| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Departments of
*
Immunology,
Pathology,
Molecular Biology,
Antibody Technology, and
¶ Cell Biology and Technology, Genentech, South San Francisco, CA 94080
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Human VE-JAM/JAM 2 and JAM 3 cDNA were isolated from a human colonic cDNA library by colony hybridization. Human IgG1 Fc fusion protein (immunoadhesins) of VE-JAM/JAM 2 (VE-JAM/JAM 2.Fc) or JAM 3 (JAM 3.Fc) was prepared, as previously described (3), and purified over a protein A column (Amersham Pharmacia Biotech, Piscataway, NJ). Identity of the purified protein was verified by N-terminal sequence analysis.
Expression cloning of JAM 3
Identification of JAM 3 was done by transiently transfecting pooled cDNA libraries encoding secreted and transmembrane proteins into COS cells grown on glass chamber slides. Twenty-four hours after transfection, VE-JAM/JAM 2.Fc was added (0.5 µg/ml) and incubated for 30 min. VE-JAM/JAM 2.Fc binding was determined. Positives were processed as previously described (4).
Ab generation
BALB/c females were immunized and boosted with 10 µg of
VE-JAM/JAM 2.Fc or 8xHis-tagged JAM 3 via footpad injections,
as previously described (5). Single clones were screened
against VE-JAM/JAM 2.Fc or 8xHis-tagged JAM 3. Selected clones were
tested for cross-reactivity against A33/JAM family members and human
IgG Fc. Clones were titrated out to single cell densities and
rescreened. Clone 12D10.2F9 (IgG1 
) was discovered to be selectively
reactive to VE-JAM/JAM 2 and not to JAM or JAM 3. Clone MaJIR1 (IgG2b)
was found to be selectively reactive to JAM 3 and not to JAM or
VE-JAM/JAM 2. Both clones were isolated and used for ascites fluid
generation (6). Abs were purified over a protein G column
(Amersham Pharmacia Biotech); protein concentration was determined
using the Pierce BCA reagent (Pierce, Rockford, IL).
Expression of VE-JAM/JAM 2 in CHO cells
VE-JAM/JAM 2 cDNA was amplified by PCR from a human colon cDNA library (Clontech Laboratories, Palo Alto, CA) using primers specific for the 5' and 3' ends of the coding sequence. The amplified fragment was gel purified, digested (BamHI and HindIII), extracted in phenol-chloroform-isoamyl alcohol (Life Technologies, Gaithersburg, MD), lyophilized, ligated into pSD5 expression vector (pSD5.huJAM), transfected into Chinese hamster ovary (CHO) cells, and selected, as previously described (7). Stable clones were screened for 12D10.2F9 reactivity; one clone, CHO-JI, found to have the highest expression, was used for characterization and further studies.
In situ hybridization
PCR primers (upper, 5'-GGGAAGATGGCGAGGAGGAG, and lower, 5'-CCAAGGCCACAAACGGAAATC) were designed to amplify a 776-bp fragment of VE-JAM/JAM 2. Primers included T7 or T3 RNA polymerase initiation sites to allow for in vitro transcription of sense or antisense probes, respectively, from the amplified products. Normal human tissues included tonsil, lymph node, spleen, kidney, urinary bladder, lung, heart, aorta, coronary artery, liver, gall bladder, prostate, stomach, small intestine, colon, pancreas, thyroid gland, skin, adrenal gland, placenta, uterus, ovary, testis, retina, brain (cerebellum, brainstem, cerebral cortex), and human fetal tissues (E12–E16 wk brain, spleen, bowel, thyroid). Tissues with chronic inflammatory disease included lungs with chronic asthma, chronic bronchopneumonia, chronic bronchitis/chronic obstructive pulmonary disease, kidneys with chronic lymphocytic interstitial nephritis, and livers with chronic inflammation and cirrhosis due to chronic hepatitis C infection, autoimmune hepatitis, or alcoholic cirrhosis. Primary human neoplasms were breast carcinoma, pulmonary squamous cell carcinoma, pulmonary adenocarcinoma, prostatic adenocarcinoma, and colonic adenocarcinoma. Tissues were fixed (4% formalin), paraffin-embedded, sectioned (3–5 µm thick), deparaffinized, deproteinated (20 µg/ml) with proteinase K (15 min at 37°C), and processed for in situ hybridization (8). Probes were 33P-UTP labeled, hybridized overnight (55°C), washed (0.1x SSC for 2 h at 55°C), dipped in NBT2 nuclear track emulsion (Eastman Kodak, Rochester, NY), exposed (4 wk at 4°C), and developed and counterstained with H&E.
Cell culture
J45 and Ramos cells were obtained from American Type Culture Collection (ATCC, Manassas, VA), and grown in conditions defined by ATCC.
Purification of peripheral blood cells
Blood was obtained from healthy adult volunteers by venous puncture and separated using Ficoll-Paque PLUS (Amersham Pharmacia Biotech) per the manufacturer’s instructions. PBMC were obtained from the interface, washed in cold PBS, lysed (with 0.2% NaCl for 30 s and neutralized with 1.6% NaCl) as needed, counted, and kept on ice at 5 x 107 cells/ml until use. By flow cytometric analysis, no contaminating platelets were observed in the purified PBMC fractions. Neutrophils were obtained from the pellet after lysis of contaminating RBCs. Neutrophils were washed in cold PBS, counted, and kept at 5 x 107 cells/ml until use on ice. To isolate peripheral blood subsets, "untouched" MACS kits (Miltenyi Biotec, Auburn, CA) were used following the manufacturer’s instructions.
