Nanomolar small molecule inhibitors for ανβ6, ανβ5, and ανβ3 integrins

Article abstract of DOI:10.1021/jm0102598

Integrin adhesion receptors frequently recognize a core amino acid sequence, Arg-Gly-Asp, in their target ligands. Inhibitors with the ability to inhibit one or a small subset of such RGD-dependent integrins have been invaluable in defining their biological function. Here, we have characterized low molecular weight inhibitors for their ability to specifically inhibit ανβ6 integrin, a fibronectin/tenascin receptor. As of yet, no nonpeptidic inhibitor of ανβ6 was known. New peptidomimetic and nonpeptidic compounds were examined in isolated integrin binding assays and in cell adhesion assays for their ability to block ανβ6, ανβ3, ανβ5, and αIIbβ3 integrins. The compounds are based on an aromatically substituted β amino acid or glutaric acid derivative as an acidic center and an aminopyridyl or guanidyl residue as a basic mimetic. We found several classes of inhibitors with different selectivities, especially mono- or biselectivity on the αν-integrins ανβ6 and ανβ3, and nanomolar activity. Furthermore, nearly all compounds are inactive on αIIbβ3. Compound 11 is the first specific, peptidomimetic inhibitor of the ανβ6 integrin receptor.

Full text of DOI:10.1021/jm0102598

J . Med. Chem. 2002, 45, 1045-1051
1045
Na n om ola r Sm a ll Molecu le In h ibitor s for rvâ6, rvâ5, a n d rvâ3 In tegr in s
Simon L. Goodman,*,† Gu¨nter Ho¨lzemann,‡ Ga´bor A. G. Sulyok,§ and Horst Kessler§
Departments of Preclinical Oncology and Medicinal Chemistry, Merck KGaA, Frankfurterstr. 250,
64271 Darmstadt, Germany, and Institute for Organic Chemistry and Biochemistry, Technische Universita¨t Mu¨nchen,
Lichtenbergstr. 4, 85747 Garching, Germany
Received J une 12, 2001
Integrin adhesion receptors frequently recognize a core amino acid sequence, Arg-Gly-Asp, in
their target ligands. Inhibitors with the ability to inhibit one or a small subset of such RGD-
dependent integrins have been invaluable in defining their biological function. Here, we have
characterized low molecular weight inhibitors for their ability to specifically inhibit Rvâ6
integrin, a fibronectin/tenascin receptor. As of yet, no nonpeptidic inhibitor of Rvâ6 was known.
New peptidomimetic and nonpeptidic compounds were examined in isolated integrin binding
assays and in cell adhesion assays for their ability to block Rvâ6, Rvâ3, Rvâ5, and RlIbâ3
integrins. The compounds are based on an aromatically substituted â amino acid or glutaric
acid derivative as an acidic center and an aminopyridyl or guanidyl residue as a basic mimetic.
We found several classes of inhibitors with different selectivities, especially mono- or biselectivity
on the Rv-integrins Rvâ6 and Rvâ3, and nanomolar activity. Furthermore, nearly all compounds
are inactive on RIIbâ3. Compound 11 is the first specific, peptidomimetic inhibitor of the Rvâ6
integrin receptor.
In tr od u ction
related integrins were also studied to demonstrate the
specificity of the new compounds. We examined the
following: RIIbâ3, the platelet fibrinogen receptor,
involved in thrombosis;38 Rvâ3 a promiscuous receptor
for vitronectin, fibronectin, fibrinogen, and other pro-
teins, up-regulated on endothelia during tumor angio-
genesis39 and on smooth muscle cells during prolifera-
tion;40 and Rvâ5 a vitronectin receptor involved in
angiogenesis of retinopathy.41
Integrins are a family of heterodimeric cell surface
receptors that regulate cell attachment and response to
the extracellular matrix.1 Integrins discriminate well
between their target ligands, so it came as a surprise
that for many integrins the recognition sequence in the
matrix protein contains a core amino acid sequence,
“Arg-Gly-Asp” (RGD). This has complicated the discov-
ery of specific inhibitors. For example, at least the five
Rv-integrins (â1, â3, â5, â6, â8), as well as integrins
RIIbâ3, R5â1, and R8â1 bind their ligands via the RGD
sequence. For some of these integrins, the specificity of
interaction with the ligand involves a supplementary
site distant from the RGD site.2 It is, therefore, a major
challenge to identify low molecular weight compounds
that can discriminate between these integrins. Never-
theless, extremely selective nonpeptidic inhibitors have
been discovered for several “RGD-dependent” integrins
including RIIbâ6,3-6 Rvâ3,7-14 R4â1,15 and R5â1,16 while
peptidic inhibitors with high specificity and selectivity
have been described for RIIbâ3,17,18 for Rvâ3,19-25 for
Rvâ6,26 and for R5â1.27
We have attempted to find small molecule inhibitors
that discriminate between subsets of the “RGD” inte-
grins implicated in human pathologies, focusing on Rvâ6
for which no small molecule inhibitors have been
described so far. Rvâ6 is a fibronectin/tenascin receptor
specific to epithelia that is up-regulated during inflam-
matory events,28 in tumor proliferation,29-31 and during
wound healing.32,33 Rvâ6 has been linked with cell
migration and proliferation,34 gelatinase B expres-
sion,35,36 and the activation of TGF-â1.37 Other closely
Integrin inhibitors are of interest in three major
areas. First, integrin inhibitors may lead to the develop-
ment of drugs. Alterations of integrin activity and
expression characterize numerous pathological condi-
tions, whose consequences may be reversed by inhibitors
of integrins. Specific inhibitors of individual integrins
or of defined integrin groups have helped clarify the
function of integrins in pathologies. Second, from the
perspective of defining the basic biological role of the
integrins, specific inhibitors can be used to investigate
the role of individual integrins at the cellular level.
Finally, from the perspective of structural analysis, the
molecular basis of ligand recognition by integrins is for
the most part mysterious, and novel structural probes
are valuable for this recognition process. However, there
are still gaps in our knowledge due to the lack of specific
inhibitors.
