| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Immunogenetics Division, University Hospital, School of Medicine, University of Buenos Aires, Argentina
Correspondence: María del Carmen Salamone, Ph.D., División Immunogenética, Hospital de Clínicas, José de San Martín, Av Córdoba 2351, 3rd Piso, (1120) Buenos Aires, Argentina. E-mail: marys{at}sinectis.com.ar
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
3- and
3'-untranslated regions of CD1C showed that
phytohemagglutinin (PHA) activation induced expression of transcripts
that encode the three isoforms (soluble, membrane, and
cytoplasmic/soluble). Immunocytochemical studies showed a specific
association of CD1c with the cell membrane and a cytoplasmic,
perinuclear distribution. Although flow-cytometric staining confirmed
the intracellular presence of CD1c, membrane expression on PHA blast
cells was not detected. We found that membrane detection of CD1c
antigen was temperature dependent. Cell surface binding of the
anti-CD1c monoclonal antibody (mAb) was consistently negative at 4 and
37°C but was detected at room temperature (18–22°C). At
physiologic temperatures, activated PBMCs showed intracellular
accumulation of the anti-CD1c mAbs, indicating that CD1c cycled between
cell surface and intracellular compartments. The CD1c exocytosis
pathway was sensitive to Brefeldin A, cytochalasin B, and
chloroquine.
Key Words: CD1 antigens • alternative splicing • cycling of CD1c molecule • CD1c isoforms • activated PBMC
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1,
2, 3, transmembrane, and cytoplasmic-3' UT [2
6
7
]. With the exception of CD1B, all CD1
genes have the same transcriptional orientation. CD1 genes
are categorized into two different classes according to the sequence of
the leader, 1, and 2 domains. Class I consists of human
CD1A, -B, -C, and the homologous rabbit
CD1B genes (genes that encode "classic CD1
antigens"); class II includes the CD1D genes from humans,
rabbits, and mice and a potentially homologous gene detected in
sheep. The human CD1E gene is equally related to both
classes I and II, as is a homologous gene identified in guinea
pigs [2
, 6
7
8
9
10
]. Classic human CD1 antigens
are expressed on most thymocytes and dendritic cells and have also been
detected in patients with various oncohematologic disorders
[5
, 11
12
13
]. CD1c has also been found in a
small number of normal mature B-lymphocytes [14
]. In
addition to thymocytes, CD1d has been located in B-lymphocytes,
hepatocytes, intestinal epithelial cells, and activated mature T
lymphocytes [15
16
17
18
]. Products of the CD1E
gene were identified recently on human dendritic cells
[19
]. The tissue distribution of the different CD1
isotypes has been a matter of controversy. Previous reports suggest
that CD1 is absent from mature human thymocytes but that high
expression of CD1a has been detected on single positive mature
thymocytes [20
]. In addition, we found cytoplasmic
expression of the classic CD1 antigens in activated peripheral blood
mononuclear cells (PBMCs) [21
]. Recent reports show that all five CD1 genes have some degree of polymorphism [22 , 23 ] and that they are part of a novel recognition system used by specialized populations of T cells. CD1 molecules act as a target for CD4- CD8- cytotoxic T lymphocytes and might restrict T-cell responses to glycolipid antigens [24 25 26 27 28 29 30 31 ]. CD1 molecules might also have functions other than their roles in the host defense against invading pathogens. For example, they might provide an alternative activation pathway for thymocytes [32 ].
