N-Hydroxyformamide peptidomimetics as TACE/matrix metalloprotease inhibitors: Oral activity via P1′ isobutyl substitution

Article abstract of DOI:10.1016/S0960-894X(01)00377-8

N-Hydroxyformamide-class metalloprotease inhibitors were designed and synthesized, which have potent broad-spectrum activity versus matrix metalloproteases and TNF-α converting enzyme (TACE). Compound 13c possesses good oral and intravenous pharmacokinetics in the rat and dog.

Full text of DOI:10.1016/S0960-894X(01)00377-8

Bioorganic & Medicinal Chemistry Letters 11 (2001) 2147–2151
N-Hydroxyformamide Peptidomimetics as
TACE/Matrix Metalloprotease Inhibitors:
Oral Activity via P10 Isobutyl Substitution
David L. Musso,* Marc W. Andersen,y Robert C. Andrews,{ Richard Austin,x
Elizabeth J. Beaudet, J. David Becherer, Dulce G. Bubacz, D. MarkBickett,
Joseph H. Chan, James G. Conway, David J. Cowan, Michael D. Gaul,k
Kimberly C. Glennon, Kevin M. Hedeen, Millard H. Lambert, M. Anthony Leesnitzer,
Darryl L. McDougald, Justin L. Mitchell,{ Marcia L. Moss,** Michael H. Rabinowitz,
Michele C. Rizzolio, Lee T. Schaller, Jennifer B. Stanford, Timothy K. Tippin,
Janet R. Warner, L. Graham Whitesell and Robert W. Wiethe
GlaxoSmithKline Research and Development, Five Moore Drive, Research Triangle Park, NC 27709, USA
Received 27 March 2001; accepted 23 May 2001
Abstract—N-Hydroxyformamide-class metalloprotease inhibitors were designed and synthesized, which have potent broad-spec-
trum activity versus matrix metalloproteases and TNF-a converting enzyme (TACE). Compound 13c possesses good oral and
intravenous pharmacokinetics in the rat and dog. # 2001 Elsevier Science Ltd. All rights reserved.
The matrix metalloprotease (MMP) family of endo-
peptidases represents a group of tightly regulated
metalloproteases mediating turnover of the extracellular
matrix proteins proteoglycan, collagen, and gelatin.1
Increased levels of certain MMPs are evident in rheu-
matoid arthritis and osteoarthritis.2 MMPs are also
present in certain cancers at the tumor epithelium, and
are implicated in metastasis.3 Given the extent of disease
association with MMPs, the pharmaceutical industry
has dedicated substantial resources to MMP inhibitor
discovery with agents at all stages of development.4
Tumor necrosis factor-a (TNF-a) is a cytokine pro-
duced by many cell types but mainly by those of mono-
cytic lineage. Elevated TNF levels are implicated in the
pathologies of rheumatoid arthritis,5 multiple sclerosis,6
type II diabetes,7 inflammatory bowel disease,8 and
other human ailments. TNF-a is unusual among the
cytokines in that it is processed from a 26 kD, mem-
brane-bound form to a 17 kD soluble form by a specific
proteolytic cleavage. A subset of known matrix metal-
loprotease inhibitors was subsequently shown to inhibit
this processing event in cells,9ꢀ11 and subsequently the
enzyme activity was characterized, cloned, and expres-
sed.12 TNF-a converting enzyme (TACE, or ADAM17)
is a member of the ADAM family of metalloproteases, a
rapidly expanding family of metalloproteases implicated
in diverse biological function. Therefore, TACE/MMP
inhibitors have been and continue to be of significant
interest in the biomedical community. The debate con-
tinues to be centered on the desired selectivity profile of
metalloprotease inhibitors. The recent clinical successes
of biologicals, such as etanercept13 and infliximab14 that
neutralize TNF suggest a selective TACE inhibitor
would be preferred. While incorporating collagenase
and/or aggrecanase inhibitory activity seems to be a
*Corresponding author. Fax: +1-919-483-6053; e-mail: dm12907@
glaxowellcome.com
yPresent address: Magellan Laboratories, PO Box 13341, Research
Triangle Park, NC 27709, USA.
{Present address: TransTech Pharma Inc., 4170 Mendenhall Oaks
Pkwy, Suite 110, High Point, NC 27265, USA.
xPresent address: Aventis Pharmaceuticals, Selectide Division, 1580 E.
