ORIGINAL ARTICLE 368

High Affinity Interaction of

Integrin a4b1 (VLA-4) and
Vascular Cell Adhesion Molecule
1 (VCAM-1) Enhances Migration
of Human Melanoma Cells Across
Activated Endothelial Cell Layers
MARTIN KLEMKE, TATJANA WESCHENFELDER,
MATHIAS H. KONSTANDIN, AND YVONNE SAMSTAG*
Institute for Immunology, University of Heidelberg, Heidelberg, Germany

The capacity of tumor cells to form metastatic foci correlates with their ability to interact with and migrate through endothelial cell layers.
This process involves multiple adhesive interactions between tumor cells and the endothelium. Only little is known about the molecular
nature of these interactions during extravasation of tumor cells. In human melanoma cells, the integrin avb3 is involved in transendothelial
migration and its expression correlates with metastasis. However, many human melanoma cells do not express b3 integrins. Therefore, it
remained unclear how these cells undergo transendothelial migration. In this study we show that human melanoma cells with different
metastatic potency, which do not express b2 or b3 integrins, express the VCAM-1 receptor a4b1. VCAM-1 is up-regulated on activated
endothelial cells and is known to promote transendothelial migration of leukocytes. Interestingly, despite comparable cell surface levels of
a4b1, only the highly metastatic melanoma cell lines MV3 and BLM, but not the low metastatic cell lines IF6 and 530, bind VCAM-1 with
high affinity without further stimulation, and are therefore able to adhere to and migrate on isolated VCAM-1. Moreover, we demonstrate
that function-blocking antibodies against the integrin a4b1, as well as siRNA-mediated knock-down of the a4 subunit in these highly
metastatic human melanoma cells reduce their transendothelial migration. These data imply that only high affinity interactions between the
integrin a4b1 on melanoma cells and VCAM-1 on activated endothelial cells may enhance the metastatic capacity of human b2/b3-
negative melanoma cells.
J. Cell. Physiol. 212: 368374, 2007. 2007 Wiley-Liss.

Tumor cell metastasis is a complex process and consists of a
series of sequential steps. Initially, tumor cells have to detach
from the primary tumor, migrate through the tissue and invade
the lymphatic system or blood vessels. As a next step,
circulating tumor cells temporarily adhere to endothelial cells
and then extravasate by infiltrating the underlying basement
membrane. Finally, cells migrate to a suitable location where
they form metastases (Stetler-Stevenson et al., 1993; Nicolson,
1988; Engers and Gabbert, 2000; Orr et al., 2000). The
capacity of tumor cells to form metastatic foci correlates
with their ability to interact with and migrate through the
endothelial cell layer (transendothelial migration). Although
much is known about the adhesive interactions of
non-hematopoetic tumor cells with components of the basal
membrane, surprisingly little is known about the mechanisms by
which non-hematopoetic tumor cells adhere to and pass
through the endothelial cell layer.
Most of the knowledge about molecular mechanisms involved
in the process of transendothelial migration comes from studies
performed with leukocytes (Ebnet et al., 1996; Worthylake and
Burridge, 2001; Alon and Feigelson, 2002; Luscinskas et al.,
2002). For migration of most leukocytes across endothelial cell
layers, interactions between the b2 integrins aLb2 (LFA-1,
CD11a/CD18) or aMb2 (Mac-1, CD11b/CD18) and their
ligand ICAM-1 are essential (Oppenheimer-Marks et al., 1991;
Reiss and Engelhardt, 1999; Wong et al., 1999; Lyck et al., 2003).
These b2 integrins have also been implicated in transendothelial
migration of certain non-hematopetic human tumor cells
(Wang et al., 2005). However, most non-hematopoetic tumor
cells do not express b2 integrins (Varner and Cheresh, 1996;
Mizejewski, 1999) and are therefore not able to bind ICAM-1 on
the endothelial cell surface. Some non-hematopoetic human
tumor cells express the avb3 integrin, which binds to the
adhesion molecule L1 on endothelial cells (Danen et al., 1995;
Saini et al., 1997; Varner and Cheresh, 1996; Mizejewski, 1999;
Voura et al., 2001). In certain human melanoma cells,
expression levels of this integrin correlate with the metastatic
potential of these cells (Albelda et al., 1990a; Saini et al., 1997;
Hofmann et al., 2000), and this integrin is a critical component
for transendothelial migration of these cells (Voura et al., 2001).
Finally human melanoma cells exist which neither express b2
nor b3 integrins, but are still highly metastatic (Van Muijen et al.,
1991a,b). These cells should therefore also be able to efficiently
undergo transendothelial migration. From leukocytes we know
that, besides b2 integrins, the a4b1 integrin and its ligand
VCAM-1 are involved in the interaction of leukocytes with
endothelial cells during transendothelial migration
(Oppenheimer-Marks et al., 1991; Chan and Aruffo, 1993;
Chuluyan and Issekutz, 1993; Hourihan et al., 1993; Weber and
Springer, 1998; Ding et al., 2001). The a4b1 integrin has a
flexible molecular structure allowing initial capture, tethering,
rolling and firm attachment of hematopoetic cells to endothelial
*Correspondence to: Yvonne Samstag, Institute for Immunology,
University of Heidelberg, Im Neuenheimer Feld 305, 69120
Heidelberg, Germany. E-mail: yvonne.samstag@urz.uni-heidelberg.de
Received 15 November 2006; Accepted 21 December 2006
DOI: 10.1002/jcp.21029

