Silicon-based metalloprotease inhibitors: Synthesis and evaluation of silanol and silanediol peptide analogues as inhibitors of angiotensin-converting enzyme

Article abstract of DOI:10.1021/ja026158w

Silanols are best known as unstable precursors of siloxane (silicone) polymers, substances generally considered stable and inert, but have the potential to mimic a hydrated carbonyl and inhibit protease enzymes. While previous testing of simple silanediol and silanetriol species as inhibitors of hydrolase enzymes found them ineffective, this study reports polypeptide mimics with a central methylsilanol [SiMeOH] or silanediol [Si(OH)₂] group and their assessment as effective transition state analogue inhibitors of the well-studied metalloprotease angiotensin-converting enzyme (ACE). Central to the synthesis strategy, phenylsilanes were employed as acid-hydrolyzable precursors of the silanol group. The N-benzoyl Leu-[SiMeOH]-Gly benzyl amides proved to be stable and readily characterized. In contrast, the Leu-[Si(OH)₂]-Gly structure was difficult to characterize, possibly because of self-association. Capping the silanediols with chlorotrimethylsilane gave a well-defined trisiloxane, demonstrating that the silanediol was monomeric. The Leu-[Si]-Gly structures were converted to Leu-[Si]-Ala analogues by enolate alkylation. Coupling of the silanol precursors with proline tert-butyl ester gave N-benzoyl Leu-[Si]-Gly-Pro and N-benzoyl Leu-[Si]-Ala-Pro tripeptide analogues. Treatment of these with triflic acid formed the corresponding methylsilanols and silanediols, all of which were monomeric. The silanediol tripeptide mimics inhibited ACE with IC₅₀ values as low as 14 nM. Methylsilanols, in contrast, were poor inhibitors, with IC₅₀ values above 3000 nM. These data, including comparisons with inhibition data from carbon analogues, are consistent with binding of the silanediols by chelation of the ACE active site zinc, whereas the methylsilanols ligate poorly.

Full text of DOI:10.1021/ja026158w

Published on Web 05/30/2002
Silicon-Based Metalloprotease Inhibitors: Synthesis and
Evaluation of Silanol and Silanediol Peptide Analogues as
Inhibitors of Angiotensin-Converting Enzyme1
Mwangi wa Mutahi, Thomas Nittoli, Luxuan Guo, and Scott McN. Sieburth*
Contribution from the Department of Chemistry, Temple UniVersity,
Philadelphia, PennsylVania 19122
Received March 11, 2002
Abstract: Silanols are best known as unstable precursors of siloxane (silicone) polymers, substances
generally considered stable and inert, but have the potential to mimic a hydrated carbonyl and inhibit protease
enzymes. While previous testing of simple silanediol and silanetriol species as inhibitors of hydrolase
enzymes found them ineffective, this study reports polypeptide mimics with a central methylsilanol [SiMeOH]
or silanediol [Si(OH)2] group and their assessment as effective transition state analogue inhibitors of the
well-studied metalloprotease angiotensin-converting enzyme (ACE). Central to the synthesis strategy,
phenylsilanes were employed as acid-hydrolyzable precursors of the silanol group. The N-benzoyl Leu-
[SiMeOH]-Gly benzyl amides proved to be stable and readily characterized. In contrast, the Leu-[Si(OH)2]-
Gly structure was difficult to characterize, possibly because of self-association. Capping the silanediol with
chlorotrimethylsilane gave a well-defined trisiloxane, demonstrating that the silanediol was monomeric.
The Leu-[Si]-Gly structures were converted to Leu-[Si]-Ala analogues by enolate alkylation. Coupling of
the silanol precursors with proline tert-butyl ester gave N-benzoyl Leu-[Si]-Gly-Pro and N-benzoyl Leu-
[Si]-Ala-Pro tripeptide analogues. Treatment of these with triflic acid formed the corresponding methylsilanols
and silanediols, all of which were monomeric. The silanediol tripeptide mimics inhibited ACE with IC50
values as low as 14 nM. Methylsilanols, in contrast, were poor inhibitors, with IC50 values above 3000 nM.
These data, including comparisons with inhibition data from carbon analogues, are consistent with binding
of the silanediols by chelation of the ACE active site zinc, whereas the methylsilanols ligate poorly.
has the only naturally occurring silicon-carbon bond.4 With
Introduction
the discovery of the useful physical and chemical properties of
Silicon is the second most abundant element and the one most
similar to carbon. This close chemical analogy has been the
inspiration for many studies of silicon as a replacement for
carbon, particularly in biologically active molecules.2 These
investigations were bolstered by the finding that organic
derivatives of silicon are devoid of the intrinsic toxicity found
with the heavier group 4 elements tin and lead.3 The strong
silicon-carbon bond, and the correspondingly stable organic
derivatives, make bioactive organosilanes all the more curious
because of their absence from the natural worldssilicon carbide
the organosilane polymers,5,6 particularly the silicones,7 and the
development of the direct process for their synthesis,8 organo-
silanes have become cheap and readily available commodities.
