Synthesis, characterization and evaluation of pro-drugs of VLA-4 antagonists

Article abstract of DOI:10.1016/j.bmcl.2004.01.062

A pro-drug strategy to identify orally efficacious VLA-4 antagonists is described. Potential pro-drugs were evaluated for their physical chemical characteristics and in vitro properties, including solubility, stability, permeability and plasma stability.

Full text of DOI:10.1016/j.bmcl.2004.01.062

Bioorganic & Medicinal Chemistry Letters 14 (2004) 1651–1654
Synthesis, characterization and evaluation of pro-drugs of
VLA-4 antagonists
Donna M. Huryn,a,* Susan Ashwell,a,y Reinhardt Baudy,a Darren B. Dressen,b
William Gallaway,a Francine S. Grant,b,{ Andrei Konradi,b Robert W. Ley,c
Susan Petusky,a Michael A. Pleiss,b Dimitri Sarantakis,a Christopher M. Semko,b
Mary M. Sherman,a,x Cesar Tiod and Lu Zhanga
aChemical & Screening Sciences, Wyeth Research, CN 8000, Princeton, NJ 08543, USA
bElan Pharmaceuticals, 800 GatewayBoulevard, South San Francisco, CA 94080, USA
cDepartment of Pharmaceutical Sciences, Unit 50, Wyeth Research, 156 Fareham Road, Gosport, UK
dDepartment of Drug Safety & Metabolism, Wyeth Research, 500 Arcola Road, Collegeville, PA 19426, USA
Received 2 December 2003; accepted 22 January 2004
Abstract—A pro-drug strategy to identify orally efficacious VLA-4 antagonists is described. Potential pro-drugs were evaluated for
their physical chemical characteristics and in vitro properties, including solubility, stability, permeability and plasma stability.
Based on this characterization, promising compounds were identified for in vivo pharmacokinetic evaluation. These studies resulted
in the identification of a pro-drug that exhibited desirable blood levels in PK studies in several different species.
# 2004 Elsevier Ltd. All rights reserved.
The disruption of cell adhesion processes through
pharmacological intervention holds significant promise
for the treatment of inflammatory diseases. Of parti-
cular interest is the integrin, VLA-4, which was first
shown to mediate cell trafficking, and more recently, has
been shown to regulate the activation of a number of
cell types.1 Several reports describe the utility of VLA-4
antagonists (mAb’s, peptides, and small molecules)
which have shown efficacy in disease models of multiple
sclerosis,2,3 asthma,4 and rheumatoid arthritis,5 among
others.6 Unfortunately, many of the small molecules
described, while exquisitely potent, suffer from low oral
bioavailability and poor PK properties.7ꢀ9 This paper
describes our efforts to apply a pro-drug strategy to
improve the PK characteristics and ‘developability’ of a
potent VLA-4 antagonist.
A series of derivatives of proline-phenylalanine which
potently disrupt VLA-4 mediated adhesion has recently
been described by us10 and others.11 Of these deriva-
tives, our attention was focused on compound 1. This
molecule inhibited VLA-4 mediated adhesion to
12
HVEC’s with an IC50=6nM.
Keywords: Cell Adhesion; Physico-chemical characterization; Pro-
drugs; VLA-4 antagonist.
* Corresponding author. Tel.: +1-732-274-4025; fax: +1-732-274-
4129; e-mail: hurynd@wyeth.com
y
Current Address: AstraZeneca Pharmaceuticals LP, 35 Gatehouse
Drive, Waltham, MA 02451, USA.
Current Address: Icos Pharmaceuticals, 2023 120th Avenue NE,
Bellevue, WA 98005, USA.
Current Address: MPI Research, 54943 N. Main Street, Mattawan,
MI 49071, USA.
{
x
Figure 1. Parent VLA-4 antagonist 1 and ester pro-drugs 2–7.
