EXPERIMENTAL CELL RESEARCH 238, 188–196 (1998)
ARTICLE NO. EX973821
The Disintegrin Eristostatin Interferes with Integrin a4b1 Function and
with Experimental Metastasis of Human Melanoma Cells
Erik H. J. Danen,*,1 Cezary Marcinkiewicz,† Ine MHA Cornelissen,* Annemieke A. van Kraats,*
Jonathan A. Pachter,‡ Dirk J. Ruiter,* Stefan Niewiarowski,† and Goos N. P. van Muijen*
*Department of Pathology, University Hospital, Nijmegen, The Netherlands; †Department of Physiology and Sol Sherry Thrombosis
Center, Temple University Medical School, Philadelphia, Pennsylvania; and ‡Department of Tumor Biology,
Schering–Plough Research Institute, Kenilworth, New Jersey
Peptides containing the integrin recognition se-
quence, RGD, can inhibit experimental metastasis of
mouse melanoma cells, but the integrin(s) affected in
these experiments is unknown. Besides ‘‘classical’’
RGD-binding integrins such as a5b1 and avb3, RGD
has been reported to bind a4b1, and mAbs to a4b1 can
inhibit melanoma metastasis. We investigated the
mode of action of the disintegrin eristostatin, an RGD-
containing peptide isolated from snake venom, in a
human melanoma experimental metastasis model.
Lung colonization following i.v. injection of MV3 cells
in nude mice was strongly inhibited by eristostatin.
MV3 cells bound FITC-eristostatin and adhered to eris-
tostatin-coated wells. This adhesion was partially in-
hibited by a GRGDSP peptide and by a4 mAb. Binding
of FITC-eristostatin to Jurkat cells and adhesion of
Jurkat (but not K562) cells to eristostatin-coated wells
further suggested that eristostatin binds a4b1, even
though, again, a4 mAb only partially inhibited adhe-
sion. Expression of a4b1 was enhanced in metastatic
melanoma cells compared to normal melanocytes and
nonmetastatic melanoma cells. Finally, eristostatin in-
hibited adhesion of both MV3 and CHOa4 cells to the
a4b1-ligand VCAM-1, while adhesion to other ligands
via other integrins was not affected. These findings
demonstrate that inhibition of melanoma cell metasta-
sis by RGD-containing peptides such as eristostatin,
may be due to interference with a4b1-VCAM binding,
in addition to inhibition of the classical RGD-binding
integrins. q 1998 Academic Press
INTRODUCTION
Cell adhesion events are critical in tumor metastasis
[1] and integrins play an important role in adhesion of
cells to other cells and to the extracellular matrix
1
To whom correspondence and reprint requests should be ad-
dressed at present address: CDBRB, Bldg 30, room 408, NIDR, Na-
tional Institutes of Health, Bethesda, MD 20892-4370. Fax: (301)-
402-0897. E-mail: edanen@yoda.nidr.nih.gov.
0014-4827/98 $25.00
Copyright q 1998 by Academic Press
All rights of reproduction in any form reserved.
AID ECR 3821 / 6i2ah$$$81 12-16-97 00:51:48
(ECM) [2]. Integrins a3b1 [3], a5b1 [4, 5], avb1 [6],
avb3 [7], avb5 [8], avb6 [9], and avb8 [10] recognize
the RGD sequence in their ligands. In addition, ligand
binding of a2b1 [11] and a4b1 [12, 13] can be inhibited
by (cyclic) RGD peptides.
Synthetic RGD peptides, nonpeptide RGD mimetics,
or RGD peptides isolated from snake venom called dis-
integrins can inhibit experimental metastasis of B16
murine melanoma cells [14–21], though it is not known
which of the RGD-binding integrins is affected in these
studies. For human melanoma, a4b1, a5b1, and avb3
are interesting candidates, since: (a) they are strongly
upregulated with melanocytic tumor progression in
situ [22–25], (b) antibodies to a4 can inhibit human
melanoma cell metastasis [26, 27], and (c) a5b1 and
avb3 have been implicated in growth [28, 29] and inva-
sion [30, 31] of human melanoma cells.
In the present study we investigate the inhibitory
effect of the disintegrin eristostatin in experimental
metastasis of a human melanoma cell line. In addition
to its previously reported effect on murine melanoma
metastasis [14], we find that eristostatin efficiently in-
hibits experimental metastasis of human melanoma
cells, and its effect may be mediated, at least in part,
by targeting integrin a4b1.
MATERIALS AND METHODS
Integrin ligands, synthetic peptides, and disintegrins. Plasma fi-
bronectin (Fn) was purchased from Sigma (St Louis, MO). A 120-
kDa chymotryptic Fn fragment containing the central cell binding
domain and synthetic peptides Gly-Arg-Gly-Asp-Ser-Pro (GRGDSP)
and Gly-Arg-Gly-Glu-Ser-Pro (GRGESP) were purchased from Life
Technologies (Gaithersburg, MD). Laminin (Ln), isolated from Engle-
breth–Holm–Swarm mouse sarcoma cells was purchased from Life
Technologies, and collagen (Co) type I, isolated from rat tail, was
provided by Dr. Eberhard Klein (Wurzburg, Germany). Recombinant
human VCAM-1, produced as described [32], was provided by Dr.
Mark Renz (Genentech, San Francisco, CA).
The cyclic peptide cRRETAWA that binds specifically to a5b1 [33]
was generated as described previously [34]. Disintegrins were iso-
lated from crude viper venoms by reverse phase HPLC as described
[35]. In purified systems, echistatin is a potent inhibitor of a5b1,
avb3, and aIIbb3, bitistatin moderately affects avb3 and aIIbb3,
188
eca
 ERISTOSTATIN INHIBITS a4b1 FUNCTION AND METASTASIS
TABLE 1
Inhibition of Experimental Metastasis of MV3 Cells
mg/mouse Exp. 1 Exp. 2
No 5/5a 4/5
Eristostatin 25 1/5 0/5
Bitistatin 25 5/5 5/5
Echistatinb 25 4/5 3/5
cRRETAWAb 1000 3/5 4/5
NKI-Sam1b 100 5/5 4/5
Clg 100 4/5 4/5
a
Number of mice with lung colonies/number of mice inoculated.
b
Reagents blocking a5b1.
and eristostatin inhibits aIIbb3 with very little effect on a5b1 or
avb3 [36–39].
Antibodies. Anti-integrin mAbs included 5E8 anti-a2 [40], pro-
vided by Dr. Richard Bankert (Buffalo, NY); P1B5 anti-a3 [41], pur-
chased from Life Technologies; HP2/1 anti-a4 [42], provided by Dr.