Hanley Blvd, Tucson, AZ 85737, USA.
kPresent address: 3-Dimensional Pharmaceuticals, 665 Stockton Drive,
Exton, PA 19341, USA.
{Present address: Sphinx Pharmaceuticals, 20 T. W. Alexander Drive,
Research Triangle Park, NC 27709, USA.
**Present address: Cognosci Inc., 2 Davis Dr., PO Box 12076,
Research Triangle Park, NC 27709, USA.
0960-894X/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(01)00377-8

2148
D. L. Musso et al. / Bioorg. Med. Chem. Lett. 11 (2001) 2147–2151
Scheme 1. (a) [Ru(R-(+)-BINAP)Cl2]2NEt3, catalytic HCl, MeOH, H2, 60 psi; (b) LDA, THF; methallyl bromide, HMPA, ꢀ78–0 ꢁC; (c) H2, Pd/C,
MeOH, 50 psi; (d) NaOH, H2O, MeOH; (e) 1-ethyl-3-(3-dimethylamino propyl)carbodiimide, CH2Cl2, O-(2-tetrahydropyranyl)hydroxylamine, 0–
25 ꢁC; (f) methanesulfonyl chloride, CH2Cl2–pyridine, 0–25 ꢁC; (g) K2CO3, acetone, reflux; (h) NaOH, H2O; (i) formic acetic anhydride, CH2Cl2–
pyridine, 0–25 ꢁC; (j) 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide, 1-hydroxybenzotriazole, N-methylmorpholine, DMF, P30-NH2; (k) 4 N HCl/
dioxane, CH2Cl2 (for PG2=tert-butoxycarbonyl) or H2, Pd/C (for PG2=benzyloxycarbonyl); (l) 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide, 1-
hydroxybenzotriazole, N-methylmorpholine, DMF; (m) acetic acid, 25 ꢁC, 3 days, or 50 ꢁC, 6–10 h for (PG1=2-tetrahydropyranyl) or H2, Pd/C (for
PG1=benzyl).
logical tactic to preserve the joint cartilage, this
approach to inhibitor development may increase the
likelihood of adverse events. The broad-spectrum
matrix metalloprotease inhibitor marimistat3 has been
reported to cause musculoskeletal pain or ‘tendonitits’
in cancer patients. While the precise mechanism behind
this clinical observation is not understood, more specific
inhibitors should allow us to better understand the
pharmacology observed in the clinic.
hydroxamic acids. Described herein are some initial
efforts at synthesis of P10 isobutyl N-hydroxyformamide
tripeptidomimetics and determination of their potential
as MMP/TACE inhibitors in vitro and in vivo.
The synthetic route to compounds of general structure
13 is shown in Scheme 1. Asymmetric reduction of a b-
ketoester17 3 is followed by alkylation of the derived
alkoxy enolate with methallyl bromide.18 Catalytic
hydrogenation of the alkylation product affords the
ester 5. Saponification of 5 to the hydroxy acid followed
by coupling with 2-tetrahydropyranyloxyamine or O-
benzylhydroxylamine hydrochloride gives the protected
N-hydroxyamide 6. Methanesulfonylation of the
hydroxyl group of 6 followed by addition of the mesy-
late to refluxing potassium carbonate in acetone affords
the b-lactam 7.19 Alternately, the transformation of 6 to
7 can be accomplished in one step via cyclodehydration
with diisopropylazodicarboxylate/triphenylphosphine in
tetrahydrofuran at 0 ꢁC. Hydrolysis of the b-lactam 7 is
followed by N-formylation to give the formamido acid
8. A suitably N-protected a-amino acid 9 is treated with
Structure–activity relationships for MMP inhibitors
such as Ro31-979015 (1) constructed on a tripeptidomi-
metic ‘right-side’ template possessing a hydroxamate or
carboxylate active-site zinc chelator are documented
(Fig. 1), but less well known are the analogous right-
side templates incorporating an N-hydroxyformamide
chelator (2). N-Hydroxyformamide chelators have been
explored for ACE/NEP inhibition with some success.16
These ‘retrohydroxamates’ were seen to offer potential
advantage in drug properties, since they might possess
lower intrinsic hydrophilicity as well as less propensity
for biological conjugation or hydrolysis relative to
Figure 1.