0960-894X/$ - see front matter # 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.01.062

1652
D. M. Huryn et al. / Bioorg. Med. Chem. Lett. 14 (2004) 1651–1654
Table 1. Physico-chemical properties of VLA-4 antagonist 1 and its pro-drugs
d
Compd
SIF solubilitya
(mg/mL)
SGF solubilitya
(mg/mL)
SIBLM solubilityb
(mg/mL)
Caco-2c
(ꢁ10ꢀ6 cm/s)
Log Doct
(pH 4.5–6.5)
Rat plasma stabilitye
(t1/2 min)
1
2
3
4
5
6
7
>20
3.3
0.629
0.0949
0.000334
0.0130
2.83
>20
0.07
4.24
6.83
4.52
2.49
2.03
7.58
0.341
0.0455
Not detectable
Not detectable
2.4
1.18
2.27–2.00
2.45–2.47
3.43–3.42
3.10–3.17
1.41–1.42
3.21–3.34
8.08
10.9
53.4
2.82
13.1
1.6
0.158
0.00697
0.119
4.65
0.000506
Not detectable
0.0240
a According to USP 23, pg. 2051.
b TenHoor, C. N.; Bakatselou, V.; Dressman, J. J. Pharm. Res. 1991, 8, 1203.
c Ref 16; Metoprolol, used as a reference of a highly permeable compound (fa=90–100), exhibited Mean Papp=2ꢁ10–6 cm/s, which corresponds to
90–100% absorption; Mannitol, used as a paracellular reference marker with fa=15, exhibited a Mean Papp=0.3ꢁ10ꢀ6 cm/s.
d LogDoctanol was determined by HPLC peak area analysis of aqueous and octanol phases after 24 h shaking.
e Stock solutions of compound was added to rat plasma, and incubated at 37ꢁ for 24 h. Aliquots were removed at timepoints 0, 10, 20, 40, and 80
min and 3, 6, and 24 h. After extraction, centrifugation, evaporation and reconstitution with mobile phase, half-lives were determined by HPLC
peak area analysis of ester and acid 1.
To evaluate its potential to be developed as an oral
agent, the physico-chemical properties of 1 were mea-
sured. Two pKa’s13 were measured at 11.0 and 2.9. The
neutral species, which exists at low and high pH values,
exhibits logP values in the range of 3–4. 14 These results
indicate 1 will be amphoteric over the critical pH range
5–8, and predicts it will be poorly absorbed.15 The
amorphous compound exhibited good to high solubility
in a variety of buffer systems (>20 mg/mLat pH 4.5–9),
and in simulated physiological fluids (Table 1). Perme-
ability in Caco-2 assays16 was measured (Mean Papp=
0.07ꢁ10ꢀ6 cm/s) and predicted poor permeability and
<50% fraction absorbed in humans. The permeability
rate was less than that observed for the paracellular
reference marker Mannitol (Mean Papp=0.3ꢁ10ꢀ6 cm/
s; 15% fraction absorbed in man). According to BCS
System, the acceptable solubility but poor permeability of
compound 1, would result in a class III classification.
for these pro-drug esters. In all cases, Caco-2 data16
indicated that the pro-drug strategy significantly
improved permeability. Mean Papp values were all
>2ꢁ10ꢀ6 cm/s indicating the potential for high per-
meability (>90% absorption). When measuring log
Doctanol’s at pH’s between 4.5 and 6.5, all compounds
had values (ranging from 1.4–3.4) that fell within, or
near the critical range of >0 and <3 that has been
correlated with good oral absorption.15 Solubility of the
esters in a variety of simulated physiological fluids
(simulated intestinal fluid, simulated gastric fluid, simu-
lated bile lipid mixture) was measured. In comparison
to the acid 1, solubility diminished in all cases. In the
case of the pivaloyl 4, dimethylpropionyloxymethyl 5,
and 2-phenoxyethyl 7 esters, solubility was extremely
poor, or undetectable.