The five CD1 genes that have already been cloned have an additional complexity that arises from the alternative splicing that they undergo [19 , 33 ]. In a previous report, we described transcripts that encode different CD1 isoforms in phytohemagglutinin (PHA)-activated human mononuclear cells [34 ]. Similarly, three transcripts encoding soluble, membrane, and cytoplasmic/soluble isoforms have been reported on CD1C-transfected cell lines [33 ]. In the present study, we investigated whether the different splicing pattern of CD1C was present in mRNA from pre-T cell lines, different transfected CD1C cell lines, and resting and activated PBMCs. Additionally, flow-cytometric analysis and immunoperoxidase staining confirmed our previous observation of the cytoplasmic expression of the CD1c molecule on activated PBMCs. However, the intracellular localization of CD1c antigen raises several questions, such as how the CD1c molecule functions during T cell activation and how the membrane expression of this glycoprotein is regulated. We found that CD1c antigens are present in the membrane of activated PBMCs and that their cell surface expression is dynamically regulated by a continuous process of internalization and exportation to the cell surface.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Cell cultures
PBMCs were obtained from normal donors with confirmed negative
serology for hepatitis B virus, human immunodeficiency virus, hepatitis
C virus, and Chagas’ disease. The cells were separated on a
Ficoll-Hypaque gradient. The mononuclear cells were washed and
resuspended to a concentration of 106/mL cells in Iscove’s
modified Dulbecco’s medium (IMDM) supplemented with 10% fetal calf
serum (FCS), gentamicin (100 µ/mL), and fungizone (20 µg/mL).
Mitogenic stimulation of PBMCs was induced by adding PHA (1 µg/mL)
(Sigma, St. Louis, MO) or phorbol myristate acetate (PMA) (Sigma) (25
ng/mL) and culturing the cells for different lengths of time at 37°C.
Additional experiments were also performed in the presence of
recombinant (r) interleukin (IL)-2 (rIL-2) (80 U/mL).
Long-term culture
PHA-activated PBMCs were cultured for as long as 15 days. After
6 days of culture, the cells were fed with IMDM supplemented with 10%
FCS, 100 µg/mL of gentamicin, and 20 µg/mL of fungizone. Then they
were re-exposed to PHA (1 µg/mL) (Sigma) and placed into long-term
culture in the presence of high concentrations of rIL-2 (80 U/mL). On
the basis of the cellular growth detected, this process was repeated
every 5 or 6 days. In these experimental conditions, we observed that
the IL-2 concentration did not become growth limiting, assuring the
exponential growth and high levels of expression of IL-2 receptor
through the duration of the culture.
Cell lines
CD1-transfected cell lines 10B3, ER1, NR7, ER3
(transfected with CD1A, -B, -C, and
-D genes, respectively), CIRC, (CIR mock transfected
with the CD1C gene), NS0, and EL4 (mouse thymoma) or
CIR/mock (human lymphoblastoid cell line), which was used as a negative
control, were maintained at 37°C in RPMI 1640 medium supplemented
with 10% bovine calf serum and 50 µg/mL of gentamicin.
CIRC-transfected cell lines were kindly provided by S. Porcelli (Albert
Einstein College of Medicine, Bronx, NY). The transfected cell lines
10B3, ER1, NR7, and ER3 were provided by C. Milstein. The pre-T MOLT-4
cell line, used as a positive control, was grown in IMDM, supplemented
similarly.
Reverse transcription
Total poly (A+) RNA was prepared from cell lines or
activated PBMCs by using guanidinium thiocyanate and
oligo(dT)-cellulose microcolumns (Pharmacia P-L Biochemicals, Uppsala,
Sweden) according to the manufacturer’s recommended protocol. Reverse
transcription (RT) was carried out in a 50-µL reaction mixture
containing RT buffer, 10 mM dithiothreitol, four deoxynucleotide
triphosphates (0.25 mM each), 20 pmol of poly-dT, and 80 U of RNasin
inhibitor (Promega, Madison, WI). After incubating for 3 min at 65°C,
2 U of reverse transcriptase were added, and the reaction mixture was
incubated for 60 min at 42°C.
