| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Cellular Biology University of Georgia, Athens, GA 30602
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Ly-6A.2 (T cell-activating protein/Sca-1) is a member of this multigene family expressed on pluripotent hemopoietic stem cells, some developing thymocytes, and a majority of mature T cells. Thymocytes overexpressing Ly-6A.2 aggregate when cultured in vitro. This aggregation is dependent on the expression of the Ly-6A.2 molecule on these cells because its is inhibited by phosphatidylinositol-specific phospholipase C (PIPLC)3 treatment (which removes Ly-6A.2) as well as by anti-Ly-6A.2 Abs (including their Fab') (13). These experiments demonstrate that Ly-6A.2 is capable of mediating cell-cell adhesion. What is still unknown is the identity of the protein that interacts with Ly-6A.2. We report generation of an Ab, 9AB2, against the putative Ly-6A.2 ligand. This Ab recognizes a 66-kDa protein with unique tissue distribution and also blocks Ly-6A.2-dependent cell-cell adhesion.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CD2-Ly-6A.2 transgenic mice backcrossed to BALB/c for 10–15
generations were used (17). Four- to 8-wk-old mice were
used in all the experiments. The human CD2 enhancer was used to express
the Ly-6A.2 transgene in Ly-6A.2 Tghigh (>50
copies/cell) and Ly-6A.2 Tglow (
3–10
copies/cell) mice. The expression of the transgene is restricted to
only T cells in both the transgenic mice (17), as expected
(18).
Antibodies
Anti-Ly-6A.2 mAbs (3A7, 1H12 (19) and 8G12 (4C9, 3E7) (20)), anti-Ly-6A/E (D7) (21), anti-ThB (53.9.2) (22), and anti-human CD2 (3pt-2H9) (23) were used in the study. Culture supernatants containing mAbs were prepared as described previously (24). Abs were purified from culture supernatant of hybridomas grown in presence of bovine serum or under serum-free conditions on a protein A- or protein G-Sepharose column (Bio-Rad, Hercules, CA). 9AB2 Ab was purified from the ammonium sulfate (0–50%) precipitation of culture supernatants on gel filtration on Bio-Gel A-1.5 m (Bio-Rad) or by fast protein liquid chromatography using Superdex 200 gel filtration column-HiLoad 26/60 (Amersham Pharmacia Biotech U.K., Aylesbury, U.K.).
Cell lines
The cell lines YH16.33, MVB2, KQ1.3A, (T-T hybridomas), AKR1G1, BW5147, E710, NFC105, EL4, R1.1, RADA (thymomas), M12, A20, ED-2, 70-Z, RAW, AJ9, WEHI-231, WEHI 164, WEHI-3, WR19.M1, RAW 309. Cr, J774R1, Moms (pre-B or mature B lymphoblastoid), WEHI (fibrosarcoma), 230-37 (pre-B lymphoblastoid), and NIH-3T3 (fibroblast) were passaged in DMEM with 4.5 g glucose/l supplemented with 10% FCS, glutamine, and antibiotics (Irvine Scientific, Santa Ana, CA).
Derivation of Chinese hamster ovary (CHO) cells overexpressing Ly-6A.2
Ly-6A.2 was overexpressed in dihydrofolate reductase (DHFR)-negative CHO cells by cotransfecting plasmids containing cDNA for Ly-6A.2 and DHFR using lipofectin (Life Technologies, Gaithersburg, MD). The expression of Ly-6A.2 was amplified by culturing transfected cells with methotrexate (Sigma, St. Louis, MO) at 0.08–0.32 µM. Selected cells were cloned by limiting dilution, and clones with the highest Ly-6A.2 expression were used for binding experiments. The CHO cell line transfected with human CD4 was obtained from Dr. C. Doyle (Duke University, Raleigh, NC) and used for control binding experiments.
Preparation of monoclonal and polyclonal Abs
Armenian hamsters were immunized with 10–20 x 106 BALB/c thymocytes. Cells were injected i.p. five or six times over a period of 3–5 mo. Three days after the last injection, the hamsters were sacrificed, and their spleen cells were fused with a B lymphoblastoid cell line, Moms (gift from Dr. Robert Schreiber, Washington University School of Medicine, St. Louis, MO), using standard protocol (25). Hybrid cells were screened for their growth in hypoxanthine-aminopterin-thymidine medium and their ability to block aggregation of Ly-6A.2 transgenic positive thymocytes. B cell hybridomas were cloned by limiting dilution to achieve monoclonality. These cell lines were adapted to grow under serum-free conditions (Calbiochem (La Jolla, CA) and Life Technologies/BRL).
