| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Ben May Institute for Cancer Research and
Department of Medicine, University of Chicago, Chicago, IL 60637;
Laboratory for Molecular Biology, Gene Center University of Munich, Munich, Germany; and
§
Laboratory of Molecular Microbiology, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, MD 20892
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and
CD3
and a lack of activation of the tyrosine kinases
p56lck and ZAP-70. (17, 20, 21).
Although there are differences, both signaling pathways converge at the
level of IL-2 gene expression, resulting in reduced IL-2 mRNA
levels and IL-2 protein production (21, 22). Impaired
signaling through the IL-2R is the key event in the induction of anergy
(23, 24). Thus, despite the different signaling events in
the costimulation-deficient vs the TCR-antagonistic pathway, both
treatments result in anergy. By comparison, a common requirement for
anergy induction through either pathway is a dependence upon an
intracellular calcium flux, because EGTA or cyclosporin A blocks anergy
induction in either model (16, 25, 26). In an attempt to understand how anergic T cells maintain their unresponsive state, several groups have analyzed the molecular defects in anergic Th1 clones. The results have been summarized in recent reviews (27, 28). The current models suggest that T cell anergy is a consequence of an alteration in the normal cascade of signaling events, ultimately leading to impaired IL-2 production. However, previous studies have not addressed a role for novel proteins that might be induced during the establishment of anergy and regulate cell function. We reasoned that the existence of proteins that are specifically induced in anergic T cells is likely, because anergy is long lasting and results in multiple changes in T cell function (27, 28). Genes associated with the anergic state could be important in the maintenance of anergy. Alternatively, these cells may express proteins that regulate cell survival or migration, because anergic T cells persist for weeks in vivo or in vitro (29, 30, 31), but are unable to home toward germinal centers (32). Thus, we employed the differential display technique (33) to compare the gene expression in anergic vs responsive murine Th1 clones and to identify novel molecules that might be crucial to the function of these cells.
In this report we describe the identification and characterization of a gene that is specifically induced in anergic T cells, namely general receptor of phosphoinositides 1 (GRP1)3 (34), a recently published nucleotide exchange factor for the small G protein ADP ribosylation factor-1 (ARF1) (35). We demonstrate that GRP1 localizes to the plasma membrane in lymphoid cells and regulates LFA-1/ICAM-1-mediated adhesion. Finally, we show that T cells expressing a retrovirally transduced GRP1 construct proliferate normally and display no altered cytokine production, but are not stable and rapidly lose GRP1 expression.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
DO11.10 mice, transgenic for an
OVA323–339-specific 
ß-TCR, were a gift from
Drs. D. Loh and K. Murphy (Washington University, St. Louis, MO)
(36). DBA/2 mice and B10.A mice were purchased at the age
of 6–8 wk from the National Cancer Institute, Frederick Cancer
Research and Developmental Center (Frederick, MD). The animals were
bred and maintained in a specific pathogen-free barrier facility.
Medium and Abs
All murine cells were maintained and stimulated in complete DMEM
containing 5% FCS. Human cell lines were maintained in supplemented
RPMI 1640 containing 10% FCS. The hybridoma secreting anti-MHC
class I (H-2Ld) mAb, 30-5-7S, was obtained from
the American Type Culture Collection (ATCC; Manassas, VA). 145-2C11, a
hamster anti-mouse CD3 mAb (37), was purified on
protein A-Sepharose from hybridoma supernatant and dialyzed in PBS. The
anti-CD3-IgG3 is a chimeric Ab combining the Ag specificity of
145-2C11 with a murine IgG3 Fc portion with weak affinity for murine Fc
receptors (38). The mAb was used as an ascites. A rabbit
anti-GRP1 reactive polyclonal antiserum was generated by HTI
Bio-Products (Ramona, CA) using recombinant GRP1 obtained by
PreScission Protease (Pharmacia, Uppsala, Sweden)-mediated cleavage of
a GST-GRP1 fusion protein harvested from transformed Escherichia
coli. The purified hyperimmune rabbit Abs were purified on protein
A-Sepharose and used for Western blotting. A purified rabbit
anti-actin serum was purchased from Sigma (St. Louis, MO). Blocking
anti-CD11a and anti-CD18 Abs MEM-30 and MEM-48
(39) were gifts from Dr. V. Horejsi (Institute of
Molecular Genetics, Prague, Czech Republic). The anti-IL-4
Ab-secreting hybridoma 11B11 (40) was purchased from the
ATCC. The Ab was used as concentrated culture supernatant. The
anti-IL-12 p40 mAb C15.6 was a gift from Dr. R. Hendricks,
University of Pittsburgh (Pittsburgh, PA). The PE-coupled
anti-murine CD4 mAb (CD4-PE), clone GK1.5, was obtained from
Southern Biotechnology Associates (Birmingham, AL).
Cells
The OVA-reactive murine Th1 clones pGL2 and pGL10 and the Th2
clones pL3 and pL104 have been described previously (41, 42). The CD8+CD4- T
cell line, 3.7.2, was provided by P. Fields (University of Chicago,
Chicago, IL). The 3.7.2 cell line was generated by multiple
restimulation of lymph node T cells from 2C TCR transgenic mice
(43) as described previously (44). The
CD8+ T cells were found by ELISA to secrete
IFN-
upon activation, but not IL-4 (data not shown).