2 0 0 7 W I L E Y - L I S S , I N C .
 TRANSMIGRATION OF MELANOMA CELLS
cells (Chan and Aruffo, 1993; Masumoto and Hemler, 1993;
Alon et al., 1995). These properties result from affinity
regulation due to inside-out signaling (Shimaoka et al., 2002; After washing the melanoma cells in 200 ml of FACS medium and
Takagi and Springer, 2002). Endothelial cells express the ligand
for the integrin a4b1, VCAM-1, on their surface only after
inflammatory stimuli, as for example TNF-a (Xu et al., 1994;
Haraldsen et al., 1996).
In this study, we demonstrate that only highly metastatic human
melanoma cells expressed the a4b1 integrin in its high affinity
conformation at the cell surface and adhered to and migrated
on VCAM-1. In contrast, human melanoma cells of low
metastatic potential expressed the a4b1 integrin in its low
affinity conformation. Moreover, transendothelial migration of
the highly metastatic melanoma cells was supported through
interaction of the integrin a4b1 with its ligand VCAM-1 on the
surface of activated endothelial cells. This suggests that highly
metastatic melanoma cells preferentially leave the blood vessels
at sites of inflammation. Therefore, blocking the a4b1 integrin
on metastatic melanoma cells or VCAM-1 on activated
endothelial cells may be a valuable approach to interfere with
tumor metastasis.
Materials and Methods
Cell culture
The human melanoma cell lines 530, IF6, BLM and MV3 (a generous gift
of Dr. van Mujien, University Hospital, Nijmengen, The Netherlands)
were grown in RPMI 1640 medium supplemented with 10% fetal
calf serum (FCS), non essential amino acids, 2 mM glutamine, 25 mM
HEPES, 100 U/ml penicillin and 100 mg/ml streptomycin at 378C in a
humidified atmosphere containing 5% CO2. HMEC-1 cells (Ades
et al., 1992) were grown in MCDB 131 medium (Invitrogen, Karlsruhe,
Germany) supplemented with 10% FCS, 1 mg/ml hydrocortisone,
10 ng/ml epidermal growth factor, 100 U/ml penicillin and 100 mg/ml
streptomycin at 378C in a humidified atmosphere containing 5% CO2.
Transfection of melanoma cells with siRNA
The siRNA against the human integrin subunit a4 (ON-TARGETplus
SMARTpool L-005189-00-0005, Human ITG4) as well as non-targeting
control siRNA (ON-TARGETplus siCONTROL Non-targeting pool
D-001810-10-05) were obtained from Dharmacon. Human melanoma
cells were transfected using the Lipofectamine 2000 (LF2000) reagent
(Invitrogen) following the manufacturers instructions. Briefly,
melanoma cells were seeded in a 24-well plate at 1
105 cells per well
and transfected with 2.5 mg (100 nM) of the respective siRNA.
Melanoma cells were incubated for 4872 h until they were analyzed
for protein knock-down.
Antibodies
The function blocking monoclonal antibody against the integrin subunit
a4 and the monoclonal antibody against VCAM-1 were from
CHEMICON. The monoclonal antibody against the integrin subunit
aL and the monoclonal antibody against ICAM-1 were from DAKO.
The directly labeled monoclonal antibodies against the integrin
subunits b1, b2, and b3 were obtained from BD Biosciences
(Heidelberg, Germany). All monoclonal antibodies used are of the IgG1
isotype. IgG1 isotype-matched control antibodies were from BD
Biosciences. The R-Phycoerythrin-conjugated donkey anti-mouse
IgG F(ab0 )2 fragments and the R-Phycoerythrin-conjugated goat
anti-human Fcg fragment specific F(ab0 )2 fragments were purchased
from Jackson ImmunoResearch (Baltimore).
Flow cytometry
For cell-surface staining, melanoma cells were released from the cell
culture dish by incubation with PBS containing 4 mM EDTA for 10 min
at 378C and washed in 200 ml FACS medium consisting of PBS
containing 1 mM CaCl2, 0.5 mM MgCl2, 5 % FCS (w/v), 0.5% BSA (w/v)
and 0.1% sodium azide. Then 2
105 melanoma cells were incubated
for 30 min at 48C in 50 ml of FACS medium containing the respective
monoclonal antibodies at the indicated concentrations. Melanoma cells
were washed twice in 200 ml FACS medium and, if the primary
antibodies were not directly labeled, incubated for 20 min at 48C in
JOURNAL OF CELLULAR PHYSIOLOGY DOI 10.1002/JCP
369
50 ml of FACS medium containing 2 mg/ml R-Phycoerythrin-conjugated
donkey anti-mouse IgG F(ab0 )2 fragments (Jackson ImmunoResearch).
then in 200 ml of PBS to remove unbound antibodies, the stained
melanoma cells were resuspended in 300 ml of PBS and analyzed by a
FACSCalibur cytometer using the CellQuest Pro software (Becton
Dickinson, Heidelberg, Germany).
Ligand-complex-based adhesion assay
Assays were performed as described (Konstandin et al., 2006) with
minor modifications. For ligand-complex-based adhesion assays,
270 ml of 20 mg/ml VCAM-1/Fc in PBS (PBS containing 1 mM CaCl2,
0.5 mM MgCl2 and 0.5% BSA) were incubated at 48C for 30 min with
86.4 ml of R-PE conjugated anti-human Fcg fragment specific
IgG F(ab0 )2 fragments to generate soluble multimeric VCAM-1/Fc
F(ab0 )2-PE complexes. Melanoma cells were washed twice with PBS.
The cell density was adjusted to 4
106 cells per ml in PBS. For each
sample 45 ml (1.8
105 cells) of this melanoma cell suspension were
mixed with 5 ml of the soluble multimeric VCAM-1/Fc F(ab0 )2-PE
complexes and incubated at 378C for the indicated time. Binding was
stopped by the addition of 1 ml 378C warm 4% PFA in PBS. Melanoma
cells were fixed for 5 min at 378C and transferred into 1 ml of ice
cold PBS containing 5% FCS and 0.5% BSA. After pelleting the
melanoma cells, bound soluble multimeric VCAM-1/Fc F(ab0 )2-PE
complexes were detected by flow cytometry using a FACSCalibur
and the CellQuest Pro software (Becton Dickinson). The percentages
and mean channel fluorescence intensities (MFI) of labeled melanoma
cells were determined.
Haptotaxis and chemotaxis assay
Transwell migration inserts (PET membrane, 8 mm pore-size, 6.4 mm
diameter, BD Biosciences) were left either untreated or the underside
was coated in a total volume of 350 ml with 20 mg/ml of recombinant
VCAM-1/Fc. After washing with migration medium (RPMI 1640
medium supplemented with 0.1% (w/v) BSA), filters were placed into
the chambers with the coated membrane side facing the lower
compartment and 600 ml of migration medium (haptotaxis) or 600 ml of
migration medium supplemented with 10% FCS (chemotaxis) were
added to the lower compartment. Melanoma cells were labeled for
15 min at 378C/5% CO2 with 1 mM CFDA-SE (Molecular Probes,
Eugene) in PBS. The labeling solution was removed and the melanoma
cells were incubated in culture medium for an additional 30 min at
378C/5% CO2. Afterwards, melanoma cells were detached from the
cell culture dish by incubation with PBS containing 4 mM EDTA for 1
0 min at 378C, pelleted, washed, and resuspended in migration
medium at a concentration of 2.5
105 cells/ml. To start the assay,
0.5
105 melanoma cells (200 ml) were added to the top of the filter.
Chambers were subsequently incubated at 378C/5% CO2 and
melanoma cells were allowed to migrate for 3 h. Then filters were
removed and all cells from the upper membrane surface were wiped
off with a cotton swab. Filters were then washed, fixed and mounted
on glass slides. Melanoma cells, which had migrated to the coated
underside of the filter were detected by confocal laser scanning
microscopy, and cells in four defined optical fields were counted
for each filter.
Transendothelial migration assay
HMEC-1 cells were seeded on the upper side of transwell migration
inserts (PET membrane, 8 mm pore-size, 6.4 mm diameter, BD
Biosciences) and grown until a confluent monolayer was established.
The integrity of the monolayer was checked microscopically and by
trypan blue exclusion. HMEC-1 cells were left either untreated or
stimulated overnight with 10 ng/ml TNF-a. After washing with
migration medium (RPMI 1640 medium supplemented with 0.1% (w/v)
BSA), 200 ml of migration medium were added to the upper
compartment and 600 ml of RPMI 1640 medium supplemented with
10% FCS were added to the lower compartment. Melanoma cells were
labeled for 15 min at 378C/5% CO2 with 1 mM CFDA-SE (Molecular
Probes) in PBS. The labeling solution was removed and the melanoma
cells were incubated in culture medium for an additional 30 min at
378C/5% CO2. Afterwards, melanoma cells were detached from the
cell culture dish by incubation with PBS containing 4 mM EDTA for 10 min
at 378C, pelleted, washed, and resuspended in migration medium at
a concentration of 2.5
105 cells/ml. To start the assay, 0.5
105
melanoma cells (200 ml) were added to the top of the filter. Chambers
370 KLEMKE ET AL.