Here, we describe the characterization of nanomolar
Rvâ6 receptor inhibitors with mixed profiles that inhibit
Rvâ6 and Rvâ3; or Rvâ6, Rvâ3, and Rvâ5; in each case,
with little effect on RIIbâ3. These inhibitors with similar
structural features allow the investigation of structure-
activity relationships for the different Rv-integrin sub-
types. Together, these inhibitors are very useful to study
the combined function of Rv-series integrins and, in
particular, provide for the first time highly active small
molecule inhibitors of Rvâ6.
* Author for correspondence. Tel: 0049 6151 726452. Fax: 0049
6151 7290452. E-mail: Goodman@merck.de.
† Department of Preclinical Oncology, Merck KGaA.
‡ Department of Medicinal Chemistry, Merck KGaA.
§ Technische Universita¨t Mu¨nchen.
10.1021/jm0102598 CCC: $22.00 © 2002 American Chemical Society
Published on Web 01/31/2002

1046 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 5
Goodman et al.
Exp er im en ta l Section
RGD peptide sequence. These mimetics possess, with
the exception of 13, an aminopyridine moiety as the
basic part of the molecule. In compound 13, the guani-
dino group itself served this role. The acidic, C-terminal
component of the molecules is formed either by an
aromatic â amino acid (8-15) or a substituted glutaric
acid group (6, 7). The glycine spacer is retained in 9-12
and 14. Aza-glycine is used in 6-8 and 13, and it is
completely omitted in 15.
Ch em istr y. The synthesis of the cyclic peptides has already
been described: N-methylated peptides 1 and 225 and aza-
glycine containing peptides 3-5.42,43 The aza-RGD mimetics
6-8 and 13 were synthesized by solid-phase synthesis on TCP
resin using the oxadiazolone route.9 Details of the syntheses
have been published elsewhere.13 Non aza compounds 9-12
and 14-15 were synthesized in solution using conventional
peptide coupling techniques.44 The aromatic â amino acid was
obtained from a reaction of the corresponding benzaldehyde
with ammonium acetate and malonic acid.13 After esterifica-
tion, Boc-Gly-OH was coupled N-terminally. The Boc group of
the dipeptide was cleaved and the resulting compound coupled
with a readily prepared pyridin-2-ylamino alkylcarboxylic acid.
The final product was obtained by saponification and purifica-
tion by preparative HPLC (for details, see Supporting Infor-
mation).
Biology. (a ) Isola ted In tegr in -Liga n d Bin d in g Assa ys.
The production of recombinant human Rvâ3 and Rvâ6 has been
described elsewhere.26,45 Recombinant human Rvâ5 was pro-
duced by similar technologies.46 Human platelet RIIbâ3 was
isolated as described.47 The inhibitory activity of the sub-
stances described here was tested in ligand inhibition assays,25
using immobilized integrin as the target, and biotinylated
human serum vitronectin (Rvâ3 and Rvâ5), fibronectin (Rvâ6),
and fibrinogen (RIIbâ3) as ligands. In brief, 96-well ELISA
plates were coated by adsorption from neutral aqueous buffers
of 1 µg mL-1 integrin. After blocking residual sites on the plate
with BSA, biotinylated ligands (1 µg mL-1) were added in the
presence or absence of serial dilutions of inhibitor, and after
incubation and washing, bound biotin was detected with
peroxidase-coupled anti-biotin antibody and TMB substrate.
IC50, the concentration of inhibitor needed to inhibit ligand
binding in the absence of inhibitor by 50%, was established
by curve fitting, and the values presented are usually the mean
of three or more such independent determinations. These
ligand binding assays distinguish reproducibly between com-
pounds with minimal activity (IC50 > 10 µM), and those with
activity here defined as low (10 µM < IC50 < 1 µM), moderate
(1 µM < IC50 < 10 nM), and high (IC50 < 10 nM). Standard
compound 2 gave mean IC50 on Rvâ3 of 3.4 nM, a value within
the experimental variability of our previously reported data.25
Standard deviations were typically of the same order as the
mean values.
(b) Cell Atta ch m en t Assa y. The role of various Rv-
integrins in initial cell attachment to extracellular matrix was
measured in a quantitative assay using lysosomal hexosamini-
dase to determine cell number.47 M21 human melanoma cells
attach to vitronectin using integrins Rvâ3 and Rvâ5, and the
Rvâ3 component can be examined by performing the assay in
the presence of the Rvâ5 blocking antibody P1F6.47 UCLAP-3
human lung adenocarcinoma cells attach to vitronectin exclu-
sively via Rvâ5 integrin, and to fibronectin exclusively via
Rvâ6, as confirmed by blocking with an RTDLXXL peptide as
described for HT-29 cells.26 In brief, the effect of inhibitors on
cell attachment to surfaces coated with vitronectin or fibronec-
tin was measured 1 h after the freshly harvested cells had
been plated into a serial dilution of the inhibitors. IC50, the
concentration of inhibitor needed to inhibit cell attachment
in the absence of inhibitor by 50%, was established by curve
fitting, and the values presented here are the mean of two such
independent determinations. Standard deviations were typi-
cally of the same order as the mean values.
Biology. We tested the activities of the novel syn-
thetic compounds in inhibiting the interaction of ligands
with isolated immobilized integrins. The results are
shown in Table 1 and visualized in Figure 2. We have
previously recognized that constraints in cyclic RGD-
peptide stereoisomers can be used to tune the integrin
inhibition profiles,19,48 as shown by comparing com-
pounds 1-5, some of which have been previously
described by ourselves. Compound 1 is a specific Rvâ3
inhibitor (IC50 24 nM) with moderate activity on Rvâ5
and low activity on Rvâ6 and RIIbâ3. The substitu-
tion of valine for alanine to furnish 2, cyclic (-Arg-
Gly-Asp-D-Phe-(NMe)Val-), EMD 121974, increased
inhibitory activity on each integrin.25 Interestingly,
the aza-Gly derivative 3, cyclic (-Arg-azaGly-Asp-D-Phe-
(NMe)Val-),42,43 also exhibits high activities and a
similar activity profile (Table 1): compound 2 [3] has
an IC50 value of 3 [10] nM on Rvâ3, 37 [40] nM on Rvâ5,
and 590 [2800] nM on RIIbâ3. Compounds 2 and 3 also
have similar activities on Rvâ6, 470 nM vs 900 nM.