Polyclonal Abs against the putative ligand were prepared by immunizing one New Zealand White rabbit with immunoprecipitated material using anti-ligand mAb (9AB2). 9AB2 immunoprecipitates were resuspended in 500 µl of PBS and emulsified with CFA before injection into the rabbit intradermally in 20 sites on the back. The first and second booster injections with 9AB2 immunoprecipitates-PBS suspension (500 µl) emulsified with IFA were given 3 and 10 wk after the primary immunization, respectively. Antiserum from the immunized rabbit was collected 2 wk after the second booster injection. Preimmune serum was collected from the same rabbit before immunization and served as the control. IgG fraction from the preimmune and postsecondary boost was purified on protein A-Affi-Prep column (Bio-Rad) used for immunoprecipitations.
Flow cytometry
Cells were stained by immunofluorescence as described previously (24). Thymocytes, Tris-NH4Cl-treated spleen cells, or cell lines were incubated with primary Ab followed by appropriate fluorochrome-conjugated second-step reagents as indicated in the figures. Five or 10 x 103 cells were analyzed on FACScan flow cytometer (Becton Dickinson (Mountain View, CA) or Coulter (Hialeah, FL)).
Immunohistochemistry
Tissues from normal or transgenic mice were removed and snap-frozen in liquid nitrogen for 15 s and then stored at -80°C until use. The frozen tissues were embedded in cryo-embedding medium, M-1 embedding matrix (Lipshaw, Pittsburgh, PA). Serial tissue sections of 5–8 µm thickness were cut at -25°C. The sections were then fixed using a graded series of acetone in water (60, 70, 80, and 90% acetone) for 3 min for each solution. The fixed tissue sections were washed three times for 15 min each time in PBS.
An avidin-biotin-peroxidase complex (Vector, Burlingame, CA) was used for immunohistologic staining of the cryosections as previously described and according to the vendor’s instructions. Biotinylated primary Abs were used for staining. In some experiments culture supernatants containing appropriate Ab (5–20 µg/ml) followed by biotinylated anti-hamster IgG (Caltag, Burlingame, CA) as secondary Ab were used. All sections were counterstained with hematoxylin (Sigma). The slides were air-dried and mounted permanently using DPX (Aldrich, Metuchen, NJ).
Cell culture and homotypic aggregation
Thymocytes and splenic cells were obtained by grinding the thymus or spleen, respectively, using frosted glass slides. For cell aggregation experiments thymocytes (1 x 106) were cultured in flat-bottom plates for 18–24 h at 37°C in a final volume of 100 µl of culture medium consisting of RPMI 1640 supplemented with 20 mM HEPES, 2 mM L-glutamine, 1 mM nonessential amino acids (Irvine Scientific, Irvine, CA), 10% heat-inactivated FCS (Sigma), 0.25 µg/ml of Fungibact (Life Technologies), and 5 x 10-5 M 2-ME. The precise culture conditions are given in the appropriate legend to each figure. To quantitate thymocytes in the aggregates, 0.4 x 106 thymocytes were cultured in 96-well flat-bottom plates for 18–24 h at 37°C in a final volume of 100 µl. Cells were gently pipetted and counted in a hemocytometer. The number of single (free) cells and the number of cells in small aggregates (of <5 cells) were counted. In some cases the percentage of cells forming large aggregates was calculated by applying the following formula: % aggregation = (1 - number of free cells + number of cells in small aggregates)/number of total cells) x 100. The experiments were conducted in duplicate. Data are representative of three independent assays.