were subsequently incubated at 378C/5% CO2 and melanoma cells TABLE 1. Cell surface expression of integrin subunits
were allowed to migrate for 5 h. Then filters were removed and all cells
from the upper membrane surface were wiped off with a cotton swab. Control a4 b1 b2 b3
Filters were then washed, fixed and mounted on glass slides. Melanoma
cells that had transmigrated to the underside of the filter were 530 2W1 39 W 9 147 W 10 2W1 2W0
IF6 3W1 22 W 1 143 W 23 3W1 4W1
detected by confocal laser scanning microscopy and cells in four BLM 3W1 36 W 14 113 W 15 2W1 2W1
defined optical fields were counted for each filter. MV3 3W1 34 W 10 184 W 18 3W1 3W1
a
Results Shown are the mean MFI (mean fluorescence intensity) values  SD of three independent
measurements. Control: secondary antibody only.
b2/b3 integrin-negative human melanoma cells with
different metastatic capacity express comparable
levels of the integrin a4b1 on their cell surface
Expression of b3 integrins is important for melanoma cell differentially adhere to VCAM-1. To this end, a novel assay that
migration and metastasis (Albelda et al., 1990a; Hofmann et al., we have recently developed (ligand-complex-based adhesion
2000). However, none of the human melanoma cell lines assay, LC-AA; Konstandin et al., 2006) was employed. This
investigated here (530, IF6, MV3, and BLM) expressed b2 or b3 method allows to measure differences in integrin affinity and
integrins (Fig. 1 and Table 1). Nevertheless, two of them (MV3 avidity at the single cell level. Melanoma cells were incubated
and BLM) are highly metastatic upon injection into nude mice with soluble multimeric VCAM-1/Fc F(ab0 )2 complexes and the
(Van Muijen et al., 1991a; van Muijen et al., 1991b) and should percentages and mean fluorescence intensities of VCAM-1/Fc-
therefore be able to efficiently undergo transendothelial complex-binding cells were determined by flow cytometry
migration. We found that all b2/b3 integrin negative melanoma (Fig. 2A). Indeed, binding of a given cell line to multimeric
cell lines analyzed in this study express the a4b1 integrin with VCAM-1 correlated with the metastatic potential of the
comparable levels on their surface (Fig. 1 and Table 1). This respective cell line. Thus, melanoma cells of low metastatic
integrin recognizes, besides fibronectin, VCAM-1 (Chan and potential (IF6 and 530) bound to VCAM-1/Fc-complexes to a
Aruffo, 1993), a cell surface protein expressed by activated
endothelial cells (Xu et al., 1994; Haraldsen et al., 1996; Lee
et al., 2001; Park et al., 2001). Since VCAM-1 is known to be
important for transendothelial migration of hematopoetic
cells (Oppenheimer-Marks et al., 1991; Chan and Aruffo, 1993;
Chuluyan and Issekutz, 1993; Hourihan et al., 1993; Weber
and Springer, 1998; Ding et al., 2001), we investigated, whether
the integrin a4b1 is involved in transendothelial migration of
these human melanoma cell lines of different metastatic
potential.

Adhesiveness of a4b1 expressing human melanoma cells

to VCAM-1 correlates with their metastatic potential
To undergo transendothelial migration, tumor cells first have to
stably adhere to the endothelial cells. As the integrin a4b1 is
subject to affinity regulation due to inside-out-signaling, we
investigated, whether the a4b1 expressing melanoma cells

Fig. 2. Adhesiveness of a4b1 expressing melanoma cells to VCAM-1

Fig. 1. b2/b3 integrin-negative human melanoma cells express
comparable levels of the integrin a4b1 on their cell surface. Cell
surface levels of the integrin subunits a4, b1, b2, and b3 were
determined by flow cytometry. Open histogram, dotted line: control
(cells stained with secondary antibody only). Open histogram, solid
line: staining with an isotype-matched control antibody. Filled
histogram: staining with the respective monoclonal antibody. Shown
is one representative experiment out of three.
correlates with their metastatic potential. A: Melanoma cells of either
low metastatic capacity (530 and IF6), or high metastatic capacity
(MV3 and BLM) were incubated at 37-C with 50 mg/ml of soluble
multimeric VCAM-1/Fc F(ab()2-PE complexes for the indicated time
points. Melanoma cells were fixed, and bound PE-labeled VCAM-1/Fc
complexes were detected by flow cytometry. B: 530, IF6, MV3, and
BLM melanoma cells were preincubated for 30 min at 4-C with 20 mg/
ml of function blocking antibodies against the integrin subunits a4 and
aL as indicated and allowed to bind soluble multimeric VCAM-1/Fc
F(ab()2-PE complexes for 20 min at 37-C. Melanoma cells were fixed,
and bound PE-labeled VCAM-1/Fc complexes were detected by flow
cytometry. Shown is one representative experiment out of two. Mean
fluorescence intensity (MFI).