D. L. Musso et al. / Bioorg. Med. Chem. Lett. 11 (2001) 2147–2151
Table 1. In vitro and in vivo activitites of compounds 1 and 13a–k
2149
0
Entry
P1
—
P2
P3
—
Cell TNFa Inhibition,
MonoMac-6 (IC50, nM)
TACE Ki
(nM)
MMP1 Ki
(nM)
MMP9 Ki
(nM)
MMP3 Ki
(nM)
Murine LPS-Induced
Plasma TNF (% inhib. po)
1
—
5716
3073
2954
245
87
156
142
20
2
181
17
2
33
21
34
29
17
98
17
28
89
24
73
310
406
2323
437
93
Inactivea
13a
13b
13c
13d
13e
13f
13g
13h
13i
CH3
CH3
H
ndb
nd
CH3
CH3
27
60
CH3
270
20
41
48
CH3
681
27
36
120
354
80
61
CH3
3964
1686
200
64
170
97
Inactive
Inactive
52
CH3
55
-C2H5
-C2H5
-C2H5
-C2H5
16
16
135
153
45
304
29
47
61
13j
92
11
42
48
13k
1414
31
55
38
62
aInactive means no effect on TNF levels.
bnd, not determined.
Table 2. 24Pharmacokinetic data for 13c
Species
Dose iv (mg/kg)
Cmax iv (ng/mL)
AUC iv (h*ng/mL)
t1/2 iv (h)
Cl(s) iv (mL/min/kg)
F (%)
Rat
Dog
3.4
1.2
2330
4530
957
2100
3.4
2.8
62
10
26
44
water soluble carbodiimide and N-hydroxybenzo-
triazole and a substituted amine to afford the amide 10,
which is deprotected to give the amino amide 11. Cou-
pling of the amino amide 11 and the acid 8 provides the
protected hydroxyformamide 12. Acidic media (for
PG1=2-tetrahydropyranyl) or hydrogenolytic (for
PG1=benzyl) deprotection of 12 followed by solvent
removal and precipitation or crystallization affords the
product metalloprotease inhibitor 13.
lagenase-1 (MMP1), gelatinase B (MMP9, 92 kD gela-
tinase), and stromelysin-1 (MMP3) according to a
standard protocol.21 Compounds were also evaluated
for inhibition of cellular release of 17 kD TNF-a from
MonoMac-6 cells according to a standard protocol.22
Lastly, compounds were assessed for in vivo inhibition
of plasma TNF-a upon oral administration at 40 mg/kg,
in lipopolysaccharide (LPS) treated mice according to a
standard protocol.23
Compounds were tested for inhibition of cell-free TACE
according to
The hydroxamate Ro31-9790 1 inhibits TACE with
respectable potency and is more potent versus the
MMPs. The N-hydroxyformamide carboxamide and N-
a
previously published protocol.20
Compounds were evaluated also for inhibition of col-

2150
D. L. Musso et al. / Bioorg. Med. Chem. Lett. 11 (2001) 2147–2151
methylamide 13a and 13b are weaker MMP inhibitors
but comparable in potency versus TACE. All three
molecules lackpotency in cells, however. The addition
of a heteroaryl substituent at P30 (13c) engenders TACE
potency in vitro and in cells, and preserves respectable
potency versus MMPs.
6. Clements, J. M.; Cossins, J. A.; Wells, G. M.; Corkill, D. J.;
Helfrich, K.; Wood, L. M.; Pigott, R.; Stabler, G.; Ward,
G. A.; Gearing, A. J.; Miller, K. M. J. Neuroimmunol. 1997,
74, 85.
7. Morimoto, Y.; Nishikawa, K.; Ohashi, M. Life Sci. 1997,
61, 795.
8. Feldman, M.; Taylor, P.; Paleolog, E.; Brennan, F. M.;
Maini, R. N. Transplant. Proc. 1998, 30, 4126.
Given heteroaryl substitution at P30, some trends are
evident from the data in Table 1. TACE and MMP
potency changes little as P1 is changed from methyl to
ethyl, as cell potency tends to increase (entries 13c/h,
13d/j). Increased lipophilicity at this place in the mole-
cule might tend to shield the polar N-hydroxy-
formamide substituent from hydration and thereby
promote enzyme binding or partitioning into membrane
lipid. Thiazole at P30 engenders better TACE and cell
potency, with little effect otherwise (entries 13h/i, 13j/k).
Butyl isomers at P20 given show uniform potency versus
TACE, but cell potency varies with the nature of
branching. tert-Butyl and s-butyl perform best in the
cell assay, while isobutyl does not. The analogous iso-
propyl compound 13e is good in the cell-based assay
while the neopentyl compound 13f is poor. Clearly
branching at the P20 position vicinal to the peptide
chain is essential for good cell-based activity. Further,
such branching promotes oral activity in the murine
LPS-induced TNF assay, as all P30 heteroaryl com-
pounds except 13f and 13g show oral activity in this
model.