cDNA amplification
After RT, cDNAs were amplified by PCR. The reaction was
performed in a 50-µL reaction mixture containing 1 µL of cDNA, 10
pmol of each 5' primer (specific for the 3 domain) and 3' primer
(specific for the 3' UT region), 10x PCR buffer, 0.5 mM
deoxynucleotide triphosphates, and 2 U of TaqPol (Promega). The
reaction was carried out for 35 cycles with the following conditions:
denaturation for 30 s at 92°C, annealing for 1 min at 55°C,
and extension for 3 min at 72°C, followed by a final extension for 10
min at 72°C. Reactions without cDNA were also performed as controls
for PCR contamination. The following CD1C oligonucleotide
primers were used: (CD1C) 5'-AGT TTA CGT AAT GAA TTC GGC ACT
AAA CAT GGT GAT ATT CTT CCT AAT-3' and 5'-GTC AAT ATC TAT GGA TCC AGA
GAA ACA ATT TAA ATG GAG AGT CGA CGG-3' [33
]. The
amplified products (6 µL) were loaded onto a 2% agarose microgel and
electrophoresed in Tris borate/EDTA electrophoresis buffer. The bands
were visualized by ethidium bromide staining under UV light.
Flow cytometry cytoplasmic analysis
Cells (106/mL) were fixed in 2% formaldehyde in
phosphate-buffered saline (PBS) for 15 min, washed, and permeabilized
with methanol for 30 min on ice. The cells were washed again with PBS
and incubated with a solution of glycine supplemented with PBS (0.1
mg/mL) for 15 min at room temperature. After washing, an indirect
immunofluorescence assay was performed by incubating the permeabilized
cells for 30 min at room temperature with appropriate amounts of L161
mAb. An IgG1 isotype control antibody was included to establish
background fluorescence. After washing, an FITC-conjugated goat
anti-mouse F(ab')2 Ig (GAM/FITC; Dako, Carpinteria, CA) was
added for 30 min on ice. The cells were then washed twice with RPMI
1640 supplemented with FCS, and immunofluorescence was analyzed by flow
cytometry with a FACStar Plus® (Becton-Dickinson, San Jose, CA). The
lymphocyte fraction, excluding monocytes and polymorphic cells, was
analyzed. The regions were gated based on forward and scatter
parameters.
Immunoperoxidase staining
Multitest slides (ICN Biomedicals Inc., Horsham, PA) were
treated with a 20% solution of poly-L-lysine (30
µL/well) (Sigma, St. Louis, MO). After overnight incubation at 4°C,
the slides were washed for 10 min in distilled water and air-dried.
PBMCs (5x104) were added to each well of coated glass
slides and incubated for 30 min at 4°C. The cells were fixed and
permeabilized as described above. Endogenous peroxidase was inhibited
by preincubation with a 0.1% solution of phenylhydrazine hydrochloride
in PBS for 30 min at 37°C. The first antibody was incubated overnight
at 4°C and then incubated for 2 h at 37°C. The slides were
washed twice in PBS for 15 min. A goat anti-mouse antibody (Sigma) was
then incubated for 60 min at 37°C. The slides were rinsed in PBS and
then incubated with a mouse monoclonal peroxidase anti-peroxidase
complex (Sigma) for 60 min at 37°C. Peroxidase activity was detected
using 3'-diaminobenzidine tetrahydrochloride (DAB, Sigma) as the
developing agent. Finally, the reaction was stopped by dilution in PBS,
and the slides were mounted for microscopic observation using buffered
glycerin.
Immunofluorescence studies
For indirect staining, 106 PBMCs were washed twice
with RPMI 1640 supplemented with 5% FCS and incubated with the
relevant mAb for 1 h at 4°C (referred to herein as the classic
immunofluorescence staining method) at room temperature (18–22° C)
or at 37°C. The cells were washed twice in cold RPMI 1640
supplemented with 5% FCS and were then stained with GAM/FITC for 15
min on ice. In all experiments, background threshold levels were set
using irrelevant mouse Ig.
For double direct staining, 106 PBMCs were incubated with M241(FITC) for 1 h at 4°C or 37°C and stained with an anti-CD3(PE) mAb for an additional 15 min on ice. In all experiments, background threshold levels were set using irrelevant mouse IgG1(FITC) and an additional fluorochrome-conjugated mAb (anti-CD3-PE).