JOURNAL OF CELLULAR PHYSIOLOGY DOI 10.1002/JCP

 TRANSMIGRATION OF MELANOMA CELLS
much lower extent and with slower kinetics than the highly
metastatic melanoma cell lines (BLM and MV3). As expected,
adhesion to multimeric VCAM-1/Fc F(ab0 )2 complexes could be
completely blocked by pre-incubation of the cells with a
function blocking antibody to the integrin subunit a4 but not
with an antibody directed against aL (Fig. 2B). Moreover, all
melanoma cell lines used did not bind to ICAM-1/Fc F(ab0 )2
complexes due to lack of b2-integrin expression (data not
shown). Together, these data show, thatdespite similar
a4b1surface expression levelsonly the highly metastatic
melanoma cell lines MV3 and BLM, but not the low metastatic
melanoma cell lines 530 and IF6, can use multimeric VCAM-1 as
a ligand for stable adhesion.
Haptotactic and chemotactic migration of the highly
metastatic a4b1 expressing human melanoma cells is
enhanced in the presence of VCAM-1
Adhesion to endothelial cells is the first step in transendothelial
migration. Then cells need to migrate on and finally through the
endothelial cell layer. For monocytes it is known that the
interaction of a4b1 with VCAM-1 facilitates lateral migration
on endothelial cells and thereby supports transendothelial
migration (Weber and Springer, 1998). We therefore
determined, whether the presence of high affinity VCAM-1
would enhance the haptotactic and chemotactic migration of
a4b1 expressing melanoma cells. Cell migration was assayed
using transwell inserts which were either left uncoated or
coated with VCAM-1/Fc before the assay was performed. For
haptotaxis experiments, no chemoattractant was added to the
lower compartment, whereas for chemotaxis assays, FCS was
added to attract the cells. The presence of VCAM-1/Fc
significantly increased the haptotactic (Fig. 3A) as well as
chemotactic (Fig. 3B) migration of the two metastatic
melanoma cell lines MV3 and BLM, which display high affinity
a4b1 on their surface. In marked contrast, the low metastatic
melanoma cell lines IF6 and 530 showed almost no migration
even in the presence of VCAM-1/Fc. Therefore, the migratory
behavior on immobilized VCAM-1 correlated with the
metastatic potential of the melanoma cell lines.
Fig. 3. Haptotactic and chemotactic migration of the highly
metastatic a4b1 expressing human melanoma cells is enhanced in the
presence of VCAM-1. The underside of cell culture transwell inserts
was either left uncoated (gray bars) or coated with 20mg/ml VCAM-1/
Fc (black bars) before the assay. To analyze haptotaxis (A), CFDA-SE
labeled 530, IF6, BLM, and MV3 melanoma cells were allowed to
migrate across cell culture inserts (8 mm pore size) for 3 h at 37-C. For
chemotactic migration (B), FCS was added to the lower
compartment of the transwell chamber before the assay was started.
The numbers of melanoma cells that migrated to the underside of the
cell culture insert were determined by confocal laser scanning
microscopy. Cells in four defined optical fields were counted. Shown
are the mean values W SD of one representative experiment out of
three.
JOURNAL OF CELLULAR PHYSIOLOGY DOI 10.1002/JCP
371
The interaction between the integrin a4b1 and
VCAM-1 enhances transendothelial migration of
human melanoma cells
To investigate, whether the high affinity interaction between
the integrin a4b1 on melanoma cells and VCAM-1 on
endothelial cells supports transendothelial migration of human
melanoma cells, immortalized human dermal microvascular
endothelial cells (HMEC-1; Kielbassa et al., 1998; Lidington
et al., 1999; Kielbassa-Schnepp et al., 2001; Eum et al., 2004)
were used. As shown by flow cytometry, unstimulated HMEC-1
cells expressed integrin b1 and ICAM-1, but not VCAM-1, on
their cell surface (Fig. 4A, upper panels). In accordance with data
published by others (Xu et al., 1994; Haraldsen et al., 1996),
upon treatment with TNF-a for 18 h, HMEC-1 cells also
expressed VCAM-1 (Fig. 4A, lower panels). In addition, the
expression level of ICAM-1 increased while the level of the
integrin b1 subunit remained unchanged.
Fig. 4. The interaction between a4b1 and VCAM-1 enhances
transendothelial migration of human melanoma cells. A: Inducible
VCAM-1 expression on HMEC-1 endothelial cells. HMEC-1 cells were
either left untreated (upper row) or stimulated with 10 ng/ml TNF-a
for 18 h (lower row). Cell surface levels of the integrin subunit b1,
VCAM-1 and ICAM-1 were determined by flow cytometry. Open
histogram: control (secondary antibody only), filled histogram:
staining with the respective monoclonal antibody. Shown is one
representative experiment out of two. B: Transendothelial migration
through a HMEC-1 endothelial cell layer. CFDA-SE labeled 530, IF6,
BLM, and MV3 melanoma cells were allowed to transmigrate for 5 h at
37-C across a confluent monolayer of HMEC-1 cells grown on the
upper side of a transwell cell culture insert (8 mm pore size). HMEC-1
cells were either left untreated or stimulated with 10 ng/ml TNF-a
for 18 h as indicated. For function blocking of integrins, melanoma
cells were preincubated with either 20 mg/ml of anti-a4 or anti-aL
antibodies for 30 min at room temperature as indicated. The
numbers of melanoma cells that migrated to the underside of the
cell culture insert were determined by confocal laser scanning
microscopy. Cells in four defined optical fields were counted.
Shown are the mean values W SD of one representative experiment
out of three.
372 KLEMKE ET AL.