9. Mohler, K. M.; Sleath, P. R.; Fitzner, J. N.; Cerretti, D. P.;
Alderson, M.; Kerwar, S. S.; Torrance, D. S.; Otten-Evans, C.;
Greenstreet, T.; Black, R. A. Nature 1994, 370, 218.
10. McGeehan, G. M.; Becherer, J. D.; Bast, R. C., Jr.; Boyer,
C. M.; Champion, B.; Connolly, K. M.; Conway, J. G.; Fur-
don, P.; Karp, S.; Kidao, S.; McElroy, A.; Nichols, J.; Pryz-
wansky, K.; Schoenen, F.; Sekut, L.; Truesdale, A.; Verghese,
M.; Warner, J.; Ways, J. Nature 1994, 370, 558.
11. Gearing, A. J. H.; Beckett, P.; Christodoulou, M.;
Churchill, M.; Clements, J.; Davidson, A. H.; Drummond,
A. H.; Galloway, W. A.; Gilbert, R.; Gordon, J.; Leber, T.;
Mangan, M.; Miller, K.; Nayee, P.; Owen, K.; Patel, S.; Tho-
mas, W.; Wells, G.; Wood, L.; Wooley, K. Nature 1994, 370,
555.
12. Moss, M. L.; Jin, S.-L. C.; Milla, M. E.; Burkhart, W.;
Carter, H. L.; Chen, W.-J.; Clay, W. C.; Didsbury, J. R.;
Hassler, D.; Hoffman, C. R.; Kost, T. A.; Lambert, M. H.;
Leesnitzer, M. A.; McCauley, P.; McGeehan, G.; Mitchell, J.;
Moyer, M.; Pahel, G.; Rocque, W.; Overton, L. K.; Schoenen,
F.; Seaton, T.; Su, J.-L.; Warner, J.; Willard, D.; Becherer,
J. D. Nature 1997, 385, 733.
13. Maini, R. N.; Breedveld, F. C.; Kalden, J. R.; Smolen,
J. S.; Davis, D.; Macfarlane, J. D.; Antoni, C.; Leeb, B.;
Elliott, M. J.; Woody, J. N.; Schaible, T. F.; Feldmann, M.
Arthritis Rheum. 1998, 41, 1552.
Compound 13c was further evaluated for its pharmaco-
kinetic properties in rat and dog (Table 2).24 Upon
intravenous and oral dosing in either species, 13c
demonstrates good half-life and bioavailability. The addi-
tion of a P30 heteroaryl substituent to the N-hydroxy-
formamide peptidomimetic structure 2 has afforded
potent inhibitors of TACE and MMPs. The nature of
the P20 and P1 substituents directly influences TNF
inhibition in cells and in the mouse in vivo. Compound
13c could form the basis for a TNF-a targeted drug
therapy acting via inhibition of TACE and further
studies will be reported in due course.
14. Weinblatt, M. E.; Kremer, J. M.; Bankhurst, A. D.; Bul-
pitt, K. J.; Fleischmann, R. M.; Fox, R. I.; Jackson, C. G.;
Lange, M.; Burge, D. J. N. Engl. J. Med. 1999, 340, 253.
15. Bottomley, K. M. K.; Borkakoti, N.; Bradshaw, D.;
Brown, P. A.; Broadhurst, M. J.; Budd, J. M.; Elliott, L.;
Eyers, P.; Hallam, T. J.; Handa, B. K.; Hill, C. H.; James, M.;
Lahm, H.-W.; Lawton, G.; Merritt, J. E.; Nixon, J. S.; Roth-
lisberger, U.; Whittle, A.; Johnson, W. H. Biochem. J. 1997,
323, 483.
16. Robl, J. A.; Simpkins, L. M.; Asaad, M. M. Bioorg. Med.
Chem. Lett. 2000, 10, 257.
17. King, S. A.; Thomson, A. S.; King, A. O.; Verhoeven,
T. R. J. Org. Chem. 1992, 57, 6689.
18. Frater, G. Helv. Chim. Acta 1979, 62, 2825.
19. Miller, M. J.; Bajwa, J. S.; Mattingly, P. G.; Peterson, K.
J. Org. Chem. 1982, 47, 4928.
Acknowledgements
20. Leesnitzer, M. A.; Bickett, D. M.; Moss, M. L.; Becherer,
J. D. In High Throughput Screening Novel Anti-Inflamma-
tories; Kahn, M., Ed.; Birkhaeuser: Basel, 2000; pp 87–100.