Bioactive Organosilanes. Investigations of organosilane
properties over the last 40 years have identified an assortment
of bioactive substances. Rational searches for bioactive organo-
silanes have been conducted along two distinct paths, random
screening and molecular design. Random screening can discover
the biological activity of structures without carbon analogy, but
by its very nature it cannot be incorporated into target-directed
design programs driven by receptor modeling or substrate-
analogy. Alternatively, the systematic replacement of silicon
for carbon in bioactive substances is an approach that can be
* To whom correspondence should be addressed. Phone: 215-204-7916.
E-mail: sieburth@astro.temple.edu.
(1) Taken, in part, from: Mutahi, A. M. Ph.D. Dissertation, SUNY Stony
Brook, 1996. Nittoli, T. Ph.D. Dissertation, SUNY Stony Brook, 1997.
(2) Fessenden, R. J.; Fessenden, J. S. AdV. Drug Res. 1967, 4, 95-132. Tacke,
R.; Wannagat, U. Top. Curr. Chem. 1979, 84, 1-75. Fessenden, R. J.;
Fessenden, J. S. AdV. Organomet. Chem. 1980, 18, 275-299. Tacke, R. In
Organosilicon and Bioorganosilicon Chemistry; Sakurai, H., Ed.; John
Wiley & Sons: New York, 1985. Barcza, S. In Silicon Chemistry; Corey,
E. R., Corey, J. Y., Gaspar, P. P., Eds.; Wiley: New York, 1988; pp 135-
144. Tacke, R.; Linoh, H. In The Chemistry of Organic Silicon Compounds;
Patai, S., Rappoport, Z., Eds.; John Wiley & Sons: New York, 1989; Vol.
2, pp 1143-1206. Sieburth, S. McN. In Designing Safer Chemicals; DeVito,
S. C., Garrett, R. L., Eds.; American Chemical Society: Washington, DC,
1996; pp 74-83. Tacke, R.; Wagner, S. A. In The Chemistry of Organic
Silicon Compounds; Rappoport, Z., Apeloig, Y., Eds.; John Wiley &
Sons: New York, 1998; Vol. 2, pp 2363-2400.
(4) For a possible exception, see: Zbik, M.; Jasieniak, M.; Smart, R. S. C.
Meteorit. Planet. Sci. 2000, 35, 943-947.
(5) Rochow, E. G. Silicon and Silicones; Springer-Verlag: New York, 1987.
(6) Miller, R. D.; Michl, J. Chem. ReV. 1989, 89, 1359-1410.
(7) Stark, F. O.; Falender, J. R.; Wright, A. P. In ComprehensiVe Organome-
tallic Chemistry; Wilkinson, G., Stone, F. G. A., Abel, E. W., Eds.;
Pergamon: New York, 1982; Vol. 2, pp 305-363. Brown, S. S.; Kendrick,
T. C.; McVie, J.; Thomas, D. R. In ComprehensiVe Organometallic
Chemistry II; Abel, E. W., Stone, F. G. A., Wilkinson, G., Eds.;
Pergamon: New York, 1995; Vol. 2, pp 111-135.
(8) Bu¨chner, W.; Schliebs, R.; Winter, G.; Bu¨chel, K. H. Industrial Inorganic
Chemistry; VCH: New York, 1989. Lewis, L. N. In The Chemistry of
Organic Silicon Compounds; Rappoport, Z., Apeloig, Y., Eds.; John Wiley
& Sons: New York, 1998; Vol. 2, pp 1581-1597.
(3) Safety of Silicone Breast Implants; Bondurant, S., Ernster, V., Herdman,
R., Eds.; National Academy Press: Washington, DC, 2000; pp 80-113.
9
10.1021/ja026158w CCC: $22.00 © 2002 American Chemical Society
J. AM. CHEM. SOC. 2002, 124, 7363-7375
7363

A R T I C L E S
Mutahi et al.
Figure 1. Biologically active organosilanes designed by analogy to carbon
compounds.
readily incorporated into design programs, and it has led to the
identification of many bioactive organosilanes. Compounds 1
and 2, Figure 1, are organosilanes designed by analogy.
Flusilazole 19 belongs to a class of antifungal sterol biosynthesis
inhibitors,10 most of which do not contain silicon. Silafluofen
211 is a distant relative of the natural pyrethrin insecticides and
a direct analogue of the carbon-based MTI-800.12 Despite
decades of effort to discover organosilanes with useful bioac-
tivity, these are the only organosilanes produced commercially
because of their biological activity. Nevertheless, ongoing
research promises that new examples of bioactive organosilanes
will continue to emerge.13
Bioactive Organosilane Design. Design of bioactive organo-
silanes, starting with a molecule of known biological activity
and replacing a carbon atom with a silicon atom, has a long
history and has taken several forms.2 The most straightforward
version of this approach involves replacement of a quaternary
carbon, to give a silicon surrounded by carbons (e.g., 1 and 2).
This takes advantage of the robust silicon-carbon bond and
avoids the metabolic instability of the silicon-hydrogen bond14
and the hydrolytic instability of most silicon-heteroatom bonds.