Francisco Sanchez-Madrid (Madrid, Spain); NKI-Sam1 anti-a5 [43],
provided by Dr. Carl Figdor (Nijmegen, The Netherlands); GoH3
anti-a6 [44], provided by Dr. Arnoud Sonnenberg (Amsterdam, The
Netherlands); 4B4 anti-b1 [45], purchased from Coulter Immunology
(Hialeah, FL); TS2/16 anti-b1 [46], provided by Dr. Francisco San-
chez-Madrid; LM142 anti-av and LM609 anti-avb3 [8], provided by
Dr. David Cheresh (La Jolla, CA); P1F6 anti-avb5 [47] purchased
from Life Technologies; SN1 anti-avb8 [10], provided by Dr. Steven
Nishimura (San Francisco, CA); and 10E5 anti-aIIb [48], provided
by Dr. Barry Coller (Stoneybrook, NY). A109 polyclonal anti-av [49]
was purchased from Life Technologies. G250 [50], an isotype-
matched (IgG2b) control mAb for NKI-Sam1, was provided by Dr.
Egbert Oosterwijk (Nijmegen, The Netherlands).
Nude mice and cell culture. BALB/c athymic nude mice (nu/nu)
were purchased from The Laboratory Breeding and Research Center
(Bomholtgaard, Ry, Denmark) and kept in cages covered with air
filters under specific pathogen-free conditions. Mice were used when
6–8 weeks old. Within a single experiment mice were sex and age
matched.
Normal human foreskin melanocytes, the nonmetastatic human
melanoma cell lines IF6 and 530 [51], and the highly metastatic
human melanoma cell lines BLM and MV3 [51, 52] were cultured
as described previously [53]. The erythroleukemic cell line K562 and
the T cell line Jurkat were cultured in RPMI medium (Flow, Irvine,
UK) supplemented with 10% fetal bovine serum and antibiotics. Con-
trol and a4-cDNA transfected CHO cells [54] were provided by Dr.
Yoshikazu Takada, La Jolla, CA.
Flow cytometry. Cells were incubated sequentially with mAbs
and fluorescein-isothiocyanate (FITC)-labeled F(ab*)2 fragments of
rabbit anti-mouse Ig antibodies (Dako, Glostrup, Denmark), washed,
and fluorescence was measured on an Epics Elite flow cytometer
(Coulter, Mijdrecht, The Netherlands). For the disintegrin binding
assay, eristostatin was labeled with FITC as described [McLane et
al., 1994]. FITC-eristostatin, separated by gel filtration, fully re-
tained its platelet aggregation inhibitory activity.
Experimental metastasis assay. Tumor cells (2 1 106) in 200 ml
0.9% NaCl in the absence or presence of mAbs or peptides were
injected intravenously (i.v.) into the lateral tail vein. Mice were killed
after 1 month, lungs were formalin-fixed and embedded in paraffin,
and H&E-stained 4-mm sections of three levels of the lungs were
microscopically examined for lung metastases. In our experience, no
other organs are colonized [51, 52].
Fisher’s Exact Test (two-sided) was used for comparison of percent-
ages of mice that developed metastases.
AID ECR 3821 / 6i2ah$$$82 12-16-97 00:51:48
189
Exp. 3 Total % P
3/5 12/15 80
1/5 2/15 13 0.0007
5/5 15/15 100 ú0.05
3/5 10/15 67 ú0.05
n.d. 7/10 70 ú0.05
n.d. 9/10 90 ú0.05
n.d. 8/10 80 ú0.05
Cell adhesion assay. Cell adhesion assays were performed as de-
scribed previously [55]. Briefly, wells were coated with the appro-
priate adhesive ligands and blocked with BSA. Subsequently, 1 1
104 51Cr-labeled cells (or, for adhesion to VCAM-1, 1 1 105 5-chloro-
methylfluorescein-labeled cells) in 50 ml DMEM/BSA were added to
the wells and incubated for 30 min at 377C. Unbound cells were
washed away, bound cells were lysed, and radioactivity or fluores-
cence of the lysate was measured. Results are presented as the mean
percentage of adherent cells from triplicate wells. For adhesion inhi-
bition studies, cells were incubated with the appropriate mAbs or
peptides for 15 min at 47C before seeding into the wells. For induction
of adhesion, the cells were preincubated with TS2/16 mAbs that in-
duce a high affinity state of b1-integrins [43, 56] for 30 min at 47C
prior to seeding in the wells.
Immunoprecipitation. Cells were labeled overnight with [35S]-
methionine (Amersham, Houten, The Netherlands) and glycopro-
teins were isolated from 0.5% NP-40 solubilized cell extracts by ad-
sorption to concanavalin A (Con A)–Sepharose (Pharmacia, Uppsala,
Sweden). Equal numbers of counts of the Con A-bound fractions
were incubated by tumbling overnight at 47C with mAbs, rabbit anti-
mouse Ig (Dako, Glostrup, Denmark), and Prot A beads (Pharmacia).
Beads were washed, resuspended in sample buffer containing 2-mer-
captoethanol, boiled, and run on SDS–PAGE.
RESULTS
Eristostatin Inhibits MV3 Experimental Metastasis
To examine the effect of eristostatin on experimental
metastasis of human melanoma cells, we injected MV3
cells in the tail vein of nude mice in the presence or
absence of eristostatin and examined lung colonization.
Incubation for 1 h in vitro with eristostatin did not
affect the viability of the cells (not shown). Eristostatin
efficiently reduced lung colonization in three indepen-
dent experiments (P Å 0.0007) (Table 1) while two
other disintegrins, bitistatin and echistatin, had no ef-
fect.
Inhibition of MV3 Experimental Metastasis by
Eristostatin Does Not Involve Blocking Integrins
a5b1 or avb3
MV3 expressed several RGD-binding integrins that
might be affected by eristostatin, including a5b1 but
not avb3 (Table 2). Eristostatin did not affect a5b1-
eca
190 DANEN ET AL.
TABLE 2
Expression of RGD-Binding Integrins on MV3 Cells
Subunit ca a2 a3 a4 a5
m.f.b 4 20 69 31 62
a
Control Ig.
b
Mean fluorescence (arbitrary units).
mediated MV3 adhesion to Fn or to a 120-kDa fragment
containing the central cell-binding domain of Fn, while
a control disintegrin, echistatin, blocked adhesion to
both ligands (Fig. 1). Moreover, when MV3 cells were
injected in the presence of reagents that bind a5b1,
lung colonization was not affected by the NKI-Sam1
a5 mAb, the a5b1-specific cyclic peptide cRRETAWA,
or the a5b1-binding disintegrin echistatin (P ú 0.05
for all treatments) (Table 1). All three reagents blocked
in vitro adhesion of MV3 to Fn and to the 120-kDa Fn
fragment (Fig. 1). Amounts of mAbs and disintegrins
were larger than those described to be effective in other
studies [14, 21, 26, 27].