The OVA-reactive clones were maintained as follows. T cells (3 x 106) were expanded on 108 irradiated (20 Gy) E-depleted splenocytes from DBA/2 mice in the presence of 0.2 mg/ml OVA (Sigma) and 12.5 U/ml recombinant human IL-2 (a gift from Cetus, San Francisco, CA). The 3.7.2 cells were expanded on 108 irradiated (20 Gy) E-depleted splenocytes from DBA/2 mice. The pigeon cytochrome c-reactive murine Th1 clone, A.E7, was expanded as previously described (45). The human cell lines HeLa (adenocarcinoma), U937 (monocytoma), Jurkat E6 (T lymphoma), and U266BL (EBV-transformed B myeloma), and the retroviral packaging cell line Phoenix-eco (http://www.stanford.edu/group/nolan/NL-phoenix.html) were obtained from the ATCC.
Anergy models
To induce anergy with the anti-CD3 mAb, T cell clones were cultured for 18–24 h in plates coated with 10 µg/ml 145-2C11, except for A.E7, which was anergized on plates coated with 1 µg/ml. After this induction period, cells were transferred into uncoated plates and incubated for a variable rest period. Control cells were incubated in uncoated plates for the same total amount of time or were activated according to the maintenance protocol.
As a separate model, T cell anergy was induced on chemically modified splenocytes as previously published (30). Briefly, irradiated (30 Gy) E-depleted B10.A splenocytes were treated with the chemical cross-linker 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (ECDI) (Sigma), washed intensively, and used to anergize A.E7 cells in the presence of 1 µM peptide Ag DASP (NH2-KKANELIAYLKQATK-COOH, a gift from Dr. Marc Jenkins, University of Minnesota, Minneapolis, MN). The APC to A.E7 ratio was 7.5:1. As a control, A.E7 cells were activated on irradiated splenocytes that had not been treated with ECDI in the presence of the DASP Ag. Twenty hours later, A.E7 cells were separated from the APC by density gradient centrifugation on Ficoll-Hypaque and rested for another day before being lysed for RNA or protein analyses.
In the third model, T cell anergy was induced on I-Ad-transfected EL-4 cells (ELAD) as described previously (46). Briefly, pGL10 cells were incubated with mitomycin C (Sigma)-treated ELAD cells in a 1:1 cell ratio in the presence of 0.1 µg/ml OVA323–339 (University of Chicago Cancer Research Center Oligopeptide Synthesis Facility). The pGL10 cells were separated from the dead APC after 20–24 h and rested for an additional 24 h. As a control, pGL10 was activated on ELAD cells that coexpressed the costimulatory ligand B7-1 (ELAD-B7) in the presence of OVA peptide. The expression of the transfected molecules on the cell surface of the APCs was verified by flow cytometry.
Finally, anergy was induced with the non-cross-linking, nonmitogenic anti-CD3-IgG3 in Th1 clones by incubating the cells for 20 h with 1 µg/ml of this mAb. The T cells were washed three times and incubated in growth medium until harvested for cell lysis or restimulation (47).
T clone proliferation assay to verify anergy induction
Cells inactivated by the various methods were evaluated in a restimulation assay on days 4–7 as follows. Rested anergic or control OVA-specific T clones (2–5 x 104) were restimulated in 96-well microtiter plates with 2–5 x 105 E- and T-depleted irradiated (20 Gy) splenocytes from DBA/2 mice in the presence of various amounts of OVA. A.E7 was stimulated at 2–5 x 104 cells/well with 2–5 x 105 E- and T-depleted irradiated (30 Gy) splenocytes from B10.A mice in the presence of various amounts of pigeon cytochrome c (Sigma). Exogenous IL-2 at 20 U/ml was added to selected wells to examine whether the T cells were viable and responsive to growth factor. Thirty-seven kilobecquerels of [methyl-3H]thymidine (Amersham, Arlington Heights, IL) was added to each well for the final 8–14 h of a 60- to 72-h restimulation. Samples were examined in triplicate and harvested, and the emission of beta irradiation was determined using the Packard TopCount System and the MicroScint 0 scintillation solution (Packard Instruments, Meriden, CT). The counts per minute reported is the mean of the triplicate samples. For most experiments described, the restimulation capacity of the anergic cells was reduced between 15- and 40-fold, except for the system using ELAD transfectants, which resulted in a 3- to 10-fold reduction in proliferative activity (data not shown).
RNA isolation and Northern blot analysis
Total RNA was isolated from live cells by applying the cell lysate to a cesium chloride gradient ultracentrifugation (48, 49) or using Trizol (Life Technologies, Gaithersburg, MD) according to the manufacturer’s recommendation. An additional final ethanol reprecipitation step was included. Tissues were derived from a B10.A mouse and lysed in Trizol. Five micrograms of total RNA was loaded onto a 1.2% agarose/1.1% formaldehyde gel along with an RNA size marker (Life Technologies). Northern blotting, hybridization with 32P-labeled cDNA, and stripping of blots were performed as described previously (50). The cDNA probe used for hybridization was a 305-bp GRP1-specific sequence from the 3'-untranslated region obtained after a digest with BamHI and XbaI. This probe did not cross-hybridize with mRNAs representing other GRP1 family members (data not shown). Equal loading and transfer of RNA were verified by UV illumination of the ethidium bromide-stained 28S and 18S ribosomal RNA on the blot. Some autoradiograms (Kodak X-OMAT, Eastman Kodak, Rochester, NY) were quantitatively evaluated using the Computing Densitometer 325E and ImageQuant software (Molecular Dynamics, Sunnyvale, CA). Images of ethidium bromide-stained 18S ribosomal RNA on the Northern blot were quantitated using a UMAX S-12 desktop scanner (Hsinchu, Taiwan) and 1D Image Analysis Software (version 2.0.2, Kodak).
