five of which established
lines. To determine whether thymocytes from mice transgenic for the
lck-Dad1 construct expressed elevated levels of Dad1 relative to
nontransgenic littermates, RNA from three lines of transgenic mice were
probed with the Dad1 cDNA (Fig. 2B). The transgenic Dad1
message, some of which is larger than endogenous message due to the
human growth hormone minigene contained within the transgenic
construct, is present in the thymus. In the peripheral immune system,
endogenous lck transcription from the proximal promoter is
down-regulated, but transgenic Dad1 message is nevertheless detectable
in spleen and lymph node RNA. In non-T cell bearing tissues, we
observed no Dad1 signal from the transgene (data not shown).
|
A), as did CD3, CD44, CD25,
and CD69 levels (data not shown). Transgenic mice also had thymus cell
numbers within the normal range and without consistent difference from
nontransgenic littermates (Fig. 3B). To confirm that
increased Dad1 levels do not grossly affect T cell development, we
looked in mice transgenic for a clonal TCR by crossing lck-Dad1 mice to
mice transgenic for the F5 TCR, directed against influenza NP and
Db (13, 15). The number of T cells
in F5/lck-Dad1 transgenic mice was not significantly different from
those of F5 littermates, again implying that an increase in Dad1 levels
has no obvious effect on T cell development (Fig. 3, B and
D). The vast majority of developing thymocytes undergoes
apoptosis due to lack of positive selection, a situation referred to as
death by neglect. Because we see normal T cell development in mice
overexpressing Dad1, these neglected cells are evidently not rescued by
Dad1. In vitro culturing of thymocytes also results in death by
neglect; we observed a similar rate of such apoptosis between
transgenic and nontransgenic thymocyte cultures (Fig. 3C).
In the periphery, lck-Dad1 mice and F5/lck-Dad1 mice also appear to
contain normal numbers of T cells with typical ratios of mature
CD4+ cells to CD8+ cells
(data not shown, Fig. 3, B and D). Therefore, in
the absence of stimuli, the immune system does not appear to be
perturbed by increasing levels of Dad1.
|
In addition to death by neglect, T cells undergo apoptosis in
response to a variety of stimuli. We asked whether overexpressed Dad1
protects T cells from some of these death-inducing signals. For
example, thymocyte apoptosis can also arise from a negative selection
signal, which is received through an autoreactive TCR. Administration
of influenza NP peptide to F5 transgenic mice mimics this signal and
causes immature F5 TCR+ thymocytes to undergo
apoptosis (13, 15). In the presence of overexpressed Dad1,
overall thymocyte numbers are increased 2- to 3-fold in
peptide-injected mice (Fig. 4,
A and B). However, this is not due to a rescue of
apoptosis-sensitive DP cells, but instead is mostly due to an increase
in CD8+ cells in the Dad1 transgenic mice (Fig. 4A, upper-right panel). Because DP cells are not
rescued, it is likely that overexpressed Dad1 does not block Ag-induced
negative selection. Instead, the increased numbers of
F5+ cells in the doubly transgenic mouse suggest
that the response of mature T cells to Ag is increased when Dad1 is
overexpressed. Consistent with this idea is the increased expansion of
F5+ mature T cells in the periphery upon
NP-injection of F5/lck-Dad1 mice vs F5 mice (Fig. 4A, lower
panels, and Fig. 4B). The greater cellularity is
probably not due to Dad1-mediated rescue from cell death, as NP-induced
activation is not accompanied by significant apoptosis in the spleen
(16).
|
Because the in vivo situation is complicated by T cell homing
during the NP response (16), we wished to examine more
precisely the activation of T cells in lck-Dad1 mice in vitro. In the
first series of experiments, cells were isolated from the spleen or
lymph nodes of littermate mice and treated with dilutions of
anti-CD3 Ab. This type of short-term culturing of naive T cells
does not induce significant amounts of apoptosis (17) and
therefore allows a direct measure of proliferative capacity. As
measured by [3H]thymidine uptake, cells from
transgenic mice had a greater capacity to proliferate than those
isolated from nontransgenic littermates (Fig. 5A). To show that the
increased proliferation was intrinsic to T cells overexpressing Dad1,
we purified T cells from the lymph node (Fig. 5B) and also
tested thymocytes, either unpurified or sorted for mature SP cells
(data not shown). In response to anti-CD3 and anti-CD28, we
again saw a greater extent of proliferation in transgenic samples.