These results indicate that eristostatin inhibits MV3
metastasis through a mechanism that does not involve
integrins a5b1 or avb3.
Eristostatin Does Not Inhibit MV3 Adhesion to
Basement Membrane Components Ln and Co
Interference with MV3 attachment to basement
membranes might be involved in the in vivo effect of
eristostatin. Therefore, we tested if eristostatin blocked
adhesion to basement membrane components Ln and
Co. Eristostatin did not interfere with a2b1/a6b1-me-
diated adhesion to Ln and a2b1-mediated adhesion to
FIG. 1. Inhibition of a5b1-mediated adhesion. MV3 cells were
allowed to adhere to wells coated with 20 mg/ml Fn (closed bars) or
Fn 120 kDa (open bars) in the absence or presence of 10 mg/ml of
the disintegrins eristostatin, echistatin, or bitistatin or 5 mg/ml NKI-
Sam1 anti-a5 mAb or G250 control Ig (clg); or 100 mg/ml of the cyclic
peptide cRRETAWA. Mean { SD of triplicate determinations for one
experiment of three is shown.
AID ECR 3821 / 6i2ah$$$82 12-16-97 00:51:48
b1 av avb3 av b 5 avb8 aIIb
107 13 2 19 4 3
Co (Fig. 2). As eristostatin-binding to these integrins,
and thus its capacity to inhibit attachment via them,
may depend on the activation state of the integrin, we
performed the same experiments in the presence of Mn
or TS2/16 stimulatory b1 mAb. However, incubation
of MV3 cells with these reagents did not induce an
inhibitory effect of eristostatin (not shown).
Thus, eristostatin does not seem to act through inter-
ference with a2b1- or a6b1-mediated adhesion events.
Eristostatin-MV3 Interaction Is Partially Inhibited by
GRGDSP and by a4 mAb
We tested if MV3 cells bound eristostatin. As shown
in Fig. 3A, FITC-labeled eristostatin bound to MV3
cells in a concentration-dependent manner. Adhesion
of MV3 cells to eristostatin could be partially inhibited
by GRGDSP while a control GRGESP peptide had no
effect, but the effect of RGD was low compared to the
complete block of adhesion to echistatin and bitistatin
(Fig. 3B). To identify the RGD-binding integrin on MV3
that was involved in adhesion to eristostatin, we incu-
bated MV3 cells with function-blocking mAbs to vari-
ous subunits of integrins known to recognize RGD. Par-
tial inhibition was observed with HP2/1 anti-a4 mAbs
but not with 4B4 anti-b1 (Fig. 3C). None of the other
mAbs directed to a2, a3, a5, avb3, avb5, nor polyclonal
anti-av, had any effect (Fig. 3C), nor did they enhance
FIG. 2. Eristostatin does not inhibit a2b1-/a6b1-mediated adhe-
sion. MV3 cells were allowed to adhere to wells coated with 20 mg/
ml Co (closed bars) or Ln (open bars) in the absence or presence of
10 mg/ml eristostatin or 5E8 anti-a2, GoH3 anti-a6, or 4B4 anti-b1.
Mean { SD of triplicate determinations for one experiment of two
is shown.
eca
 ERISTOSTATIN INHIBITS a4b1 FUNCTION AND METASTASIS
FIG. 3. MV3 binds eristostatin. (A) MV3 cells were incubated
with increasing concentrations of FITC-labeled eristostatin and flu-
orescence was measured. Nonspecific fluorescence with 1 mg FITC-
BSA was less than 0.5 fluorescence units. (B) MV3 cells were allowed
to adhere to wells coated with 25 mg/ml of disintegrins, in the absence
(open bars) or presence of 1 mg/ml GRGDSP (hatched bars) or GRG-
ESP (closed bars). (C) MV3 cells were allowed to adhere to wells
coated with 25 mg/ml eristostatin in the absence or presence of 5E8
anti-a2, P1B5 anti-a3, HP2/1 anti-a4, NKI-Sam1 anti-a5, 4B4 anti-
b1, LM609 anti-avb3, P1F6 anti-avb5 mAbs, or A109 polyclonal
anti-av. One experiment of two is shown. For B and C mean { SD
of triplicate determinations is shown.
inhibition in combination with HP2/1 (not shown). MV3
does not express a4b7 (not shown).
These results show that the interaction of MV3 cells
with eristostatin is partially RGD-dependent and may
involve binding to a4b1.
Eristostatin Binds a4b1, an Integrin That Is
Upregulated in Metastatic Melanoma Cells
To further investigate if eristostatin binds a4b1, we
incubated Jurkat cells (which express predominantly
a4b1 and trace amounts of a5b1; not shown) with
FITC-eristostatin. FITC-eristostatin bound to Jurkat
cells in a concentration-dependent manner (Fig. 4A).
In addition, when Jurkat cells were stimulated with 1
AID ECR 3821 / 6i2ah$$$82 12-16-97 00:51:48
191
mM manganese, they adhered to eristostatin, whereas
K562 (which expresses exclusively a5b1; not shown)
did not (Fig. 4B). In line with the findings for MV3,
adhesion of Jurkat cells was partially reduced by the
HP2/1 anti-a4 mAb (Fig. 4B). In contrast to MV3, for
Jurkat cells the 4B4 anti-b1 mAb inhibited adhesion
as well.
As eristostatin binding to a4b1 may be involved in
the observed effect of eristostatin, we investigated if
a4b1 expression was related to the metastatic potential
of melanocytic cells. We determined expression and
synthesis of the a4 subunit in normal human melano-
cytes and nonmetastatic (IF6 and 530) and highly met-
astatic melanoma cells (BLM and MV3). Synthesis
(Fig. 5A) and surface expression (Fig. 5B) of the a4
subunit were absent or low in melanocytes, IF6, and
530, and high in BLM and MV3.
FIG. 4. Eristostatin binds a4b1. (A) Jurkat cells were incubated
with increasing concentrations of FITC-labeled eristostatin and flu-
orescence was measured. Nonspecific fluorescence with 1 mg FITC-
BSA was less than 0.5 fluorescence units. (B) Jurkat (closed bars)
and K562 cells (open bars) in the absence or presence of 1 mM manga-
nese (Mn) were allowed to adhere to wells coated with 25 mg/ml
eristostatin in the absence or presence of HP2/1 anti-a4, NKI-Sam1
anti-a5, or 4B4 anti-b1. Mean { SD of triplicate determinations for
one experiment of two is shown.
eca
192 DANEN ET AL.

sis by the disintegrin eristostatin may not only involve

blocking the ‘‘classical’’ RGD-binding integrins. We
show that eristostatin inhibits experimental metasta-
sis of MV3 human melanoma cells and interferes with
a4b1-VCAM-1 interaction rather than blocking a5b1
or avb3 ligand binding.