Protein conjugation
VE-JAM/JAM 2.Fc, human IgG1, or JAM 3.Fc were biotinylated with 200 µg of EZ-Link sulfo-NHS-LC-biotin (Pierce) per 1 mg of protein in PBS for 30 min at room temperature. Biotinylation was quenched with the addition of (final concentration) 200 mM Tris, pH 8, and incubated for 30 min at room temperature. Biotinylated proteins were then dialyzed extensively against PBS and concentrated to a concentration of 2 mg/ml with Centricon-10 microconcentrators (Millipore, Bedford, MA).
Alexa-488 (Molecular Probes, Eugene, OR) protein conjugation kit was used per the manufacturer’s instructions for the conjugation of Alexa-488 onto VE-JAM/JAM 2.Fc or human IgG1.
Isolation of VE-JAM/JAM 2-binding cells by magnetic sorting
Biotinylated VE-JAM/JAM 2.Fc or human IgG1 were incubated (1 h
at 4°C) with PBMC (10 µg/107) in SerF buffer
(10% FBS (v/v; Life Technologies) plus 0.1%
NaN3 (w/v; Sigma-Aldrich, St. Louis, MO) in HBSS
without phenol red or sodium bicarbonate (HBSS+;
Life Technologies) buffered with 10 mM HEPES (Life Technologies), pH
7.4). Cells were washed in SerF buffer and resuspended at 80
µl/107 cells. Streptavidin magnetic beads
(Miltenyi Biotec) were added at 20 µl/107 cells
and incubated for 15 min at 4°C, washed with SerF buffer, resuspended
at 500 µl/108 cells, and passed over a positive
selecting MACS column (Miltenyi Biotec), per the manufacturer’s
instructions. Positively selected cells were eluted per the
manufacturer’s instructions, washed with SerF buffer, and analyzed by
flow cytometry for surface CD Ags at 2 x
105 cells per condition. In all experiments
conducted with this method, a large discrepancy was observed in the
recovery of VE-JAM/JAM 2.Fc-binding cells vs human IgG1-binding cells.
Whereas only 
25% of total applied cells can be recovered in the
VE-JAM/JAM 2.Fc conditions, >40% of total applied cells can be
recovered in the human IgG1 conditions. This discrepancy can only be
attributed to the function of the FcR, as the presence of excess
nonbiotinylated human IgG1 cannot cross-compete biotinylated human
IgG1, but can only establish a steady state binding of nonbiotinylated
vs biotinylated human IgG1. For VE-JAM/JAM 2.Fc, the nonbiotinylated
human IgG1 is competing for the FcR, leaving the VE-JAM/JAM 2 portion
of the Fc-fusion protein free to interact with cell surface receptors.
Because of this, all data are presented as percentage positive,
representing the percentage of positively stained cells in a total of
2 x 105 cells collected per staining
condition for flow cytometry.
Flow cytometry and FACS sorting
Cells for use in flow cytometric analysis were blocked for 30 min at 4°C with SerF buffer and stained with Abs to CD3, CD4, CD8, CD14, CD19, or CD56, conjugated to either FITC, PE, or CyChrome (BD PharMingen, San Diego, CA). For sorting, cells were incubated (30 min at 4°C) with Alexa-488-conjugated human IgG1 or VE-JAM/JAM 2.Fc (10 µg/106 cells) in a modified SerF buffer (SerF buffer with 5 µg/ml anti-CD16 Ab 3G8 (BD PharMingen) and 20 µg/ml human IgG1 (Calbiochem, San Diego, CA)), washed, and sorted on an Elite ESP (Beckman Coulter, Miami, FL). In these conditions, Alexa-488-conjugated human IgG1 was used as background. For competition assays, the competitor (20 µg/106 cells) was mixed with the cells for 20 min at room temperature in SerF buffer before Alexa-488-conjugated VE-JAM/JAM 2 or human IgG were introduced. The cells were then washed and analyzed by flow cytometry.
Adhesion/ELISA
For all assays, microtiter wells (NUNC Maxisorb 96-well plates; VWR, Scientific Products, Brisbrane, CA) were coated with conditions at 50 µl/well (in HBSS+), 10 µg/ml for 2 h at room temperature, unless otherwise noted. For adhesion assays, 50 µl of 10 µg/ml goat anti-human IgG1 Fc-specific Ab was first coated and blocked before the addition of conditions in binding/blocking buffer (BBB; HBSS+ containing 10% (v/v) FBS) for 1 h at room temperature before the addition of coating condition. Cells (5 x 106 cells/ml in BBB) were treated (10 min, 37°C/5% CO2) with 5 µg/ml 2',7'-bis-(2-carboxyethyl)-5 (and -6)-carboxyfluorescein, acetoxymethyl esther (BCECF AM) (Molecular Probes), washed, and allowed to adhere to coated wells (2 x 105 cells/well in BBB) for 1 h at 37°C/5% CO2. Plates were read on a SpectraMax fluorescence plate reader (Molecular Devices, Sunnyvale, CA) for total applied fluorescence, gently washed three times (by aspiration with a 28-gauge needle), and read for total adherent fluorescence. Percentage of adherence was calculated using the following equation: ((total fluorescence of adherent)/(total fluorescence of applied)) x 100. Blank wells consisted of BBB-coated wells exposed to BCECF AM-labeled J45 cells. Values obtained from the blank wells (percentage of adherence) were subtracted from all experimental conditions to derive a final value. For ELISA, the plates were blocked after condition coating with BBB for 30 min at room temperature and incubated with binding conditions for 1 h at room temperature. For conditions requiring EDTA, a modified BBB (HBSS without calcium and magnesium containing EDTA instead of the normal HBSS+) was used through the experiment. Plates were washed three times, incubated with 1 µg/ml streptavidin HRP (Pierce) for 30 min at room temperature, and assessed via color development using the tetramethylbenzidine substrate (Kirkegaard & Perry Laboratories, Gaithersburg, MD) and read on the ThermoMax Microplate Reader (Molecular Devices).