In order to enhance the permeability of this important
compound, a pro-drug strategy was initiated. While a
number of linkages were considered, preliminary data
suggested esters as most promising. A series of esters
(Fig. 1) were prepared including alkyl (ethyl 2, isopropyl
3, pivaloyl 4), and substituted alkyl esters (2,2-dime-
thylpropionyloxymethyl 5, methoxyethyoxyl ethyl 6, 2-
phenoxyethyl 7). The synthesis of these derivatives and the
parent acid is described in Scheme 1. 1-Methylpyrazole-
4-sulfonyl chloride and l-(5,5-dimethyl)thiaproline were
treated with K2CO3 in water to provide the sulfonamide
8.17 Coupling of 8 with the appropriate tyrosine ester
(R=ethyl, isopropyl, ethoxyethoxylmethyl and phenox-
yethyl, t-butyl) afforded the dipeptides of general struc-
ture 9.18 Treatment of 9 with dimethyl carbamoyl
chloride and K2CO3 generated esters 2, 3, 6, 7 or 10. To
prepare the acid 1, 10 was treated with trifluroacetic acid.
The acid was used as starting material to generate ester
5 via treatment of chlroromethyl pivalate and K2CO3.19
Finally, transesterification of isopropyl ester 3 with piva-
lic alcohol and titaniumisopropoxide afforded ester 4.
Scheme 1. Synthesis of acid 1 and ester pro-drugs 2–7. (i) 1-methyl-
pyrazole-4-sulfonyl chloride, potassium carbonate; (ii) l-tyrosine ester
hydrochloride, N-methyl morpholine, 1-hydroxybenzotriazole, 1-(3-
dimethylaminopropyl)-3-ethylcarbodiimide; (iii) dimethyl carbamyl
chloride, potassium carbonate; (iv) trifluoracetic acid; (v) potassium
iodide, chloromethylpivalate, potassium carbonate; (vi) neopentyl
alcohol, titanium(IV) isopropoxide.
To evaluate their suitability, the esters were evaluated
for their physical properties—particularly solubility in
simulated physiological fluids, Caco-2 permeability, log
D and plasma stability. Table 1 lists the data generated

D. M. Huryn et al. / Bioorg. Med. Chem. Lett. 14 (2004) 1651–1654
1653
To determine that the esters could indeed be cleaved in
References and notes
plasma to generate active parent, rat plasma stability
studies were carried out. As expected, simple esters were
rapidly cleaved with half lives in the 2–10 min range.
Pivaloyl ester 4 exhibited significant stability (t1/2=53.4
min). The prolonged stability of this compound in rat
plasma was predicted to correlate to lower levels of
active acid in plasma, and even lower levels in higher
species, and therefore the compound was removed from
further consideration.
1. For a recent review see: Yusuf-Makagiansar, H.; Ander-
son, M. E.; Yakovleva, T. V.; Murray, J. S.; Siahaan, T. J.
Med. Res. Rev. 2002, 22, 146.
2. Yednock, T. A.; Cannon, C.; Fritz, L. C.; Sanchez-
Madrid, F.; Steinman, L.; Karin, N. Nature 1992, 356, 63.
3. Piraino, P. S.; Yednock, T. A.; Freedman, S. B.; Messer-
smith, E. K.; Pleiss, M. A.; Vandevert, C.; Thorsett, E. D.;
Karlik, S. J. Journal of Neuroimmunology 2002, 131, 147.
4. Lin, K.; Ateeq, H. S.; Hsiung, S. H.; Chong, L. T.; Zim-
merman, 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. Journal of Medicinal Chemistry 1999, 42,
920.
5. Seiffge, D. J. Rheumatology 1996, 23, 2086.
6. Tilley, J. W.; Sidduri, A. Drugs of the Future 2001, 26,
985.