The consequences of this substitution, beyond the generation
of a novel and often proprietary structure, are a subtle increase
in the molecular volume, a perturbation of the local electronics,
and a small increase in the molecular lipophilicity relative to
the original carbon structure.11
Figure 2. Hydrolysis of peptide bonds by protease enzymes, and silanols
that could mimic intermediate 4.
species. This type of carbon substitution, however, has been an
unrealized design strategy. In principle, a silanediol can mimic
a hydrated carbonyl group. Tetrahedral, hydrated sp3 carbonyls
are energetically disfavored over their trigonal, unhydrated sp2
counterparts.15 Conversely, trigonal silicon species are highly
unstable relative to the tetrahedral alternative.16 In the context
of hydrolytic enzyme substrates, replacement of a carbonyl
group with a silanol or silanediol would create a transition state
analogue and a potential enzyme inhibitor. In terms of binding
energetics, a tetrahedral silanediol would not require the
energetically disfavored hydration needed by the equivalent
ketone inhibitor.
Protease Inhibitors. Metallo (and aspartic) protease enzymes
catalyze the addition of a molecule of water to an amide of 3,
stabilizing the intermediate 4 through chelation of the active
site zinc (or hydrogen bonding with aspartic proteases) (Figure
2). Productive collapse of 4 leads to acid 5 and amine 6. Central
to the design of a protease inhibitor is the choice of a
nonhydrolyzable isostere (transition state analog) of the tetra-
hedral hydrated carbonyl intermediate 4. Transition state
analogues are a cornerstone of enzyme inhibitor design17 and
the basis of several successful pharmaceutical agents. Their uses
in the treatment of hypertension and for control of the AIDS
virus are most notable.17,18
As an alternative to the replacement of a stable carbon atom,
silicon also has the potential to emulate an unstable carbon
(9) Moberg, W. K.; Basarab, G. S.; Cuomo, J.; Liang, P. H. In Pesticide Science
and Biotechnology; Greenhalgh, R., Roberts, T. R., Eds.; Blackwell
Scientific: Boston, 1986; pp 57-60.
(10) Hoffman, H. L.; Ernst, E. J.; Klepser, M. E. Expert Opin. InVest. Drugs
2000, 9, 593-605.
The fit of an isostere at an enzyme active site stems from the
complementarities of the active site and the isostere, with
hydrogen bonding, Coulombic attractions, and hydrophobic
surfaces as important components.19 “Transition state analogues”
that have effectively been deployed as protease inhibitors include
reduced amides, carbinols, and geminal diols, amino alcohols,
ketones, and aldehydes (including hydrated forms). Non-carbon-
based structures include boron-, sulfur-, and phosphorus-based
groups.
(11) Sieburth, S. McN.; Manly, C. J.; Gammon, D. W. Pestic. Sci. 1990, 28,
289-307.
(12) Udagawa, T.; Numata, S.; Oda, K.; Shiraishi, S.; Kodaka, K.; Nakatani,
K. In Recent AdVances in the Chemistry of Insect Control; Janes, N. F.,
Ed.; Royal Society of Chemistry: London, 1985; pp 193-204.
(13) Tacke, R.; Merget, M.; Bertermann, R.; Bernd, M.; Beckers, T.; Reissmann,
T. Organometallics 2000, 19, 3486-3497. Reginato, G.; Mordini, A.;
Valacchi, M. Tetrahedron Lett. 1998, 39, 9545-9548. Yoo, B. R.; Suk,
M. Y.; Han, J. S.; Yu, Y. M.; Hong, S. G.; Jung, I. N. Pestic. Sci. 1998,
52, 138-144. Gotteland, J. p.; Dax, C.; Halazy, S. Bioorg. Med. Chem.
Lett. 1997, 7, 1153-1156. Gotteland, J. P.; Delhon, A.; Junquero, D.; Oms,
P.; Halazy, S. Bioorg. Med. Chem. Lett. 1996, 6, 533-538. Heinonen, P.;
Sipila¨, H.; Neuvonen, K.; Lo¨nnberg, H.; Cockcroft, V. B.; Wu¨rster, S.;
Virtanen, R.; Savola, M. K. T.; Salonen, J. S.; Savola, J. M. Eur. J. Med.
Chem. 1996, 31, 725-729. Parker, R. A.; Barnhart, R. L.; Chen, K. S.;
Edwards, M. L.; Matt, J. E., Jr.; Rhinehart, B. L.; Robinson, K. M.; Vaal,
M. J.; Yates, M. T. Bioorg. Med. Chem. Lett. 1996, 6, 1559-1562. Van
Dorsselaer, V.; Schirlin, D.; Marchal, P.; Weber, F.; Danzin, C. Bioorg.
Chem. 1996, 24, 178-193. Altamura, M.; Giammaruco, M.; Taddei, M.;
Ulivi, P. J. Org. Chem. 1995, 60, 8403-8406. Anderson, W. K.; Kasliwal,
R.; Houston, D. M.; Wang, Y.-s.; Narayanan, V. L.; Haugwitz, R. D.;
Plowman, J. J. Med. Chem. 1995, 38, 3789-3797. Danzin, C.; Collard,
J.-N.; Marchal, P.; Schirlin, D. Bioorg. Med. Chem. Lett. 1995, 5, 2363-
2366. Gotteland, J.-P.; Brunel, I.; Gendre, F.; De´sire´, J.; Delhon, A.;
Junque´ro, D.; Oms, P.; Halazy, S. J. Med. Chem. 1995, 38, 3207-3216.