Immunoprecipitation and Western blotting
For biotinylated conditions, cells were first washed in HBSS+ before being biotinylated (200 µg/106 cells) with sulfo-NHS-LC-biotin (30 min at 4°C). Cells were washed with TBS (30 min at 4°C) to quench the biotinylation.
Cells were lysed (108 cells/ml) with lysis buffer (HBSS+ containing 1% Triton X-100 and 1 Complete-Mini EDTA free protease inhibitor tablet (Roche Biochemicals, Indianapolis, IN) per 7 ml of lysis buffer) for 30 min at 4°C. Lysates were spun at 22,000 x g (1 h at 4°C) and 0.2 µm filtered. Lysates were precleared (2 h at 4°C) with 5 µl/106 cells of recombinant protein A beads (Amersham Pharmacia Biotech). Cleared lysates were 0.2 µm filtered and incubated (2 h at 4°C) with 5 µg/106 cells of either VE-JAM/JAM 2.Fc or human IgG1, conjugated to protein A matrix using the ImmunoPure Protein A IgG Plus Orientation kit (Pierce). Beads were pelleted and washed with lysis buffer and denatured by the addition of 15 µl/106 cells of nonreducing SDS sample buffer (standard sample buffer with 2 mM iodoacetamide, but without DTT or 2-ME) and boiled for 3 min at 100°C.
Samples (at 15 µl/lane) were resolved on a 4–20% Bio-Rad Tris-HCl Ready Gel (Bio-Rad, Hercules, CA) and transferred onto 0.2-µm Protran nitrocellulose membrane (Schleicher & Schuell, Keene, NH) at 100 mA for 2 h at 4°C. Blots were blocked for 1 h in Blotto (TBS containing 5% nonfat milk and 0.05% Tween 20; Bio-Rad). For biotinylated samples, HRP-conjugated streptavidin (Pierce) was used at 0.5 µg/ml for 30 min at room temperature. For nonbiotinylated samples, MaJIR1 was used at 10 µg/ml in Blotto and incubated for 1 h at 25°C before the application of 1 µg/ml HRP-conjugated goat anti-mouse IgG (Caltag Laboratories, Burlingame, CA) in Blotto for 30 min at room temperature. Blots were washed thoroughly with TTBS (TBS containing 0.05% Tween 20) and developed with the ECL Plus reagent (Amersham Pharmacia Biotech) before exposing onto Kodak BioMax ML film and development with Kodak M35A X-OMAT Film Processor (Eastman Kodak).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Screening expressed sequence tag databases for homologies to JAM and A33-Ag, we identified a full-length cDNA that encodes alternative splice forms of 298 or 312 aa, differing in the C terminus. The shorter of the two forms is the recently described VE-JAM (1), or JAM-2 (2), termed in this study as VE-JAM/JAM 2. Similar to human JAM (huJAM) (9, 10, 11) and A33-Ag (12), VE-JAM/JAM 2 is predicted to possess two potential Ig-like loops in the extracellular domain: a short single-transmembrane domain and a short cytoplasmic tail (1, 2).
Localization of VE-JAM/JAM 2 on normal tissue
Expression of VE-JAM/JAM 2 mRNA was evaluated by in situ hybridization. In the evaluated human fetal tissues (E12–16 wk brain, spleen, bowel, and thyroid), VE-JAM/JAM 2 expression was predominantly endothelial. More specifically, VE-JAM/JAM 2 was present only in the vascular endothelium of small and large vessels (excluding capillaries), in mesenteric vessels, mural vessels of the bowel wall, and small vessels of the developing mesenteric lymph nodes and thyroid (data not shown). Expression was absent in spleen.
Expression of VE-JAM/JAM 2 mRNA in normal adult human tissue was
restricted. Expression was found in the endothelium of HEV in tonsils
and lymph nodes (data not shown) (1), the spermatogenic
cells of the epithelium in the testicular seminiferous tubules (Fig. 1
, I and J), and
the intermediate trophoblasts of the placenta (data not shown).
|
The expression of VE-JAM/JAM 2 was more extensive in tissues with
chronic inflammatory diseases. Lung tissues included lungs with chronic
asthma, chronic bronchopneumonia, and chronic bronchitis/chronic
obstructive pulmonary disease. Kidney tissues included kidney with
chronic lymphocytic interstitial nephritis. Also examined were livers
with chronic inflammation and cirrhosis due to chronic hepatitis C
infection, autoimmune hepatitis, or alcoholic cirrhosis. In biopsies of
lung with chronic bronchopneumonia (Fig. 1
, A–F),
VE-JAM/JAM 2 was expressed in the endothelium of small-, medium-, and
large-caliber arterioles present within, or immediately adjacent to,
foci of lymphocytic inflammation. This VE-JAM/JAM 2 expression was not
observed in normal lungs (Fig. 1, G–H). Kidney with chronic
lymphocytic interstitial nephritis and liver with chromic lymphocytic
hepatitis were also examined. While VE-JAM/JAM 2 expression was again
restricted to the endothelium of arterioles in and adjacent to sites of
lymphocytic inflammation, no VE-JAM/JAM 2 expression was found in the
liver specimen (data not shown), suggesting VE-JAM/JAM 2 mRNA
expression may not be inducible in all inflammatory tissue types.