Binding of 9AB2 Ag-expressing cell to CHO cells overexpressing Ly-6A.2 protein
Ly-6A.2 or human CD4-expressing CHO cell lines were plated in flat-bottom 24-well plates at 0.5 x 105 cells/well in complete MEM (Life Technologies) supplemented with 0.25 µg/ml penicillin-streptomycin, fungizone, 10% heat-inactivated dialyzed FCS, and methotrexate 1 day before the binding assay. Ly-6A.2 and CD4-CHO cells were incubated with 0.6 x 106 9AB2high or 9AB2low thymocytes for 2 h at 37°C in a final volume of 500 µl. The unbound thymocytes were gently washed with 0.1 M PBS, and 100 µl of 1x trypsin-EDTA (Life Technologies) was added to each well followed by incubation at 37°C for 5 min. Cells were gently pipetted, and only thymocytes (4–5 times smaller than CHO cells) were counted. The effects of Abs was determined at a 1/5 dilution of an appropriate Ab supernatant. In some experiments thymocytes or Ly-6-CHO cells were preincubated with the Abs before the binding assay to increase their blocking efficiency.
Immunoprecipitation
Lymphoid cells were lysed in immunoprecipitation buffer
containing 10 mM Tris-HCl, 150 mM NaCl, 1 mM PMSF, 10 mM iodoacetamide,
and 1% octylglucoside. Other peptide inhibitors included were
leupeptin, pepstatin, chymostatin, antipapain, and
N-
-p-tosyl-L-lysine
chloromethyl ketone at 1 mg/ml. Immunoprecipitation and SDS-PAGE were
performed as described previously with some modifications
(26). Cell lysates were precleared with IgSorb (The Enzyme
Center, Malden, MA) followed by the rabbit anti-mouse Ig-cyanogen
bromide-Sepharose. Immunoprecipitation was conducted with or without
relevant Abs in the form of complexes with rabbit anti-mouse
Ig-cyanogen bromide-Sepharose or protein A-Sepharose (Bio-Rad). The Ab
complexes were generated by incubating rabbit anti-mouse
Ig-Sepharose or protein A-Sepharose (20 µl of swollen gel, 1 mg/ml)
with 50 µl of the appropriate Ab containing supernatants for 45 min
on ice before immunoprecipitation. Immunoprecipitates were analyzed on
14–18% SDS-PAGE under reducing and nonreducing conditions and were
transferred to nitrocellulose membrane. Western blots were performed
with appropriate Abs (see Fig. 5
) followed by incubation with goat
anti-hamster biotin or goat anti-rat biotin. The blots were
incubated with neutriavidin-HRP conjugate and visualized using a
Supersignal chemiluminescence kit (Pierce, Rockford, IL). The exposure
times were 30 s to 2 h. The Kaleidoscope prestained m.w.
markers used were myosin (201,000), ß-galactosidase (116,000), BSA
(85,000), carbonic anhydrase (42,400), soybean trypsin inhibitor
(32,600), lysozyme (18,000), and aprotinin (7,200). The arrow on the
gels indicates the 66-kDa 9AB2 immunoprecipitate. In some experiments
20 x 106 thymocytes were biotinylated
using sulfo-NHS-Biotin according to the vendor’s instructions
(Pierce). The biotinylated cells were lysed in immunoprecipitation
buffer with 1% Nonidet P-40. 9AB2 immunoprecipitates from these
lysates were run on 18% SDS-PAGE and transferred to nitrocellulose
followed by blotting with strepavidin-conjugated alkaline phosphatase.
The blots were developed with p-nitrophenyl phosphate (Life
Technologies) substrate for 5–30 min.
|
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The purpose of this study was to generate mAbs against Ly-6A.2
ligand. To obtain anti-Ly-6A.2 ligand Abs, supernatants from a
number of B cell hybridomas were tested for the ability to inhibit
homotypic aggregation of Ly-6A.2 Tghigh
thymocytes. As reported previously Ly-6A.2 participates in the
cell-cell adhesion of Ly-6A.2 transgenic thymocytes (Fig. 1A) (13). About
70–80% of the Ly-6A.2 transgenic thymocytes participated in
aggregation, and about 5% of nontransgenic thymocytes aggregate (Fig. 1E). A hamster Ab 9AB2 blocked the aggregation of
Ly-6A.2 transgenic thymocytes (Fig. 1, C and E).