Differential display of reverse transcribed RNA by PCR (DDRT-PCR)
The DDRT-PCR was performed as previously described (33, 51) with the following modifications. Aliquots of 0.5 µg of
total RNA were reverse transcribed with the T11VN
primer at a concentration of 2.5 µM. The T11VN
primer used for PCR amplification was end labeled with T4
polynucleotide kinase (Life Technologies), using
[-33P]TP (Amersham) (51)
instead of including a radioactive nucleotide in the PCR. The PCR
conditions were changed to 1.9 mM MgCl2, 20 µM
dNTP, 2.5 µM of each primer, 50 U/ml Taq polymerase, and 10% of the
cDNA product. The arbitrary primers were decamers. The decamer
sequences were chosen randomly; the only restriction was a 50% GC
content. The PCR cycling parameters were 2 min at 94°C followed by 40
cycles (30 s at 94°C, 2 min at 40°C, and 30 s 72°C) and a
final extension for 5 min at 72°C. Seven percent of each amplificate
was submitted to long distance denaturing PAGE. The primers used for
reamplification of the eluted DNA were NotI site tagged to
efficiently facilitate subcloning of the products. Reamplification
conditions were otherwise as previously described (51).
Primers were synthesized in the University of Chicago Oligonucleotide
Synthesis Facility or were purchased from Integrated DNA Technologies
(Coralville, CA).
Cloning of full-length murine GRP1
The DDRT-PCR reamplificates were cloned into the pBluescript II
SK+ phagemide (Stratagene, La Jolla, CA), and
plasmid DNA from several independent bacterial colonies was sequenced
as well as used to probe Northern blots in an effort to verify the
anergy association. At the time of GRP1 cloning, there was no match in
the nucleotide databases; thus, the GRP1 DDRT-PCR product was used to
screen an NIH-3T3 cDNA library (
ZAP II, Stratagene), and two
independent plaques were isolated and converted into the phagemide
state. Nested deletions of the cDNA clones were sequenced, and
gene-specific nested primers were designed to obtain missing 5'
sequence information by RACE from NIH-3T3
poly(A)+ RNA with the Marathon cDNA amplification
kit and the long distance, high fidelity KlenTaq (Clontech, Palo Alto,
CA) according to the manufacturer’s recommendation. A full-length cDNA
clone was assembled from the largest RACE clone obtained and was
sequenced with overlapping primers. Additional RACE clones were
obtained and sequenced to verify the coding sequence. The complete gene
sequence including the previously unpublished 3'-untranslated region is
available from GenBank under accession number AF084221.
Chromosomal localization of GRP1
Fluorescence in situ chromosomal hybridization to human PHA-stimulated peripheral blood lymphocytes arrested in metaphase was performed as described previously (52). The biotin-labeled GRP1 probe was a human full-length cDNA clone in the pcDNA I/Amp vector (Invitrogen, Carlsbad, CA). This clone was isolated and provided by Dr. Donald Staunton (ICOS, Bothel, WA). Its coding sequence is identical with the GenBank entry for ARNO3, accession number AJ223957. Chromosomes were identified by staining with 4,6-diamidino-2-phenylindoledihydrochloride. Images were obtained using a Zeiss Axiophot microscope (New York, NY) coupled to a cooled charge-coupled device camera. Separate images of 4,6-diamidino-2-phenylindoledihydrochloride-stained chromosomes and the hybridization signal were merged using image analysis software (NU200 and Image 1.57).
For murine chromosome localization GRP1 was mapped by analysis of the progeny of the genetic cross: (NFS/N x Mus spretus) x M. spretus or C58/J (53). Progeny of these crosses have been typed for >1200 markers distributed on all 19 autosomes and the X chromosome. Recombination was calculated according to the method of Green (54), and genes were ordered by minimizing the number of recombinants. The GRP1 probe was a 305-bp GRP1-specific sequence from the 3'-untranslated region obtained after a digest with BamHI and XbaI.
The SDS-PAGE and Western analysis
Cells were lysed in a 0.5% Triton X-100 (Sigma) containing reducing lysis buffer, and 40–70 µg of the total lysate protein, as determined with the Bio-Rad DC kit (Bio-Rad, Hercules, CA), was subjected to SDS-PAGE and subsequently transferred to polyvinylidene fluoride membrane (51). PonceauS staining was performed to control for uniform protein loading. The blots were immunostained with the rabbit anti-GRP1 antiserum and developed with an anti-rabbit HRP-coupled antiserum and the ECL chemiluminescence substrate (Amersham). In some experiments, the blots were reprobed with anti-actin Abs after stripping in a reducing SDS buffer to control for uniform protein loading. Some exposures were quantitated using the Computing Densitometer 325E and the ImageQuant Software (Molecular Dynamics).
Recombinant vaccinia virus
The GRP1, ARNO, or cytohesin-1 cDNA was cloned into the pcIgTKG vaccinia virus expression vector (55). Infection with the recombinant virus directed the expression of chimeric proteins with amino-terminal Ig domains in the infected cells.
Immunofluorescence
HeLa, U937, Jurkat E6, and U266 cells were infected with recombinant vaccinia viruses. Eight hours postinfection cells, were placed on poly-L-lysine-covered microscope slides for 1 h in a humidified chamber at 37°C and processed as described previously (56), except that a 3% (w/v) paraformaldehyde in PBS solution was used.
Cells were examined using a confocal laser scanning microscope (Leica TCS-NT system, Leica) attached to a Leica DM IRB inverted microscope with a PLAPO 63 x 1.32 oil immersion objective. Confocal images were collected as 512 x 512 pixel files and processed with the help of the Photoshop program (Adobe).
Adhesion assay
Jurkat E6 cells were infected with recombinant vaccinia viruses.
Five to six hours after infection, cells were labeled with 12 µg/ml
bisbenzimide H33342 fluorochrome trihydrochloride (Calbiochem, La
Jolla, CA) for 30 min at 37°C, collected by centrifugation,
resuspended in HBSS, and delivered to maxisorp 96-well plates (NUNC,
Roskilde, Denmark) at 1.5 x 105/well.