Localization of VE-JAM/JAM 2 in primary neoplasms
The expression of VE-JAM/JAM 2 mRNA was also examined in a number
of primary neoplasms (breast carcinoma, pulmonary squamous cell
carcinoma, pulmonary adenocarcinoma, prostatic adenocarcinoma, colonic
adenocarcinoma). Discrete VE-JAM/JAM 2 expression was seen in the
endothelium of small- and medium-caliber arterioles in the following
tissue samples: colonic adenocarcinoma (five of six cases, data not
shown), testicular carcinomas (two of two cases; Fig. 2, A and B),
pulmonary adenocarcinoma (three of five cases; Fig. 2, C and
D), and mammary adenocarcinoma (one of three cases; Fig. 2, E and F). The expression of VE-JAM/JAM 2 is
highly selective, as illustrated on a low magnification survey or a
breast carcinoma (Fig. 2, G and H, *) with
adjacent normal breast tissue (Fig. 2, G and H,
denoted by an arrow). The expression of VE-JAM/JAM 2 is observed only
in vessels adjacent to the tumor site (Fig. 2, G and
H, tumor denoted by arrowheads); no VE-JAM/JAM 2 staining
could be observed in the normal breast tissue.
|
Having identified VE-JAM/JAM 2 expression on certain
isolated endothelial cell types, we hypothesized VE-JAM/JAM 2 as a
leukocyte trafficking/adhesion molecule. To examine this, we generated
a biotinylated VE-JAM/JAM 2/human IgG fusion protein (VE-JAM/JAM 2.Fc)
and isolated the VE-JAM/JAM 2-interacting peripheral blood leukocytes
using streptavidin-conjugated magnetic beads. Isolated cells were then
examined for surface CD-Ag expression. Comparing results obtained with
the biotinylated VE-JAM/JAM 2.Fc with those obtained with biotinylated
human IgG, four cell populations stood out. They were
CD3+ cells (2.39% for human IgG, 20.99% for
VE-JAM/JAM 2), CD8+ cells (1.78% for human
IgG, 6.68% for VE-JAM/JAM 2), CD19+ cells
(4.42% for human IgG, 9.66% for VE-JAM/JAM 2), and
CD56+ cells (6.69% for human IgG, 36.89% for
VE-JAM/JAM 2). From these results, we conclude VE-JAM/JAM 2 to interact
with peripheral blood T cells (CD3+), B cells
(CD19+), and NK cells
(CD56+). To confirm VE-JAM/JAM 2 binding to
at least one of these cell types, isolated
CD56+ NK cells were examined for VE-JAM/JAM 2.Fc
interactions by flow cytometry using Alexa-488-conjugated VE-JAM/JAM 2
(VE-JAM/JAM 2.Fc.488). As shown in Fig. 3A,
CD56+ NK cells interacted specifically with
VE-JAM/JAM 2.Fc.488 (Fig. 3A, solid line; the gray peak
represents the Alexa-488-conjugated human IgG (HI.488) control). This
interaction was reduced to near baseline by the addition of excess
unlabeled VE-JAM/JAM 2 (Fig. 3A, gray line), confirming that
CD56+ NK cells interacted with VE-JAM/JAM 2.Fc in
a specific fashion. Next we examined peripheral blood cells for
VE-JAM/JAM 2 expression using mAb 12D10.2F9, a mAb generated against
VE-JAM/JAM 2 extracellular domain that has no observable
cross-reactivity with JAM (Fig. 3B). No significant
expression of VE-JAM/JAM 2 was found on any of the isolated peripheral
blood cell populations. huJAM, detected by mAb 10A5 (9),
was expressed on all peripheral blood cell populations, confirming
previous reports (Fig. 3C, upper half of table)
(9, 10, 11). We then proceeded to further define VE-JAM/JAM
2-binding peripheral blood cells. Using VE-JAM/JAM 2.488, we began
sorting VE-JAM/JAM 2-binding peripheral blood cells by FACS sorting;
HI.488 was used as a negative control. Roughly 30% of total applied
lymphocytes were found to interact with VE-JAM/JAM 2; of those cells,
12.6% were CD3+ T cells, 32.4% were
CD8+ T cells, and 50.4% were
CD56+ NK cells. Surprisingly, no
CD19+ B cells were found in any significance;
monocyte contamination (CD14+ cells) was <1%
(Fig. 3C). Of the CD56+ NK cells,
22.4% expressed CD3 and 40.2% expressed CD8. Of the
CD8+ T cells, 23.5% expressed CD3 and 73.2%
expressed CD56. Thus, VE-JAM/JAM 2 can interact specifically with
classical NK cells (CD56+), NK/T cells
(CD56+CD3+), and cytolytic
T cells
(CD56+CD3+CD8+).