As reported previously anti-Ly-6A.2 Abs (Fig. 1, B and
E), but not anti-CD48 (5-8A10; control hamster Ab; Fig. 1, D and E) blocked the homotypic aggregation of
Ly-6A.2 transgenic thymocytes. These results indicate the specificity
of this inhibition. Moreover, we have previously demonstrated that a
number of Abs directed to other T cell surface proteins tested in this
assay (for example, blocking anti-CD2, anti-LFA-1, and
anti-ICAM-I Abs) failed to block aggregation (13).
|
The ability of the 9AB2 mAb to inhibit aggregation of thymocytes
expressing a high level of Ly-6A.2 suggested that the mAb either
recognized the Ly-6A.2 molecule itself or its putative ligand on
thymocytes. Therefore, we first tested expression of the 9AB2 Ag on
normal and Ly-6A.2 Tghigh thymocytes by
immunofluorescence followed by flow cytometry as shown in Fig. 2. The expression of 9AB2 Ag was very low
on the surface of wild-type nontransgenic thymocytes (Fig. 2). In
contrast, transgene-positive thymocytes showed a high level of
expression (Fig. 2). This observation came as a surprise, because
nontransgenic thymocytes were the source of Ag for the production of
9AB2 hybridoma. We next examined the expression of the 9AB2 Ag on cells
that form spleen, lymph node, bone marrow, and peritoneal exudate. The
majority of cells from these tissues expressed very low levels of 9AB2
Ag (similar to nontransgenic thymocytes; Fig. 2). About 13–17% of
spleen cells stained positively (mean fluorescence intensity, 30;
negative control mean fluorescence intensity, 5) with 9AB2 Ab (data not
shown). Treatment of splenic or lymph node cells with either Con A or
cocktail of rat Con A supernatants (mixture of cytokines) or IFN-ß
did not induce the expression of 9AB2 Ag (data not shown). In contrast,
Ly-6A.2 and MHC class I expression increased under these conditions
(data not shown).
|
To gain insight into the tissue expression of 9AB2 Ag we tested for
expression on tissue sections from thymus, spleen, and lymph node. 9AB2
Ag was detected in the spleen of BALB/c mice (Fig. 3). This staining invariably clustered
around blood vessels or sinusoids in the red pulp (Fig. 3E).
In no case was the staining observed in the white pulp, including the
marginal zone and the T cell-rich area, periarteriolar lymphoid sheath
(PALS; Fig. 3E). Analysis of splenic cells by flow cytometry
did not reveal 9AB2-expressing cells, therefore suggesting that either
9AB2-expressing cells in the spleen are present in low numbers or are
comprised of stromal cells in the spleen that cannot be extracted into
cell suspension by grinding with frosted slides. The absence of
staining with control Ab indicates that the binding of 9AB2 is specific
(Fig. 3A). Moreover, an excess of nonbiotinylated 9AB2 Ab
blocked the staining (Fig. 3F). In contrast, Abs against
TCR-ß specifically stained the PALS area in the spleen, and 9AB2
mAb did not block this staining (Fig. 3D). These studies
also indicate that the staining patterns with 9AB2 and anti-Ly-6A/E
are different (compare B and E in Fig. 3). Ab to
Ly-6A/E stained the blood vessels in both the red and white pulp areas
(Fig. 3B). 9AB2 did not bind cells in the T cell-rich area,
PALS, even though immunofluorescence and flow cytometric analysis
showed slight expression on T cells from the thymus (Fig. 2
). It is
possible that we are unable to detect expression on T lymphocytes in
tissue sections due to insensitivity of our detection system. Ly-6A.2
transgenic mice showed similar results (Fig. 4A). More importantly, 9AB2 Ab
did not stain the transgenic T lymphocytes in the PALS that overexpress
the Ly-6A.2 protein (compare Fig. 4, A and
B).
|
|
Staining of thymic tissue sections with 9AB2 Ab also produced staining
of clusters of cells primarily localized around the blood vessels in
the cortex of nontransgenic thymus (Fig. 4C). Thymic medulla
did not show staining with 9AB2 (Fig. 4C). Tissue sections
from transgenic mice showed strong staining all over the thymus (Fig. 4E). Clustered staining was also observed in the tissue
sections from Ly-6A.2 transgenic thymus with 9AB2 Ab (Fig. 4E). The majority of cells staining with 9AB2 Ab were
developing T lymphocytes, consistent with the high level of staining on
thymocytes when stained in suspension and analyzed by flow cytometer
(Fig. 3). Anti-Ly-6A/E Ab stained cells in the medulla as well as blood
vessels (Fig. 4D). This Ab showed staining throughout the
thymus in Ly-6A.2 transgenic mice as expected (Fig. 4F).