Before adhesion, plates were coated with goat anti-human IgG Ab
(Fc specific) at 0.85 µg/well for 90 min at 25°C, blocked with
1% (w/v) BSA/PBS, incubated with culture supernatants from COS cells
secreting ICAM-1 receptor-globulin fusion protein (55),
and subsequently used in the assay. Where indicated, cells were
incubated with 40 nM PMA 0.5 h before the adhesion assay. To
determine whether the binding was LFA-1 specific, replicate samples
were set up in the presence of the blocking anti-LFA-1 Abs MEM-30
and MEM-48 (39). Cells were then allowed to adhere for
1 h at 37°C, and unbound cells were carefully washed off three
times with 300 µl of HBSS. Bound cells were assayed in 100 µl 2%
(v/v) formaldehyde/PBS using a fluorescence plate reader (Cytofluor II,
PerSeptive Diagnostics, Cambridge, MA). The signal of 1.5 x
105 cells/well at 490 nm corresponds to 100%
adhesion. Each value is the mean of two determinations conducted in
triplicate.
Retroviral transduction of lymph node cells
The retroviral vector pLXIN (Clontech, Palo Alto, CA), was modified to yield pLXIE by replacing the neomycin resistance gene with the gene encoding enhanced green fluorescence protein (eGFP). Into the EcoRI site of this resulting vector, the 5'-untranslated region and the coding region of GRP1 were cloned to yield pLGIE, which allowed expression of both GRP1 and eGFP from a single, bicistronic transcript in transduced cells. To produce retrovirus-containing supernatant, Phoenix-eco packaging cells were transfected with pLGIE or the control vector pLXIE using the CaPO4 transfection method (51). The supernatant was filtered through a 0.2-µm pore size, syringe filter to separate the virus from contaminating Phoenix-eco cells.
T cells from pooled inguinal, brachial, axillary, superficial cervical,
and mesenteric lymph node cells of a DO11.10 mouse were activated on
OVA-pulsed, irradiated (20 Gy) isogenic spleen feeder cells. The T
cells were cultured normally to avoid skewing toward a particular
pathway, purposefully skewed toward Th1 development using 2 ng/ml
recombinant murine IL-12 (a gift from David Presky, Hoffmann-La Roche,
Nutley, NJ) and 10 µg/ml anti-IL-4 mAb, 11B11, or purposefully
skewed toward Th2 development using recombinant murine IL-4 (BioSource
International, Camarillo, CA) and 10 µg/ml of anti-IL-12 p40 mAb.
Three days later, the activated T cells were infected with
retrovirus-containing supernatant in 15 µg/ml polybrene and cultured
for another 4 days in medium with IL-2. On day 7, live cells were
recovered upon centrifugation on Ficoll/metrizoate, stained with
CD4-PE, and sorted for eGFP and CD4 expression using a
FACStarPlus (Becton Dickinson, San Jose, CA). For
assessment of the population’s stability, the sorted cells were
restimulated under skewing conditions and sorted again 7 days later.
These cells were maintained thereafter by weekly stimulation on
OVA-pulsed spleen feeder cells in the presence of exogenous IL-2 (10
U/ml). The stability of eGFP expression over time was examined 7 days
after each stimulation by flow cytometry, gating on propidium
iodide-negative cells.
Proliferation assay and cytokine ELISA
To examine proliferation and cytokine production, 10,000 retrovirally transduced and sorted cells were restimulated on irradiated splenic feeder cells in the presence of OVA peptide. Cytokine production was measured 23 h later from culture supernatant by ELISA (Endogen, Woburn, MA). [3H]TdR incorporation was analyzed on day 3.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We were interested in differences in the genetic program expressed
in anergic vs responsive Th1 clones, reasoning that anergic Th1 clones
might express a special gene program to maintain their unresponsiveness
over the weeks that anergy has been shown to persist (31).
Thus, two murine Th1 clones, pGL2 and pGL10, were exposed to
immobilized anti-CD3 mAb to induce anergy or were treated as
indicated in Fig. 1
to yield a population
that was responsive to a subsequent mitogenic stimulus. Two Th1 clones
were used to minimize the chance of isolating clone-specific artifacts.
The anti-CD3 protocol successfully induced functional
unresponsiveness in both Th1 clones, while the control cells remained
responsive to restimulation on day 7 (Fig. 1A). As an
additional control, anti-CD3-treated pGL2 cells were cultured in
the presence of exogenous IL-2 to prevent anergy induction. As shown in
Fig. 1A (Ci), this treatment resulted
in a partial restoration of cell function, which is in agreement with
previously published results (23). The RNA from all five
anergic and control samples was subjected to DDRT-PCR analysis
(33). One thousand and two hundred different PCR
reactions, using 12 different poly(T)-anchored primers and 20 different
arbitrary primers, were analyzed. Fig. 1B depicts a typical
result obtained with two different primer combinations. PCR products
differing in expression level between the two anergic and three control
samples were isolated, reamplified, cloned, and sequenced. Among the 64
PCR isolates with different expression in anergic vs control cells, 26
were selectively present in control cells, while 38 were selectively
expressed in anergic cells. Further analyses revealed that most of the
PCR products were either false positives or represented mRNAs that
could not be detected using the less sensitive Northern blot analysis.
One PCR product, however, detected a mRNA species in Northern blots
that was not present in control cells but was induced in anergic Th1
clones (asterisk in Fig. 1B; see Figs. 2 and 3).