|
To address the difference seen between the VE-JAM/JAM
2-interacting cells obtained by MACS purification and FACS sorting,
namely that of the CD19+ B
cells, purified B cells were generated. These B cells were analyzed for
VE-JAM/JAM 2 interaction by flow cytometry and were concluded to be
negative (11.56% for HI.488 vs 14.86% for VE-JAM/JAM 2.Fc.488,
representative of four independent experiments; data not shown). This
finding would suggest that the result obtained through MACS
purification is sensitive to cell surface FcR and may represent the
error margin inherent in the purification system. Furthermore,
neutrophils and CD14+ monocytes were reexamined
and concluded negative for VE-JAM/JAM 2 interaction (neutrophils,
13.1% for HI.488 vs 13.02% for VE-JAM/JAM 2.Fc.488; monocytes,
12.82% for HI.488 vs 10.65% for VE-JAM/JAM 2.Fc.488, representative
of six different experiments, data not shown). Also examined were PBDCs
from Clonetics (San Diego, CA). These peripheral blood dendritic cells
were found to interact with VE-JAM/JAM 2 (Fig. 3DI,
VE-JAM/JAM 2.Fc.488 in solid line, HI.488 in shaded histogram). Taken
together, VE-JAM/JAM 2-interacting cell types are as follows:
CD56+ NK cells,
CD56+CD3+ NK/T cells,
CD56+CD3+CD8+
cytolytic T cells, and PBDCs. Much like the peripheral cell types
examined, PBDCs had no detectable VE-JAM/JAM 2 expression (Fig. 3DII, mouse IgG in filled histogram, 12D10.2F9 in gray
line).
VE-JAM/JAM 2 can serve as an adhesive ligand for J45 cells
In the previously published JAM 2 report, T cell lines were
demonstrated to interact with JAM 2-transfected CHO cells
(2). We attempted to duplicate this by using J45, a
CD3+ T cell line. J45 cells have no detectable
surface expression of VE-JAM/JAM 2 (data not shown). By flow cytometry,
J45 cells were shown to bind VE-JAM/JAM 2 in solution (Fig. 4A: HI.488 in filled
histogram, VE-JAM/JAM 2.Fc.488 in solid line, VE-JAM/JAM 2.Fc.488 plus
mAb 12D10.2F9 in gray line). Using plate-based adhesion assays,
J45 cells were found to adhere to VE-JAM/JAM 2 in an Ab-sensitive
fashion (Fig. 4B, human IgG, open bar; VE-JAM/JAM 2.Fc,
filled bar; VE-JAM/JAM 2.Fc plus mAb 12D10.2F9, shaded bar). This
adhesive event is efficiently inhibited by mAb 12D10.2F9.
|
Using J45 cells, we began to isolate the cell surface receptor for
VE-JAM/JAM 2 binding. By first surface biotinylating, then lysing and
immunoprecipitating with an Fc-cross-linked VE-JAM/JAM 2.Fc protein A
matrix, we were able to obtain a single avidin-reactive band of 40
kDa (Fig. 4C, lane 2). This band was not seen in
conditions immunoprecipitated with an Fc-cross-linked human IgG/protein
A matrix (Fig. 4C, lanes 1 and 3), nor
in VE-JAM/JAM 2-immunoprecipitating conditions from the non-VE-JAM/JAM
2-binding B cell line Ramos (Fig. 4C, lane
4).
Identification of the VE-JAM/JAM 2 receptor, JAM 3
While attempting to bulk purify the 37-kDa VE-JAM/JAM
2-immunoprecipitated band, we chanced upon a protein that was
identified as a VE-JAM/JAM 2-interacting protein through expression
cloning. This protein, found to be a previously cloned molecule termed
JAM 3 (accession number NP_113658), is a member of the A33/JAM family
of Ig superfamily proteins. Like huJAM, A33, and VE-JAM/JAM 2, JAM 3
contains a single predicted transmembrane domain, two putative Ig
loops, and a short predicted cytoplasmic C terminus (Fig. 5A). Abs against a
6xHis-tagged JAM 3 extracellular domain were raised; one particular
clone, in this study termed MaJIR1 (mouse anti-JAM 3), was singled
out for its inhibitory activity.
|
, B and
C). VE-JAM/JAM 2 interactions with JAM 3 in this format were
also blocked by the anti-VE-JAM/JAM 2 Ab 12D10.2F9 (data not
shown).
JAM 3 expression on J45 cells and CD56+ NK cells
was confirmed by functional assays and bolstered by
immunoprecipitation/Western blotting experiments. The
immunoprecipitation/Western blotting experiments confirmed JAM 3 as the
40-kDa VE-JAM/JAM 2-interacting band that was described in J45 cells
above (Fig. 6A, lane
6; lane 5 represents the human IgG/protein A control).
PBMC treated in a similar fashion also yield JAM 3 as the VE-JAM/JAM
2-interacting protein (Fig. 6A, lane 2;
lane 1 representing the human IgG/protein A control).