Taken together, our results indicate that the 9AB2 Ag shows a unique
pattern of staining localized around blood vessels that is distinct
from the expression of Ly-6A.2 molecule.
Ag recognized by 9AB2 mAb
To precisely identify the 9AB2 Ag we first conducted
immunoprecipitation with 9AB2 and then immunoblotted with the same Ab.
A distinct protein with a molecular mass of 66 kDa was
immunoprecipitated and immunoblotted with 9AB2 under nonreducing (Fig. 5A) as well as reducing (data
not shown) conditions. This recognition was specific, because
9AB2-immunoprecipitated material was not recognized by a control
hamster Ab (HP25; Fig. 5A). In contrast, an anti-Ly-6A/E
Ab, D7, immunoprecipitated and immunoblotted a 10-kDa protein as
expected (Fig. 5B). Immunoprecipitation and Western blotting
conducted with HP25 (hamster; Fig. 5A) and KJ126 (rat; Fig. 5B) did not recognize the 66- and 10-kDa proteins,
respectively. Similar results were obtained with cell line 1C11,
expressing 9AB2 (T-T hybridomas generated from Ly-6A.2 transgenic T
cells; data not shown). We also observed coimmunoprecipitation of a
10-kDa protein with 9AB2 Ab, and our Western blot analysis demonstrates
it to be Ly-6A/E (Fig. 5B). These experiments suggest that
the 9AB2 Ab recognizes a 66-kDa protein under nonreducing and reducing
conditions. Moreover, this 9AB2 Ag associates with the 10-kDa Ly-6A/E
molecule.
To identify the ligand and support the above studies, rabbit polyclonal
Abs against 9AB2 Ag was generated by injecting 9AB2-immunoprecipitated
material as a source of Ag into a rabbit. Immunoprecipitation with
anti-9AB2 polyclonal Ab followed by blotting with 9AB2 revealed the
presence of a 66-kDa protein (Fig. 6).
This recognition was specific, because immunoprecipitation with
preimmune Abs followed by immunoblotting with 9AB2 did not recognize
the 66-kDa protein. Furthermore, 5-8A10, a control hamster Ab, did not
recognize the 66-kDa protein (Fig. 6). The 9AB2 Ag expression is
sensitive to treatment with PIPLC, and a 66-kDa protein can be
immunoprecipitated from the supernatant of PIPLC-treated
9AB2-expressing cells (Fig. 7). Taken
together, these experiments demonstrate that 9AB2 recognizes a 66-kDa
protein that is GPI anchored to the cell surface.
|
|
We have generated CHO cells lines overexpressing Ly-6A.2 protein
using methotrexate/DHFR amplification. Ly-6A.2 is highly expressed on
the transfected cell lines (Fig. 8A), and untransfected CHO
cells do not express this protein (data not shown).
|
A). This system has allowed us to
directly test the heterotypic binding of 9AB2-expressing thymocytes to
adherent CHO cells overexpressing Ly-6A.2 that do not express the
putative ligand. We examined the binding of
9AB2high and 9AB2low
(differing in expression by >25-fold; see Fig. 2) cells to the
Ly-6A.2-CHO cells. Fig. 8 shows that about 80% of thymocytes
expressing high levels of 9AB2 Ag bind to Ly-6A.2-CHO cells (open bar,
upper left panel). This binding was inhibited by
anti-Ly-6A.2 (filled bar) as well as 9AB2 (anti-putative
ligand; bar filled with diamonds) Abs and PI-PLC (data not shown).