Walkup, R. D.; Cole, D. C.; Whittlesey, B. R. J. Org. Chem. 1995, 60,
2630-2634. Damour, D.; Barreau, M.; Dutruc-Rosset, G.; Doble, A.; Piot,
O.; Mignani, S. Bioorg. Med. Chem. Lett. 1994, 4, 415-420. Ge´hanne, S.;
Giammaruco, M.; Taddei, M.; Ulivi, P. Tetrahedron Lett. 1994, 35, 2047-
2048. Liebner, F.; Bankwitz, U.; Ru¨hlmann, K. Liebigs Ann. Chem. 1994,
145-150.
Missing from the hydrolase inhibitor literature are organosi-
lanes, and in particular silanols; several reasons for this absence
are easily identified. Silanols, silanediols, and especially silan-
(15) Lead references: Akahoshi, F.; Ashimori, A.; Sakashita, H.; Yoshimura,
T.; Imada, T.; Nakajima, M.; Mitsutom, i. N.; Kuwahara, S.; Ohtsuka, T.;
Fukaya, C.; Miyazaki, M.; Nakamura, N. J. Med. Chem. 2001, 44, 1286-
1296. LaPlante, S. R.; Bonneau, P. R.; Aubry, N.; Cameron, D. R.; Deziel,
R.; Grand-Maitre, C.; Plouffe, C.; Tong, L.; Kawai, S. H. J. Am. Chem.
Soc. 1999, 121, 2974-2986.
(16) Apeloig, Y. In The Chemistry of Organic Silicon Compounds; Patai, S.,
Rappoport, Z., Eds.; Wiley: New York, 1989; Vol. 1, pp 57-225.
(17) Schramm, V. L. Annu. ReV. Biochem. 1998, 67, 693-720.
(18) Leung, D.; Abbenante, G.; Fairlie, D. P. J. Med. Chem. 2000, 43, 305-
341.
(19) Silverman, R. B. The Organic Chemistry of Drug Design and Drug Action;
Academic: New York, 1992.
(14) Fessenden, R. J.; Hartman, R. A. J. Med. Chem. 1970, 13, 52-54.
9
7364 J. AM. CHEM. SOC. VOL. 124, NO. 25, 2002

Silanol Protease Inhibitors
A R T I C L E S
carbonyl toward hydration and stabilizes the resulting tetrahedral
intermediate, leading to hydrolysis of the amide bond.27 Active
site-directed metalloprotease inhibitors have a functional group
to bind the active site zinc ion. These groups can be zincophilic
(thiols), can use a combination of chelation and Coulombic
attraction to bind the metal (carboxylate and phosphinate), or
it can simply be a metal chelator. Metal chelators can be
hydroxamic acids or hydrated ketones (e.g., 10). The work
described here illustrates the first use of a silanediol (11) as a
transition state analogue, anticipated to emulate the hydrated
ketone 10.
Geminal diols, hydrated aldehydes and ketones, have been
known as enzyme inhibitors for some time, and they can be
effective metalloprotease inhibitors. Almquist’s ketone inhibitor
10,28 shown in Figure 3 as the energetically disfavored hydrate,
presumably acts by chelating the active site zinc. This compound
is an excellent ACE inhibitor, with an IC50 value of 1.0 nM.
This ketone was also an ideal starting point for design of a
silanol protease inhibitor, because the steric environment of the
zinc-chelating group, flanked by substituents on two sides,
would inhibit self-condensation. On the other hand, the influence
of the nearby amide groups on the silanediol could not be readily
predicted because there are only scattered examples of silanols
with γ-carbonyl groups. Amide carbonyls capable of intramo-
lecularly coordinating a silicon, to form a five-membered ring
and a pentavalent silicon, will do so when the silicon carries a
halogen.29,30 For silanols (one OH), significant coordination of
a nearby carbonyl oxygen has only been found when the silanol
is protonated.31,32
Silanol Properties. The potential for silanols to function as
metalloprotease inhibitors hinges, in part, on the ability of the
oxygens to ligate or chelate metals. Silanol oxyanions are useful,
bulky monodentate ligands for metals,33 but we are aware of
only two crystal structures that show chelation by an O-Si-O
group. The permethyl cycloheptasiloxane 12 was serendipitously
found to bind potassium,34 Figure 4. More relevant to the
metalloprotease is structure 13 in which tert-butyl silicates
Figure 3. Angiotensin-converting enzyme (ACE) generates angiotensin
II, raising blood pressure. Almquist’s ketone inhibitor 10 (shown hydrated)
and silanediol analogue 11.
etriols are notoriously unstable toward self-condensation. Indeed,
the organosilicon industry was founded largely on the production
of silanols, particularly dimethylsilanediol 7, and their self-
condensation to give silicone polymers, prized for their fluid
and thermal stability properties.5 Moreover, in the two instances
in which silanols were tested as hydrolase inhibitors, they proved
to be completely ineffective. Galardy and Kortylewicz, in a
survey of second and third row element-based tetrahedral
structures, tested an aqueous solution of dimethylsilanediol 7
as an inhibitor of angiotensin-converting enzyme (ACE)20 and
found no evidence for inhibition. More recently, Curley and
Pratt prepared phenylacetamidomethylsilanetriol 8 and evaluated
it as an inhibitor of a â-lactamase. Again, no evidence for
enzyme inhibition was observed.21
While the ability of silanediols to act as enzyme inhibitors
was unsubstantiated, examples of stable, difficult to polymerize
dialkylsilanediols are plentiful.22 Sterically hindering alkyl
groups inhibit oligomer formation.23 For example, while dimeth-
ylsilanediol 7 is intrinsically unstable toward self-condensation,
diethylsilanediol can be isolated and stored in pure form, and
diisobutylsilanediol 9 is a liquid crystal.24 These, and the nearly
100 other characterized silanediols, suggested that silanol-based
peptidomimetics could be well-behaved chemical species.