For the analysis of transendothelial migration, HMEC-1 cells

were grown to confluency on transwell inserts and VCAM-1
expression was induced by TNF-a as described (Xu et al., 1994;
Haraldsen et al., 1996). Melanoma cells were allowed to
transmigrate for 5 h across the endothelial cell layer. As shown
in Figure 4B, only the highly metastatic melanoma cell lines MV3
and BLM displayed considerable transendothelial migration.
Transmigration already occurred through unstimulated,
VCAM-1 negative HMEC-1 cells. It could be significantly
enhanced by induction of VCAM-1 expression on the surface of
the HMEC-1 cells through TNF-a treatment (Fig. 4B). To
confirm the role of integrin a4b1/VCAM-1 interaction in this
increase in transendothelial migration, melanoma cells were in
parallel incubated with a function blocking antibody against the
integrin a4 subunit. This treatment indeed reduced
transmigration to the level observed with unstimulated HMEC-
1 cells. Note that an unrelated antibody against integrin aL had
no effect (Fig. 4B). These data demonstrate that the enhanced
transmigration upon TNF-a treatment relies solely on the
interaction of high affinity a4b1 with VCAM-1.
siRNA mediated knock-down of the integrin subunit a4
reduces transendothelial migration of a4b1 expressing
human melanoma cells
To confirm the role of the a4b1 integrin in enhanced
transendothelial migration across TNF-a activated endothelial
cells by an independent experimental approach, we performed
an siRNA-mediated knock-down of the integrin subunit a4 on Fig. 5. siRNA mediated knock-down of the integrin subunit a4
the two highly metastatic melanoma cell lines MV3 and BLM. reduces transendothelial migration of a4b1 expressing melanoma
Treatment with an integrin a4 specific siRNA reduced the cell cells. A: Knock-down of the integrin subunit a4. MV3 and BLM
surface levels of the integrin subunit a4 by around 70% as melanoma cells were transfected with 100 nM of either a non-
targeting control siRNA (open histogram with black line), or an
compared to a non-targeting control siRNA, whereas the cell a4-specific siRNA (filled histogram). The cell surface levels of the
surface levels of the integrin subunit b1 did not change (Fig. 5A). integrin subunits a4 and b1 were determined by flow cytometry 72 h
In accordance with the experiments performed with function post-transfection. Open histogram with dotted line: secondary
blocking antibodies, treatment of MV3 and BLM cells with the antibody only. B: Transendothelial migration of siRNA-treated
melanoma cells. CFDA-SE-labeled BLM and MV3 melanoma cells
a4 specific siRNA reduced transmigration across TNF-a pretreated with either a non-targeting control siRNA or an
stimulated HMEC-1 cells down to the level observed with a4-specific siRNA were allowed to transmigrate for 5 h at 37-C across
unstimulated HMEC-1 cells (Fig. 5B). These data clearly a confluent monolayer of HMEC-1 cells grown on the upper side of a
demonstrate that the interaction of a4b1 and VCAM-1 transwell cell culture insert (8 mm pore size). Before the
transmigration assay was performed, HMEC-1 cells were either left
enhances transendothelial migration of human melanoma cells untreated or stimulated with 10 ng/ml TNF-a for 18 h as indicated to
which express the a4b1 integrin its high affinity state. induce VCAM-1 expression. The numbers of melanoma cells that
migrated to the underside of the cell culture insert were determined
by confocal laser scanning microscopy. Cells in four defined optical
Discussion fields were counted. Shown are the mean values W SD of one
representative experiment out of two.
In human melanoma, increased expression of the a4b1 integrin
correlates with tumor progression and metastasis (Hart et al.,
1991; Moretti et al., 1993; Schadendorf et al., 1993; Schadendorf
et al., 1995). However, the underlying mechanism has not been
investigated. In this study we demonstrate, that the interaction melanoma cells remain to be investigated. Haptotactic as well as
of the integrin a4b1 with the adhesion molecule VCAM-1 chemotactic migration on VCAM-1 also correlated strictly with
enhances transendothelial migration of human melanoma cells the metastatic potential of the melanoma cells. The two low
across activated endothelial cells. metastatic melanoma cell lines 530 and IF6 displayed hardly any
All melanoma cells used in this study are devoid of b2 and b3 haptotactic or chemotactic migration even in the presence of
integrins, but express the a4b1 integrin at comparable levels. VCAM-1. In contrast, the two highly metastatic melanoma cell
As this integrin is subject to affinity regulation by inside-out lines MV3 and BLM displayed considerable haptotactic and
signaling (Shimaoka et al., 2002; Takagi and Springer, 2002), we chemotactic migration in the presence of VCAM-1.
investigated whether the melanoma cell lines with different Interestingly, enhanced migration upon presence of VCAM-1
metastatic capacity used in this study would bind soluble was most prominent in the haptotaxis migration assay.
VCAM-1 with similar affinities. Interestingly, the two highly Therefore, especially in the absence of chemotactic signals, the
metastatic melanoma cell lines MV3 and BLM bound VCAM-1 presence of a4b1 in the high affinity conformation may decide
with much higher affinity than the 530 and IF6 melanoma cell whether tumor cells undergo enhanced transendothelial
lines which have low metastatic potential. This finding is likely migration across activated VCAM-1 expressing endothelium.
due to constant activation of signaling pathways that lead to Transendothelial migration of the highly metastatic melanoma
affinity upregulation of the a4b1 integrin in the two melanoma cell lines MV3 and BLM occurred already through unstimulated,
cell lines of high metastatic potential. A role for PKC and VCAM-1 negative endothelial cells. As the melanoma cells used
intracellular calcium was described to be involved in a4b1 in this study are devoid of b2-integrins and do therefore not
dependent adhesion and cell spreading of human melanoma bind to ICAM-1, transmigration through unstimulated
cells (Saini et al., 1997; Seller and Hart, 2000). However, the endothelial cells is likely mediated via the interaction of other
exact signaling pathways that regulate a4b1 affinity in human adhesion molecules than VCAM-1 or ICAM-1. Possible