VE-JAM/JAM 2.Fc protein A matrix showed no cross-reactivity to MaJIR1
in the 40-kDa region (Fig. 6A, lane 7). JAM 3
was also identified as the adhesion mediator in VE-JAM/JAM 2-dependent
J45 adhesion, as anti-JAM 3 was able to inhibit VE-JAM/JAM
2-mediated J45 adhesion (Fig. 6B). JAM 3 was also shown to
be the VE-JAM/JAM 2 receptor on CD56+ PBLs, as
excess free JAM 3 succeeded in inhibiting VE-JAM/JAM 2 binding by
CD56+ PBLs (Fig. 6C, histograms as
labeled). In similar experiments, anti-JAM 3 was also shown to be
able to inhibit VE-JAM/JAM 2 binding to CD56+
PBLs (data not shown).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In this study, we give a brief description pertaining to VE-JAM/JAM 2 localization by in situ hybridization in both normal and diseased human tissue. In normal fetal tissue, VE-JAM/JAM 2 is present in the vascular endothelium of small and large vessels, while in normal adult tissue, VE-JAM/JAM 2 is found in the endothelium of tonsillar and lymphatic HEV, as previously reported (1, 2). We also identified VE-JAM/JAM 2 mRNA expression the spermatogenic cells of the epithelium in testicular seminiferous tubules, and in the intermediate trophoblasts of placenta for the first time. Also for the first time, we observe VE-JAM/JAM 2 expression in tumor and chronically inflamed samples. VE-JAM/JAM 2 was found in the endothelium of small-, medium-, and large-caliber arterioles within or around foci of lymphocytic inflammation in some tissues (of the tissues examined, VE-JAM/JAM 2 expression as described was seen in lung and kidney but not in liver). In samples of primary neoplasm, VE-JAM/JAM 2 expression was seen in the endothelium of small- and medium-caliber arterioles within or around tumor foci (breast carcinoma, pulmonary squamous cell carcinoma, pulmonary adenocarcinoma, prostatic adenocarcinoma, colonic adenocarcinoma). These findings confirm VE-JAM/JAM 2 localization (1, 2), furthering the understanding of VE-JAM/JAM 2 localization beyond that which was previously reported, and demonstrate, for the first time, epithelial localization of VE-JAM/JAM 2.
Because VE-JAM/JAM 2 message was observed by in situ hybridization in
the aforementioned inflammatory tissues, we had attempted to stimulate
VE-JAM/JAM 2 protein expression by cytokine/chemokine treatment of
cultured endothelial cells (HUVECs). The following cytokines/chemokines
were tested: TNF-
, TGF-
, IL-1, IL-4, IFN-
, Exodus-2,
stromal cell-derived factor-1, stromal cell-derived
factor-1, and GM-CSF. No significant up-regulation of surface
VE-JAM/JAM 2 protein expression was observed on HUVECs treated for up
to 72 h (our unpublished observations). Such results led us
to conclude that the cytokines/chemokines used cannot up-regulate
VE-JAM/JAM 2 on the surface of HUVECs in a cultured, in vitro setting.
However, as the responsiveness of endothelial cells to
cytokines/chemokines can be drastically different in an in vitro vs an
in vivo setting, the lack of VE-JAM/JAM 2 protein up-regulation in
vitro is not unexpected. Such data may point to a more complicated
regulatory pathway that needs to take place in vivo before VE-JAM/JAM 2
protein up-regulation can be observed.
Also described for the first time is the interaction of VE-JAM/JAM 2
with discrete subsets of PBLs. VE-JAM/JAM 2, not detected on any
subsets of peripheral blood cells by flow cytometry, was found to
interact specifically with CD56+ NK cells,
CD56+CD3+ NK T cells,
CD56+CD3+CD8+
cytolytic T cells, and PBDCs. The VE-JAM/JAM 2 lymphocyte interactions,
taken together with the in situ localization data presented above, make
a strong argument for VE-JAM/JAM 2 as a molecule important for
lymphocyte trafficking to and from secondary lymphoid tissues. These
findings also support the potential role of VE-JAM/JAM 2 as a
participant in immune surveillance and Ag-dependent immune responses.
These suppositions were further supported by the findings describing
VE-JAM/JAM 2 as an adhesion mediator for J45 T cells. By
immunoprecipitating with a VE-JAM/JAM 2.Fc protein A matrix, we were
able to further identify a single 40-kDa cell surface molecule on
J45 cells as the VE-JAM/JAM 2-interacting protein. This 40-kDa J45
cell surface molecule was confirmed to be JAM 3 through
immunoprecipitation/Western blotting experiments. This led us to
hypothesize and prove, through both protein/protein and protein/cell
assays, that JAM 3 was the receptor for VE-JAM/JAM 2-dependent adhesion
on J45 cells and the VE-JAM/JAM 2-binding receptor on
CD56+ PBLs.
Certain observations from the experimental results gave some insight into the interaction between VE-JAM/JAM 2 and JAM 3. Also synthesized were 8xHis-tagged VE-JAM/JAM 2 fusion proteins. However, regardless of where the tag was applied, the 8xHis-tagged version of the VE-JAM/JAM 2 fusion proteins was only reactive to mAb 12D10.2F9 but did not interact with JAM 3. Similarly constructed JAM 3/8xHis-tagged fusion proteins were fully functional, binding to JAM 2.Fc. Also generated were Fab-like fragments of the VE-JAM/JAM 2.Fc. These, again, were reactive to mAb 12D10.2F9 but did not interact with JAM 3. The two observations stated above can suggest that VE-JAM/JAM 2 is required to be at least a dimer for proper function, an observation previously seen with huJAM (9, 14). This hypothesis can also explain why VE-JAM/JAM 2.Fc is required to be cross-linked/captured by a secondary Ab before it can support J45 cells adhesion.