Deletion of the N-methyl group in 4 resulted in a loss
of activity on Rvâ5. Compound 4 is highly specific for
integrin Rvâ3 with a selectivity of 2 orders over Rvâ5
and 3 orders over Rvâ6. Activity for RIIbâ3 was strongly
increased in compound 5 by exchanging the D,L-stere-
ochemistry of the valine and phenylalanine groups,
while the activity on Rvâ3 is still in the low nanomolar
range. These variations in cyclic peptides confirmed that
subtle alterations in stereochemistry could produce
drastic alterations in both activity and selectivity of the
integrin inhibitors.
Nonpeptidic azacarba-derivatives, compounds 6 and
7, exhibit high activity for Rvâ3 and high selectivity
toward Rvâ5. These compounds show only modest
activity on Rvâ6 and are inactive on both Rvâ5 and
RIIbâ3. Switching first the methylene for an NH group,
compound 8, and then the NH of aza-glycine for a
methylene, obtaining glycine containing compound 9,
an increase of activity to Rvâ6 and a decrease of Rvâ3/
Rvâ5 selectivity is observed. With these variations, it
is possible to switch Rvâ3/Rvâ6 selectivity in a modest
way, as shown in the selectivity diagram (Figure 2).
Scientists of Monsanto/Searle first described Rvâ3 in-
tegrin antagonists with the 3,5-dichloro substitution
pattern (compounds 7, 8, and 9) in 1998.50
Modification of the phenyl ring also produced drastic
alterations in integrin inhibitory activity. The replace-
ment of the 3,5-dichloro groups with a 3-trifluo-
romethoxy function, compound 10, resulted in slight loss
of activity on Rvâ3, and a gain of almost 2 orders of
magnitude on Rvâ5, while remaining the activities for
RIIbâ3 (inactive) and Rvâ6 (subnanomolar). Derivati-
zation of the phenyl ring at the 4-position with naphthyl
or phenyl rings, compounds 11 and 12, resulted in a 2
orders of magnitude decrease of inhibitory activity to
Resu lts
Ch em istr y. The structures of the compounds de-
scribed in this paper are shown in Figure 1. A large
number of different cyclic pentapeptides containing the
RGD sequence have been synthesized in our group.21,25,48
For example, we used N-methylated amino acids25,49 and
exchanged glycine for aza-glycine in some peptides.42,43
To obtain peptidomimetics, we synthesized com-
pounds incorporating features that are present in the

Rv-Integrin Inhibitors
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 5 1047
F igu r e 1. Structural formulas of compounds described in text. Lower case letters represent D-amino acids. NMe ) N-methyl.
Ta ble 1. Potent and Selective Rvâ6 Inhibitors in Isolated
our knowledge the first Rvâ3/Rvâ6 mixed inhibitor with
subnanomolar activity. As suggested by the loss of Rvâ5
activity between compounds 10 and 11, the activity was
apparently influenced by the bulk of the function,
derivatizing the phenyl ring (compounds 6, 11, 12).
Although the Rvâ6 inhibitory activity was relatively
insensitive to modifications of the phenyl ring, it was
sensitive to alterations at the basic end of the molecule.
Replacement of the pyridin-2-ylamino by a guanyl
function, compound 13, resulted in a loss of activity on
Rvâ6, IC50 460 nM, and a Rvâ3/Rvâ5 mixed inhibitor
with modest selectivity on Rvâ6 was obtained. Activity
on RIIbâ3 was slightly enhanced. The 3-nitro-phenyl
derivative, compound 14, in comparison to compound
10, had similar activity on Rvâ5 and Rvâ3 and lost 1
order of magnitude activity on Rvâ6, while the RIIbâ3
activity remained unchanged. As shown in Figure 2,
compound 14 was thus a strong Rvâ3, Rvâ5, and Rvâ6
mixed inhibitor. The necessity of the glycine amide
function was demonstrated by replacing this unit by an
alkyl chain, compound 15, which had lost all activity
for Rvâ5 and little activity for Rvâ6 but had gained more
than 1 order of magnitude activity for RIIbâ3. The
Receptor Binding Assaya
IC50 (nM) on integrin
Rvâ6
(st. dev.)
Rvâ5
(st. dev.)
Rvâ3
(st. dev.)
RIIbâ3
(st. dev.)
compd
nb
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
2400 (700)
470 (310)
900 (500)
6000 (2100)
530 (220) >10000 -
53 (25)
44 (12)
1.7 (1.4)
0.4 (0.4)
0.15 (0.3)
0.04 (0.07)
0.6 (0.5)
460 (110)
4 (4)
662 (570)
37 (23)
40 (48)
24 (16)
3 (2)
6000 (4000)
590 (490)
2800 (1800)
3200 (3300)
45 (58)
3000 (640)
>10000 -
3
6
2
2
2
2
2
4
3
3
6
3
2
3
3
10 (5)
3 (0.7)
3 (0.8)
4 (3.3)
6 (4)
64 (41)
53 (63)
14 (7)
8 (5)
540 (98)
8700 (1900)
5700 (200)
2500 (3500)
350 (300)
39 (40)
5170 (2200)
2670 (3600) 0.45 (0.57)
26 (23)
42 (43)
1600 (1350)
>10000 -
>10000 -
>10000 -
4230 (3080)
4050 (3300)
1000 (1500)
7300 (1290)
580 (725)
40 (25)
13 (15)
53 (41)
61 (23)
a
The concentrations (nM) of compounds, numbered as in text,
necessary to inhibit ligand binding to 50% of control, the IC50, on
the isolated recombinant human integrins Rvâ3 and Rvâ5 (ligand
vitronectin), Rvâ6 (ligand fibronectin), and platelet RIIbâ3 (ligand
b
fibrinogen) are shown. n ) number of experiments.
Rvâ5, while leaving the activity against Rvâ6 (high) and
RIIbâ3 (low) essentially unchanged. Compound 12 is to

1048 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 5
Goodman et al.
Ta ble 2. Potent and Selective Rvâ6 Inhibitors in Cell Adhesion
Assaysa
IC50 (µM) for inhibition of cell attachment
UCLAP-3/FN
compd Rvâ6 (st. dev.) Rvâ5 (st. dev.)