After the cells were stained, they were washed in PBS, and then antigen expression was measured by flow cytometry using a FACStar Plus or an Ortho Cytoron Absolute (Raritan, NJ) flow cytometer. A propidium iodide exclusion gate was preset to ensure that only viable cells were acquired. Analysis was performed on the lymphocyte fraction, excluding monocytes and polymorphic cells. The regions were gated based on forward and scatter parameters. These controls allowed us to establish the optimal cutoff channels for y and x axes for each population analyzed.
Enriched T cell fraction
The PBMCs were first depleted of adherent cells by incubation on
plastic petri dishes. After 2 h of culture in RPMI 1640 medium at
37°C, nonadherent cells were removed, and T cells were enriched by
two cycles of negative selection with magnetic beads coated with
anti-HLA-DR mAb to remove the B cells. The purity of these fractions,
as checked by fluorescein-activated cell sorter (FACS) analysis, was
94%. The percentage of monocytes and B cells was 
3%, so we
preserved these minimal contaminations of non-T cells because they
contributed to optimal PHA stimulation. In some experiments, the PBMCs
were first stimulated with PHA for 72 h, and then the
CD3+ cells were sorted in a FACStar Plus flow cytometer.
Endocytosis assays
PBMCs were pretreated with either 1 µg/mL of Brefeldin A
(BFA), 5 ng/mL of cytochalasin B (CCB), or 25 mM of chloroquine. After
1 h at 37°C, the first antibody was added. To measure the uptake
of anti-CD1c or anti-CD71 antibodies, pretreated or untreated cells
were incubated with the unconjugated antibody for 1 h at 4 or
37°C. The cells were washed and then fixed with solution A of the Fix
and Perm Kit (Caltag). After solution A was removed, the cells were
washed and permeabilized with solution B of the same kit according to
the manufacturer’s instructions and stained with GAM/FITC for 15 min
at room temperature. Immunofluorescence was analyzed by flow cytometry
with a FACStar Plus or an Ortho Cytoron Absolute flow cytometer.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
3 and 3' untranslated
segment of the CD1C gene. As previously reported
[33
], the NR7 CD1C-transfected cell line and
the pre-T cell line MOLT-4, used as positive controls, showed three
bands that encoded the soluble (band 1), membrane (band 2), and
cytoplasmic/soluble (band 3) isoforms (Fig. 1A
, lanes a and b). In contrast, the human lymphoblastoid cell line
CIR/mock, transfected with the same gene, showed predominant expression
of band 2 over band 3 (Fig. 1A
, lane c). These results suggest that the
CD1C transcripts have a complex mechanism of regulation that
could be tissue specific.
|
54% by 2 days after mitogenic
stimulation (data not shown). cDNA obtained from those cells
(N=4) also amplified bands 2 and 3 (Fig. 1B
, lane b). The
expression of the transcript encoding the soluble product became
detectable 72–96 h after PHA activation (N=9; Fig. 1B
,
lanes c and d), but all three bands were barely visible 6 days after
activation (Fig. 1B
, lane e). CD1C transcript expression was also evaluated in a 15-day long-term culture. It is known that to maintain exponential growth, expression of high levels of IL-2 receptor and nonlimiting concentrations of IL-2 are needed [35 ]. For this purpose, the cells were reexposed to PHA every 5 or 6 days (1 µg/mL) and cultured in the presence of 80 U/mL of rIL-2. Under these conditions, CD25 and HLA-DR expression remained stable during the 15 days of culture in >85% of cells. By the end of the long-term culture (day 15), all three CD1C transcripts were clearly up-modulated (data not shown).