JOURNAL OF CELLULAR PHYSIOLOGY DOI 10.1002/JCP

 TRANSMIGRATION OF MELANOMA CELLS
candidates are PECAM-1 or MCAM, which are expressed by
unstimulated endothelial cells (Newman et al., 1990; Albelda
et al., 1990b; Bardin et al., 1996) as well as by melanoma cells
(Luca et al., 1993; Lutzky et al., 2006), and which are known to
be involved in the process of leukocyte transmigration (Petri
and Bixel, 2006). VCAM-1 is expressed by endothelial cells only
upon activation by inflammatory stimuli like TNF-a or
interferon-g (Xu et al., 1994; Haraldsen et al., 1996; Lee et al.,
2001; Park et al., 2001). Accordingly, the transendothelial
migration of highly metastatic human melanoma cells was
enhanced upon treatment of endothelial cells with TNF-a.
Treatment of the melanoma cells with an integrin a4b1
(VCAM-1 receptor) function blocking antibody completely
inhibited the enhanced transmigration seen after TNF-a
treatment of the endothelial cells, while addition of an
unrelated antibody had no effect. Similar results were obtained
by siRNA-mediated down-regulation of the integrin a4 subunit
on the melanoma cells. Interestingly, the VCAM-1-dependent
increase in transmigration of melanoma cells was completely
blocked by siRNA mediated knock-down of the integrin subunit
a4, although expression of the a4 integrin subunit was not
completely down-regulated in these cells. Therefore it is likely
that a critical level of a4b1 integrin expression is required for
efficient transendothelial migration of melanoma cells. This
question could be addressed by the generation of subclones
derived from a metastatic melanoma cell line which express
different levels of the integrin a4b1 in the high affinity
conformation.
As endothelial cells express VCAM-1 on their surface only after
inflammatory stimuli, as for example TNF-a, it is tempting to
speculate, that integrin a4b1 expressing tumor cells might
preferentially leave the blood vessels at sites of inflammation, as
it is known for leukocytes (Ebnet et al., 1996; Worthylake and
Burridge, 2001; Alon and Feigelson, 2002; Luscinskas et al.,
2002). Recent data have indeed shown that inflammation might
be a critical component of tumor progression, and that many
cancers arise from sites of infection, chronic irritation and
inflammation (for a review see Coussens and Werb (2002)).
Tumor-activated macrophages are a significant component of
inflammatory infiltrates in tumor tissues. Although these cells
may kill tumor cells following activation by for example IL-2 and
IL-12 (Brigati et al., 2002; Tsung et al., 2002), they also produce
cytokines that potentiate tumor progression (Schoppmann
et al., 2002). During the development of human malignant
melanoma, tumor-infiltrating activated macrophages produce
TGF-b, TNF-a, and IL-1a (Torisu et al., 2000). Through
induction of VCAM-1 expression on endothelial cells, these
cytokines likely enhance tumor cell migration thereby favoring
tumor cell dissemination. Data obtained in mouse tumor
models support this hypothesis. Thus, pre-treatment of mice
with TNF-a or interleukin 1 leads to an increase in pulmonary
metastases upon injection of melanoma cells (Okahara et al.,
1994; Garofalo et al., 1995; Higashiyama et al., 1996). This
enhanced pulmonary metastasis could be inhibited by either
blocking the integrin a4b1 with a monoclonal antibody on the
melanoma cells before injection into TNF-a or interleukin 1
treated mice (Okahara et al., 1994; Garofalo et al., 1995;
Higashiyama et al., 1996), or by administering a monoclonal
antibody against VCAM-1 (Higashiyama et al., 1996).
An additional important aspect arises from our findings: CpG-
ODN as proinflammatory factors in combination with
preactivated tumor-specific T cells have been successfully used
for immunotherapy of cancer in a mouse model (Garbi et al.,
2004). CpG-ODN treatment leads to the up-regulation of
ICAM-1 and VCAM-1 on endothelial cells, which in-turn leads
to the enhanced recruitment of effector T-cells into the tumor
tissue and finally to tumor rejection (Garbi et al., 2004).
However, it should be kept in mind that such treatment might
not be suitable for a4b1 expressing tumors as in this case it
JOURNAL OF CELLULAR PHYSIOLOGY DOI 10.1002/JCP
373
might favor the transendothelial migration not only of effector
T-cells but also of a4b1 expressing tumor cells thereby
promoting tumor metastasis.
The data presented in this study have also implications for gene
expression profiling of tumor cells using cDNA microarrays.
Often cDNA microarrays are used to compare the expression
levels of mRNAs between non-tumor and tumor cells
(Gottschlich et al., 2006), or between non-metastatic and
metastatic tumor cells (Otsuka et al., 2001), in order to identify
genes which are selectively up- or down-regulated in only one
type of cell. In this study, we used two melanoma cell lines of
high (MV3 and BLM) and two melanoma cell lines of low
(530 and IF6) metastatic potential. All of these melanoma cell
lines expressed comparable levels of a4b1. By cDNA
microarray analysis, expression of this integrin would,
therefore, not be associated with a specific phenotype.
However, using a novel assay (ligand-complex-based adhesion
assay; Konstandin et al., 2006) which measures affinity plus
avidity of the a4b1 integrin to VCAM-1 on the single cell level,
we were able to show that, despite similar expression levels, the
binding capacity of a4b1 to VCAM-1 is much lower in
melanoma cells with low metastatic potential as compared to
melanoma cells with high metastatic potential.
In conclusion, a4b1 integrin expression on melanoma cells
leads to enhanced transendothelial migration across VCAM-1
expressing activated endothelial cells only if this integrin is
present in the high affinity/avidity state. These data
demonstrate that rather than mere determination of a4b1
integrin expression (Hart et al., 1991; Moretti et al., 1993;
Schadendorf et al., 1993; Schadendorf et al., 1995), the
measurement of the affinity plus avidity of a4b1 to VCAM-1
might be used as a predictive marker for metastasis.
Consequently, blocking the a4b1 integrin on tumor cells and/