Bypassing cell-surface signaling by stimulating T cells with PMA and
calcium ionophore results in comparable activation of transgenic and
nontransgenic T cells (Fig. 5C), consistent with the idea
that the effect of increased Dad1 expression acts at the level of
cell-surface signaling via N-glycosylated proteins.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
gene and the
ubiquitously expressed Dad1 gene has been conserved through evolution
between the avian and mammalian genomes (2). As the
receptor that recognizes foreign Ags in the context of the MHC, the
well-studied TCR is central in defining the functionality of the T cell
that expresses it. Therefore, it is not surprising that TCR is
required for the normal development of ß T cells, and TCR gene
expression is tightly regulated during this development. TCR is
first expressed at low levels in immature
CD4+CD8+ DP thymocytes. The
mature CD4+CD8- or
CD4-CD8+ SP thymocytes
instead express high levels of TCR. In the absence of TCR, the
majority of T cells arrest at the DP stage (18),
demonstrating that a requisite process during T cell development is the
proper expression of TCR upon positive selection.
Compared with TCR, much less is known about the function and regulation
of Dad1. Originally identified based on its loss-of-function phenotype
in a hamster cell line (6), this so-called "defender
against apoptotic death" was subsequently shown to have homology to a
subunit of the OST enzyme complex and to be required for proper levels
of N-linked glycosylation (8). Subsequent
biochemical studies have firmly placed Dad1 in the OST complex
(7, 9). Because embryos lacking Dad1 arrest before
development of the immune system (N. A. Hong, unpublished data),
nothing is known about the requirement for Dad1 in T cells. However,
its roles in N-linked glycosylation and potentially
apoptosis, along with its proximity to regulatory elements controlling
TCR expression, have prompted us to investigate Dad1 regulation and
function in developing and mature T cells.
We have found that Dad1 expression in the thymus precedes that of
TCR, being detectable in the earliest
CD4-CD8- DN compartment
of immature thymocytes. However, once TCR expression begins in DP
cells, we see that Dad1 is similarly up-regulated during the transition
from DP to SP. It is likely that this complex gene regulation involves
the sequences that separate the two genes. This region of DNA has LCR
activity and drives proper expression of TCR in T cells
(5). When a portion of this DNA is linked to a ß-globin
promoter/enhancer construct, ubiquitous expression is observed
(14). Furthermore, the chromatin encompassing these
sequences contains DNase I hypersensitive sites that are present in all
tissues, in distinct patterns for each cell type, strengthening the
idea that these regulatory sequences are involved in Dad1 expression
(1). Therefore, it will be very interesting to identify
the mechanisms acting at these sequences that allow differential
expression of TCR and Dad1, especially in B cells and DN T cells
where somatic rearrangement at other loci occurs while TCR remains
inaccessible. Investigating the promoter region of Dad1 may shed light
on these questions, and identifying tissue- and stage-specific
transacting factors that bind to the LCR may provide additional
insight.
We hypothesized that the up-regulation of Dad1 between the DP and SP stages of T cell development could have functional significance for at least two reasons. First, the lower levels of Dad1 in DP cells could correlate with the propensity of DP cells to undergo apoptosis, as has been shown for another anti-apoptosis molecule, bcl-2 (19, 20, 21, 22, 23). Alternatively, the higher levels of Dad1 in SP may indicate that, similar to TCR, Dad1 is important for mature T cell function. Although a T cell-specific knock-out of the Dad1 gene would definitively address these possibilities, the transgenic mice described here that overexpress Dad1 in both developing and mature T cells ask whether Dad1 is sufficient to influence either T cell apoptosis or function. First, we have yet to find a pronounced effect on apoptosis of immature thymocytes. Overexpression of the anti-apoptosis genes bcl-2 or bcl-xL leads to greater survival of thymocytes, both in vivo which results in an enlarged population of mature T cells and in vitro in response to apoptosis-inducing stimuli (23, 24, 25, 26). In contrast, lck-Dad1 mice exhibit relatively normal T cell development, and transgenic thymocytes do not have a striking survival advantage when cultured or challenged with self-Ag. In addition, we have observed normal apoptotic responses of transgenic thymocytes to in vitro anti-Fas stimulation and dexamethasone (data not shown). Although these are negative results and are therefore subject to various caveats, in general, our findings with mouse Dad1 contrast with the situation in C. elegans, where overexpression of Dad1 via a heat-shock construct was shown to rescue cells normally destined for apoptosis (10). It is possible that this disparity is due to the different cell types analyzed, and that the regulation of apoptosis is not the primary role of Dad1 in every cell context. This is consistent both with the observation that not all programmed cell deaths were averted in the heat-shock-Dad1 C. elegans (10), and with the fact that mouse embryos lacking Dad1 exhibit increased apoptosis but only in certain regions (N. A. Hong and A. Winoto, unpublished data). The effect of changes in levels of Dad1 will thus likely depend on the presence of particular glycoproteins that are sensitive to N-linked glycosylation levels for expression and/or function. The roles of these glycoproteins will then determine the outcome of Dad1 regulation.