We next evaluated whether the expression of the CD1C transcripts on PHA-activated cells was related to the presence of these transcripts on mature activated T lymphocytes. Resting PBMCs depleted of adherent cells and B cells were used as the source of a T cell-enriched population. CD1C transcript expression was analyzed using this T cell-enriched population, both in resting and in 48-h-activated cells. As depicted in Figure 1C (lane a), the transcripts encoding the membrane and cytoplasmic/soluble isoforms (bands 2 and 3) observed in the resting enriched T cells were clearly up-regulated 48 h after PHA activation (Fig. 1C , lane b). To further exclude the possibility that CD1C transcripts were amplified from contaminating non-T cells, the PBMCs were first stimulated with PHA for 72 h; then the CD3+ cells were sorted, and the presence of CD1C transcripts was analyzed. Under these experimental conditions, the three CD1C transcripts were also amplified (data not shown), which indicates that all three CD1C transcripts could be induced on activated T cells.
Detection of CD1c on permeabilized resting and activated PBMCs
To confirm the cytoplasmic expression of CD1c, which was
previously demonstrated by immunocytochemistry [21
], we
developed a flow-cytometric protocol to detect CD1c expression. For
this purpose, we tested fixers and permeabilizing agents on control
cell lines for different periods and temperatures. We found that
introduction of 2% formaldehyde and methanol at 4°C preserved the
expression of CD1c without significantly altering the flow cytometry
forward and side scatter parameters. In addition, cell permeabilization
with methanol did not affect the binding of L161 antibody to the NR7
and MOLT-4 cell lines (Fig. 2A
). As expected, the CD1A (10B3)-, CD1B (ER1)-, and
CD1D (ER3)-transfected cell lines and NS0- and the
EL4-permeabilized cell lines were nonreactive with the L161 mAb (data
not shown). After establishing the flow-cytometric conditions, we
investigated CD1c expression on resting and activated PBMCs. Resting
permeabilized PBMCs from five of seven unrelated donors showed small
amounts of CD1c (Fig. 2B)
. However, transcripts encoding the membrane
and the cytoplasmic/soluble isoforms (bands 2 and 3) were also
amplified from the two donors that were CD1c-negative by flow
cytometry. The number of CD1c-positive cells observed was higher than
expected if the antibody had recognized only B cells or dendritic
cells. The flow-cytometric studies performed on the enriched T cell
fraction confirmed the presence of low levels of cytoplasmic CD1c
expression on resting T cells (Fig. 2C)
. However, 48–72 h after
mitogenic stimulation, PBMCs from all seven donors had higher
expression of CD1c (Fig. 2B)
. The results were similar for the
48-h-activated T cell-enriched population (data not shown). Although
CD1c protein expression was up-regulated after T cell activation,
flow-cytometric analysis did not establish whether the expression was
restricted to the cytoplasm.
|
|
Membrane detection of CD1c on activated PBMCs
Although the immunocytochemical staining was compatible with the
presence of CD1c at the cell membrane, during the classic
flow-cytometric analysis, in which the first antibody was incubated at
4°C, we consistently failed to detect the CD1c membrane expression.