The sequence of the PCR product that represented only the very 3' 300
nucleotides of the gene’s 3'-untranslated region did not match any
database entry. Its corresponding mRNA was 3.5–4 kb as judged by
Northern blot analysis (data not shown). The full-length cDNA was
subsequently isolated and cloned as described in Materials and
Methods.
|
|
C). A
similar expression pattern was observed in nonlymphoid tissues derived
from a recombinase-activating gene-2-deficient mouse. These results
suggest that GRP1 expression observed in the lung was due not to
infiltrating lymphocytes but, most likely, to tissue-specific
fibroblasts, because various fibroblast lines were found to express
GRP1 at high levels constitutively (data not shown). Chromosomal localization of GRP1
Mouse Grp1 was mapped by analysis of the progeny of
interspecific crosses. PvuII digestion of parental DNAs
produced Grp1 fragments of 1.9 and 2.6 kb in NFS/N and
M. spretus, respectively (data not shown). Inheritance of
the variant fragments was followed in two sets of progeny, and linkage
to distal chromosome 5 markers was detected. Gene order and
recombination fractions are given with recombinational distances as
follows: (centromere) Zp3 -5/94 (5.3 ± 2.3) -
Ccnbl-rs1/Grp1 -5/89 (6.5 ± 2.3) -
Atrc1 (telomere). The closest linkage was detected with
Ccnbl-rs1 with no recombinants in 86 mice, suggesting that
at the upper limit of the 95% confidence level these genes are within
3.4 centimorgans. The mouse map location of Grp1 places it
between regions of homology to human chromosomes 7 and 13, suggesting
that the human homologue could map to either of these chromosomes.
Indeed, fluorescence in situ hybridization of a full-length human GRP1
cDNA probe resulted in specific labeling of human chromosome 7 (Fig. 1D). Labeling of 7p21–22 was observed on four (two cells),
three (15 cells), or two (eight cells) chromatids of the chromosome 7
homologues in 25 cells examined. Of 75 signals observed, 69 (92%) were
located at 7p21–22. Of these, three (4%) signals were located at
7p21, two (3%) signals were located at the junction of 7p21–22, and
64 (93%) signals were located at 7p22. Six single background signals
were observed at other chromosomal sites. Thus, Grp1 is
localized to chromosome band 7p22. No immunological disorder of unknown
etiology is associated with this locus according to the Online
Mendelian Inheritance in Man gene map
(http://www.ncbi.nlm.nih.gov/Omim). Another family member
cytohesin-1/B2-1, was previously assigned to human chromosome 17qter
(61). Thus, the gene family is not clustered at one
chromosomal location. The human and mouse nomenclature committees are
currently resolving the nomenclature of this family. Therefore, the
gene symbol Grp1, which is used in this report, is
preliminary.
The GRP1 is a gene induced in anergic T cells
We identified GRP1 with the DDRT-PCR in an anergy model based on a
single time point following clonal Th1 inactivation using a mAb against
CD3 in the absence of costimulation (Fig. 1B). To
determine the time course of GRP1 mRNA and protein expression, a
temporal analysis of GRP1 using the anti-CD3 model of anergy was
performed. The results obtained by Northern and Western blotting
confirmed that GRP1 was selectively induced in anergic Th1 clones (Fig. 2). The GRP1 mRNA was barely detectable, and protein was undetectable
in resting Th1 clones before stimulation, and no significant induction
of mRNA or protein occurred during the first 12 h of anti-CD3
exposure. Under continued exposure to anergizing conditions, maximal
expression of GRP1 mRNA was observed at 
48 h, and although the
signal declined, GRP1 mRNA was still expressed at 5-fold over
background levels at 7 days (Fig. 2, A and B).
The GRP1 protein levels also reached a maximum around 48 h and
continued to be expressed at near maximum levels (91%) at 7 days (Fig. 2, C and D). Thus, GRP1 protein was expressed at
a high level concurrent with the functional state of anergy. Although
the polyclonal anti-GRP1 serum used in our Western blots reacted
with multiple family members, the signal detected in the T cell clones
represents GRP1, because 1) no protein was detected in nonanergized Th1
clones, although mRNA for multiple family members was observed (Fig. 2
and data not shown); and 2) Northern blots hybridized with a GRP1 probe
that detected four mRNA species revealed that only the 3.5- to 4-kb
signal, representing GRP1, is up-regulated during anergy induction.
Thus, GRP1 seems to be the only member of the family that is
significantly induced upon clonal inactivation.
To verify the association between GRP1 and anergy using alternate
costimulation-dependent anergy models, we analyzed GRP1 mRNA expression
in A.E7 cells anergized on chemically modified APC (30).
This modification is believed to leave Ag presentation by the APC
intact, but to destroy its costimulation capacity. The restimulation
proliferation assay confirmed that anergy was successfully induced in
the cells exposed to the chemically modified APC (data not shown). The
GRP1 mRNA was induced at 48 h in A.E7 cells stimulated with
chemically modified APC (Fig. 3A, sample E) and remained
elevated at 7 days, whereas cells stimulated with untreated,
peptide-pulsed APC (sample U) did not express GRP1 mRNA above
background levels (sample Ø) at the 7 day point. Thus, GRP1 was
induced and persisted specifically in the T cells anergized by two
distinct approaches.
|
B). Unique induction of GRP1 in Th1 and Tc1 cells
The results described above were limited to analyses of murine Th1
clones. Th2 clones have been reported to be resistant to the induction
of anergy (13). Therefore, GRP1 expression was examined in
anergy-resistant Th2 clones stimulated with anti-CD3 under the
anergizing conditions used for the Th1 studies. In accordance with
previous reports, the Th2 clones failed to become anergized upon
anti-CD3 treatment (Fig. 4B). Furthermore, GRP1 was not
induced in the Th2 clones, in contrast to the Th1 clones that clearly
up-regulated GRP1 protein 48 h after the CD3 stimulus (Fig. 4A). The Th2 clones did, however, express low amounts of
GRP1 protein constitutively. Similar anergy studies were performed with
the 3.7.2 cytolytic CD8+ T cell clone, which
secreted IFN- upon mitogenic activation (Tc1). These cells could be
anergized with the anti-CD3 Ab (Fig. 4B). When anergic
and control 3.7.2 cells were analyzed for GRP1 protein expression, a
significant increase above basal levels of GRP1 expression in the
anergic cells was observed (Fig. 4A). Thus, the
up-regulation of GRP1 correlated with the capacity of T cells to become
anergized.
|
It is clear from the previous studies that GRP1 was induced in T
cells anergized by methods that fail to provide T cell costimulation.
