Substituted tetrahydrofuroyl-1-phenylalanine derivatives as potent and specific VLA-4 antagonists

Article abstract of DOI:10.1016/S0960-894X(02)00211-1

A series of substituted tetrahydrofuroyl-1-phenylalanine derivatives was prepared and evaluated as VLA-4 antagonists. Substitution of the α carbon of the tetrahydrofuran with aryl groups increased the specificity for VLA-4 versus α₄β₇ while amide substitution increased the potency of the series without increasing the specificity. Substitution of the β carbon of the tetrahydrofuran with keto or amino groups slightly improved the specificity for VLA-4 versus α₄β₇ but with a significant loss in binding affinity for VLA-4.

Full text of DOI:10.1016/S0960-894X(02)00211-1

Bioorganic & Medicinal Chemistry Letters 12 (2002) 1501–1505
Substituted Tetrahydrofuroyl-1-phenylalanine Derivatives as
Potent and Specific VLA-4 Antagonists
George A. Doherty,a,* Ginger X. Yang,a Edite Borges,a Linda L. Chang,a
Malcolm MacCoss,a Sharon Tong,a Usha Kidambi,b Linda A. Egger,b
Ermenegilda McCauley,b Gail Van Riper,b Richard A. Mumford,b John A. Schmidtb
and William K. Hagmanna
aDepartment of Medicinal Chemistry, Merck Research Laboratories, Rahway, NJ 07065, USA
bDepartment of Immunology & Rheumatology Research, Merck Research Laboratories, Rahway, NJ 07065, USA
Received 11 December 2001; accepted 8 March 2002
Abstract—A series of substituted tetrahydrofuroyl-1-phenylalanine derivatives was prepared and evaluated as VLA-4 antagonists.
Substitution of the a carbon of the tetrahydrofuran with aryl groups increased the specificity for VLA-4 versus a4b7 while amide
substitution increased the potency of the series without increasing the specificity. Substitution of the b carbon of the tetra-
hydrofuran with keto or amino groups slightly improved the specificity for VLA-4 versus a4b7 but with a significant loss in binding
affinity for VLA-4. # 2002 Elsevier Science Ltd. All rights reserved.
VLA-4 (a4b1; CD49d/CD29; ‘very late antigen-4’) is a
key cell surface integrin present on leukocytes except
neutrophils and platelets that binds vascular cell adhe-
sion molecule-1 (VCAM-1) on endothelial cell surfaces
and leads to leukocyte infiltration to extravascular tis-
sue. Antibodies against VLA-4 have been shown to
block leukocyte infiltration and prevent tissue damage
in inflammatory disease models of asthma,1 multiple
sclerosis,2 rheumatoid arthritis (RA),3 and inflamma-
tory bowl disease (IBD).4 Orally active small molecule
inhibitors of VLA-4 might therefore serve as useful
agents in the treatment of these diseases.
integrin a4b7 (7.1 nM). Although this integrin may itself
have therapeutic value, a more selective compound was
desired.6 In an effort to further increase the potency of
this series, reduce plasma protein binding and improve
the specificity for VLA-4, a systematic study of sub-
stitution at the a- and b-carbons of the tetrahydro-
furan ring was undertaken.
In a previous communication,5 we disclosed the identi-
fication of 1 and 2 as potent VLA-4 antagonists. In the
rhesus monkey, compound 2 exhibited excellent oral
bioavailability and low plasma clearance (F=51%;
Clp=2.9 mL/kg/min). When the binding assay was run
in the presence of 5% plasma, a 10-fold shift in the IC50
was observed (0.35 nM vs 3.61 nM) indicating a high
level of protein binding, a factor which may be respon-
sible for the lower plasma clearance number. Com-
pound 2 also showed good affinity for the closely related
The alkyl substituted tetrahydrofuroic acids (3–5) were
prepared via simple alkylation of tetrahydrofuroic acid
methyl ester. The alkanol and alkyl amine substituted
derivatives were prepared as shown in Scheme 1. The
lithium enolate of tetrahydrofuroic benzyl ester was
alkylated with the tert-butyl dimethyl silyl protected
iodoalkanols which provided the alkylated products in
modest yield. After removal of the benzyl ester, PyBOP
coupling with the dimethoxybiphenylalanine tert-butyl
ester7 produced the coupled products as mixtures of
diastereomers. Tetrabutylammonium fluoride deprotec-
tion of the silyl ether gave the alcohol, which was
*Corresponding author. Tel.: +1-732-594-3686; fax +1-732-594-
5966; e-mail: george_doherty@merck.com
0960-894X/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(02)00211-1