Most CD1 molecules have an intracytoplasmic domain responsible for
rapid internalization of CD1c protein to the endocytic compartment
[36
, 37
]. It is also known that at 20°C,
late endosomes no longer fuse with newly formed endosomes, resulting,
at the level of the cell membrane, in an increase in certain molecules
located in the coated pits [38
]. Because we could not
detect expression of CD1c, we considered the hypothesis that CD1c might
be rapidly endocytosed after the antibody is bound to the cell membrane
of the activated PBMC, and we explored the possibility that low
temperatures might increase its surface expression. We incubated
resting or activated PBMCs for 1 h with anti-CD1c mAb at different
temperatures. After washing in cold medium, the cells were stained with
GAM/FITC. The cells stained at 4°C or 37°C were consistently
negative. In contrast, a clear displacement of mean fluorescence
intensity (MFI) was detected when the cells were stained at room
temperature (18–22°C) (Fig. 4A
). This temperature-dependent staining was observed in three of
six resting PBMCs analyzed. Those three positive cells showed a
down-modulation of CD1c membrane expression after 30 h of PHA
activation, with reexpression at 48 h of culture (Fig. 4A)
. The
samples from the remaining three donors, which were negative at time
zero, started to express membrane CD1c 48 h after activation. In
contrast, the binding of L161 to MOLT-4 pre-T (Fig. 4B)
and NR7 (Fig. 4C)
cell lines did not show significant variations at room temperature
or 37°C. Different mechanisms could account for these results. The
exocytosis of CD1c from the intracellular stores to the cell membrane
might occur normally at room temperature or at 37°C. This pathway
might be blocked at 4°C. The negative results obtained at physiologic
temperatures might indicate that CD1c molecules that had reached the
cell membrane were rapidly endocytosed. This endocytosis was at least
partially blocked at room temperature, allowing the detection of the
antigen at the cell membrane. Assuming that the CD1c molecules were
endocytosed, we expected that by using a conjugated anti-CD1c mAb
(direct labeling) we could detect the traffic of the CD1c antigen
through the cell membrane of the activated T cells. For this purpose,
activated PBMCs were incubated for 1 h at 4 or 37°C with a
different anti-CD1c mAb (M241 mAb) conjugated with FITC. After 1 h, the cells were labeled with PE-conjugated anti-CD3 mAb. This
procedure showed that, in CD3+ cells, there was clear
displacement of the MFI on cells stained at 37°C (Fig. 4D)
,
suggesting that at physiologic temperatures the CD1c-anti-CD1c complex
was endocytosed from the cell membrane.
|
|
To establish the saturation time of the intracellular stores, activated PBMCs were incubated with the L161 mAb for different periods before being fixed and permeabilized. At the earliest time tested, 10 min after its addition, the CD1c-L161 mAb complex was already detected within the cells. Samples fixed and permeabilized 60 min after addition of L161 mAb showed a clear displacement of MFI compared with samples incubated for only 10 min, indicating an accumulation of CD1c molecules in the intracellular stores. This level of accumulation persisted for an additional hour (120 min total). After that time, the detection of the CD1c-L161 mAb complex started to decrease, presumably because of its transport to a proteolytic compartment (Fig. 5F) .
Trafficking and turnover of CD1c protein on activated PBMCs
The results described in the previous section suggested to us that
CD1c molecules might cycle between the intracellular stores and the
cell membrane in a continuous exocytosis/endocytosis process. To
investigate the exocytosis pathway involved in the membrane expression
of the CD1c antigen, we tested the effects of different inhibitory
agents such as BFA, chloroquine, and CCB on the exocytosis of the CD1c
molecule. Recent reports describe the presence of CD1c molecules in the
recycling endosomes and late endocytic compartments of dendritic cells
[39
, 40
]. Assuming that CD1c would exit to
the cell surface via the recycling endosomes, we expected that the
cycling would not be affected severely by BFA. To test this hypothesis,
we incubated the activated cells for 1 h with 1 µg/mL of BFA and
then incubated the cells with the anti-CD1c mAb or with anti-CD71 (a
marker for early endosomes) for an additional 1 h at 37°C. The
cells were washed, fixed, permeabilized, and stained with GAM/FITC.
Under those conditions, BFA treatment prevented the detection of CD1c,
but it did not affect the expression of CD71 (Fig. 6
, part I, row A vs. B). Those results clearly showed that BFA did
not affect the recycling endosomes. We concluded that this early
compartment was not involved in the CD1c exocytosis pathway on
activated mature T cells.