UCLAP-3/VN
M21/VN+P1F6
Rvâ3 (st. dev.)
nb
1
2
37 (7.1)
10 (9)
1.3 (0.6)
0.4 (0.6)
n.d.c -
3.9 (2.5)
0.4 (0.3)
n.d.c -
2
4
2
2
2
2
4
2
2
2
2
2
2
2
2
3
n.d.c -
4
8.4 (7.9)
2.0 (0.1)
3.8 (0.4)
1.6 (1.1)
0.3 (0.13)
0.3 (0.03)
0.03 (0.02)
0.2 (0.1)
0.1 (0.1)
9.6 (10)
2.1 (0.4)
31 (9.9)
24 (14)
18 (6.4)
5 (2.3)
1.6 (0.1)
15 (7.1)
0.97 (0.5)
20 (7.8)
12 (3.3)
16 (7.8)
24 (1.4)
3 (0.57)
9.5 (5.0)
>50 -
5
6
7
8
9
2 (0.2)
10
11
12
13
14
15
27 (30)
20 (0.7)
26 (2.1)
43 (1.4)
1.2 (0.4)
21 (1.4)
0.09 (0.06)
2.1 (0.6)
6.9 (0.07)
>50 -
F igu r e 2. Inhibitor selectivity diagram. The selectivity of the
highly active and selective RGD-mimetics described in the text
is shown (black dots). Selectivity varies from compounds with
high selectivity (dark gray areas), to those with biselectivity
(gray/slightly gray areas), and to nonselective compounds with
selectivity lower than 10 (white area). The scale shows the
nanomolar IC50 for isolated receptor inhibition (for calculation
of the dot-position, see Appendix).
a
The concentration of compounds (PM), numbered as in text,
necessary to inhibit cell attachment to extracellular matrix coated
substrates to 50% of control, the IC50. The sensitivity of the Rvâ6
was measured by inhibiting UCLAP-3 cells on fibronectin (FN),
that of the Rvâ5 integrin by inhibiting UCLAP-3 cells on vitronec-
tin (VN), and that of the Rvâ3 integrin by inhibiting M21 cells on
vitronectin (VN), in the presence of monoclonal antibody P1F6.
n ) number of experiments. c n.d. ) not determined.
b
methyl group of the aminopyridin has no significant
influence on Rvâ5 activity, as shown by comparison with
compound 10.
inhibitory activities at the cellular level are some 2-3
orders of magnitude lower than at the level of the
receptor for Rvâ3 inhibitors,8 and similar loss of activity
was seen for the Rvâ5 and Rvâ6 receptor inhibitors at
the cell level. The compounds 6, 7, 13, and 15 showed
anomalous behavior at the cellular level, having a low
but measurable activity, while at the receptor level they
were inactive at the level of detection. The reason for
this is not clear, but may be related to nonlinearities in
ELISA-based receptor assays as used in this study.51
To visualize the different selectivities of the inhibitors
described here for the Rv-integrin receptor subtypes, we
have summarized the data in a graphical representation
(Figure 2). In this ternary selectivity diagram, each
cornersin analogy to a phase diagramsrepresents one
of the Rv-integrin subtypes Rvâ3, Rvâ5, or Rvâ6. The
selectivities are the ratios of the activities of one
compound for two integrin receptor subtypes, thus, each
compound has three selectivity values, which are plotted
into the diagram. Consequently, every compound is
represented by one spot in the triangle. Compounds
near the triangular corners (dark gray areas) are
monoselective (compounds 4, 5, 9-11), those near the
edges (gray areas) are biselective (compounds 1-3, 6-8,
and 12), and the compounds in the middle of the triangle
(slightly gray or white area) do not differentiate between
Rvâ3, Rvâ5, and Rvâ6 (compounds 13-15). The selec-
tivities of nearly all compounds toward the platelet
receptor, RIIbâ3, are >500 over Rvâ6, Rvâ5, and Rvâ3
(exceptions are compounds 5 and 15).
Discu ssion
Integrins convey a variety of signals from the extra-
cellular matrix to the cytoplasm. Specific subgroups of
the integrin family are differentially expressed or dif-
ferentially activated in pathological over normal condi-
tions. Integrins represent therapeutically interesting
target molecules, such as Rvâ3 and RIIbâ3. Rvâ5 and
Rvâ6 are also differentially expressed but their precise
functions and possible therapeutic relevance are as yet
unknown. These four integrins have a partially overlap-
ping ligand specificity, and a conserved amino acid
sequence, RGD, within the target ligands is at the core
of their recognition sites. Nevertheless, non-RGD pep-
tides have been discovered with high specificity and
inhibitory activity for RIIbâ3 and Rvâ6.17,26 Peptidic
inhibitors have several disadvantages as low molecular
weight inhibitors including their instability to serum
degradation, which can be circumvented by use of small
cyclic peptides,52 and poor pharmacokinetics. In this
study, we describe a series of peptidomimetic inhibitors
with high affinity on Rvâ6 integrin. Furthermore, we
report about the first specific, low molecular weight
Rvâ6 integrin receptor antagonist, compound 11. To-
gether with certain inhibitors that have adjunct activity
on the integrins Rvâ3 or Rvâ5, we show that a set of
integrin specific inhibitors can be designed around a
conserved (aza)-glycine phenyl-â-amino acid backbone.