Angiotensin-Converting Enzyme (ACE). Captopril,25 the
first pharmaceutical product designed to inhibit a protease,
unequivocally demonstrated that protease inhibition could be
an effective approach to therapeutic intervention. Inhibition of
ACE by captopril prevents conversion of angiotensin I to
angiotensin II (Figure 3), resulting in lower blood pressure.
Many other protease targets have been identified, and the
invention of protease inhibitors continues to be a prominent
avenue for drug design.18,26
(27) Auld, D. S. Struct. Bonding 1997, 89, 29-50.
(28) Almquist, R. G.; Crase, J.; Jennings-White, C.; Meyer, R. F.; Hoefle, M.
L.; Smith, R. D.; Essenburg, A. D.; Kaplan, H. R. J. Med. Chem. 1982,
25, 1292-1299.
(29) Bassindale, A. R.; Borbaruah, M. J. Chem. Soc., Chem. Commun. 1993,
352-353.
(30) For examples of beta-silyl carbonyl compounds with crystal structures in
which the silicon has at least one heteroatom, see: Struchkov, Y. T.;
Ovchinnikov, Y. E.; Shipov, A. G.; Baukov, Y. I. Russ. Chem. Bull. 1995,
44, 1705-1710. Otter, J. C.; Adamson, C. L.; Yoder, C. H.; Rheingold,
A. L. Organometallics 1990, 9, 1557-1562. Onan, K. D.; McPhail, A. T.;
Yoder, C. H.; Hillyard, R. W. J. Chem. Soc., Chem. Commun. 1978, 209-
210.
(31) For examples of silanols with γ-carbonyl groups, see: Aldehydes, Shaw,
J. T.; Woerpel, K. A. Tetrahedron 1997, 53, 16597-16606. Nguyen, P.
T.; Palmer, W. S.; Woerpel, K. A. J. Org. Chem. 1999, 64, 1843-1848.
Ketones, Brueckmann, R.; Maas, G. Chem. Ber. 1987, 120, 635-641.
Polniaszek, R. P.; Dillard, L. W. J. Org. Chem. 1992, 57, 4103-4110.
Probert, G. D.; Harding, R.; Whitby, R. J.; Coote, S. J. Synlett 1997, 1371-
1374. Franz, A. K.; Woerpel, K. A. J. Am. Chem. Soc. 1999, 121, 949-
957. Esters, Hale, M. R.; Hoveyda, A. H. J. Org. Chem. 1992, 57, 1643-
1645. Fleming, I.; Winter, S. B. D. J. Chem. Soc., Perkin Trans. 1 1998,
2687-2700. Amides, Barrett, A. G. M.; Head, J.; Smith, M. L.; Stock, N.
S.; White, A. J. P.; Williams, D. J. J. Org. Chem. 1999, 64, 6005-6018.
(32) Two crystal structures of an amide carbonyl coordinated to a pentavalent
silicon carrying a protonated silanol have appeared: Pogozhikh, S. A.;
Zamyshlyaeva, O. A.; Kramarova, E. P.; Antipin, M. Y.; Ovchinnikov, Y.
E.; Baukov, Y. I. Russ. Chem. Bull. 1999, 48, 1595-1596. Bassindale, A.
R.; Parker, D. J.; Taylor, P. G.; Auner, N.; Herrschaft, B. J. Chem. Soc.,
Chem. Commun. 2000, 565-566.
ACE is a metalloprotease, with a zinc ion at the active site,
one of four major classes of protease enzymes categorized by
their active site functionality.18 This zinc ion activates the amide
(20) Galardy, R. E.; Kortylewicz, Z. P. Biochem. J. 1985, 226, 447-454.
(21) Curley, K.; Pratt, R. F. J. Am. Chem. Soc. 1997, 119, 1529-1538.
(22) Lickiss, P. D. AdV. Inorg. Chem. 1995, 42, 147-262.
(23) Lebrun, J. J.; Porte, H. In ComprehensiVe Polymer Science; Allen, G.,
Bevington, J. C., Eds.; Pergamon: New York, 1991; Vol. 5, pp 593-609.
(24) Eaborn, C.; Hartshorne, N. H. J. Chem. Soc. 1955, 549-555. Buttrus, N.
H.; Eaborn, C.; Hitchcock, P. B.; Saxena, A. K. J. Organomet. Chem. 1985,
284, 291-297.
(25) Ondetti, M. A.; Rubin, B.; Cushman, D. W. Science (Washington, D.C.)
1977, 196, 441-444. Rich, D. H. In ComprehensiVe Medicinal Chemistry;
Hansch, C., Sammes, P. G., Taylor, J. B., Eds.; Pergamon: New York,
1990; Vol. 2, pp 391-441.
(33) LaPointe, R. E.; Wolczanski, P. T.; Van Duyne, G. D. Organometallics
1985, 4, 1810-1818. Miller, R. L.; Lawler, K. A.; Bennett, J. L.;
Wolczanski, P. T. Inorg. Chem. 1996, 35, 3242-3253.