21. Recombinant catalytic domains of MMP-1 and full length
active MMP-3 were expressed and purified from Escherichia
coli. Enzymes were refolded in 200 mM NaCl, 50 mM Tris,
5 mM CaCl2, 10 mM ZnSO4 and 0.01% Brij 35, pH 7.6 for 1 h
prior to the assay. The catalytic domains of proMMP-9 were
purified from the media of baculovirus infected T. ni cells.
Assays were run in a total volume of 0.180 mL assay buffer
containing 200 mM NaCl, 50 mM Tris, 5 mM CaCl2, 10 mM
ZnSO4 and 0.01% Brij 35, pH 7.6. MMP-1, MMP-3, and
MMP-9 concentrations were adjusted to 0.5, 0.05, 5, and
0.1 nM, respectively. Enzymes were pre-incubated with inhi-
bitor for 20 min at room temperature and the reactions were
initiated with the addition of the fluorogenic substrate, Dnp-
Pro-Cha-Gly-Cys(Me)-His-Ala-Lys(Nma)–NH2. Dose respon-
ses were generated using 11 point 3-fold dilutions of the inhi-
bitor. Product was measured using excitation and emission
wavelenghts of 343 and 450 nm, respectively.
The authors thankElizabeth Sugg and Stephen Frye for
their advice during the course of this work.
References and Notes
1. Schwarz, M. A.; Van Wart, H. E. Prog. Med. Chem. 1992,
29, 271.
2. Docherty, A. J. P.; Murphy, G. Ann. Rheum. Dis. 1990, 49,
469.
3. Nelson, A. R.; Fingleton, B.; Rothenberg, M. L.; Matri-
sian, L. M. J. Clin. Oncol. 2000, 18, 1135.
4. Michaelides, M. R.; Curtin, M. L. Curr. Pharm. Des. 1999,
5, 787.
5. Moreland, L. W.; Baumgartner, S. W.; Schiff, M. H.; Tin-
dall, E. A.; Fleischmann, R. M.; Weaver, A. L.; Ettlinger,
R. E.; Cohen, S.; Koopman, W. J.; Mohler, K. N. Engl. J.
Med. 1997, 337, 141.

D. L. Musso et al. / Bioorg. Med. Chem. Lett. 11 (2001) 2147–2151
2151
22. Human mono mac-6 cells in RPMI 1640 media with 10%
fetal bovine serum (FBS) were preincubated for 10 min with
compounds and then stimulated with 10 ng/mL phorbol 12-
myristate 13-acetate (Sigma, #P-8139) and 30 ng/mL LPS
(Sigma, #L2630) and TNF measured in the media at 2 h by
ELISA kit (R&D Systems, Minneapolis, MN, USA,
#DTA50).
23. Test compounds are formulated in 0.2 mL of PBS and
0.1% Tween 80 and given orally via gavage 10 min prior to
LPS administration. C3/hen female mice are injected intra-
peritoneally with 200 mg/kg LPS (E. coli, Serotype 0111:B4,
Sigma Chemical Co, St. Louis, MO, USA) in PBS and sacri-
ficed 90 min later by CO2 asphyxiation. Blood is immediately
taken from the caudal vena cava and plasma prepared and
frozen at ꢀ80 ꢁC. Plasma concentrations of TNF are measured
by ELISA (Genzyme Co., Cambridge MA, USA).
24. Compound 13c was formulated in a 2.5–5% (w/v) leci-
thin/water emulsion and administered iv via a cannula or
orally via a feeding tube to male Lewis rats or Beagle dogs.
Blood was collected at selected timepoints over a 24 h period
following dosing. Plasma was prepared and then treated with
at least two volumes of methanol or acetonitrile to remove
plasma protein. The extracts were subsequently assayed for
compound 13c using HPLC/MS. The area under the plasma
concentration versus time curve (AUC) was determined by the
linear trapezoidal method and includes the area resulting from
extrapolation along the elimination phase to time infinity.
Oral bioavailability (F, %) was determined from the ratio of
AUCoral to AUCiv. A crossover design was used for dog
studies such that iv and oral doses were administered to the
same animal with a weekbetween dosings in order to allow for
compound washout. Values are means of at least two animals.