MV3 cells lack avb3, and even though avb3 is
thought to play an important role in melanoma metas-
tasis [reviewed in 57], the fact that MV3 can metasta-
size in the absence of this integrin [58] confirms earlier
reports for other melanoma cell lines [31, 59]. Several
findings suggest that a5b1 may be important for inva-
sion and growth of melanoma cells as well. In benign
melanocytic lesions a5b1 is absent while it can be de-
tected in melanomas, with increased expression in ad-
vanced stages of tumor progression [22–24]. Further-

FIG. 5. Expression of a4b1 is upregulated in highly metastatic

melanoma cells. (A) Melanocytes, IF6, 530, BLM, and MV3 were
labeled o/n with [35S]methionine, lysed, and glycoproteins were iso-
lated on Con A–Sepharose. Equal numbers of Con A-bound counts
were used for immunoprecipitation with HP2/1 anti-a4. (B) Melano-
cytes (MCT), nonmetastatic melanoma cells (IF6 and 530) and highly
metastatic melanoma cells (BLM and MV3) were incubated with
HP2/1 anti-a4 or control Ig (dotted graph), followed by FITC-labeled
second antibody and fluorescence was measured.

These results confirm that eristostatin binds a4b1,

and show that expression of this integrin is related to
the metastatic potential of human melanoma cells.

Eristostatin Inhibits Binding to Immobilized VCAM-1

As eristostatin binding to MV3 seemed to involve
integrin a4b1, we next investigated if eristostatin af-
fected a4b1-mediated MV3 adhesive events. Indeed,
binding of MV3 cells to immobilized VCAM-1 was in-
hibited by eristostatin, while a control disintegrin
echistatin had almost no effect (Fig. 6A). To confirm
these findings in another system, we tested the ability
of eristostatin to interfere with binding of a4 cDNA-
transfected CHO cells to VCAM-1. Control CHO cells
did not bind to VCAM-1 while CHOa4 cells bound
weakly. Similar to the findings for MV3, eristostatin
partially inhibited this adhesion while echistatin had
no effect (Fig. 6B).
These results demonstrate that binding of eristo-
statin to a4b1 inhibits cell attachment to VCAM-1, sug-
gesting that interference with a4b1-VCAM-1 interac-
tion may play a role in inhibition of MV3 lung coloniza- FIG. 6. Eristostatin inhibits a4b1-mediated cell binding to
tion by eristostatin in vivo. VCAM-1. (A) MV3 cells were allowed to adhere to wells coated with
0.5 mg recombinant VCAM-1 in the absence or presence of increasing
concentrations of eristostatin (closed circles) or echistatin (open cir-
DISCUSSION cles). (B) Control or a4 cDNA-transfected CHO cells were allowed to
adhere to wells coated with 0.5-mg recombinant VCAM-1 in the ab-
The findings from this study show that the mode of sence or presence of 50 mg/ml disintegrin as indicated. Mean { SD
action of the reported inhibition of melanoma metasta- of three experiments in duplicate is shown.

AID ECR 3821 / 6i2ah$$$82 12-16-97 00:51:48 eca

 ERISTOSTATIN INHIBITS a4b1 FUNCTION AND METASTASIS
more, triggering a5b1 can stimulate invasion of avb3-
negative C8161 melanoma cells [31] and a5b1-binding
to Fn stimulates in vitro melanoma cell proliferation
[29]. We have previously shown that a5b1 in a highly
active state can bind RGD with similar high potency as
avb3 [55], suggesting that a5b1 might be a candidate
integrin for the reported inhibition of melanoma metas-
tasis by RGD-containing peptides such as eristostatin.
In fact, there is one previous report on inhibition of
human melanoma cell metastasis by a disintegrin [60],
in which contortrostatin was found to bind a5b1 on
M24met cells, indicating that blocking a5b1 might
have been involved in its in vivo effect. Our findings
demonstrate that such a mechanism is unlikely for eri-
stostatin as mAbs to a5b1 do not inhibit binding of
MV3 to immobilized eristostatin. Moreover, NKI-
Sam1, cRRETAWA, and echistatin do not affect the
development of MV3 lung metastases, while they all
bind a5b1 [33, 37, 43] and inhibit MV3 adhesion to Fn
which is exclusively mediated by a5b1 [55]. Thus,
taken together, our data show that inhibition of MV3
experimental melanoma metastasis by eristostatin
does not involve blocking a5b1 or avb3.
In addition to interference with normal cell attach-
ment, binding of RGD-containing peptides to integrins
can result in the induction of signal transduction
events (‘‘outside-in signaling’’) [2]. These may be im-
portant in the interpretation of the in vivo effects ob-
served with such reagents. However, there is evidence
that monomeric disintegrins such as eristostatin do not
activate outside-in signaling events. Thus, it has been
demonstrated that eristostatin does not induce tyro-
sine phosphorylation upon binding to platelets,
whereas dimeric disintegrins such as contortrostatin
do [61]. We have also observed the absence of tyrosine
phosphorylation in platelets upon binding of mono-
meric disintegrins echistatin and eristostatin [S. Nie-
wiarowski, unpublished]. In addition, we have shown
that adhesion of endothelial cells to immobilized mono-
meric disintegrins does not promote cell spreading [38],
again demonstrating the inability of these peptides to
induce outside-in signaling. Together these findings
show that the action of eristostatin probably does not
involve outside-in signaling.
There are several reports indicating the importance
of a4b1 for melanoma metastasis. MAbs to a4 have
been reported to inhibit TNFa- or IL1-augmented mu-
rine melanoma cell metastasis [26, 27], and increased
expression of a4b1 is related to tumor progression and
poor prognosis in melanoma patients [22, 23, 25]. In
addition, a4b1 expression correlates with the meta-
static potential of various tumor cell lines [62] and it
was recently shown that transfection of a4b1-cDNA in
K562 or CHO cells induces experimental bone metasta-
sis in mice [63]. Our finding that increased a4b1 ex-
pression is related to the metastatic potential of human
AID ECR 3821 / 6i2ah$$$83 12-16-97 00:51:48
193
melanoma cell lines extends these reports. Transfec-
tion of a4b1 cDNA has been shown to inhibit spontane-
ous but not experimental metastasis of a murine mela-
noma cell line [64], suggesting that a4b1 inhibits early
stages but facilitates late stages of melanoma metasta-
sis. An obvious role for a4b1 in the late stage of mela-
noma metastasis would be in binding of tumor cells to
VCAM-1 on endothelial cells in the blood vessel wall
previous to extravasation.