However, the results did not distinguish between a relationship of GRP1
with the absence of CD28 signaling vs the anergic state per se. T cell
inactivation with partial TCR agonists has been shown to be a
fundamentally different method to anergize T cells. Partial agonists
induce anergy by altering early events in TCR signal transduction
proximal to the TCR complex, even in the presence of costimulation
(16). We have shown previously that a nonmitogenic
anti-CD3-IgG3 Ab, incapable of being cross-linked on
costimulation-competent APC, triggers a partial intracellular signal
that induces anergy in Th1 clones similar to that observed with
peptide-based partial TCR agonists (47, 62). As shown in
Fig. 5A, the nonmitogenic
anti-CD3-IgG3 Ab induced anergy in the pGL10 clone.
Unresponsiveness occurred in the treated pGL10 cells even if
costimulation was provided by splenic APC (Fig. 5C).
Moreover, as in the costimulation-deficient anergy models, the GRP1
protein expression was up-regulated by this T cell inactivating
protocol (Fig. 5, B and D). Thus, GRP1 expression
was specifically associated with the T cell unresponsiveness in two
independent models of anergy induction.
|
To begin to understand the intracellular site of GRP1 function, we
compared the subcellular localizations of overexpressed GRP1, ARNO, and
cytohesin-1 in various cell lines using confocal immunofluorescence
microscopy. Recombinant vaccinia virus-derived Ig fusion proteins were
used because they had previously been found to be very useful in these
types of analyses for three reasons. First, the amino-terminal tagging
does not interfere with known functions of cytohesin-1 (55, 56). Second, these chimeric proteins could be visualized with
the same highly specific reagent, an Ab directed against the Fc portion
of human IgG. Finally, as shown in Fig. 6C, similar expression levels
could be achieved for each Ig fusion protein, excluding the possibility
that differences in compartmentalization observed for the three Ig
fusion proteins would be due to different expression levels. Analysis
of the subcellular distribution revealed two important features of
GRP1. First, GRP1 displayed a cell type-specific localization pattern.
In nonlymphoid HeLa (Fig. 6Ac) and U937 cells (Fig. 6Af), GRP1 showed diffuse cytoplasmic appearance with only a
small degree of plasma membrane staining. In fact, the fusion protein
seemed to accumulate in intracellular vesicular compartments. In
contrast, in the lymphoid cell lines, U266 (Fig. 6Ai) and
Jurkat E6 (Fig. 6Bg), GRP1 almost exclusively localized to
the plasma membrane. This GRP1-specific staining pattern in the
lymphoid cells was not observed with the other family members.
Cytohesin-1 was detected in the cytoplasm and at the plasma membrane in
all cell lines analyzed (Fig. 6A, a, d, and
g, and Fig. 6Bc), while ARNO showed mostly
diffuse cytoplasmic staining (Fig. 6A, b and h,
and Fig. 6Be), except for the immature macrophage cell line
U937 (Fig. 6Ae) where the fusion protein had a punctuate
appearance in the membrane. The second unique characteristic of GRP1
was its exclusive localization to the lymphocyte cortex (Fig. 6Bg), similar to the distribution observed with plasma
membrane-associated actin (Fig. 6B, a and
h). This feature was not shared by cytohesin-1 (Fig. 6B, c and d) or ARNO (Fig. 6B, e and f), which could both be
found in the cytoplasm as well. In Jurkat cells, the cytoplasm to
nucleus ratio was sufficiently large to conclude that GRP1 was largely
absent from the cytoplasm and exclusively localized to the cortex (Fig. 6B).
|
Human cytohesin-1 was shown to promote CD18-mediated adhesion upon
overexpression in Jurkat cells (55). We wanted to know
whether GRP1 could affect adhesion in a similar manner, especially
after our observation that GRP1 localizes to the plasma membrane in
Jurkat cells (Fig. 6B). Jurkat E6 cells, expressing
Ig-fusion proteins of cytohesin-1, GRP1, or ARNO, were examined in
adhesion assays as described previously (55). As shown in
Fig. 6D, Jurkat cells, which expressed a control protein
(the isolated Ig domains), adhered minimally to ICAM-1. Upon phorbol
ester treatment the adhesiveness of these cells increased 4- to 5-fold
as previously described (60). The GRP1 or cytohesin-1
overexpression induced 6- and 8-fold increases in specific adhesion to
ICAM-1, respectively, whereas ARNO had no significant effect. The
observed adhesion to ICAM-1 was reduced to background levels if a pair
of blocking anti-LFA-1 Abs was included into the assay (data not
shown), indicating that the adhesion was LFA-1 specific. Thus, GRP1 and
cytohesin-1 induced LFA-1-mediated lymphocyte adhesion, which
corresponds well to the localization of these proteins to the plasma
membrane in lymphoid cells (Fig. 6). These results were also consistent
with empirical observations that homotypic adhesion as well as adhesion
to tissue culture dishes increased in anergic Th1 clones (data not
shown). In addition, we observed a specific interaction of the
cytoplasmic domain of CD18 with GRP1 in a yeast two-hybrid assay (data
not shown), which was performed as described previously for cytohesin-1
(55), indicating that GRP1 and CD18 can interact
directly.