Of the RGD-dependent integrins used here, Rvâ3 is
usually considered the most promiscuous in its sub-
To examine the relationship between activity at the
receptor level with that in a cellular environment, we
also tested the efficacy of the compounds for their ability
to perturb initial cell attachment mediated by Rvâ3,
Rvâ5, or Rvâ6 integrins. The results are summarized
in Table 2. In general, compounds found active at the
receptor level on Rvâ6 also inhibited Rvâ6-dependent
cell attachment of UCLAP-3 carcinoma cells to fibronec-
tin, and compounds more active at the cellular level
tended to rank more active at the receptor level, while
compounds inactive at the receptor level tended to be
weak inhibitors of cell attachment. For example, com-
pound 10 also inhibited attachment over these integrins
at the cellular level with a similar activity profile (IC50:
Rvâ6, 30 nM; Rvâ5, 27 µM; Rvâ3, 24 µM). The specific
Rvâ6 receptor blocker compound 11 also had only weak
activity on Rvâ5 on cells (IC50: Rvâ3, 3 µM; Rvâ6, 200
nM; Rvâ5, 20 µM). However, as previously noted, the

Rv-Integrin Inhibitors
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 5 1049
strate specificity, closely followed by RIIbâ3. Rvâ5 is a
relatively specific vitronectin receptor, while Rvâ6 binds
primarily to fibronectin with some binding activity
reported for tenascin-C. We predicted, therefore, that
it should be possible to make a series of inhibitors that
recognize one or more of the different Rv-integrin
receptor subtypes. The inhibitors we found fall into this
class. We found one peptidomimetic inhibitor, phenyl-
naphthalene containing compound 11, favoring integrin
Rvâ6 with subnanomolar activity. The activity of this
compound was nearly 2 orders of magnitude higher on
Rvâ6 over Rvâ3, with almost 5 orders of magnitude
smaller activity on both Rvâ5 and RIIbâ3. This com-
pound is to our knowledge the first specific, low molec-
ular weight Rvâ6 inhibitor that is described in the
literature. The selectivity is similar to the monospecific
peptidic inhibitors of Rvâ6 with the general structure
RT/GDLXXLX.26 The selectivity toward the Rvâ5 inte-
grin receptor can be obtained by incorporation of a bulky
function at the phenyl ring, compounds 6, 11, 12, and
improved with the peptidomimetic azacarba moiety,
compounds 6 and 7.
integrin receptor (IC50 0.04 nM). As well as providing a
starting point for further research on structure-activity
relationships, these molecules provide useful structural
information for the design of therapies targeting Rv-
series integrins.
Ack n ow led gm en t. We thank Dr. R. Dunker and Dr.
M. Frech (Merck, GBT) for fermentation and purifica-
tion of recombinant integrins, and Ms. C. Maddock, D.
Hahn, and J . Welge for excellent technical assistance.
We also thank A. Schro¨der, B. Cordes, and M. Kranawet-
ter for excellent technical assistance in synthesis and
analysis of the compounds. The authors thank Dr. S.
Glaser for his scientific contribution to the graphical
presentation of the biological data, and Dr. G. Zischin-
sky for access to unpublished studies on related struc-
tures. This work was supported by the ‘Deutsche
Forschungsgemeinschaft’ and the ‘Fonds der Chemis-
chen Industrie’.
Ap p en d ix
For the calculation of the position XB ) (x1, x2) of the
compounds (black dots) in the selectivity diagram, we
used the following equations
On the other hand, we found bispecific compounds
with interesting profiles. A similar profile to compound
11, with 1 order of magnitude less activity both on Rvâ6
and Rvâ3, was seen in the aza-dichlorophenyl compound
8 and its peptidic-linked analogue compound 9. An even
more active biselective Rvâ6/Rvâ3 inhibitor was found
in compound 12, with similarly low activity on Rvâ5 and
RIIbâ3. Furthermore, compound 14 represents a mixed,
triselective Rvâ6/Rvâ3/Rvâ5 inhibitor with nanomolar
activity for these integrin receptors.
E ) R(bx - AB)2 + â(bx - BB)2 + γ(bx - CB)2
∂E
∂x1
) 2R(x - a ) + 2â(x - b ) + 2γ(x - c )
0
0
)
1
1
1
1
1
1
!
∂E
∂x2
) 2R(x - a ) + 2â(x - b ) + 2γ(x - c )
)
2
2
2
2
2
2
!
At the cellular level, compared with the isolated
receptor assays, the specificity of the compounds was
usually retained, but with a loss of activity as measured
by IC50. We hypothesize that the loss of activity is due
to the need to inhibit a large fraction of the receptors
on a cell before attachment is lost. Cooperativity and
clustering of integrins at the cell surface can also
increase apparent local receptor affinity. The isolated
receptor assay, by contrast, involves the inhibition of
what is more nearly a simple bimolecular interaction,
and nonlinear effects can enhance apparent IC50 in such
assays.51
R‚a1 + â‚b1 + γ‚c1
R + â + γ
w x1 )
and
x2 )
R‚a2 + â‚b2 + γ‚c2
R + â + γ
with the coordinates for the integrin receptors AB, BB, CB
(cornerpoints) and the activity factors R, â, γ:
3
x
3 3
1
AB )
,
,
R )
(Rvâ3-integrin)
(Rvâ5-integrin)
(Rvâ6-integrin)
(
)
2
2
IC
x
50
3
Rvâ6, Rvâ3, and Rvâ5 integrins activate different
intracellular signaling pathways, so the ability to specif-
ically target desired combinations of these receptors
may be therapeutically relevant. Rvâ6 is expressed in
proliferating keratinocytes33 and in the invasive front
of certain carcinomas,30 and it may stimulate release
of MMP935,36 or help activate latent TGF-â1,37 while
Rvâ3 is up-regulated on neoangiogenic blood vessels,53
where it may promote cell migration54 and induce
apoptosis.55 Clearly, identifying the precise role of the
integrin in such diseases awaits the availability of
highly active, specific, and well-defined inhibitors,
inhibitors such as are described here.
1
BB ) (0, 0), â )
IC
x
50
3
1
x
CB ) ( 3, 0), γ )
IC
x
50
Su p p or tin g In for m a tion Ava ila ble: Additional experi-
mental details. This material is available free of charge via
the Internet at http://pubs.acs.org.
Refer en ces
(1) Hynes, R. O. Integrins: Versatility, Modulation, and Signaling
in Cell Adhesion. Cell 1992, 69, 11-25.
(2) Nagai, T.; Yamakawa, N.; Aota, S.; Yamada, S. S.; Akiyama, S.
K.; Olden, K.; Yamada, K. M. Monoclonal antibody characteriza-
tion of two distant sites required for function of the central cell-
binding domain of fibronectin in cell adhesion, cell migration,
and matrix assembly. J . Cell Biol. 1991, 114, 1295-1305.
(3) Hartman, G. D.; Egbertson, M. S.; Halczenko, W.; Laswell, W.
L.; Duggan, M. E.; Smith, R. L.; Naylor, A. M.; Manno, P. D.;
Lynch, R. J .; Zhang, G.; et al. Non-peptide fibrinogen receptor
antagonists. 1. Discovery and design of exosite inhibitors. J . Med.