Similar inhibition of binding was observed with either 9AB2
Ab-preincubated 9AB2high thymocytes or Ly-6A.2
Ab-preincubated Ly-6CHO cells when used for binding experiments (data
not shown). In these later experiments excess unbound Abs were absent
in the binding experiments. These observations indicate that Ly-6A.2
expressed on Ly-6-CHO cells binds to 9AB2 expressed on thymocytes. The
lack of inhibition with the isotype control, anti-CD48 (5-8A10;
hatched bar) indicates the specificity of the inhibition. Moreover,
CD4-CHO cells do not bind 9AB2high thymocytes
(Fig. 8, lower left panel). These results
strongly suggest that 9AB2 Ag specifically interacts with Ly-6A.2
protein. |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
80 kDa. This immunoreactivity suggests that
the 66- and 80-kDa proteins are products of different splice products
of the same gene. Alternatively, both the proteins may share an epitope
recognized by 9AB2 mAb. We cannot completely rule out the possibility
that the 66-kDa protein is a degradative product of the 80-kDa protein.
We think that this possibility is unlikely for two major reasons.
First, 66-kDa protein, and not the 80-kDa molecule, is the major
product immunoprecipitated by 9AB2 Ab. Second, we have used a cocktail
of multiple protease inhibitors in our immunoprecipitation buffer that
prevents protein degradation. Additional experiments, including
analysis of the cDNA clone encoding 9AB2 protein, are needed to address
this issue. Our experiments indicate that the 9AB2 Ab blocks Ly-6A.2-dependent cell-cell adhesion of thymocytes. More importantly, our direct cell-binding and Ab inhibition experiments with Ly-6A.2-overexpressing CHO cells and 9AB2-expressing thymocytes indicate that 9AB2 Ag is the putative ligand for Ly-6A.2. Coimmunoprecipitation of Ly-6A.2 with 66-kDa protein with our 9AB2 Ab supports these observations. This latter observation does not directly address the issue of whether Ly-6A.2 binding to its putative ligand occurs within the same (intracellular) or the adjacent (intercellular) cell. Taken together these observations suggest that 9AB2 Ag is one of the candidate ligands for Ly-6A.2.
T cells expressing 4- to 5-fold lower expression of Ly-6A.2 do not undergo homotypic aggregation. Based on these observations we suggest that the affinity of Ly-6A.2 to its putative ligand is low. It is surprising that we observe coimmunoprecipitation of Ly-6A.2 with the putative ligand when it is immunoprecipitated with 9AB2, because detergents are known to disrupt the receptor-ligand interactions. It is also possible that the presence of detergents may increase the affinity of receptor for its ligand by oligomerizing either the receptor or the ligand. One such example is the GPI-PLC that oligomerizes in the presence of detergents and thereby increases its substrate activity (29).
9AB2 Ag is expressed in both the red pulp of the spleen and the cortex of the thymus of nontransgenic mice. The high expression of 9AB2 observed on clusters of cells around blood vessels in the spleen and the thymus is intriguing. The identity of cells in these clusters around blood vessels in the normal spleen and thymus is currently unknown. Expression of 9AB2 Ag around blood vessels suggests that the binding of Ly-6A.2 to its putative ligand may be involved in migration of mature and/or activated T lymphocytes from the lymphoid organs to the blood. Noteworthy is the property of another member of the Ly-6 gene family, Ly-6C, that seems to directly bind the high endothelial venules in the lymph nodes (16). Moreover, cross-linking of this protein up-regulates other adhesion proteins (ß1 and ß2 integrins) (16). 9AB2 Ab does not stain cells in the germinal follicles in the spleen or the lymph node. These results suggest that expression of 9AB2 Ag is unique and distinct from the expression of Ly-6A.2.
Tissue sections from nontransgenic mice showed more staining with 9AB2 Ab than that observed with lymphoid cells and cell lines by flow cytometric analysis using the same Ab. Very low surface expression of 9AB2 on splenic cells in suspension may suggest an additional intracellular staining observed with splenic sections. Alternatively, it is possible that cell types that stained in the tissue sections were not enriched or were poorly represented in the cell preparation obtained after grinding the spleen and thymus tissues. In contrast, high expression of 9AB2 and Ly-6A.2 is observed in the thymus and spleen of Ly-6A.2 transgenic mice by both immunohistochemistry and flow cytometry, as expected.