With the findings reported in this work, we believe we have furthered the understanding of VE-JAM/JAM 2 function. In this study, we have identified a candidate receptor/ligand pair that may be involved in the process of the T lymphocyte, NK cell, and dendritic cell trafficking and recruitment to sites of inflammation.
|
Footnotes |
|---|
2 Abbreviations used in this paper: JAM, junctional adhesion molecule; BBB, binding/blocking buffer; PBDC, peripheral blood, GM-CSF/IL-4-derived dendritic cell; BCECF AM, 2',7'-bis-(2-carboxyethyl)-5 (and -6)-carboxyfluorescein, acetoxymethyl esther; CHO, Chinese hamster ovary; HEV, high endothelial venule; huJAM, human JAM; VE-JAM, vascular endothelial-JAM.
Received for publication June 28, 2001. Accepted for publication December 12, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  B. J. Rabquer, Y. Hou, F. Del Galdo, G. Kenneth Haines III, M. L. Gerber, S. A. Jimenez, J. R. Seibold, and A. E. Koch The proadhesive phenotype of systemic sclerosis skin promotes myeloid cell adhesion via ICAM-1 and VCAM-1 Rheumatology, July 1, 2009; 48(7): 734 - 740. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Praetor, J. M. McBride, H. Chiu, L. Rangell, L. Cabote, W. P. Lee, J. Cupp, D. M. Danilenko, and S. Fong Genetic deletion of JAM-C reveals a role in myeloid progenitor generation Blood, February 26, 2009; 113(9): 1919 - 1928. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A.-C. Luissint, P. G. Lutz, D. A. Calderwood, P.-O. Couraud, and S. Bourdoulous JAM-L-mediated leukocyte adhesion to endothelial cells is regulated in cis by {alpha}4{beta}1 integrin activation J. Cell Biol., December 15, 2008; 183(6): 1159 - 1173. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Sugano, M. Takeuchi, A. Hirata, H. Matsushita, T. Kitamura, M. Tanaka, and A. Miyajima Junctional adhesion molecule-A, JAM-A, is a novel cell-surface marker for long-term repopulating hematopoietic stem cells Blood, February 1, 2008; 111(3): 1167 - 1172. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Galkina and K. Ley Vascular Adhesion Molecules in Atherosclerosis Arterioscler Thromb Vasc Biol, November 1, 2007; 27(11): 2292 - 2301. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. F. Bradfield, S. Nourshargh, M. Aurrand-Lions, and B. A. Imhof JAM Family and Related Proteins in Leukocyte Migration (Vestweber Series) Arterioscler Thromb Vasc Biol, October 1, 2007; 27(10): 2104 - 2112. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. F. Bradfield, C. Scheiermann, S. Nourshargh, C. Ody, F. W. Luscinskas, G. E. Rainger, G. B. Nash, M. Miljkovic-Licina, M. Aurrand-Lions, and B. A. Imhof JAM-C regulates unidirectional monocyte transendothelial migration in inflammation Blood, October 1, 2007; 110(7): 2545 - 2555. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Sakaguchi, M. Nishimoto, S. Miyagi, A. Iwama, Y. Morita, N. Iwamori, H. Nakauchi, H. Kiyonari, M. Muramatsu, and A. Okuda Putative "Stemness" Gene Jam-B Is Not Required for Maintenance of Stem Cell State in Embryonic, Neural, or Hematopoietic Stem Cells. Mol. Cell. Biol., September 1, 2006; 26(17): 6557 - 6570. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Lamagna, P. Meda, G. Mandicourt, J. Brown, R. J.C. Gilbert, E. Y. Jones, F. Kiefer, P. Ruga, B. A. Imhof, and M. Aurrand-Lions Dual Interaction of JAM-C with JAM-B and {alpha}M{beta}2 Integrin: Function in Junctional Complexes and Leukocyte Adhesion Mol. Biol. Cell, October 1, 2005; 16(10): 4992 - 5003. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Corada, S. Chimenti, M. R. Cera, M. Vinci, M. Salio, F. Fiordaliso, N. De Angelis, A. Villa, M. Bossi, L. I. Staszewsky, et al. Junctional adhesion molecule-A-deficient polymorphonuclear cells show reduced diapedesis in peritonitis and heart ischemia-reperfusion injury PNAS, July 26, 2005; 102(30): 10634 - 10639. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Zen, Y. Liu, I. C. McCall, T. Wu, W. Lee, B. A. Babbin, A. Nusrat, and C. A. Parkos Neutrophil Migration across Tight Junctions Is Mediated by Adhesive Interactions between Epithelial Coxsackie and Adenovirus Receptor and a Junctional Adhesion Molecule-like Protein on Neutrophils Mol. Biol. Cell, June 1, 2005; 16(6): 2694 - 2703. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Aurrand-Lions, C. Lamagna, J. P. Dangerfield, S. Wang, P. Herrera, S. Nourshargh, and B. A. Imhof Junctional Adhesion Molecule-C Regulates the Early Influx of Leukocytes into Tissues during Inflammation J. Immunol., May 15, 2005; 174(10): 6406 - 6415. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Chavakis, T. Keiper, R. Matz-Westphal, K. Hersemeyer, U. J. Sachs, P. P. Nawroth, K. T. Preissner, and S. Santoso The Junctional Adhesion Molecule-C Promotes Neutrophil Transendothelial Migration in Vitro and in Vivo J. Biol. Chem., December 31, 2004; 279(53): 55602 - 55608. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Zen, B. A. Babbin, Y. Liu, J. B. Whelan, A. Nusrat, and C. A. Parkos JAM-C Is a Component of Desmosomes and a Ligand for CD11b/CD18-mediated Neutrophil Transepithelial Migration Mol. Biol. Cell, August 1, 2004; 15(8): 3926 - 3937. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. D. van Buul and P. L. Hordijk Signaling in Leukocyte Transendothelial Migration Arterioscler Thromb Vasc Biol, May 1, 2004; 24(5): 824 - 833. [Abstract] [Full Text]