Chem. 1992, 35, 4640-4642.
In conclusion, we describe here for the first time
highly active small molecule inhibitors of integrin Rvâ6.
These compounds have nanomolar IC50 on Rvâ6 at the
receptor level combined with nanomolar activity on Rvâ3
and/or on Rvâ5, while having an IC50 over 10 µM on
RIIbâ3. With compound 11, we have obtained a specific
inhibitor of Rvâ6 with subnanomolar activity on this

1050 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 5
Goodman et al.
(4) Bondinell, W. E.; Keenan, R. M.; Ali, F. E.; Allen, A. C.; Brosse,
C. W. D.; Eggleston, D. S.; Erhard, K. F.; Haltiwanger, R. C.;
Huffman, W. F.; Hwang, S.-M.; J akas, D. R.; Koster, P. F.; Ku,
T. W.; Lee, C. P.; Nichols, A. J .; Ross, S. T.; Samanen, J . M.;
Valocik, R. E.; Vasko-Moser, J . A.; Venslavsky, J . W.; Wong, A.
S.; Yuan, C.-K. Design of a Potent and Orally Active Nonpeptide
Platelet Fibrinogen Receptor (GPIIb/IIIa) Antagonist. Bioorg.
Med. Chem. Lett. 1994, 2, 897-908.
(5) Egbertson, M. S.; Chang, C. T.-C.; Duggan, M. E.; Gould, R. J .;
Halczenko, W.; Hartman, G. D.; Laswell, W. L.; J r., J . J . L.;
Lynch, R. J .; Manno, P.; Naylor, A. M.; Prugh, J . D.; Ramjit, D.
R.; Sitko, G. R.; Smith, R. S.; Turchi, L. M.; Zhang, G. Non-
Peptide Fibrinogen Receptor Antagonists. 2. Optimization of a
Tyrosine Template as a Mimic for Arg-Gly-Asp. J . Med. Chem.
1994, 37, 2537-2551.
(6) Samanen, J . M.; Ali, F. E.; Barton, L. S.; Bondinell, W. E.;
Burgess, J . L.; Callahan, J . F.; Calvo, R. R.; Chen, W.; Chen, L.;
Erhard, K.; Feuerstein, G.; Heys, R.; Hwang, S.-M.; J akas, D.
R.; Keenan, R. M.; Ku, T. W.; Kwon, C.; Lee, C.-P.; Miller, W.
H.; Newlander, K. A.; Nichols, A.; Parker, M.; Peishoff, C. E.;
Rhodes, G.; Ross, S.; Shu, A.; Simpson, R.; Takata, D.; Yellin,
T. O.; Uzsinskas, I.; Venslavsky, J . W.; Yuan, C.-K.; Huffman,
W. F. Potent, Selective, Orally Active 3-Oxo-1,4-benzodiazepine
GPIIb/IIIa Integrin Antagonists. J . Med. Chem. 1996, 39, 4867-
4870.
(7) Keenan, R. M.; Miller, W. H.; Kwon, C.; Ali, F. E.; Callahan, J .
F.; Calvo, R. R.; Hwang, S.-M.; Kopple, K. D.; Peishoff, C. E.;
Samanen, J . M.; Wong, A. S.; Yuan, C.-K.; Huffman, W. F.
Discovery of Potent Nonpeptide Vitronectin Receptor Antago-
nists. J . Med. Chem. 1997, 40, 2289-2292.
(19) Aumailley, M.; Gurrath, M.; Miller, G.; Calvete, J .; Timpl, R.;
Kessler, H. Arg-Gly-Asp constrained within cyclic pentapeptides.
Strong and selective inhibitors of cell adhesion to vitronectin
and laminin fragment P1. FEBS Lett. 1991, 291, 50-54.
(20) Gurrath, M.; Mu¨ller, G.; Kessler, H.; Aumailley, M.; Timpl, R.
Conformation/activity studies of rationally designed potent anti-
adhesive RGD peptides. Eur. J . Biochem. 1992, 210, 911-921.
(21) Haubner, R.; Gratias, R.; Diefenbach, B.; Goodman, S. L.;
J onczyk, A.; Kessler, H. Structural and Functional Aspects of
RGD-Containing Cyclic Pentapeptides as Highly Potent and
Selective Integrin Rvâ3 Antagonists. J . Am. Chem. Soc. 1996,
118, 7461-7472.
(22) Bach, A. C., II; Espina, J . R.; J ackson, S. A.; Stouten, P. F. W.;
Duke, J . L.; Mousa, S. A.; DeGrado, W. F. Type II′ to Type I
P3-Turn Swap Changes Specificity for Integrins. J . Am. Chem.
Soc. 1996, 118, 293-294.
(23) Burgess, K.; Lim, D. Synthesis and Solution Conformation of
Cyclo[RGDRGD]: A Cyclic Peptide with Selectivity for the Rvâ3
Receptor. J . Med. Chem. 1996, 39, 4520-4526.
(24) Tran, T.-A.; Mattern, R.-H.; Zhu, Q.; Goodman, M. A Novel RGD
Containing Dodecapeptidomimetic which exhibits Selective Bind-
ing to the Rvâ3 Receptor. Bioorg. Med. Chem. Lett. 1997, 7, 997-
1002.
(25) Dechantsreiter, M. A.; Planker, E.; Matha¨, B.; Lohof, E.; Ho¨lze-
mann, G.; J onczyk, A.; Kessler, H. N-Methylated Cyclic RGD
Peptides as Highly Active and Selective Rvâ3 Integrin Antago-
nists. J . Med Chem. 1999, 42, 3033-3040.
(26) Kraft, S.; Diefenbach, B.; Mehta, R.; J onczyk, A.; Luckenbach,
G. A.; Goodman, S. L. Definition of an Unexpected Ligand
Recognition Motif for Rvâ6 Integrin. J . Biol. Chem. 1999, 274,
1979-1985.
(8) Nicolaou, K. C.; Trujillo, J . I.; J andeleit, B.; Chibale, K.;
Rosenfeld, M.; Diefenbach, B.; Cheresh, D. A.; Goodman, S. L.