We were surprised to observe the lack of 9AB2 Ag expression on a number
of lymphoid and myloid cell lines tested. Moreover, the expression of
the ligand was not induced by IFN-ß or IFN-
or supernatant from
cultures of rat splenocytes stimulated with Con A, which contains a
mixture of cytokines (data not shown). The constitutive and induced
expression of 9AB2 Ag is different from that of Ly-6A.2. We were also
surprised to observe high levels of 9AB2 Ag on the surface of
thymocytes and cell lines overexpressing Ly-6A.2 compared with either
nontransgenic thymocytes/cell lines or even cell lines that expressed
4- to 5-fold less Ly-6A.2. It is unclear why high Ly-6A.2 expression
results in concomitant expression of the putative ligand. It is
possible that overexpression of Ly-6A.2 in T cells facilitates
induction of the ligand. These effects observed in the Ly-6A.2
transgenic thymocytes are not due to abnormal expression or mutations
in the Ly-6A.2 transgene (data not shown).
Higher expression of 9AB2 on Ly-6A.2 transgenic thymocytes initially raised the suspicion that 9AB2 may be directed to Ly-6A.2 protein. Our data demonstrate that 9AB2 is not directed against Ly-6A.2. 1) 9AB2 immunoprecipitates and immunoblots a protein that migrates on SDS-PAGE as a 66-kDa protein distinct from Ly-6A.2; in addition, 9AB2 does not recognize Ly-6A.2 in the Western blots. 2) Ly-6A.2-expressing cell lines do not stain with this Ab using immunofluorescence and flow cytometric analysis. 3) Immunohistochemistry of spleen, thymus, and lymph nodes indicates distinct expression patterns of Ly-6A.2 and 9AB2 Ag. 4) Contrary to the Ly-6 proteins, stimulation through the Ag receptor or by IFNs does not alter the expression of 9AB2 Ag (data not shown).
Our direct binding studies using overexpressed Ly-6A.2 in CHO cells
with 9AB2-expressing cells and the property of 9AB2 and
anti-Ly-6A.2 Abs to block this cell-cell adhesion (Fig. 8),
strongly suggest that 9AB2 Ag binds Ly-6A.2. Therefore, it was
surprising that Ly-6A.2-IgG fusion protein was unable to inhibit
Ly-6A.2-dependent cell-cell adhesion (data not shown). Consistent with
these findings, we were also unable to detect specific binding of
Ly-6A.2-IgG fusion protein to cells expressing 9AB2 Ag by
immunofluorescence (data not shown). The absence of direct binding of
the Ly-6A.2-IgG fusion protein to 9AB2high cells
and its inability to block homotypic aggregation suggest that the
affinity of Ly-6A.2 for its ligand may be low, and simply the
overexpression of Ly-6A.2 on cells might overcome this low affinity
binding. Other explanations may also exist. For example, Ly-6A.2
expressed on the cell surface may be qualitatively different from the
purified fusion protein with respect to binding to the putative ligand.
One such difference may arise due to the lack of the GPI anchor on the
Ly-6A.2-IgG fusion protein. Consistent with this speculation is the
observation that an anti-Ly-6A.2 Ab is unable to immunoblot
PIPLC-treated Ly-6A.2, but successfully recognizes GPI-intact Ly-6A.2
molecule (A. English et al., unpublished observations). These
observations are supported by a recent report that suggests that the
GPI anchor is critical for binding of some anti-Thy-1 Abs to Thy-1
molecule (30). It is also possible that Ly-6A.2 is part of
the receptor complex on the cell surface, and its binding to 9AB2 Ag is
dependent on other cell surface proteins. Alternatively, binding of
Ly-6A.2 to its ligand may involve aggregation or clustering of Ly-6A.2
protein on the cell surface that cannot be achieved by the Ly-6A.2
fusion protein in solution.