The GRP1-transduced T cells do not maintain ectopic gene expression
Jurkat T cells are easy to transfect and have been used
successfully in the past to study signal transduction and adhesion.
However, the autonomous proliferation of Jurkat cells interferes with
studies directed toward analyzing the role of GRP1 in cell
proliferation triggered by TCR-mediated antigenic stimulation.
Therefore, TCR-transgenic lymph node CD4+ T cells
from DO11.10 mice were transduced with GRP1 using a retroviral
expression strategy. The goal was to generate Th1 cells that
overexpressed GRP1 and to access their proliferative capacity and
cytokine production upon antigenic stimulation. To achieve this, the
lymph node T cells were activated under Th1 skewing conditions,
transduced, and sorted by FACS for expression of the coexpressed marker
eGFP. Because eGFP was expressed from a bicistronic RNA, preceded by
GRP1, eGFP expression correlated with GRP1 expression in the transduced
T cells (Fig. 7C). The
FACS-purified T cells were examined for continued expression of
eGFP/GRP1 after Ag stimulation and cell expansion. A selective loss of
green fluorescence was observed in the GRP1 population. Three weeks
after the sort, the fraction of eGFP-positive cells among the
GRP1-transduced cells was only 21% (Fig. 7A). Furthermore,
the mean fluorescence intensity of eGFP in the GRP1-transduced
population declined as well (Fig. 7, A and B),
suggesting that the proliferating cells could only tolerate low levels
of GRP1 expression. In contrast, cells transduced with a control vector
containing eGFP remained >70% eGFP positive over the course of the
experiment. The same high percentage of the cells remained eGFP
positive when a retroviral construct containing GRP1 cDNA in an
antisense orientation was used (data not shown). Thus, it seemed
unlikely that the observed instability of the GRP1-transduced
population was an artifact caused by genomic imprinting of the
integrated retroviral construct. Rather, these results indicated
that there existed strong negative selection pressures in T cells
expressing the GRP1 transgene. The mechanism of the observed
instability of the GRP1-transduced population was further addressed by
examining whether ectopic GRP1 expression altered the cells’ ability
to proliferate and/or produce cytokines. Control and GRP1-transduced
DO11.10 T cells were sorted into eGFP+ and
eGFP-deficient cells and restimulated with OVA peptide. No difference
in [3H]TdR incorporation was observed among the
various groups, including GRP1-transduced cells
(eGFP+ or eGFP-) and
control transduced T cells (Fig. 7D). Similar results were
observed with T cells skewed either toward the Th1 or the Th2 pathway.
Th2 cells, however, incorporated more [3H]TdR,
presumably because they produced significant amounts of IL-4 (data not
shown). In addition, there was no difference in ectopic expression of
GRP1 on IL-2 production (Fig. 7E), IL-4 in the Th2 cells
(data not shown) or IFN- in the Th1 cells (data not shown).
Increased IL-2 production was observed in all cells that did not
express eGFP, indicating that the retroviral vector affected
cytokine production (Fig. 7E). The lack of difference in
cytokine production by GRP1-transduced cells compared with that by a
vector control was confirmed by intracellular cytokine staining, gating
on the eGFP-positive cells (data not shown).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Historically, T cell anergy is induced by interfering with costimulatory signals delivered by APC. CD28/B7 interactions have been shown to be the primary pathway to regulate IL-2 production, leading to unresponsiveness (15). In this report we demonstrate that GRP1 is up-regulated in anergic T cells generated following stimulation in the absence of CD28 engagement. However, our initial results could not distinguish between the relationship between GRP1 and CD28 vs anergy per se. Therefore, GRP1 induction was subsequently investigated in a distinct model of anergy induction that results from altered TCR-mediated signals in the presence of potent CD28-mediated costimulation. The GRP1 was induced, concurrent with unresponsiveness, in this model as well. Interestingly, the induction of anergy in both systems requires a calcium flux (47, 73). In this regard, cyclosporin A prevented GRP1 expression in anti-CD3-treated Th1 clones (data not shown), again correlating this genetic marker with the anergic phenotype.
The strong correlation between anergy and GRP1 induction suggests a
functional relationship. For instance, up-regulation of GRP1 in the
anergic T cells might lead to an inability to respond to a mitogenic
stimulus. However, overexpression of GRP1 following retroviral
transduction of T cells was not sufficient to affect the cell’s
ability to proliferate, produce cytokines, or undergo AICD (Fig. 7). We
did observe, however, a reproducible instability of the GRP1-expressing
population compared with that of the control. The expression of ectopic
GFP was rapidly lost during culture. Thus, GRP1 could be important for
the anergic cell’s stability. This effect may be dependent on other as
yet unidentified cofactors, such as the GRP1 interaction partners ARF1
or cytohesin-1.
Unlike activated T cells, anergic cells cannot home to the marginal
zone of the germinal centers to interact with activated B cells
(32). It is conceivable that GRP1 binds LFA-1 (Fig. 6) and
affects LFA-1-mediated adhesion, thus interfering with migration of the
anergic T cell. In this scenario, CD28 is not merely a costimulatory
signal, but is also a regulator of migration; the prevention of GRP1
induction (Fig. 3B) could affect LFA-1-mediated adhesion,
and the ability of CD28 costimulation to abrogate expression of the T
cell chemokine lymphotactin (74, 75) would interfere with
the recruitment of additional T cells. Adoptive transfer experiments
with T cells derived from GRP1-deficient mice will be required to
further address this issue.