We have several clues that eristostatin binds a4b1
and interferes with its function. First, FITC-eristos-
tatin binds to Jurkat cells. Second, Jurkat but not K562
cells adhere to eristostatin and this adhesion is par-
tially inhibited by a4 mAbs. Third, binding of a4-trans-
fected CHO cells to immobilized VCAM-1 is inhibited
by eristostatin. At present some aspects of the eristos-
tatin-a4b1 interaction remain puzzling. As cyclic RGD
peptides have been reported to interfere with a4b1 li-
gand binding, the relatively poor inhibitory activity of
the GRGDSP peptide seems surprising [12, 65]. This
may be explained by high-avidity binding of eristos-
tatin, but it may also suggest that the RGD-loop in
eristostatin is not the only region involved in binding
to MV3 cells. Furthermore, as there is direct evidence
that RGDS binding to a4b1 (induced by stimulatory
b1 mAb) can be blocked by HP2/1 [13], it is surprising
that HP2/1 is relatively ineffective to block binding of
eristostatin to Jurkat and MV3 cells. Again, this may
be explained by the fact that the RGD-loop is not the
only region in eristostatin involved in binding to these
cells. Studies on recombinant eristostatin and its mu-
tants will be required to provide insight in these ques-
tions. It appears that biological activities of disinte-
grins depend on several structural features including
amino acids adjacent to RGD, the structure of the RGD-
loop, the pattern of SÅS bridges, and the C-terminal
domain [37, 66, 67].
The fact that this novel a4b1-binding property of
eristostatin is relevant for its interaction with MV3 is
illustrated by the observations that binding of eristos-
tatin to MV3 cells is partially inhibited by HP2/1 a4
mAb and that eristostatin blocks MV3 adhesion to im-
mobilized VCAM-1. Therefore, we propose that the in-
hibition of experimental metastasis by eristostatin in
vivo may involve interference with a4b1-VCAM-1 bind-
ing as well. Recent findings in another system support
this idea. We have observed that eristostatin adminis-
tered every other day intraperitoneally at a dose of 3 mg
in nonobese diabetic (NOD) mice delayed lymphocyte-
infiltration of pancreatic capillaries, resembling the ef-
fect of a4b1 mAb [68]. Furthermore, 200 nM eristo-
statin was found to reduce in vitro adhesion of splenic
lymphocytes from NOD mice to frozen sections of pan-
creatic tissue from the same animals [69].
In these experiments, as well as those with MV3
cells, the biological activities of eristostatin are signifi-
eca
194 DANEN ET AL.
cantly higher than its in vitro effect. For instance, 20–
40 mg of eristostatin causes only partial inhibition of
MV3 cell adhesion to immobilized VCAM-1 in vitro
while this amount is sufficient to prevent lung coloniza-
tion in vivo. Therefore, it is quite possible that the in
vivo activity of eristostatin involves other mechanisms
as well. Morris et al. [18] observed that eristostatin
inhibits liver metastasis of B16F1 murine melanoma
cells and suggested that eristostatin exerted its major
effect by regulating the outgrowth of melanoma cells in
the liver tissue after extravasation. We have previously
found that eristostatin at a dose of 0.5 mg inhibits
B16F10 murine melanoma cell lung colonization [14].
Since eristostatin is a potent inhibitor of in vitro plate-
let aggregation by B16F10 cells, we suggested that dis-
ruption of platelet-melanoma aggregates could explain
its anti-metastatic effect. The formation of B16F10-
platelet aggregates is mediated by binding of fibrinogen
to platelet-aIIbb3 and melanoma cell-avb3. As induc-
tion of in vitro platelet aggregation could not be ob-
served for MV3 (not shown) and as these cells lack
avb3 such a mechanism is unlikely in this experimen-
tal system. It is noteworthy that B16F10 also express
high levels of a4b1 (as determined with an antibody
to mouse a4 chain; Marcinkiewicz and Niewiarowski,
unpublished observations). Thus, blocking this integrin
may have contributed to the antimetastatic effect with
B16 cells as well.
In conclusion, our findings demonstrate that (a) in
addition to earlier findings in the B16F10 model sys-
tem, eristostatin inhibits experimental metastasis of
human melanoma cells, (b) eristostatin binds integrin
a4b1, and (c) eristostatin inhibits melanoma cell adhe-
sion to immobilized VCAM-1. Thus, besides inhibition
of the ‘‘classical’’ RGD-binding integrins, its in vivo ef-
fect may involve interference with a4b1-VCAM bind-
ing. The potent inhibition of experimental metastasis
of both murine and human melanoma cells by eristos-
tatin suggests that it might be a useful tool in the de-
sign of experimental therapies targeted at RGD-depen-
dent cell adhesive events.
We thank Drs. Richard Bankert, Barry Coller, Carl Figdor, Eber-
hard Klein, Steven Nishimura, Egbert Oosterwijk, Mark Renz, Fran-
cisco Sanchez-Madrid, Arnoud Sonnenberg, and Yoshikazu Takada
for kindly providing reagents. This study was supported by Dutch
Cancer Society grant NUKC 91-09, NIH Grants HL45486 and DE
11844, and the American Heart Association, Southeastern Pennsyl-
vania Chapter.
REFERENCES
1. Cancer Surveys (1995). ‘‘Cell Adhesion and Cancer’’ (I. Hart,
Ed.), Vol. 24.
2. Hynes, R. O. (1992). Integrins: Versatility, modulation, and sig-
nalling in cell adhesion. Cell 69, 11–25.
3. Elices, M. J., Urry, L. A., and Hemler, M. E. (1991). Receptor
AID ECR 3821 / 6i2ah$$$83 12-16-97 00:51:48
function for the integrin VLA-3: Fibronectin, collagen, and lami-
nin binding are differentially influenced by ARG-GLY-ASP pep-
tide and by divalent cations. J. Cell Biol. 112, 169–181.
4. Akiyama, S. K., and Yamada, K. M. (1985). The interaction of
plasma fibronectin with fibroblastic cells in suspension. J. Biol.
Chem. 260, 4492–4500.
5. Pytela, R., Pierschbacher, M. D., and Ruoslahti, E. A. (1985).
Identification and isolation of a 140 Kd cell surface glycoprotein
with properties expected of a fibronectin receptor. Cell 40, 191–
198.