(34) Churchill, M. R.; Lake, C. H.; Chao, S.-H. L.; Beachley, O. T., Jr. J. Chem.
Soc., Chem. Commun. 1993, 1577-1578.
(26) Ripka, A. S.; Rich, D. H. Curr. Opin. Chem. Biol. 1998, 2, 441-452.
9
J. AM. CHEM. SOC. VOL. 124, NO. 25, 2002 7365

A R T I C L E S
Mutahi et al.
Scheme 1. The Phenyl-Silicon Bond Is Stable to TFA but
Hydrolyzes Rapidly with TfOH
and the two O-P-C angles are not identical (107° and 110°).
The combination of C-X-C bond angles and C-X bond
lengths (longest for X ) Si) results in a distance between the
carbons attached to X that is shortest for X ) C and longest
for X ) Si. The effect of these structural variations on the
binding of an inhibitor to an active site will depend, naturally,
on the specific enzyme in question.
Figure 4. Crystal structures of metals chelated by a siloxane and by
silicates.
Results and Discussion
A Silanol Protecting Group. Synthesis of a silanol or
silanediol within a peptide backbone required the identification
of a suitable silanol protecting group. Among the available
options was the use of silyl ethers and readily oxidized silicon
substituents such as hydride.41 The correlation of silyl ether
stability and steric hindrance42 led us to reject this option as
too substrate dependent and potentially troublesome. Silicon
dihydrides were also rejected as being too air sensitive and
potentially reactive with nucleophiles. On the other hand, the
lability of unsaturated carbon substituents seemed to make them
a more appropriate choice. In the hierarchy of acid-sensitive
carbon appendages of silicon,43 allyl, aryl, and vinyl groups
attached to silicon are readily cleaved by acid and other
electrophiles, with phenyl less acid sensitive than allyl and more
acid sensitive than vinyl groups. Moreover, diarylsilanes can
be commercially available, and the sensitivity of the aryl-silicon
bond to acid can be modulated by the use of electron-donating
and -withdrawing substituents.44
In a preliminary study of phenylsilane as a silanol precursor,
we prepared ethoxy(diisobutyl)phenylsilane 15 (Scheme 1), with
the expectation that this would closely mirror the second
phenylsilane cleavage of a dialkyldiphenylsilane. The ease of
acidic cleavage of unsaturated groups from silicon can be
dependent on the electronegativity of the silicon substituents.45
The ideal reactivity level for the phenyl silane bond would be
stability under conditions used for removal of common peptide
protecting groups such as the tert-butoxycarbonyl (Boc) group,
that is, stable to trifluoroacetic acid (TFA) but labile under more
strongly acidic conditions used for peptide cleavage from a solid
support,46 so that the protected silanediol could participate in
polypeptide synthesis schemes. We were therefore pleased to
find that overnight treatment of 15 with neat TFA at ambient
temperature led only to exchange of the ethoxy group, to give
14. In contrast, treatment with trifluoromethanesulfonic acid
(triflic acid, TfOH) in TFA at 0 °C for 10 min led to the
quantitative isolation of the liquid crystalline diisobutylsilanediol
Figure 5. Comparison of dicyclohexylmethanol, dicyclohexylsilanediol,
and dicyclohexylphosphinic acid crystal structures.
chelate zinc.35 The caveat with regard to this latter structure,
however, is that the chelating groups are silicate-based oxyan-
ions derived from silanols that are more acidic than the
dialkylsilanediols studied here because of the four electrone-
gative groups around silicon. Thus, the ability of a neutral
silanediol to act as a metal chelator remained to be tested.
Silicon is not only larger than carbon, but it also adopts
slightly different bond angles. The effect of a trialkylsilyl group
on the preferred geometry of an attached oxygen has been noted
by others.36 The crystal structures in Figure 5, dicyclohexyl-
carbinol,37 dicyclohexylsilanediol,38 and dicyclohexylphosphinic
acid,39 provide points of comparison for carbon, silicon, and
phosphorus structures. Phosphinic and related phosphorus-based
acids, notably based on a second row element like the silanols
described here, can be potent metalloprotease inhibitors.40
Comparing these three related structures, the C-X-C bond
angle decreases smoothly from X ) C (117°) to X ) Si (113°)
to X ) P (109°), while the O-X-C and the O-X-O bond
angles are nearly identical for X ) C (110°, 105°) and X ) Si
(110°, 106°), but the O-P-O bond angle is much larger (114°),
(41) Adam, W.; Mitchell, C. M.; Saha-Moeller, C. R.; Weichold, O. J. Am.
Chem. Soc. 1999, 121, 2097-2103.
(35) Su, K.; Tilley, T. D.; Sailor, M. J. J. Am. Chem. Soc. 1996, 118, 3459-
3468.
(42) Greene, T. W.; Wuts, P. G. M. ProtectiVe Groups in Organic Synthesis,
(36) Shambayati, S.; Blake, J. F.; Wierschke, S. G.; Jorgensen, W. L.; Schreiber,
S. L. J. Am. Chem. Soc. 1990, 112, 697-703.
3rd ed.; John Wiley & Sons: New York, 1999.