In summary, we have shown that GRP1 is an integrin regulatory protein and a potential marker for clonal anergy, which is synthesized de novo upon inactivation, a finding that makes GRP1 different from all previously reported molecules associated with anergy. Because GRP1-transduced cells proliferate and produce cytokines normally, but adhere spontaneously to ICAM-1, GRP1 induced in anergic cells may affect LFA-1-dependent T cell responses, including adhesion.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Jeffrey A. Bluestone, Ben May Institute for Cancer Research, University of Chicago, 5841 South Maryland Avenue, MC 1089, Chicago, IL 60637. E-mail address:
3 Abbreviations used in this paper: GRP1, general receptor of phosphoinositides 1; AICD, activation-induced cell death; ARF1, ADP-ribosylation factor-1; ARNO, ARF nucleotide binding site opener; DDRT-PCR, differential display of reverse transcribed RNA by PCR; ELAD, I-Ad-transfected EL-4 cells; ECDI, 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide; eGFP, enhanced green fluorescence protein.
Received for publication May 3, 1999. Accepted for publication October 18, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and lack of zap70 recruitment in APL-induced T cell anergy. Cell 79:913.[Medline]
phosphorylation without ZAP-70 activation induced by TCR antagonists or partial agonists. Science 267:515.c chain of the IL-2 receptor. Science 266:1039. receptors suppresses transplantation responses while minimizing acute toxicity and immunogenicity. J. Immunol. 155:1544.[Abstract]
Lß2 integrin/LFA-1 binding to ICAM-1 induced by cytohesin-1, a cytoplasmic regulatory molecule. Cell 86:233.[Medline]
This article has been cited by other articles:
|
|
 |
|
  I. A. Parish, S. Rao, G. K. Smyth, T. Juelich, G. S. Denyer, G. M. Davey, A. Strasser, and W. R. Heath The molecular signature of CD8+ T cells undergoing deletional tolerance Blood, May 7, 2009; 113(19): 4575 - 4585. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. L. Sukiennicki and D. J. Fowell Distinct Molecular Program Imposed on CD4+ T Cell Targets by CD4+CD25+ Regulatory T Cells J. Immunol., November 15, 2006; 177(10): 6952 - 6961. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M.-B. Poirier, G. Hamann, M.-E. Domingue, M. Roy, T. Bardati, and M.-F. Langlois General Receptor for Phosphoinositides 1, a Novel Repressor of Thyroid Hormone Receptor Action that Prevents Deoxyribonucleic Acid Binding Mol. Endocrinol., August 1, 2005; 19(8): 1991 - 2005. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. I. Mead, Y. Zheng, C. N. Manzotti, L. C. A. Perry, M. K. P. Liu, F. Burke, D. J. Powner, M. J. O. Wakelam, and D. M. Sansom Exocytosis of CTLA-4 Is Dependent on Phospholipase D and ADP Ribosylation Factor-1 and Stimulated during Activation of Regulatory T Cells J. Immunol., April 15, 2005; 174(8): 4803 - 4811. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. E. Harris, K. D. Bishop, N. E. Phillips, J. P. Mordes, D. L. Greiner, A. A. Rossini, and M. P. Czech Early Growth Response Gene-2, a Zinc-Finger Transcription Factor, Is Required for Full Induction of Clonal Anergy in CD4+ T Cells J. Immunol., December 15, 2004; 173(12): 7331 - 7338. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Heissmeyer and A. Rao E3 Ligases in T Cell Anergy--Turning Immune Responses into Tolerance Sci. Signal., July 13, 2004; 2004(241): pe29 - pe29. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. A. Calderwood Integrin activation J. Cell Sci., March 1, 2004; 117(5): 657 - 666. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Venkateswarlu Interaction Protein for Cytohesin Exchange Factors 1 (IPCEF1) Binds Cytohesin 2 and Modifies Its Activity J. Biol. Chem., October 31, 2003; 278(44): 43460 - 43469. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Tzachanis, L. J. Appleman, A. A. F. L. van Puijenbroek, A. Berezovskaya, L. M. Nadler, and V. A. Boussiotis Differential Localization and Function of ADP-Ribosylation Factor-6 in Anergic Human T Cells: A Potential Marker for Their Identification J. Immunol., August 15, 2003; 171(4): 1691 - 1696. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Tang, T. P. Cheng, D. Agnello, C.-Y. Wu, B. D. Hissong, W. T. Watford, H.-J. Ahn, J. Galon, J. Moss, M. Vaughan, et al. Cybr, a cytokine-inducible protein that binds cytohesin-1 and regulates its activity PNAS, February 20, 2002; (2002) 52712999. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Ermann, V. Szanya, G. S. Ford, V. Paragas, C. G. Fathman, and K. Lejon CD4+CD25+ T Cells Facilitate the Induction of T Cell Anergy J. Immunol., October 15, 2001; 167(8): 4271 - 4275. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S Liu, D. Calderwood, and M. Ginsberg Integrin cytoplasmic domain-binding proteins J. Cell Sci., January 10, 2000; 113(20): 3563 - 3571. [Abstract] [PDF]
|
|
|
|
|
|
L. Zeitlmann, P. Sirim, E. Kremmer, and W. Kolanus Cloning of ACP33 as a Novel Intracellular Ligand of CD4 J. Biol. Chem., March 16, 2001; 276(12): 9123 - 9132. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. S. Harris, T. M. McIntyre, S. M. Prescott, and G. A. Zimmerman The Leukocyte Integrins J. Biol. Chem., July 28, 2000; 275(31): 23409 - 23412. [Full Text] [PDF]
|
|
|
|
|
|
P. Tang, T. P. Cheng, D. Agnello, C.-Y. Wu, B. D. Hissong, W. T. Watford, H.-J. Ahn, J. Galon, J. Moss, M. Vaughan, et al. Cybr, a cytokine-inducible protein that binds cytohesin-1 and regulates its activity PNAS, March 5, 2002; 99(5): 2625 - 2629. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