A distinct feature of some Ly-6 proteins is their regulated expression during the development of hemopoietic cells. For example, Ly-6A.2 is expressed on stem cells (31, 32) as well as on a CD44+ subset of CD4-CD8-CD3- T cells in the thymus (17). Ly-6A.2 is absent on the majority of immature thymocytes, including the CD4+CD8+ subset, but is re-expressed on mature T cells in the thymus and spleen (33). The biological significance of the regulated developmental expression of Ly-6 proteins is unclear. The observations that dysregulation of expression of Ly-6A.2 results in a block of development at the CD4-CD8- cell stage in the thymus suggests that this developmental regulation is important. Considering the adhesion properties of Ly-6A.2 (13) and other Ly-6 proteins (14, 15, 16), we hypothesize that one of the functions of Ly-6 is to hold thymocytes in the appropriate thymic region where thymic selection occurs. Therefore, down-regulation of Ly-6A.2 expression will aid in loss of adhesion and movement of maturing T cells toward the medulla. Alternatively, it is also possible that Ly-6A.2 participates in signaling in immature T cells during their development. This later possibility is consistent with signaling property of Ly-6 proteins (2, 3).
A body of data on cell lines suggests that TCR stimulation is impaired in T cell hybridomas that have lost Ly-6A.2 expression due to mutations (34). Other experiments with normal T cells or T cell clones, in which Ly-6A.2 expression was extinguished by either antisense oligonucleotides (35) or Ly-6A.2 cDNA (36), respectively, support these findings. These previously published observations indicate that expression of Ly-6A.2 is critical for activation through the Ag receptor. Contrary to the above studies, CD4+ T cells from Ly-6A mutant mice are modestly hyper-responsive to anti-CD3 Ab compared with their wild-type littermates (37). In addition, T cells from Ly-6A.2-null mice have a prolonged proliferative response to Ag stimulation. Why Ly-6A.2 protein has both activating (as data with T-T hybridoma under in vitro culture conditions indicate) and inhibitory (Ly-6A mutant mice) roles in the Ag-specific response is unknown. One possibility is that the engagement of Ly-6A.2 ligand with overexpressed Ly-6A.2 may determine the outcome of the Ag-specific response. For example, interaction of Ly-6A.2 with its ligand may result in an inhibitory response, whereas a lack of this interaction may have the opposite effect. Alternatively, there may exist multiple ligands for Ly-6A.2 protein that can potentially produce different outcomes on interaction with Ly-6A.2. Identification of these ligands and studying their interaction with Ly-6A.2 are critical for understanding the biology of the Ly-6 family of proteins.
Some studies have suggested a role for Ly-6A/E expression in tumorigenicity (27) and metastases (28). One human homologue of the mouse Ly-6 gene, prostate stem cell Ag, is known to be prostate specific and therefore can be a potential target for diagnosis as well as therapy (38). What remains unclear is whether the expression of Ly-6 molecules contributes to tumor progression. Identification of a ligand(s) for the members of the Ly-6 multigene family may provide insight into this important issue.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Anil Bamezai, 615 Biological Sciences Building, University of Georgia, Athens, GA 30602.
3 Abbreviations used in this paper: PIPLC, phosphatidylinositol-specific phospholipase C; CHO, Chinese hamster ovary; DHFR, dihydrofolate reductase; PALS, periarteriolar lymphoid sheath.
Received for publication January 27, 2000. Accepted for publication July 12, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  C. L. Epting, J. E. Lopez, X. Shen, L. Liu, J. Bristow, and H. S. Bernstein Stem cell antigen-1 is necessary for cell-cycle withdrawal and myoblast differentiation in C2C12 cells J. Cell Sci., December 1, 2004; 117(25): 6185 - 6195. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Y. Ito, C. Y. J. Li, A. Bernstein, J. E. Dick, and W. L. Stanford Hematopoietic stem cell and progenitor defects in Sca-1/Ly-6A-null mice Blood, January 15, 2003; 101(2): 517 - 523. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. L. Pflugh, S. E. Maher, and A. L. M. Bothwell Ly-6 Superfamily Members Ly-6A/E, Ly-6C, and Ly-6I Recognize Two Potential Ligands Expressed by B Lymphocytes J. Immunol., November 1, 2002; 169(9): 5130 - 5136. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. C. Henderson, M. M. Kamdar, and A. Bamezai Ly-6A.2 Expression Regulates Antigen-Specific CD4+ T Cell Proliferation and Cytokine Production J. Immunol., January 1, 2002; 168(1): 118 - 126. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Gebauer, H. von Melchner, and T. Beckers Genomewide Trapping of Genes that Encode Secreted and Transmembrane Proteins Repressed by Oncogenic Signaling Genome Res., November 1, 2001; 11(11): 1871 - 1877. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