6. Vogel, B. E., Tarone, G., Giancotti, F. G., Gailit, J., and Ruos-
lahti, E. (1990). A novel fibronectin receptor with an unexpected
subunit composition (avb1). J. Biol. Chem. 265, 5934–5937.
7. Pytela, R., Pierschbacher, M. D., and Ruoslahti, E. A. (1985). A
125/115 Kda cell surface receptor specific for vitronectin inter-
acts with the arginine-glycine-aspartic acid adhesion sequence
derived from fibronectin. Proc. Natl. Acad. Sci. USA 82, 5766–
5770.
8. Cheresh, D. A., Smith, J. W., Cooper, H. M., and Quaranta, V.
(1989). A novel vitronectin receptor integrin (avbx) is responsi-
ble for distinct adhesive properties of carcinoma cells. Cell 57,
59–69.
9. Busk, M., Pytela, R., and Sheppard, D. (1992). Characterization
of the integrin avb6 as a fibronectin-binding protein. J. Biol.
Chem. 267, 5790–5796.
10. Nishimura, S. L., Sheppard, D., and Pytela, R. (1994). Integrin
avb8: Interaction with vitronectin and functional divergence of
the b8 cytoplasmic domain. J. Biol. Chem. 269, 28708–28715.
11. Cadarelli, P. M., Yamagata, S., Taguchi, I., Gosrcan, F., Chiang,
S. L., and Lobl, T. (1992). The collagen receptor a2b1 from H-
63 and HT1080 cells interacts with a cyclic RGD peptide. J.
Biol. Chem. 267, 23159–23164.
12. Cadarelli, P. M., Cobb, R. R., Nowlin, D. M., Scholz, W., Gorc-
san, F., Moscinski, M., Yasuhara, M., Chiang, S., and Lobl,
T. J. (1994). Cyclic RGD peptide inhibits a4b1 interaction with
connecting segment 1 and vascular cell adhesion molecule. J.
Biol. Chem. 269, 18668–18673.
13. Sanchez-Apparicio, P., Dominguez-Gimenez, C., and Garcia-
Pardo, A. (1994). Activation of the a4b1 integrin through the
b1 subunit induces recognition of the RGDS sequence in fibro-
nectin. J. Cell Biol. 126, 271–279.
14. Beviglia, L., Stewart, G. J., and Niewiarowski, S. (1995). Effect
of four disintegrins on the adhesive and metastatic properties
of B16F10 melanoma cells in a murine model. Oncol. Res. 7, 7–
20.
15. Hardan, I., Weiss, L., Hershkoviz, R., Freenspoon, N., Alon, R.,
Cahalon, L., Reich, S., Slavin, S., and Lider, O. (1993). Inhibi-
tion of metastatic cell colonization in murine lungs and tumor-
induced morbidity by non-peptidic arg-gly-asp mimetics. Int. J.
Cancer 55, 1023–1028.
16. Humphries, M. J., Olden, K., and Yamada, K. M. (1986). A syn-
thetic peptide from fibronectin inhibits experimental metastasis
of murine melanoma cells. Science 233, 467–470.
17. Humphries, M. J., Yamada, K. M., and Olden, K. (1988). Inves-
tigation of the biological effects of anti-cell adhesive synthetic
peptides that inhibit experimental metastasis of B16-F10 mu-
rine melanoma cells. J. Clin. Invest. 81, 782–790.
18. Morris, V. L., Schmidt, E. E., Koop, S., MacDonald, I. C., Grat-
tan, M., Khokha, R., McLane, M. A., Niewiarowski, S., Cham-
bers, A. F., and Groom, A. C. (1995). Effects of the disintegrin
eristostatin on individual steps of hematogenous metastasis.
Exp. Cell Res. 219, 571–578.
19. Saiki, I., Iida, J., Murata, J., Ogawa, R., Nishi, N., Sugimura,
K., Tokura, S., and Azuma, I. (1989). Inhibition of the metasta-
eca
 ERISTOSTATIN INHIBITS a4b1 FUNCTION AND METASTASIS
sis of murine malignant melanoma by synthetic polymeric pep-
tides containing core sequences of cell-adhesive molecules. Can-
cer Res. 49, 3815–3822.
20. Sheu, J. R., Lin, C. H., Chung, J. L., Teng, C. M., and Huang,
T. F. (1992). Triflavin, an Arg-Gly-Asp-containing antiplatelet
peptide inhibits cell-substratum adhesion and melanoma cell-
induced lung colonization. Jpn. J. Cancer Res. 83, 895–893.
21. Soszka, T., Knudsen, K. A., Beviglia, L., Rossi, C., Poggi, A.,
and Niewiarowski, S. (1991). Inhibition of murine melanoma
cell-matrix adhesion and experimental metastasis by albola-
brin, an RGD-containing peptide isolated from the venom of
Trimeresurus albolabris. Exp. Cell Res. 196, 6–12.
22. Albelda, S. M., Mette, S., Elder, D., Stewart, R., Damjanovich,
L., Herlyn, M., and Buck, C. A. (1990). Integrin distribution in
malignant melanoma: association of the b3 subunit with tumor
progression. Cancer Res. 50, 6757–6764.
23. Danen, E. H. J., Ten Berge, P. J. M., Van Muijen, G. N. P., Van’t
Hof-Grootenboer, A. E., Brocker, A. B., and Ruiter, D. J. (1994).
Emergence of a5b1 fibronectin- and avb3 vitronectin-receptor
expression in melanocytic tumor progression. Histopathology
24, 249–256.
24. Schadendorf, D., Gawlik, C., Haney, U., Ostmeier, H., Suter,
L., and Czarnetzki, B. M. (1993). Tumor progression and metas-
tasis in vivo correlates with integrin expression on melanocytic
tumors. J. Pathol. 170, 429–434.
25. Schadendorf, D., Heidel, J., Gawlick, C., Suter, L., and Czarnet-
zki, B. M. (1995). Association with clinical outcome of expres-
sion of VLA-4 in primary cutaneous malignant melanoma as
well as P-selectin on intratumoral vessels. J. Natl. Cancer Inst.
87, 366–371.
26. Garofalo, A., Chirivi, R. G. S., Foglieni, C., Pigott, R., Mortarini,
R., Martin-Padura, I., Anichini, A., Gearing, A. J., Sanchez-Ma-
drid, F., Dejana, E., and Giavazzi, R. (1995). Involvement of
the very late antigen 4 integrin on melanoma in interleukin 1-
augmented experimental metastasis. Cancer Res. 55, 414–419.