Design, Synthesis and Biological Evaluation of Nonpeptide
Integrin Antagonists. Bioorg. Med. Chem. 1998, 6, 1185-1208.
(9) Gibson, C.; Goodman, S. L.; Hahn, D.; Ho¨lzemann, G.; Kessler,
H. Novel Solid-Phase Synthesis of Azapeptides and Azapeptoides
via Fmoc-Strategy and Its Application in the Synthesis of RGD-
Mimetics. J . Org. Chem. 1999, 64, 7388-7394.
(10) Lark, M. W.; Stroup, G. B.; Hwang, S. M.; J ames, I. E.; Rieman,
D. J .; Drake, F. H.; Bradbeer, J . N.; Mathur, A.; Erhard, K. F.;
Newlander, K. A.; Ross, S. T.; Salyers, K. L.; Smith, B. R.; Miller,
W. H.; Huffman, W. F.; Gowen, M. Design and Characterization
of Orally Active Arg-Gly-Asp Peptidomimetic Vitronectin Recep-
tor Antagonist SB 265123 for Prevention of Bone Loss in
Osteoporosis. J . Pharmacol. Exp. Ther. 1999, 291, 612-617.
(11) Mousa, S. A.; Lorelli, W.; Mohamed, S.; Batt, D. G.; J adhav, P.
K.; Reilly, T. M. Rvâ3 Integrin Binding Affinity and Specificity
of SM256 in Various Species. J . Cardiovasc. Pharmacol. 1999,
33, 641-646.
(12) Gibson, C.; Sulyok, G. A. G.; Hahn, D.; Goodman, S. L.;
Ho¨lzemann, G.; Kessler, H. Nonpeptidic Rvâ3 Integrin Antago-
nist Libraries: On-Bead Screening and Mass Spectrometric
Identification without Tagging. Angew. Chem. Int. Ed. Engl.
2001, 40, 165-169.
(13) Sulyok, G. A. G.; Gibson, C.; Goodman, S. L.; Ho¨lzemann, G.;
Wiesner, M.; Kessler, H. Solid-Phase Synthesis of a Nonpeptide
RGD Mimetic Library: New Selective Rvâ3 Integrin Antago-
nists. J . Med. Chem. 2001, 44, 1938-1950.
(27) Koivunen, E.; Wang, B.; Ruoslahti, E. Isolation of a highly
specific ligand for the R5â1 integrin from a phage display library.
J . Cell Biol. 1994, 124, 373-380.
(28) Huang, X. Z.; Wu, J . F.; Cass, D.; Erle, D. J .; Corry, D.; Young,
S. G.; Farese, R. V., J r.; Sheppard, D. Inactivation of the integrin
â6 subunit gene reveals a role of epithelial integrins in regulat-
ing inflammation in the lung and skin. J . Cell Biol. 1996, 133,
921-928.
(29) Agrez, M.; Chen, A.; Cone, R. I.; Pytela, R.; Sheppard, D. The
Rvâ6 integrin promotes proliferation of colon carcinoma cells
through a unique region of the â6 cytoplasmic domain. J . Cell
Biol. 1994, 127, 547-556.
(30) Breuss, J . M.; Gallo, J .; DeLisser, H. M.; Klimanskaya, I. V.;
Folkesson, H. G.; Pittet, J . F.; Nishimura, S. L.; Aldape, K.;
Landers, D. V.; Carpenter, W.; et al. Expression of the â6
integrin subunit in development, neoplasia and tissue repair
suggests a role in epithelial remodeling. J . Cell Sci. 1995, 108,
2241-2251.
(31) J ones, J .; Watt, F. M.; Speight, P. M. J . Oral. Pathol. Med. 1997,
26, 63-68.
(32) Clark, R. A.; Ashcroft, G. S.; Spencer, M. J .; Larjava, H.;
Ferguson, M. W. Reepithelialization of normal human excisional
wounds is associated with a switch from Rvâ5 to Rvâ6 integrins.
Br. J . Dermatol. 1996, 135, 46-51.
(33) Haapasalmi, K.; Zhang, K.; Tonnesen, M.; Olerud, J .; Sheppard,
D.; Salo, T.; Kramer, R.; Clark, R. A.; Uitto, V. J .; Larjava, H.
Keratinocytes in human wounds express Rvâ6 integrin. J . Invest.
Dermatol. 1996, 106, 42-48.
(34) Huang, X.; Wu, J .; Spong, S.; Sheppard, D. The integrin Rvâ6 is
critical for keratinocyte migration on both its known ligand,
fibronectin, and on vitronectin. J . Cell Sci. 1998, 111, 2189-
2195.
(35) Niu, J .; Gu, X.; Turton, J .; Meldrum, C.; Howard, E. W.; Agrez,
M. Integrin-mediated signaling of gelatinase B secretion in colon
cancer cells. Biochem. Biophys. Res. Commun. 1998, 249, 287-
291.
(36) Agrez, M.; Gu, X.; Turton, J .; Meldrum, C.; Niu, J .; Antalis, T.;
Howard, E. W. The Rvâ6 integrin induces gelatinase B secretion
in colon cancer cells. Int. J . Cancer 1999, 81, 90-97.
(37) Munger, J . S.; Huang, X.; Kawakatsu, H.; Griffiths, M. J . D.;
Dalton, S. L.; Wu, J .; Pittet, J .-F.; Kaminski, N.; Garat, C.;
Matthay, M. A.; Rifkin, D. B.; Sheppard, D. The Integrin Rvâ6
Binds and Activates Latent TGF-â1: A Mechanism for Regulat-
ing Pulmonary Inflammation and Fibrosis. Cell 1999, 96, 319-
328.
(38) Naik, U. P.; Parise, L. V. Structure and function of platelet
aIIb33. Curr. Opin. Hematol. 1997, 4, 317-322.
(39) Varner, J . A.; Cheresh, D. A. Tumor angiogenesis and the role
of vascular cell integrin Rvâ3. Important Adv. Oncol. 1996, 69-
87.
(40) Choi, E. T.; Engel, L.; Callow, A. D.; Sun, S.; Trachtenberg, J .;
Santoro, S.; Ryan, U. S. Inhibition of neointimal hyperplasia by
blocking Rvâ3 integrin with a small peptide antagonist Gpen-
GRGDSPCA. J . Vasc. Surg. 1994, 19, 125-134.