|
|
|
|
|
|
K. J. Mandell, I. C. McCall, and C. A. Parkos Involvement of the Junctional Adhesion Molecule-1 (JAM1) Homodimer Interface in Regulation of Epithelial Barrier Function J. Biol. Chem., April 16, 2004; 279(16): 16254 - 16262. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Raschperger, U. Engstrom, R. F. Pettersson, and J. Fuxe CLMP, a Novel Member of the CTX Family and a New Component of Epithelial Tight Junctions J. Biol. Chem., January 2, 2004; 279(1): 796 - 804. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Ebnet, A. Suzuki, S. Ohno, and D. Vestweber Junctional adhesion molecules (JAMs): more molecules with dual functions? J. Cell Sci., January 1, 2004; 117(1): 19 - 29. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Ebnet, M. Aurrand-Lions, A. Kuhn, F. Kiefer, S. Butz, K. Zander, M.-K. M. z. Brickwedde, A. Suzuki, B. A. Imhof, and D. Vestweber The junctional adhesion molecule (JAM) family members JAM-2 and JAM-3 associate with the cell polarity protein PAR-3: a possible role for JAMs in endothelial cell polarity J. Cell Sci., October 1, 2003; 116(19): 3879 - 3891. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Hirabayashi, M. Tajima, I. Yao, W. Nishimura, H. Mori, and Y. Hata JAM4, a Junctional Cell Adhesion Molecule Interacting with a Tight Junction Protein, MAGI-1 Mol. Cell. Biol., June 15, 2003; 23(12): 4267 - 4282. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. E. Prota, J. A. Campbell, P. Schelling, J. C. Forrest, M. J. Watson, T. R. Peters, M. Aurrand-Lions, B. A. Imhof, T. S. Dermody, and T. Stehle Crystal structure of human junctional adhesion molecule 1: Implications for reovirus binding PNAS, April 29, 2003; 100(9): 5366 - 5371. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Takai and H. Nakanishi Nectin and afadin: novel organizers of intercellular junctions J. Cell Sci., January 1, 2003; 116(1): 17 - 27. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. A. Johnson-Leger, M. Aurrand-Lions, N. Beltraminelli, N. Fasel, and B. A. Imhof Junctional adhesion molecule-2 (JAM-2) promotes lymphocyte transendothelial migration Blood, September 18, 2002; 100(7): 2479 - 2486. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. A. Cunningham, J. M. Rodriguez, M. P. Arrate, T. M. Tran, and T. A. Brock JAM2 Interacts with alpha 4beta 1. FACILITATION BY JAM3 J. Biol. Chem., July 26, 2002; 277(31): 27589 - 27592. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) and can be neutralized by the addition of
0.5 µg/well mAb 12D10.2F9 (
). Data are representative of four
independent experiments; error bars represent the SD of
n = 4 samples. C, Biotinylated J45
(lanes 1 and 2) or Ramos (lanes
3 and 4) cells were lysed and immunoprecipitated
against an Fc-cross-linked VE-JAM/JAM 2.Fc protein A matrix (lanes 2 and
4) or an Fc-cross-linked human IgG1/protein A matrix
(lanes 1 and 3). Samples were run on a
4–20% SDS-PAGE gel and blotted with streptavidin HRP. As can be seen
in lane 2, a unique 
, Putative signal cleavage site; •,
conserved extracellular cysteines. The transmembrane domain is
underlined; dotted lines overlie the potential
N-glycosylation sites. B, Plate-bound JAM
3.Fc or huJAM.Fc were exposed to biotinylated VE-JAM/JAM 2.Fc
(VE-JAM/JAM 2.Fc biotin, used at specified microgram per milliliter
concentrations) in the presence of 0.25 µg/well mouse IgG or the
anti-JAM 3 Ab MaJIR1. Streptavidin HRP was used to detect; results
were obtained from enzymatic color change assay. JAM 3.Fc-coated wells
demonstrated VE-JAM/JAM 2.Fc biotin binding (
); huJAM.Fc-coated
wells (
) did not. Mouse IgG (•) had no effect, while the
anti-JAM 3 Ab MaJIR1 (
) inhibited VE-JAM/JAM 2 binding. Data are
representative of four independent experiments; error bars represent
the SD in an n = 4 condition. C,
VE-JAM/JAM 2.Fc were captured onto a plate, and biotinylated JAM 3.Fc
(JAM 3.Fc biotin) was used, at specified µg/ml concentrations, to
examine VE-JAM/JAM 2 and JAM 3 interaction. Mouse IgG or MaJIR1 were
used at 0.5 µg/ml. The anti-JAM 3 Ab MaJIR1 was again able to
inhibit the VE-JAM/JAM 2 and JAM 3 interaction. Data are representative
of three independent experiments; error bars represent the SD in an
n = 4 condition.