27. Okahara, H., Yagita, H., Miyake, K., and Okumura, K. (1994).
Involvement of very late activation antigen 4 (VLA-4) and vas-
cular cell adhesion molecule 1 (VCAM-1) in tumor necrosis fac-
tor a enhancement of experimental metastasis. Cancer Res. 54,
3233–3236.
28. Felding-Habermann, B., Mueller, B. M., Romerdahl, C. A., and
Cheresh, D. A. (1992). Involvement of integrin av gene expres-
sion in human melanoma tumorigenicity. J. Clin. Invest. 89,
2018–2022.
29. Mortarini, R., Gismondi, A., Santoni, A., Parmiani, G., and Ani-
chini, A. (1992). Role of the a5b1 integrin receptor in the prolif-
erative response of quiescent human melanoma cells to fibro-
nectin. Cancer Res. 52, 4499–4506.
30. Seftor, R. E. B., Seftor, E. A., Gehlsen, K. R., Stetler-Stevenson,
W. G., Brown, P. D., Ruoslahti, E., and Hendrix, M. J. C. (1992).
Role of the avb3 integrin in human melanoma cell invasion.
Proc. Natl. Acad. Sci. USA 89, 1557–1561.
31. Seftor, R. E. B., Seftor, E. A., Stetler-Stevenson, W. G., and
Hendrix, M. J. C. (1993). The 72 kDa type IV collagenase is
modulated via differential expression of avb3 and a5b1 inte-
grins during human melanoma cell invasion. Cancer Res. 53,
3411–3415.
32. Renz, M. E., Chiu, H. H., Jones, S., Fox, J., Kiru, K. J., Presta,
L. G., and Fong, S. (1994). Structural requirements for adhesion
of soluble recombinant vascular adhesion molecule 1 to a4b1.
J. Cell Biol. 125, 10241–10245.
33. Koivunen, E., Wang, B., and Ruoslahti, E. (1993). Isolation of
a highly specific ligand for the a5b1 integrin from a phage
display library. J. Cell Biol. 124, 373–380.
AID ECR 3821 / 6i2ah$$$84 12-16-97 00:51:48
195
34. Pachter, J. A., Zhang, R., and Mayer-Ezell, R. (1995). Scintilla-
tion proximity assay to measure binding of soluble fibronectin
to antibody-captured a5b1 integrin. Anal. Biochem. 230, 101–
107.
35. Williams, J. A., Rucinski, B., Holt, J. C., and Niewiarowski, S.
(1990). Elegantin and albolabrin purified peptides from viper
venoms; homologies with the RGDS domain of fibrinogen and
von Willebrand factor. Biochim. Biophys. Acta 1039, 81–89.
36. McLane, M. A., Kowalska, M. A., Silver, L., Shattil, S. J., and
Niewiarowski, S. (1994). Interaction of disintegrins with the
aIIbb3 receptor on resting and activated human platelets. Bio-
chem. J. 301, 429–436.
37. Pfaff, M., McLane, A. M., Beviglia, L., Niewiarowski, S., and
Timpl, R. (1994). Comparison of disintegrins with limited varia-
tion in the RGD loop in their binding to purified integrins aI-
Ibb3, avb3 and a5b1 and in cell adhesion inhibition. Cell Adhe-
sion Comm. 2, 491–501.
38. Juliano, D., Wang, Y., Marcinkiewicz, C., Rosenthal, L. A.,
Stewart, G. J., and Niewiarowski, S. (1996). Disintegrin inter-
action with avb3 integrin on human umbilical vein endothelial
cells: Expression of ligand induced binding site on b3 subunit.
Exp. Cell Res. 225, 132–142.
39. Marcinkiewicz, C., Rosenthal, L. A., Mosser, B. M., Kunicki,
T. J., and Niewiarowski, S. (1996). Immunological characteriza-
tion of eristostatin and echistatin binding sites on aIIbb3 and
avb3 integrins. Biochem. J. 317, 817–825.
40. Zylstra, S., Chen, F. A., Ghosh, S. K., Repasky, E. A., Rao, U.,
Takita, H., and Bankert, R. B. (1986). Membrane associated
gp160 identified on human lung tumor by a monoclonal anti-
body. Cancer Res. 48, 2768–2773.
41. Wayner, E. A., and Carter, W. G. (1987). Identification of multi-
ple cell adhesion receptors for collagen and fibronectin in hu-
man fibrosarcoma cells possessing unique a and common b sub-
units. J. Cell Biol. 105, 1873–1884.
42. Sanchez-Madrid, F., De Landazuri, M. O., Morago, G., Cebrian,
M., Acevedo, A., and Bernabeu, C. (1986). VLA-3: A novel poly-
peptide association within the VLA molecular complex: Cell
distribution and biochemical characterization. Eur. J. Immu-
nol. 16, 1343–1349.
43. Van de Wiel van Kemenade, E., van Kooyk, Y., de Boer, A. J.,
Huijbens, R. J. F., Weder, P., van de Kasteele, W., Melief,
C. J. M., and Figdor, C. G. (1992). Adhesion of T and B lympho-
cytes to extracellular matrix and endothelial cells can be regu-
lated through the b subunit of VLA. J. Cell Biol. 117, 461–470.
44. Sonnenberg, A., Janssen, H., Hogervorst, F., Calafat, J., and
Hilgers, J. (1987). A complex of platelet glycoproteins Ic and
IIa identified by a rat monoclonal antibody. J. Biol. Chem. 264,
13745–13750.
45. Morimoto, C., Letvin, N. L., Boyd, A. W., Hagan, M., Brown,
H. M., Kornacki, M. M., and Schlossman, S. F. (1985). The isola-
tion and characterization of the human helper inducer T cell
subset. J. Immunol. 134, 3762–3769.
46. Hemler, M. E., Sanchez-Madrid, F., Flotte, T. J., Kremsky,
A. M., Burakoff, S. J., Bhan, A. K., Springer, T. A., and Strom-
inger, J. L. (1984). Glycoproteins of 210,000 and 130,000 m.w.
on activated T cells: Cell distribution and antigenic relation to
components on resting cells and T cell lines. J. Immunol. 132,
3011–3018.
47. Wayner, E. A., Orlando, R. A., and Cheresh, D. A. (1991). Inte-
grins avb3 and avb5 contribute to cell attachment to vitronec-
tin but differentially distribute on the cell surface. J. Cell Biol.
113, 919–929.
48. Coller, B. S., Peerschke, E. I., Scudder, L. E., and Sullivan, C. A.
(1983). A murine monoclonal antibody that completely blocks
eca
196 DANEN ET AL.
the binding of fibrinogen to platelets produces a thrombas-
thenic-like state in normal platelets and binds to glycoproteins
IIb and/or IIIa. J. Clin. Invest. 72, 325–338.