(14) Ho¨lzemann, G. Recent advances in Rvâ3 integrin inhibitors.
IDrugs 2001, 4, 72-81.
(15) Lin, K. C.; Ateeq, H. S.; Hsiung, S. H.; Chong, L. T.; Zimmerman,
C. N.; Castro, A.; Lee, W. C.; Hammond, C. E.; Kalkunte, S.;
Chen, L. L.; Pepinsky, R. B.; Leone, D. R.; Sprague, A. G.;
Abraham, W. M.; Gill, A.; Lobb, R. R.; Adams, S. P. Selective,
tight-binding inhibitors of integrin ta4pl that inhibit allergic
airway responses. J . Med. Chem. 1999, 42, 920-934.
(16) Cue, D.; Southern, S. O.; Southern, P. J .; Prabhakar, J .; Lorelli,
W.; Smallheer, J . M.; Mousa, S. A.; Cleary, P. P. A nonpeptide
integrin antagonist can inhibit epithelial cell ingestion of Strep-
tococcus pyogenes by blocking formation of integrin R5â1-
fibronectin-M1 protein complexes. Proc. Natl. Acad. Sci. U.S.A.
2000, 97, 2858-2863.
(17) Kloczewiak, M.; Timmons, S.; Lukas, T. J .; Hawiger, J . Platelet
receptor recognition site on human fibrinogen. Synthesis and
structure-function relationship of peptides corresponding to the
carboxy-terminal segment of the γ-chain. Biochemistry 1984, 23,
1767-17.
(18) Peishoff, C. E.; Ali, F. E.; Bean, J . W.; Calvo, R.; D’Ambrosio, C.
A.; Eggleston, D. S.; Hwang, S. M.; Kline, T. P.; Koster, P. F.;
Nichols, A.; Powers, D.; Romoff, T.; Samanen, J . M.; Stadel, J .;
Vasko, J . A.; Kopple, K. D. Investigation of Conformational
Specificity at GPIIb/IIIa: Evaluation of Conformationally Con-
strained RGD Peptides. J . Med. Chem. 1992, 35, 3962-3969.

Rv-Integrin Inhibitors
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 5 1051
(41) Friedlander, M.; Brooks, P. C.; Shaffer, R. W.; Kincaid, C. M.;
Varner, J . A.; Cheresh, D. A. Definition of Two Angiogenic
Pathways by Distinct Rv Integrins. Science 1995, 270, 1500-
1502.
(42) Wermuth, J . Thesis, Technische Universita¨t Mu¨nchen, 1996.
(43) Schmitt, J . S. Thesis, Technische Universita¨t Mu¨nchen, 1998.
(44) Ho¨lzemann, G.; Goodman, S.; Jonczyk, A.; Sta¨hle, W. W00048996;
Merck Patent GmbH; 2000.
(50) Carron, C. P.; Meyer, D. M.; Pegg, J . A.; Engleman, V. W.;
Nickols, M. A.; Settle, S. L.; Westlin, W. F.; Ruminski, P. G.;
Nickols, G. A. A Peptidomimetic Antagonist of the Integrin Rvâ3
inhibits leyding Cell Tumor Growth and the Development of
Hypercalcemia of Malignancy. Cancer Res. 1998, 58, 1930-1935.
(51) Tangemann, K.; Engel, J . Demonstration of nonlinear detection
in ELISA resulting in up to 1000-fold too high affinities of
fibrinogen binding to integrin RIIbâ3. FEBS Lett. 1995, 358,
179-181.
(52) Kessler, H. Peptide Conformations, 19. Conformation and
Biological Activity of Cyclic Peptides. Angew. Chem. Int. Ed.
Engl. 1982, 21, 512-523.
(53) Brooks, P. C.; Clark, R. A. F.; Cheresh, D. A. Requirement of
Vascular integrin Rvâ3 for Angiogenesis. Science 1994, 264,
569-571.
(54) Filardo, E. J .; Brooks, P. C.; Deming, S. L.; Damsky, C.; Cheresh,
D. A. Requirement of the NPXY motif in the integrin â3 subunit
cytoplasmic tail for melanoma cell migration in vitro and in vivo.
J Cell Biol. 1995, 130, 441-450.
(55) Brooks, P. C.; Montgomery, A. M. P.; Rosenfeld, M.; Reisfeld, R.
A.; Hu, T.; Klier, G.; Cheresh, D. A. Integrin Rvâ3 Antagonists
Promote Tumor Regression by Inducing Apoptosis of Angiogenic
Blood Vessels. Cell 1994, 79, 1157-1164.
(45) Mehta, R. J .; Diefenbach, B.; Brown, A.; Cullen, E.; J onczyk,
A.; Gussow, D.; Luckenbach, G. A.; Goodman, S. L. Transmem-
brane-truncated Rvâ3 integrin retains high affinity for ligand
binding: evidence for an ‘inside-out’ suppressor? Biochem. J .
1998, 330, 861-869.
(46) Mehta, R. J . Unpublished results.
(47) Mitjans, F.; Sander, D.; Adan, J .; Sutter, A.; Martinez, J . M.;
J aggle, C. S.; Moyano, J . M.; Kreysch, H. G.; Piulats, J .;
Goodman, S. L. An anti-Rv-integrin antibody that blocks integrin
function inhibits the development of a human melanoma in nude
mice. J . Cell Sci. 1995, 108, 2825-2838.
(48) Haubner, R.; Finsinger, D.; Kessler, H. Stereoisomeric peptide
libraries and peptidomimetics for designing selective inhibitors
of the Rvâ3 integrin for a new cancer therapy. Angew. Chem.,
Int. Ed. Engl. 1997, 36, 1374-1389.
(49) Dechantsreiter, M. A.; Matha¨, B.; J onczyk, A.; Goodman, S. L.;
Kessler, H. Synthesis and Conformational Studies of N-methy-
lated Cyclic RGD-peptides. In Peptides 1996 (Proc. 24th Euro-
pean Peptide Symposium, September 8-13, 1996, Edinburgh,
Scotland); Ramage, R., Epton, R., Eds.; Mayflower Scientific
Ltd.: England, 1998, pp 329-330.
J M0102598