(43) Eaborn, E. J. Organomet. Chem. 1975, 100, 43-57. Uhlig, W. J.
Organomet. Chem. 1993, 452, 29-32. Fleming, I.; Dunogue`s, J.; Smithers,
R. Org. React. 1989, 37, 57-575.
(37) Sgarabotto, P.; Ugozzoli, F. Acta Crystallogr. 1988, C44, 671-673.
(38) Buttrus, N. H.; Eaborn, C.; Hitchcock, P. B.; Lickiss, P. D.; Taylor, A. D.
J. Organomet. Chem. 1986, 309, 25-33.
(44) Eaborn, C. J. Chem. Soc. 1956, 4858-4864. Eaborn, C.; Pande, K. C. J.
Chem. Soc. 1960, 1566-1571.
(39) Aslanov, L. A.; Sotman, S. S.; Rybakov, V. B.; Elepina, L. G.; Nifantev,
E. E. J. Struct. Chem. 1979, 20, 646-647.
(45) Mayr, H.; Hagen, G. J. Chem. Soc., Chem. Commun. 1989, 91-92.
(46) Bodanszky, M. Peptide Chemistry: A Practical Textbook; Springer-
Verlag: New York, 1988.
(40) Dive, V.; Lucet-Levannier, K.; Georgiadis, D.; Cotton, J.; Vassiliou, S.;
Cuniasse, P.; Yiotakis, A. Biochem. Soc. Trans. 2000, 28, 455-460.
9
7366 J. AM. CHEM. SOC. VOL. 124, NO. 25, 2002

Silanol Protease Inhibitors
A R T I C L E S
Scheme 2. Synthesis of Silanol and Silanediol Precursors 24a and 24b (Yields Shown Are for Diphenylsilane Series b)
Scheme 3. Hydrolysis of One Diastereomer of 24a Gives the Silanol as an Inseparable Mixture of Diastereomersa
a Capping the silanols 25 and silanediol 27 gave siloxanes 26 and 28 (the diastereomeric identity of 26a and 26b is not known).
9.47 On the basis of these experimental results, dichlorodiphe-
nylsilane 16b (Scheme 2) was chosen as starting material for
the silanediol synthesis.
by asymmetric reduction of the ketone, precedented in the work
of Buynak52 and Soderquist.53 Mesylation of the alcohol
followed by displacement of the mesylate with sodium azide
in DMF gave the R-azido silane in 84% overall yield from 20a.
Reduction of the azide with lithium aluminum hydride gave
the corresponding amine, which was coupled with benzoyl
chloride. Purification of the amide gave 22a in 74% yield from
the azide. Oxidative cleavage of the terminal alkene using
Weinreb’s mixture of Jones reagent and catalytic osmium
tetroxide54 was followed by coupling with benzylamine using
DEC and HOBt. Purification gave diamide 24a in 69% yield
from alkene 22a. The diastereomers of diamide 24a were
separated chromatographically and characterized individually.
Silanol Dipeptide Mimics. Before attempting to prepare the
silanediol tripeptide mimic 11, a dipeptide model system was
prepared in two forms, containing central methylsilanol and
silanediol groups. As a further simplifying step, Phe-[Si]-Gly
analogues (25 and 27, Scheme 3) were the first goals, instead
of Phe-[Si]-Ala systems. Elimination of the alanine methyl group
would leave a single stereogenic center in the silanediol
precursor 24b and two diastereomers for the methylsilanol
precursor 24a, Scheme 2. Nevertheless, these would be the most
complex silanediols ever prepared. The synthesis of 24a began
with conversion of dichloro(methyl)phenylsilane 16a to the more
reactive and yet more easily handled difluorosilane 17a, by
stirring with aqueous HF following the method of Eaborn.48
This difluoride was treated with 4-bromomagnesium-1-butene
to give 18a, followed by 2-lithio-1,3-dithiane, and the resulting
19a was isolated in 50% overall yield from 16a. Deprotonation
of 19a with n-butyllithium and alkylation with isobutylbromide
gave 20a in quantitative yield.49 Following the acylsilane
methodology of Brook50 and Corey,51 the dithiane 20a was
hydrolyzed with aqueous, unbuffered mercuric chloride. The
resulting ketone was reduced with lithium aluminum hydride
to give alcohol 21a as a mixture of diastereomers. A potential
advantage of an acylsilane intermediate was the potential for
absolute stereocontrol of the R-aminosilane stereogenic center
Dichlorodiphenylsilane 16b was taken through an identical
sequence to give 24b. The yields for each of these steps are
shown in Scheme 2.
The key hydrolysis reactions, taking 24a to silanol 25 and
24b to silanediol 27, were performed using an excess of triflic
acid in methylene chloride for 10 min at 0 °C, followed by
addition of ammonium hydroxide (Scheme 3). The diastereo-
mers of methylphenyl 24a were subjected to the hydrolysis
individually. The resulting methylsilanol 25, isolated in 81%
yield, was homogeneous by TLC, but was clearly a 3:2 mixture
of diastereomers by NMR. Although the diastereomers of silanol
25 could not be separated chromatographically, treatment of
25 with chlorotrimethylsilane and triethylamine55 gave two
chromatographically separable disiloxanes 26a and 26b in a ratio
(47) We thank Maria (Brito) Joshi for performing these experiments as an
undergraduate research project.