|
In addition, if the secretory pathway is involved, the blockade of that pathway should result in intracellular accumulation of CD1c, similar to that observed with MHC class I molecules. We added 1µg/mL of BFA at 30, 46, and 47 h and 47 h 30 min after PHA activation. All samples were collected 48 h after PHA addition. By this means, we analyzed the effect of BFA incubated for 18 h, 120 min, 60 min, and 30 min, respectively. In none of those incubation times was the amount of CD1c detected higher than that observed in the cells not treated with BFA [Fig. 6 , part II; see BFA (18 h) vs. BFA (-)] . In contrast, a clear MHC class I cytoplasmic accumulation was observed with the longer incubation time [Fig. 6 , part II; see MHC class I BFA (18 h) vs. BFA (-)]. We speculated that for cells to maintain a constant amount of CD1c, this molecule might be subject to a complex regulatory mechanism with cycles of accumulation and degradation.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
We found differences in the expression of CD1C mRNA and CD1c protein. The cytoplasm and membrane transcripts detected in resting PBMCs did not always correlate with protein expression. In samples from some individuals, resting PBMCs had undetectable levels of protein expression despite the presence of CD1 transcripts. However, we cannot rule out that low levels of some CD1c isoforms were not detected in our experimental conditions.
Although CD1C gene products were clearly expressed in the cytoplasm of resting and activated T cells, detection of CD1c on the cell membrane remained elusive for many years. Flow-cytometric analysis showed that detection of CD1c on the cell surface of mature T cells is temperature dependent. That observation suggested to us that after the binding of the antibody, the CD1c molecule could undergo a continuous cycle of endocytosis and recycling to the cell membrane. We confirmed that hypothesis by showing that two different anti-CD1c mAbs were internalized without accumulation at the cell membrane. Although at physiological temperatures the uptake of CD1c-anti-CD1c mAb complexes was detected with both L161 and M241 mAbs, the MFIs observed were clearly different. Those results require separate explanations. First, the antibodies were used under different experimental conditions. Whereas M241 was used in a direct immunofluorescence assay, the L161 binding was detected by using indirect immunofluorescence. This could explain the greater MFI detected with the second antibody. Second, although these two antibodies recognize the same or closely related epitopes [3 ], only L161 was reported to be capable of modifying the intracellular Ca2+ concentration on pre-T cell lines [32 ], suggesting that there might be at least differential functional features for those antibodies.
Previous reports showed that endocytosis and recycling of lymphocyte surface molecules are involved in activation and/or immunoregulation. For instance, the CD3/T cell receptor complex, the surface Ig on B cells, and the transferrin receptor are internalized by lymphoid cells [43 44 45 46 ]. A group of those molecules is internalized only after binding with specific antibodies or ligands (e.g., CD5) [43 , 46 ]. Our experimental conditions cannot exclude the possibility that endocytosis of CD1c was mediated through the cross-linking of the bivalent first antibody or an additional signal triggered by the antibody-CD1c interaction.
There are several intracellular routes described for many membrane molecules after endocytosis. For instance, 90% of transferrin receptors are rapidly recycled back to the cell membrane via early endosomes [47 ]. Other molecules are recycled more slowly via a pathway involving the trans-Golgi complex (i.e., manose-6-phosphate receptor or the CD3/T cell receptor [48 , 49 ]), and some molecules are routed to lysosomes and degraded after internalization (i.e., epidermal growth factor receptor [50 ]). Our findings suggest that CD1c could be included in the latter group because the rate of uptake of the L161 mAb was not linear with time. The rapid saturation of the intracellular compartment by the CD1c-antiCD1c mAb complex, followed by a decay 2 h after its internalization, suggests that a proteolytic compartment could be involved. The cellular mechanisms that determine the endocytosis or exocytosis of the CD1c antigen on activated T cells are not understood, and further experiments are necessary to establish which stimuli are responsible for removing this protein from the membrane or from the intracellular stores. We observed differences in the endocytic ability between the pre-T-cell line and the mature activated T lymphocytes that could be related to a different interaction between CD1c and cell-surface coated pits.