49. Susuki, S., Argraves, W. S., Pytela, R., Arai, H., Krusius, T.,
Pierschbacher, M. D., and Ruoslahti, E. (1986). cDNA and
amino acid sequence of the cell adhesion receptor recognizing
vitronectin reveal a transmembrane membrane and homologies
with other adhesion protein receptors. Proc. Natl. Acad. Sci.
USA 83, 8614–8618.
50. Oosterwijk, E., Ruiter, D. J., Hoedemaker, P. J., Pauwels,
E. K. J., Jonas, U., Zwartendijk, J., and Warnaar, S. O. (1986).
Monoclonal antibody G250 recognizes a determinant present in
renal cell carcinoma and absent from normal kidney. Int. J.
Cancer 38, 489–494.
51. Van Muijen, G. N. P., Cornelissen, I. M. H. A., Jansen, C. F. J.,
Figdor, C. G., Johnson, J. P., Brocker, E.-B., and Ruiter, D. J.
(1991). Antigen expression of metastasizing and non-metasta-
sizing human melanoma cells xenografted into nude mice. Clin.
Exp. Metast. 9, 259–272.
52. Van Muijen, G. N. P., Jansen, C. F. J., Cornelissen, I. M. H. A.,
Smeets, D. F. C. M., Beck, J. L. M., and Ruiter, D. J. (1991). Es-
tablishment and characterization of a human melanoma cell
line (MV3) which is highly metastatic in nude mice. Int. J.
Cancer 48, 85–91.
53. Danen, E. H. J., Van Muijen, G. N. P., Van de Wiel-van Keme-
nade, E., Jansen, K. F. J., Ruiter, D. J., and Figdor, C. G. (1993).
Regulation of integrin-mediated adhesion to laminin and colla-
gen in human melanocytes and in non-metastatic and highly
metastatic human melanoma cells. Int. J. Cancer 54, 315–321.
54. Kamata, T., Puzon, W., and Takada, Y. (1995). Identification
of putative ligand binding sites of the integrin a4b1 (VLA4,
cd49/cd29). Biochem. J. 305, 945–951.
55. Danen, E. H. J., Aota, S. I., Van Kraats, A., Yamada, K. M.,
Ruiter, D. J., and Van Muijen, G. N. P. (1995). Requirement for
the synergy site for cell adhesion to fibronectin depends on the
activation state of integrin a5b1. J. Biol. Chem. 270, 21612–
21618.
56. Aroyo, A. G., Garcia-Pardo, A., and Sanchez Madrid, F. (1993).
A high affinity conformation state on VLA integrin heterodim-
ers induced by an anti-b1 chain monoclonal antibody. J. Biol.
Chem. 268, 9863–9868.
57. Nip, J., and Brodt, P. (1995). The role of the integrin vitronectin
receptor, avb3 in melanoma metastasis. Cancer Metast. Rev.
14, 241–252.
58. Danen, E. H. J., Jansen, K. F. J., Van Kraats, A. A., Cornelis-
sen, I. M. H. A., Ruiter, D. J., and van Muijen, G. N. P. (1995).
av-Integrins in human melanoma: gain of avb3 and loss of avb5
Received April 18, 1997
Revised version received August 18, 1997
AID ECR 3821 / 6i2ah$$$84 12-16-97 00:51:48
are related to tumor progression in situ but not to metastatic
capacity of cell lines in nude mice. Int. J. Cancer 61, 491–496.
59. Boucherke, H., Benchaibi, M., Berthier-Vergnes, O., Lizard, G.,
Bailly, M., and McGregor, J. L. (1994). Two human melanoma
cell-line variants with enhanced in vivo tumor growth and meta-
static capacity do not express the b3 integrin subunit. Eur. J.
Biochem. 220, 485–491.
60. Trikha, M., De Clerck, Y. A., and Markland, F. S. (1994). Con-
tortrostatin, a snake venom disintegrin, inhibits b1 integrin-
mediated human metastatic melanoma cell adhesion and blocks
experimental metastasis. Cancer Res. 54, 4993–4998.
61. Clark, E. A., Trikha, M., Markland, F. S., and Brugge, J. S.
(1994). Structurally distinct disintegrins contortrostatin and
multisquamatin differentially regulate platelet tyrosine phos-
phorylation. J. Biol. Chem. 269, 21940–21943.
62. Bao, L., Pigott, R., Matsumura, Y., Baban, D., and Tarin, D.
(1993). Correlation of VLA-4 integrin expression with meta-
static potential in various human tumor cell lines. Differentia-
tion 52, 239–246.
63. Matsuura, N., Puzon-McLaughlin, W., Irie, A., Morikawa, Y.,
Kakudo, K., and Takada, Y. (1996). Induction of experimental
bone metastasis in mice by transfection of integrin a4b1 into
tumor cells. Am. J. Pathol. 148, 55–61.
64. Qian, F., Vaux, D. L., and Weissman, I. (1994). Expression of
the integrin a4b1 on melanoma cells can inhibit the invasive
stage of metastasis formation. Cell 77, 335–347.
65. Nowlin, D. M., Gorcsan, F., Moscinski, M., Chiang, S., Lobl,
T. J., and Cardarelli, P. M. (1993). A novel cyclic pentapeptide
inhibits a4b1 and a5b1 integrin-mediated cell adhesion. J. Biol.
Chem. 268, 20352–20359.
66. Marcinkiewicz, C., Vijay-Kumars, S., McLane, M. A., and Nie-
wiarowski, S. (1996). Significance of RGD loop and C-terminal
domain of echistatin for recognition of aIIbb3 and avb3 inte-
grins and expression of ligand induced binding site. Blood
88(Suppl 1), 1110. [Abstract]
67. McLane, M. A., Vijay-Kumar, s., Marcinkiewicz, C., Calvete,
J. J., and Niewiarowski, S. (1996). Importance of the structure
of RGD-containing loop in the disintegrins echistatin and eris-
tostatin for recognition of aIIbb3 and avb3 integrins. FEBS
Lett. 391, 139–143.
68. Marcinciewicz, C., Brando, C., Danen, E. H. J., Marcinkiewicz,
and Niewiarowski, S. (1997). Disintegrin eristostatin inhibits
lymphocyte infiltration in NOD mice by inhibiting interaction
of a4b1 integrin with VCAM-1. FASEB J. 11. [Abstract No.
602]
69. Marcinkiewicz, C., Brando, C., Juliano, D., Goldman, B., and
Niewiarowski, S. (1996). Disintegrins inhibit interaction be-
tween lymphoid and endothelial cells. FASEB. J. 10. [Abstract
No. 3546]
eca