7. Gutteridge, C. E.; de Laszlo, S. E.; Kamenecka, T. M.;
McCauley, E.; van Riper, G.; Mumford, R. A.; Kidambi,
U.; Egger, L. A.; Tong, S.; Hagmann, W. K. Bioorganic &
Medicinal Chemistry Letters 2003, 13, 885.
Compounds 5 and 7 were de-prioritized due to lack of
solubility under conditions that mimic the GI tract.
Based on a suggested minimum thermodynamic solubility
of 50 mg/mL15b esters 2, 3, and 6 (aqueous solubi-
lity=0.473 mg/mL, 0.069 mg/mL and 5.38 mg/mL,
respectively) were predicted to be the most promising for
oral delivery, however, 6 existed as an amorphous solid,
and it too was dropped from further consideration.
Due to their acceptable solubility, good Caco-2 perme-
ability, short half-life in rat plasma, and optimal logD,
the ethyl (2) and isopropyl (3) esters were chosen for
further evaluation. Both compounds were dosed orally
at 30 mg/kg to male rats and mean blood concentra-
tions of parent and active acid were measured.20 In the
case of 2, significant concentrations of acid parent were
detected at time points of 5 min though 24 h, with a
peak level of 0.31 mg/mLat 15 min. No ester was
detected. When the iso-propyl ester 3 was dosed, both
ester and acid were detected at all times points (5 min to
24 h). The ester was quantitated between time points of
5 min and 8 h, albeit at consistently low levels (0.001–
0.004 mg/mL). The acid was quantitated at acceptable
levels (0.014–0.382 mg/mL) from 5 min (0.2 mg/mL)
through 4 h (0.014 mg/mL), with a peak level of 0.38 mg/
mLat 15 min. Similar results in another animal species
(results not shown) indicate that the pro-drug optimi-
zation was not species specific. These concentrations of
active parent are significantly higher than the IC50 for
VLA-4 inhibition, and were therefore predicted to result
in vivo efficacy in disease models mediated by VLA-4.
The translation of these blood levels into oral efficacy in
animal models of disease of iso-propyl and ethyl ester
pro-drugs will be reported elsewhere.
8. 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.; Wilson, 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. Bioorganic & Medicinal
Chemistry Letters 2001, 11, 2709.
9. Chen, L.; Trilles, R.; Miklowski, D.; Huang, T.-N.; Fry,
D.; Campbell, R.; Rowan, K.; Schwinge, V.; Tilley, J. W.
Bioorganic & Medicinal Chemistry Letters 2002, 12, 1679.
10. (a) Kubrak, D.; Cannon, C.; Giberson, J.; Leeson, P. D.;
Mann, C. W.; Vandevert, C.; Yang, C.; Kreft, A.
Abstracts of Papers—American Chemical Society (2000),
220th MEDI-139. (b) Ashwell, S.; Banker, A. L.; Bicksler,
J. J.; Dressen, D. B.; Cannon, C.; Giberson, J.; Grant, F.
F.; Konradi, A. W.; Leeson, P. D.; Lombardo, L. J.;
Pleiss, M. A.; Sarantakis, D.; Thompson, T.; Thorsett, E.
D.; Vandevert, C.; Yang, C. Abstracts of Papers—Amer-
ican Chemical Society (2000), 220th MEDI-138 (c) Sar-
antakis, D.; Bicksler, J. J.; Cannon, C.; Dressen, D. B.;
Giberson, J.; Grant, F. F.; Konradi, A. W.; Kreft, A.
Leeson, P. D.; Lombardo, L. J.; Mann, C. W.; Pleiss, M.
A.; Sabalski, J. E.; Thorsett, E. D.; Vandevert, C.; Yang,
C. Abstracts of Papers—American Chemical Society
(2000), 220th MEDI-137 (d) Sarantakis, D.; Baudy, R. B.;
Bicksler, J. J.; Cannon, C.; Dressen, D. B.; Giberson, J.;
Grant, F. F.; Konradi, A. W.; Kreft, A.; Kubrak, D.;
Leeson, P. D.; Lombardo, L. J.; Mann, C. W.; Pleiss, M.