The short cytoplasmic tails of CD1b, CD1c, and CD1d contain a tyrosine-based motif (YXXZ, where Y is tyrosine, X is any amino acid, and Z is a bulky hydrophobic residue) [28 , 36 , 37 ]. Sequences corresponding to this motif are known to interact with at least two adapter protein (AP) complexes, AP-1 and AP-2. AP-1 directs proteins bearing the YXXZ motif into vesicles of the trans-Golgi network (TGN), and AP-2 at the cell membrane directs proteins carrying that motif into the clathrin-coated pits [51 ]. Two pathways have been proposed for the trafficking of CD1b in antigen-presenting cells [28 , 36 ]. The first pathway begins with transport from the TGN to the surface membrane. Subsequently, interaction of the AP-2 complex with the cytoplasmic tail motif of CD1b causes localization into clathrin-coated pits and internalization to endosomal compartments. In the second proposed pathway, CD1b interacts with the AP-1 complex, which mediates sorting of proteins from the TGN directly to MHC class II compartments [28 , 36 ]. In dendritic cells, it has been reported that CD1c is also trafficked through the late endosomes/lysosomes, where, in a way similar to that of CD1b, they might pick glycolipid antigens in an acidified environment. However, the intracellular distribution of CD1c is not completely identical with that of CD1b [52 ].
Recent findings suggest that the CD1c isotype is distributed broadly throughout the endocytic system and is expressed in recycling endosomes and late endocytic compartments [39 , 40 ]. Several differences exist between the intracellular trafficking proposed for CD1c on dendritic cells and our results on mature activated T cells. First, dendritic cells express high levels of CD1c on the plasma membrane but show low expression in the intracellular vesicle [52 ]. Conversely, the amount of CD1c detected on the cell surface of activated mature PBMCs is always lower than the level detected in the intracellular compartment. Second, CD1c is partially detected in the early endosomes on dendritic cells [39 , 40 ]. In the present study, we showed that BFA treatment completely abolished CD1c cycling without affecting the expression of CD71 antigen, which cycled from the early endosomes. These results suggest that CD1c has not accumulated in the early endosomes of the activated T cells. In addition, after treatments with different inhibitor agents, CD1c did not show a behavior compatible with exocytosis via the secretory pathway. Considered together, our present results suggest that, on activated PBMCs, CD1c is exported from the intracellular stores by an atypical pathway that could be interrupted by BFA, which does not involve the early endosomes. BFA was reported to disturb not only the transport of newly synthesized molecules on their way to the cell membrane (secretory pathway) but also the binding of adapter proteins to the TGN, which also affects the sorting to the intracellular compartments [53 ]. This drug also affects the vesicles involved in the transport between the MHC class II compartment and the cell membrane by blocking the recruitment of coat components necessary for the formation of proper intermediate vesicles and their transport to other cellular organelles [54 ]. We showed that the CD1c cycling on activated T cells was abolished not only by BFA but also by lysosomotropic drugs such as chloroquine and CCB. Additionally, although there was clear MHC class I cytoplasmic accumulation with longer incubation with BFA, the cytoplasmic amount of CD1c detected was higher in the BFA-untreated cells at all times tested. Although direct evidence for the predominant route taken by CD1c is not yet available, our data indicate that the principal pathway involved in the exocytosis of CD1c on mature T cells is mediated through a vesicular, acidic, intermediate BFA-sensitive route. CD1c might be sorted from the TGN directly to an intracellular store and then could migrate to the cell membrane. From there, CD1c-anti-CD1c complexes were internalized back to the proteolitic compartment.
The expression of CD1c molecules on mature T cells is tightly regulated. In these activated T cells, CD1c has a primary intracellular localization. At physiologic temperatures, this molecule seems to undergo a rapid process of exocytosis and endocytosis. The striking similarities between CD1c and other molecules involved in activation signals support the assumption that expression of CD1c on mature T cells might have functional regulatory implications. Our data provide experimental support for the hypothesis that, in addition to the capacity to act as an antigen-presenting molecule for glycolipids, CD1c might act in T cells as an immune regulatory molecule. The mechanism of this putative function remains to be elucidated.
|
ACKNOWLEDGEMENTS |
|---|
Received January 25, 2001; revised April 23, 2001; accepted April 24, 2001.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
ß+ T cells Nature 372,691-694[Medline]
) chains from -ß dimers in B cells J. Exp. Med. 162,1371-1376
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