1502
G. A. -Doherty et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1501–1505
Scheme 1. (a) LDA, HMPA, I(CH)nOTBS; (b) Pd/C, H2, NEt3, MeOH; (c) PyBOP, DIPEA, CH2Cl2; (d) TBAF, THF; (e) TFA, CH2Cl2; (f)
MeOTf; (g) oxalyl chloride, DMSO, NEt3; (h) HNMe2, NaBH(OAc)3, CH2Cl2.
converted to the methyl ether and also to the dimethyl-
amino derivatives. The tert-butyl ester was cleaved and
the diastereomeric products separated via prep TLC.
The more polar diastereomer proved to be the more
active isomer. The stereochemistry of the more active
isomer was assigned as (R) in analogy to 1, which was
rigorously proven by X-ray crystallography.8
decrease in activity in the presence of 5% plasma pro-
tein (0.35 nM vs 0.61 nM). This reduction in protein
binding also resulted in a significant loss in rat
pharmacokinetics (F=1.9%; Clp=81.1 mL/kg/min).
The aryl substituted tetrahydrofurans 15–23 were
prepared from the reaction of potassium cyanide with
4-chlorobutyrophenones followed by hydrolysis the
resulting cyanides (Scheme 2). Coupling of the acids
with biphenylalanine ester produced the coupled pro-
ducts as mixtures of diastereomers which were separated
by prep TLC after hydrolysis of the ester.
Alkyl substitution (3–5) did little to increase the potency
or specificity of the series (Table 1). Dimethyl amine
(12–14), methyloxy (9–11), and hydroxy (6–9) were
appended to the a-position of the tetrahydrofuran with
a two, three and four carbon spacer. The three carbon
spacer was optimal with activity greatly tailing off as the
chain length was increased. The ethers were more potent
than the alcohols and amines, but were only equipotent
to methyl, with no improvement in specificity. The
introduction of the ether functionality did reduce
plasma protein binding, with 10 showing only a 2-fold
The phenyl substituted derivative 15 (Table 2) showed
equal potency to the methyl derivative 1, but was start-
ing to show an increased specificity for a4b1 (26ꢀ vs
12ꢀ). Substitution of the meta or para position
increased the specificity even more. The trifluoromethyl
group proved to be the most effective substituent for
Table 1. Inhibition of VLA-4a and a4b7b by alkyl substituted tetrahydrofuroyl-1-phenylalanine derivatives
Compd
R
IC50 (nM)
Ratio
b7/b1
21
Compd
R
IC50 (nM)
Ratio
b7/b1
40
a
a,c
b
a
a,c
b
a4b1
a4b1
a4b7
a4b1
a4b1
a4b7
2
3
4
5
6
7
8
CH3
0.34
0.48
0.7
3.61
1.46
7.46
4.02
7.59
3.36
870
7.09
9
10
11
12
13
14
0.7
1.99
0.61
1.27
19.5
—
16.1
7.8
13.8
29
9
0.35
0.62
22
35
5
6.1
22.3
16.3
35.3
22.0
80.0
67.0
226.5
1.15
1.56
2.12
19
10
17
1
15.5
9.18
16.3
7
22.9
14
255
245
aVCAM-Ig.
bMAdCAM-Ig.
cPlus 5% human plasma.