A.; Sze, J.; Thorsett, E. D.; Vandevert, C.; Yang, C.
Abstracts of Papers—American Chemical Society (2000),
220th MEDI-136 (e) Dappen, M. S.; Dressen, D. B.;
Grant, F. S.; Konradi, A. W.; Pleiss, M. A.; Semko, C.
M.; Thorsett, E. D.; Freedman, S. B.; Holsztynska, E. J.;
Quinn, K. P.; Ashwell, S.; Banker, A. L.; Baudy, R. B.;
Bicksler, J. J.; Giberson, J.; Leeson, P. D.; Lombardo, L.
J.; Sarantakis, D. Abstracts of Papers—American Che-
mical Society (2000), 220th MEDI-135. (f) Thorsett, E.
D.; Dappen, M. S.; Dressen, D. B.; Ellingboe, J. W.;
Grant, F. S.; Jacobson, M.; Kincaid, S. L.; Konradi, A.
W.; Kreft, A.; Lombardo, L. J.; Mann, C. W.; Nunn, D.;
Pleiss, M. A.; Sabalski, J. E.; Sacchi, K.; Sarantakis, D.;
Semko, C. M. Abstracts of Papers—American Chemical
Society (2000), 220th MEDI-134 (g) Semko, C. M.; Dres-
sen, D. B.; Grant, F. S.; Konradi, A.W.; Pleiss, M. A.;
In summary, a pro-drug strategy was employed to
impart desirable physico-chemical and PK properties to a
potent VLA-4 antagonist. By evaluation of their physical
properties, pro-drugs with the most promising properties
(solubility, permeability, logD, plasma stability) could
be prioritized for further testing. These predictions were
confirmed in in vivo PK studies. The ability to rapidly
evaluate a series of pro-drugs using in vitro assays can
be used to identify the most promising compounds on
which to perform more elaborate in vivo assays.
Acknowledgements
The authors would like to thank Jeff Kennedy, Ted
Yednock, and Steve Freedman for scientific discussion,
and Uday Jain for editorial contributions.

1654
D. M. Huryn et al. / Bioorg. Med. Chem. Lett. 14 (2004) 1651–1654
Thorsett, E. D.; Freedman, S. B.; Holsztynska, E J.;
Quinn, K. P.; Yednock, T. Abstracts of Papers—Amer-
ican Chemical Society (2000), 220th MEDI-133.
16. (a) Artursson, P. J. Pharm. Sci. 1990, 79, 476. (b) Arturs-
son, P.; Karlsson, J. Biochem Biophys Res. Comm. 1991,
175, 880.
11. Tilley, J. W. Expert Opin. Ther. Patents 2002, 12, 991.
12. Results of T. Yednock
17. Cupps, T. L.; Boutin, R. H.; Rapoport, H. J. Org. Chem.
1985, 50, 3972.
13. pKa’s were determined potentiometrically over pH 2-12
using a GlpKa Titrator, manufactured by Sirius Analy-
tical Instruments.
14. LogP values were determined potentiometrically using
octanol as the partitioning solvent.
15. (a) Navia, M. A.; Chaturvedi, P. R. DDT 1996, 1, 179. (b)
Lipinski, C. A.; Lombardo, F.; Dominy, B. W.; Feeney,
P. J. Adv. Drug Delivery Reviews 1997, 23, 3.
18. Windridge, G. C.; Jorgensen, E. C. J. Am. Chem. Soc.
1971, 93, 6318.
19. Hu, M. W.; Singh, P.; Ullman, E. F. J. Org. Chem. 1980,
45, 1711.
20. Each compound was dosed in 45% propyleneglycol: 45%
PEG400: 10% Tween 80 to 6 male rats; Mean values were
determined using an average of 3 animals; the limit of
quantitation was 0.5 ng/mL