In the case of T cells, we have found that the process of activation is
sensitive to altered levels of Dad1. We have observed a striking
increase in the ability of mature T cells to respond to TCR stimuli in
lck-Dad1 mice. In vivo, this results in much greater numbers of
activated splenic T cells in F5/lck-Dad1 mice, suggesting that levels
of Dad1 are sufficient to influence T cell response. In vitro,
anti-TCR stimulation similarly results in a greater extent of
proliferation of T cells from lck-Dad1 mice compared with those from
nontransgenic littermates. We can therefore say that Dad1 serves a
complementary role to TCR in T cells by enhancing the ability of T
cells to react when TCR is activated. Having the two genes linked in
expression would then be an efficient mechanism for ensuring proper T
cell responses once maturation is complete. Intriguingly, a gene
complex that encodes a family of proteases strongly expressed in
activated cytotoxic T cells is also linked to the TCR/ genes, but
on the opposite end of the locus from Dad1 (27).
If we bypass TCR signaling with PMA and calcium ionophore when
stimulating T cells from lck-Dad1 mice, proliferation is equivalent to
wild type, as would be predicted if Dad1 is acting by increasing
N-linked glycosylation of cell-surface molecules. There are
at least two potential molecular mechanisms by which increased
N-linked glycosylation could result in stronger T cell
activation. lck-Dad1 mice may exhibit enhanced expression of relevant
proteins: for instance, TCR has been shown to require
N-linked glycosylation for optimal cell-surface expression
(28). Although we have yet to find a consistent increase
in levels of known N-linked glycosylated signaling molecules
such as TCR itself, increased N-glycosylation could
stabilize receptor complexes and lead to prolonged TCR signaling
without having a discernible effect on steady-state cell-surface
protein levels. Alternatively, more N-glycosylation could
influence the function of various glycoproteins through other
mechanisms, perhaps by increasing adhesiveness or ease of triggering.
It is interesting to consider that increased levels of
O-linked glycosylation have precisely the opposite effect on
T cell activation, leading to a dampened proliferative response in mice
expressing the core 2
ß-1,6-N-acetylglucosaminyltransferase (C2GnT) under the
same lck promoter used in our study (29). How the two
types of protein modification by carbohydrate attachment can lead to
opposite biological effects will be very interesting to explore,
especially if one can establish whether this is a pleiotropic effect on
many relevant proteins or instead is accomplished through effects on a
small class of signaling molecules. We favor the idea that it will be a
relatively small group of pivotal proteins regulated by glycosylation,
as altering either N- or O-linked glycosylation
appears to have a very specific effect on mature T cell activation and
not T cell development. The lck-Dad1 mice will be useful tools
in dissecting these molecular details in T cells, and
furthermore illustrates a striking phenomenon that may relate to the
conserved linkage between two nonhomologous co-regulated genes, Dad1
and TCR.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Current address: Hooper Foundation, University of California, San Francisco, CA 94143-0552.
3 Address correspondence and reprint requests to Dr. Astar Winoto, Division of Immunology and Cancer Research Laboratory, Department of Molecular and Cell Biology, University of California, Berkeley, CA 94720-3200. E-mail address:
4 Abbreviation used in the paper: LCR, locus control region; Dad1, defender against apoptotic death-1; OST, oligosaccharyltransferase; NP, nucleoprotein; DN, double negative; DP, double positive; SP, single positive.
Received for publication April 6, 1999. Accepted for publication June 8, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
/ locus impairs T-cell development and reveals the presence of the nearby antiapoptosis gene Dad1. Mol. Cell. Biol. 17:2151.[Abstract]
chain constant region. Immunogenetics 46:376.[Medline]
/ locus. Immunity 1:207.[Medline]
and ß block thymocyte development at different stages. Nature 360:225.[Medline]
- and -chain genes on mouse chromosome 14. Genomics 1990:252.
This article has been cited by other articles:
|
|
 |
|
  J. Gomos-Klein, F. Harrow, J. Alarcon, and B. D. Ortiz CTCF-Independent, but Not CTCF-Dependent, Elements Significantly Contribute to TCR-{alpha} Locus Control Region Activity J. Immunol., July 15, 2007; 179(2): 1088 - 1095. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Magdinier, T. M. Yusufzai, and G. Felsenfeld Both CTCF-dependent and -independent Insulators Are Found between the Mouse T Cell Receptor {alpha} and Dad1 Genes J. Biol. Chem., June 11, 2004; 279(24): 25381 - 25389. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. D. Ortiz, F. Harrow, D. Cado, B. Santoso, and A. Winoto Function and Factor Interactions of a Locus Control Region Element in the Mouse T Cell Receptor-{alpha}/Dad1 Gene Locus J. Immunol., October 1, 2001; 167(7): 3836 - 3845. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Virtaneva, F. A. Wright, S. M. Tanner, B. Yuan, W. J. Lemon, M. A. Caligiuri, C. D. Bloomfield, A. de la Chapelle, and R. Krahe Expression profiling reveals fundamental biological differences in acute myeloid leukemia with isolated trisomy 8 and normal cytogenetics PNAS, January 30, 2001; 98(3): 1124 - 1129. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