G. A. -Doherty et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1501–1505
1503
increasing specificity, but due to its large molecular
weight and high hydrophobicity, chlorine was chosen as
the optimal substituent.
Substitution of the a-carbon of the tetrahydrofuran ring
with acyl groups provided a significant increase in
potency and a significant decrease in protein binding,
but without a significant gain in specificity (Table 4).
Amide substitution 31–33 proved to be better than ester
30 or ketone 35, with the best amide being the morpho-
line amide 31. The homologue compound 34 was sig-
nificantly less potent. Unfortunately, the rat pKfor 31
(Clp=87 mL/min/kg, F=8.6%) was quite poor and an
increase in specificity was not seen (a4b7/a4b1=16ꢀ). In
an attempt to merge the potency increase and protein
binding decrease of the amide series with the good pK
profile and specificity of the aryl substituted series,
thiazole and benzoxazole substitution were examined.
These were synthesized from acid 29 via standard het-
erocyclic chemistry.9 Unfortunately, neither the ben-
zoxazole 36 nor the thiazole 37 showed the potency
enhancements seen in the amide series, although the
benzoxazole exhibited remarkable specificity (a4b7/
a4b1=897ꢀ).
The dimethoxybiphenylalanine derivative 24 (Table 3)
showed the predicted 10-fold increase in potency (a4b1
IC50=4.09 nM vs 0.55 nM) compared with 20 and
maintained good specificity (a4b7/a4b1=95ꢀ). In the rat
this compound exhibited moderate clearance and excel-
lent bioavailability (Clp=19.3 mL/min/kg, F=100%).
Unfortunately, the good pharmacokinetics were
accompanied by a highly level of protein binding. As
seen with 2, a 9-fold shift in activity was seen when the
binding assay was run with 5% plasma (0.55 nM vs
4.75 nM). To address the issue of protein binding, polar
groups were appended onto the phenyl ring. Nitro sub-
stituents were well tolerated, but interestingly, its
reduction to the corresponding amine (27) led to a sig-
nificant loss in potency without any improvement in
protein binding. The dimethoxy analogue 28 was also
synthesized with the anticipation that these substituents
would reduce protein binding, but unfortunately this ana-
logue showed a significant reduction in binding affinity.
Substitution of the b carbon of the tetrahydrofuran ring
was briefly explored. As outlined in Scheme 4, the first
step
toward
3-keto,
5-alkyl-substitued
tetra-
The acyl substituted derivatives 30–35 were prepared
next using a similar protocol to the alkyl series (Scheme 3).
Tetrahydro-2-furoic acid methyl ester was acylated with
benzyl chloroformate followed by hydrogenolysis to the
acid 29. After coupling to the amino acid, the methyl
ester was hydrolyzed and converted to the amides.
hydrofuranoyls was the rhodium acetate insertion of the
acyclic diazo compound. It was found that in order to
b
Table 3. Inhibition of VLA-4a and a4b7 by aryl substituted tetra-
hydrofuroyl-1-phenylalanine derivatives
Scheme 2. (a) KCN, MeOH; (b) KOH, ethylene glycol, 200 ꢁC.
Compd
R
IC50 (nM)
Ratio
b7/b1
Table 2. Inhibition of VLA-4a and a4b7 by aryl substituted tetra-
b
a
a,c
b
a4b1
a4b1
a4b7
hydrofuroyl-1-phenylalanine derivatives
24
25
26
27
28
4-Cl
3-NO2,4-Cl
3-NO2
3-NH2
3,5-Dimethoxy
0.55
0.47
0.23
0.92
1.48
4.75
2.41
0.95
5.63
5.77
52.2
45.6
19.6
26.6
—
95
97
85
29
—
aVCAM-Ig.
bMAdCAM-Ig.
cPlus 5% human plasma.
Compd
R
IC50 (nM)
Ratio
b7/b1
a
a,c
b
a4b1
a4b1
a4b7
15
16
17
18
19
20
21
22
23
H
4-t-Bu
4-F
4-CF3
4-Br
4-Cl
3-CF3
3,4-Me
3,5-CF3
2.2
1.8
4.3
5.31
3.62
4.09
4.65
6.18
4.62
—
29
48
120
53
46
290
34
1105
56.4
99.5
45.7
26
55
11
95
64
67
70
60
504
232
275
327
374
499
108
aVCAM-Ig.
Scheme 3. (a) LDA, HMPA, ClCO2Bn; (b) Pd/C, H2, MeOH; (c)
PyBOP, DIPEA, amino acid, CH2Cl2; (d) NaOH, MeOH; (e) PyBOP,
ROH or R2NH; (f) TFA, CH2Cl2.
bMAdCAM-Ig.
cPlus 5% human plasma.

1504
G. A. -Doherty et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1501–1505
Table 4. Inhibition of VLA-4a and a4b7b by acyl and heteroaryl substituted tetrahydrofuroyl-1-phenylalanine derivatives
Compd
R
R0
IC50 (nM)
Ratio
b7/b1
Compd
R
R0
IC50 (nM)
Ratio
b7/b1
a
a,c
b
a
a,c
b
a4b1
a4b1
a4b7
a4b1
a4b1
a4b7
1
H
H
Me
3.9
60.6
19
45.3
12
13
2
OCH3
OCH3
Me
0.34
0.67
3.6
—
7.09
21
22
30
4.06
0.54
51.1
35
11.24
31
32
33
34
H
H
H
H
1.0
1.0
4.42
—
8.49
16
39
17
11
36
37
OCH3
OCH3
0.42
0.31
2.4
—
376.9
14.5
897
47
0.59
23.2
1.33
22.95
11.0
23.15
aVCAM-Ig.
bMAdCAM-Ig.
cPlus 5% human plasma.
Table 5. Inhibition of VLA-4a and a4b7 binding by 3-substituted
b
tetrahydrofuroyl-1-phenylalanine derivatives
Compd
R
IC50 (nM)
Ratio
b7/b1
a
a,c
b
a4b1
a4b1
a4b7
2
0.3
3.6
1.32
—
7.1
24
35
38
39
40
0.7
2.6
3.4
24.5
Scheme 4. (a) {Rh(OAc)2}2, benzene, 1,2-DCE, 80–100 ꢁC; (b)
K2CO3, MeI, MeCN, reflux; (c) TFA, DCM, rt; (d) PyBOP, DIEA,
DCM, rt; (e) p-OMe-benzyl amine, NaBH(OAc)3, acetic acid, DCM,
rt; (f) Pd(OH)2, HCOꢂ2 (NH4)+, MeOH, reflux.
149
57
further derivatize the keto compound, one must meth-
ylate the 2-position of the tetrahydrofuran. This was
accomplished by treating the insertion reaction product
with anhydrous potassium carbonate and iodomethane
in acetonitrile. After hydrolysis of the tert-butyl ester
and subsequent amide coupling, the desired 3-keto
compound 38 was obtained. The 3-amino derivatives 39
14.9
733
225
PMB, p-methoxybenzyl.
aVCAM-Ig.
bMAdCAM-Ig.
cPlus 5% human plasma.