or blocking of VCAM-1 on endothelial cells may be a valuable
approach to interfere with tumor metastasis.
Literature Cited
Ades EW, Candal FJ, Swerlick RA, George VG, Summers S, Bosse DC, Lawley TJ. 1992.
HMEC-1: Establishment of an immortalized human microvascular endothelial cell line.
J Invest Dermatol 99:683690.
Albelda SM, Mette SA, Elder DE, Stewart R, Damjanovich L, Herlyn M, Buck CA. 1990a.
Integrin distribution in malignant melanoma: Association of the beta 3 subunit with tumor
progression. Cancer Res 50:67576764.
Albelda SM, Oliver PD, Romer LH, Buck CA. 1990b. EndoCAM: A novel endothelial cell-cell
adhesion molecule. J Cell Biol 110:12271237.
Alon R, Feigelson S. 2002. From rolling to arrest on blood vessels: Leukocyte tap dancing on
endothelial integrin ligands and chemokines at sub-second contacts. Semin Immunol
14:93104.
Alon R, Kassner PD, Carr MW, Finger EB, Hemler ME, Springer TA. 1995. The integrin VLA-4
supports tethering and rolling in flow on VCAM-1. J Cell Biol 128:12431253.
Bardin N, Frances V, Lesaule G, Horschowski N, George F, Sampol J. 1996. Identification of
the S-Endo 1 endothelial-associated antigen. Biochem Biophys Res Commun 218:210216.
Brigati C, Noonan DM, Albini A, Benelli R. 2002. Tumors and inflammatory infiltrates: Friends
or foes? Clin Exp Metastasis 19:247258.
Chan PY, Aruffo A. 1993. VLA-4 integrin mediates lymphocyte migration on the inducible
endothelial cell ligand VCAM-1 and the extracellular matrix ligand fibronectin. J Biol Chem
268:2465524664.
Chuluyan HE, Issekutz AC. 1993. VLA-4 integrin can mediate CD11/CD18-independent
transendothelial migration of human monocytes. J Clin Invest 92:27682777.
Coussens LM, Werb Z. 2002. Inflammation and cancer. Nature 420:860867.
Danen EH, Aota S, van Kraats AA, Yamada KM, Ruiter DJ, van Muijen GN. 1995. Requirement
for the synergy site for cell adhesion to fibronectin depends on the activation state of
integrin alpha 5 beta 1. J Biol Chem 270:2161221618.
Ding Z, Xiong K, Issekutz TB. 2001. Chemokines stimulate human T lymphocyte
transendothelial migration to utilize VLA-4 in addition to LFA-1. J Leukoc Biol 69:458466.
Ebnet K, Kaldjian EP, Anderson AO, Shaw S. 1996. Orchestrated information transfer
underlying leukocyte endothelial interactions. Annu Rev Immunol 14:155177.
Engers R, Gabbert HE. 2000. Mechanisms of tumor metastasis: Cell biological aspects and
clinical implications. J Cancer Res Clin Oncol 126:682692.
Eum SY, Lee YW, Hennig B, Toborek M. 2004. VEGF regulates PCB 104-mediated stimulation
of permeability and transmigration of breast cancer cells in human microvascular
endothelial cells. Exp Cell Res 296:231244.
Garbi N, Arnold B, Gordon S, Hammerling GJ, Ganss R. 2004. CpG motifs as
proinflammatory factors render autochthonous tumors permissive for infiltration and
destruction. J Immunol 172:58615869.
Garofalo A, Chirivi RG, Foglieni C, Pigott R, Mortarini R, Martin-Padura I, Anichini A, Gearing
AJ, Sanchez-Madrid F, Dejana E, et al. 1995. Involvement of the very late antigen 4 integrin
on melanoma in interleukin 1-augmented experimental metastases. Cancer Res 55:414
419.
374 KLEMKE ET AL.
Gottschlich S, Ambrosch P, Cordes C, Gorogh T, Schreiber S, Hasler R. 2006. Gene
expression profiling of head and neck squamous cell carcinoma using cDNA microarrays.
Int J Oncol 29:605613.
Haraldsen G, Kvale D, Lien B, Farstad IN, Brandtzaeg P. 1996. Cytokine-regulated expression
of E-selectin, intercellular adhesion molecule-1 (ICAM-1), and vascular cell adhesion
molecule-1 (VCAM-1) in human microvascular endothelial cells. J Immunol 156:2558
2565.
Hart IR, Birch M, Marshall JF. 1991. Cell adhesion receptor expression during melanoma
progression and metastasis. Cancer Metastasis Rev 10:115128.
Higashiyama A, Watanabe H, Okumura K, Yagita H. 1996. Involvement of tumor necrosis
factor alpha and very late activation antigen 4/vascular cell adhesion molecule 1 interaction
in surgical-stress-enhanced experimental metastasis. Cancer Immunol Immunother
42:231236.
Hofmann UB, Westphal JR, Waas ET, Becker JC, Ruiter DJ, van Muijen GN. 2000.
Coexpression of integrin alpha(v)beta3 and matrix metalloproteinase-2 (MMP-2)
coincides with MMP-2 activation: Correlation with melanoma progression. J Invest
Dermatol 115:625632.
Hourihan H, Allen TD, Ager A. 1993. Lymphocyte migration across high endothelium is
associated with increases in alpha 4 beta 1 integrin (VLA-4) affinity. J Cell Sci 104:1049
1059.
Kielbassa K, Schmitz C, Gerke V. 1998. Disruption of endothelial microfilaments selectively
reduces the transendothelial migration of monocytes. Exp Cell Res 243:129141.
Kielbassa-Schnepp K, Strey A, Janning A, Missiaen L, Nilius B, Gerke V. 2001. Endothelial
intracellular Ca2 release following monocyte adhesion is required for the
transendothelial migration of monocytes. Cell Calcium 30:2940.
Konstandin MH, Sester U, Klemke M, Weschenfelder T, Wabnitz GH, Samstag Y. 2006.
A novel flow-cytometry-based assay for quantification of affinity and avidity changes of
integrins. J Immunol Methods 310:6777.
Lee YW, Kuhn H, Hennig B, Neish AS, Toborek M. 2001. IL-4-induced oxidative stress
upregulates VCAM-1 gene expression in human endothelial cells. J Mol Cell Cardiol 33:83
94.
Lidington EA, Moyes DL, McCormack AM, Rose ML. 1999. A comparison of primary
endothelial cells and endothelial cell lines for studies of immune interactions. Transpl
Immunol 7:239246.
Luca M, Hunt B, Bucana CD, Johnson JP, Fidler IJ, Bar-Eli M. 1993. Direct correlation between
MUC18 expression and metastatic potential of human melanoma cells. Melanoma Res
3:3541.
Luscinskas FW, Ma S, Nusrat A, Parkos CA, Shaw SK. 2002. Leukocyte transendothelial
migration: A junctional affair. Semin Immunol 14:105113.
Lutzky VP, Carnevale RP, Alvarez MJ, Maffia PC, Zittermann SI, Podhajcer OL, Issekutz AC,
Chuluyan HE. 2006. Platelet-endothelial cell adhesion molecule-1 (CD31) recycles and
induces cell growth inhibition on human tumor cell lines. J Cell Biochem 98:13341350.
Lyck R, Reiss Y, Gerwin N, Greenwood J, Adamson P, Engelhardt B. 2003. T-cell interaction