G. A. -Doherty et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1501–1505
1505
and 40 were also explored. These were prepared by
reductive amination of the 38 followed by hydro-
genolysis to remove the p-methoxybenzyl group. Bind-
ing data for VLA-4 and a4b7 are shown in Table 5.
Substitution at the b-carbon resulted in some improve-
ment in specificity. However, the improved specificity
was achieved at the expense of decreased potency in
VLA-4 binding affinity.
Adams, S. P. J. Med. Chem. 1999, 42, 920. (b) Lin, K.-C.;
Castro, A. C. Curr. Opin. Chem. Biol. 1998, 2, 453.
2. (a) Tubridy, N.; Behan, P. O.; Capildeo, R.; Chaudhuri, A.;
Forbes, R.; Hawkins, C. P.; Hughes, R. A. C.; Palace, J.;
Sharrack, B.; Swingler, R.; Young, C.; Moseley, I. F.; Mac-
Manus, D. G.; Donoghue, S.; Miller, D. H. Neurology 1999,
53, 466. (b) Keszthelyi, E.; Karlik, S.; Hyduk, S.; Rice, A.;
Gordon, G.; Yednock, T.; Horner, H. Neurology 1996, 47,
1053.
3. Seiffge, D. J. Rheumatol. 1996, 23, 2086.
In summary, a-substituted tetrahydrofuroyl-1-phenyla-
lanine derivatives were examined as VLA-4 antagonists.
The introduction of alkyl groups on the tetrahydrofuran
ring did not effect either the potency or the specificity.
Amide substitution produced significantly more potent
inhibitors of VLA-4 but with a deterioration of clear-
ance and oral bioavailability. The most promising com-
pounds were the aryl and heteroaryl substituted
derivatives. These derivatives maintained the potency
seen in the lead compound, but with improved specificity
and good pharmacokinetics. Substitution at the b-carbon
resulted in some improvement in specificity, but with an
unacceptable decrease in binding affinity for VLA-4.
4. Podolsky, D. K.; Lobb, R.; King, N.; Benjamin, C. D.;
Pepinsky, B.; Sehgal, P.; deBaumont, M. J. Clin. Invest. 1993,
92, 372.
5. Yang, G. X.; Chang, L. L.; Truong, Q.; Doherty, G. A.;
Magriotis, P. A.; de Laszlo, S. E.; Li, B.; MacCross, M.;
Kidambi, U.; Egger, L. A.; McCauley, E.; Van Riper, G.;
Mumford, R. A.; Schmidt, J. A.; Hagmann, W. K. Bioorg.
Med. Chem. Lett. 2002, 12, 1497.
6. Lin, L. S.; Lanza, T.; McCauley, E.; Van Riper, G.;
Kidambi, U.; Cao, J.; Egger, L. A.; Mumford, R.; Schmidt,
J. A.; MacCoss, M.; Hagmann, W. K. Bioorg. Med. Chem.
Lett. 2002, 12, 133.
7. Hagmann, W. K.; Durette, P. L.; Lanza, T.; Kevin, N. J.;
de Laszlo, S. E.; Kopka, I. E.; Young, D.; Magriotis, P. A.; Li,
B.; Lin, L. S.; Yang, G.; Kamenecka, T.; Chang, L. L.; Wil-
son, J.; MacCoss, M.; Mills, S. G.; Van Riper, G.; McCauley,
E.; Egger, L. A.; Kidambi, U.; Lyons, K.; Vincent, S.; Stearns,
R.; Colletti, A.; Teffera, J.; Tong, S.; Fenyk-Melody, J.;
Owens, K.; Levorse, D.; Kim, P.; Schmidt, J. A.; Mumford,
R. A. Bioorg. Med. Chem. Lett. 2001, 11, 2709.
References and Notes
1. (a) 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. A.;
8. Wollweber, H.; Horstmann, H.; Meng, K. Eur. J. Med.
Chem.-Chimica Ther. 1976, 11, 159.
9. Katritzky, A. R., Ed. Comprehensive Heterocyclic Chem-
istry II. Elsevier Science: New York, 1996; Vol. 3.