with ICAM-1/ICAM-2 double-deficient brain endothelium in vitro: The cytoplasmic tail of
endothelial ICAM-1 is necessary for transendothelial migration of T cells. Blood 102:3675
3683.
Masumoto A, Hemler ME. 1993. Multiple activation states of VLA-4. Mechanistic differences
between adhesion to CS1/fibronectin and to vascular cell adhesion molecule-1. J Biol Chem
268:228234.
Mizejewski GJ. 1999. Role of integrins in cancer: Survey of expression patterns. Proc Soc Exp
Biol Med 222:124138.
Moretti S, Martini L, Berti E, Pinzi C, Giannotti B. 1993. Adhesion molecule profile and
malignancy of melanocytic lesions. Melanoma Res 3:235239.
Newman PJ, Berndt MC, Gorski J, White GCII, Lyman S, Paddock C, Muller WA. 1990.
PECAM-1 (CD31) cloning and relation to adhesion molecules of the immunoglobulin gene
superfamily. Science 247:12191222.
Nicolson GL. 1988. Organ specificity of tumor metastasis: Role of preferential adhesion,
invasion and growth of malignant cells at specific secondary sites. Cancer Metastasis Rev
7:143188.
Okahara H, Yagita H, Miyake K, Okumura K. 1994. Involvement of very late activation antigen
4 (VLA-4) and vascular cell adhesion molecule 1 (VCAM-1) in tumor necrosis factor alpha
enhancement of experimental metastasis. Cancer Res 54:32333236.
Oppenheimer-Marks N, Davis LS, Bogue DT, Ramberg J, Lipsky PE. 1991. Differential
utilization of ICAM-1 and VCAM-1 during the adhesion and transendothelial migration of
human T lymphocytes. J Immunol 147:29132921.
JOURNAL OF CELLULAR PHYSIOLOGY DOI 10.1002/JCP
Orr FW, Wang HH, Lafrenie RM, Scherbarth S, Nance DM. 2000. Interactions between
cancer cells and the endothelium in metastasis. J Pathol 190:310329.
Otsuka M, Kato M, Yoshikawa T, Chen H, Brown EJ, Masuho Y, Omata M, Seki N. 2001.
Differential expression of the L-plastin gene in human colorectal cancer progression and
metastasis. Biochem Biophys Res Commun 289:876881.
Park HJ, Lee YW, Hennig B, Toborek M. 2001. Linoleic acid-induced VCAM-1 expression in
human microvascular endothelial cells is mediated by the NF-kappa B-dependent pathway.
Nutr Cancer 41:126134.
Petri B, Bixel MG. 2006. Molecular events during leukocyte diapedesis. FEBS J 273:4399
4407.
Reiss Y, Engelhardt B. 1999. T cell interaction with ICAM-1-deficient endothelium in vitro:
Transendothelial migration of different T cell populations is mediated by endothelial ICAM-
1 and ICAM-2. Int Immunol 11:15271539.
Saini A, Seller Z, Davies D, Marshall JF, Hart IR. 1997. Activation status and function of the
VLA-4 (alpha4beta1) integrin expressed on human melanoma cell lines. Int J Cancer
73:264270.
Schadendorf D, Gawlik C, Haney U, Ostmeier H, Suter L, Czarnetzki BM. 1993. Tumour
progression and metastatic behaviour in vivo correlates with integrin expression on
melanocytic tumours. J Pathol 170:429434.
Schadendorf D, Heidel J, Gawlik C, Suter L, Czarnetzki BM. 1995. Association with clinical
outcome of expression of VLA-4 in primary cutaneous malignant melanoma as well as P-
selectin and E-selectin on intratumoral vessels. J Natl Cancer Inst 87:366371.
Schoppmann SF, Birner P, Stockl J, Kalt R, Ullrich R, Caucig C, Kriehuber E, Nagy K, Alitalo K,
Kerjaschki D. 2002. Tumor-associated macrophages express lymphatic endothelial
growth factors and are related to peritumoral lymphangiogenesis. Am J Pathol
161:947956.
Seller Z, Hart IR. 2000. Investigation of signal transduction pathways involved in melanoma
cell spreading. Indian J Exp Biol 38:211221.
Shimaoka M, Takagi J, Springer TA. 2002. Conformational regulation of integrin structure and
function. Annu Rev Biophys Biomol Struct 31:485516.
Stetler-Stevenson WG, Aznavoorian S, Liotta LA. 1993. Tumor cell interactions with the
extracellular matrix during invasion and metastasis. Annu Rev Cell Biol 9:541573.
Takagi J, Springer TA. 2002. Integrin activation and structural rearrangement. Immunol Rev
186:141163.
Torisu H, Ono M, Kiryu H, Furue M, Ohmoto Y, Nakayama J, Nishioka Y, Sone S, Kuwano M.
2000. Macrophage infiltration correlates with tumor stage and angiogenesis in human
malignant melanoma: possible involvement of TNFalpha and IL-1alpha. Int J Cancer 85:182
188.
Tsung K, Dolan JP, Tsung YL, Norton JA. 2002. Macrophages as effector cells in interleukin
12-induced T cell-dependent tumor rejection. Cancer Res 62:50695075.
Van Muijen GN, Cornelissen LM, Jansen CF, Figdor CG, Johnson JP, Brocker EB, Ruiter DJ.
1991. Antigen expression of metastasizing and non-metastasizing human melanoma cells
xenografted into nude mice. Clin Exp Metastasis 9:259272.
van Muijen GN, Jansen KF, Cornelissen IM, Smeets DF, Beck JL, Ruiter DJ. 1991.
Establishment and characterization of a human melanoma cell line (MV3) which is highly
metastatic in nude mice. Int J Cancer 48:8591.
Varner JA, Cheresh DA. 1996. Integrins and cancer. Curr Opin Cell Biol 8:724730.
Voura EB, Ramjeesingh RA, Montgomery AM, Siu CH. 2001. Involvement of integrin
alpha(v)beta(3) and cell adhesion molecule L1 in transendothelial migration of melanoma
cells. Mol Biol Cell 12:26992710.
Wang HS, Hung Y, Su CH, Peng ST, Guo YJ, Lai MC, Liu CY, Hsu JW. 2005. CD44 cross-
linking induces integrin-mediated adhesion and transendothelial migration in breast cancer
cell line by up-regulation of LFA-1 (alpha L beta2) and VLA-4 (alpha4beta1). Exp Cell Res
304:116126.
Weber C, Springer TA. 1998. Interaction of very late antigen-4 with VCAM-1 supports
transendothelial chemotaxis of monocytes by facilitating lateral migration. J Immunol
161:68256834.
Wong D, Prameya R, Dorovini-Zis K. 1999. In vitro adhesion and migration of T lymphocytes
across monolayers of human brain microvessel endothelial cells: regulation by ICAM-1,
VCAM-1, E-selectin and PECAM-1. J Neuropathol Exp Neurol 58:138152.
Worthylake RA, Burridge K. 2001. Leukocyte transendothelial migration: Orchestrating the
underlying molecular machinery. Curr Opin Cell Biol 13:569577.
Xu Y, Swerlick RA, Sepp N, Bosse D, Ades EW, Lawley TJ. 1994. Characterization of
expression and modulation of cell adhesion molecules on an immortalized human dermal
microvascular endothelial cell line (HMEC-1). J Invest Dermatol 102:833837.