(48) Eaborn, C. J. Chem. Soc. 1952, 2846-2849.
(49) The use of benzyl bromide in this step led to the expected product, but it
could not be taken through to the final target and was abandoned.
(50) Brook, A. G.; Duff, J. M.; Jones, P. F.; Davis, N. R. J. Am. Chem. Soc.
1967, 89, 431-434.
(51) Corey, E. J.; Seebach, D.; Freedman, R. J. Am. Chem. Soc. 1967, 89, 434-
436.
(52) Buynak, J. D.; Strickland, J. B.; Lamb, G. W.; Khasnis, D.; Modi, S.;
Williams, D.; Zhang, H. M. J. Org. Chem. 1991, 56, 7076-7083.
(53) Soderquist, J. A.; Anderson, C. L.; Miranda, E. I.; Rivera, I.; Kabalka, G.
W. Tetrahedron Lett. 1990, 31, 4677-4680.
(54) Henry, J. R.; Weinreb, S. M. J. Org. Chem. 1993, 58, 4745.
(55) Brook, M. A.; Henry, C.; Jueschke, R.; Modi, P. Synlett 1993, 97-104.
Henry, C.; Brook, M. A. Tetrahedron 1994, 50, 11379-11390.
9
J. AM. CHEM. SOC. VOL. 124, NO. 25, 2002 7367

A R T I C L E S
Mutahi et al.
Scheme 4. Introduction of the Methyl Group by Alkylation of the Trianion of 23, Coupling with Proline, and Hydrolysis
of 3:2 (note that the diastereomeric identity of 24a, 26a, and
26b has not been determined).
more difficult to characterize. The integration of aromatic and
aliphatic absorbances in the proton NMR spectrum indicated
that the two phenyl groups attached to silicon had been cleaved,
but the resolution was poor, suggesting either intermolecular
association of the silanediols or oligomerization to give a
complex mixture of diastereomers. The identity of this species
was finalized by treatment with chlorotrimethylsilane/triethyl-
amine55 to give trisiloxane 28. This trisiloxane was isolated in
58% yield from the diphenylsilane 24b, nearly identical to the
two-step yield of disiloxanes 26 from 24a.
Protolytic cleavage of a phenylsilane would be expected to
invert stereochemistry at silicon on the basis of Corriu’s studies
with phenoxide nucleophiles,56,57 but the effect of a protonated
carbon nucleofuge and the acidic reaction conditions were
difficult to assess.58 Electrophilic reactions of Si-chiral allyl
silanes have been investigated, but the fate of the silicon chirality
appears to have gone undetermined.59 While inseparable, the
diastereomers of 25 were otherwise well defined spectroscopi-
cally.
Silanol Tripeptide Mimics. With the successful synthesis
and hydrolysis of the model systems 24, we turned our attention
to tripeptide mimics suitable for enzyme inhibition studies
(Scheme 4). Beginning with acids 23a and 23b, a DCC/HOBt-
mediated coupling with the tert-butyl ester of L-proline pro-
ceeded without incident to give 29a (58%) and 29b (78%).
Silane 29a was a mixture of four diastereomers, and 29b was
a mixture of two diastereomers. Each of these diastereomeric
mixtures was subjected to triflic-acid-mediated hydrolysis of
the phenylsilane bonds and tert-butyl ester cleavage: an excess
of triflic acid in methylene chloride at 0 °C for 25 min, followed
by dilution with ammonium hydroxide to a pH of 5-6, and
then addition of 6 N HCl to give a final pH of 1. The ammonium
hydroxide treatment was to ensure that any internal nucleophiles,
such as the amides, were hydrolyzed from the silicon center.
The silanol 30 and silanediol 31 solutions were then concentrated
and purified by HPLC.
The inseparability of the silanol 25 isomers may be because
of a facile epimerization of the silanol stereogenic center. In
his seminal studies of silanol chirality, Tacke found that
enantiomerically pure silanols were stable in aprotic solvents,
but in an aqueous environment racemization occurred within
minutes,60 and diastereomerically pure silanols were found to
rapidly epimerize at silicon in an aqueous environment.61 It is
intriguing to consider that the proximal amides in 25 may act
as internal nucleophiles and thereby participate in the epimer-
ization of the silanol. Solvents such as DMF are known to
racemize stereogenic silicon centers by coordination of the
carbonyl oxygen to silicon followed by pseudorotation of the
trigonal bipyramid intermediate.57,58 The rapid epimerization of
silanols may provide some advantage in drug design, as control
of this stereogenic center could therefore be ignored. In
bioassays of silanols, Tacke found that the stereochemistry of
the organosilanes had a lesser effect than that of the corre-
sponding carbinols.60
To introduce a methyl group adjacent to the carboxylic acid,
the trianion of 23 was employed. Three equivalents of LDA
was expected to remove the acid proton, the amide proton, and
then generate the acid enolate. Addition of iodomethane
generated R-methyl 32. Coupling of 32 with proline tert-butyl
ester using DIC and HOBt produced esters 33. Hydrolysis of
the phenylsilanes 33a and 33b with triflic acid, in a manner
identical to that employed for 29a and 29b, proceeded smoothly,
and the products 34 and 35 were isolated by HPLC. NMR data
for all of the proline amides were complicated by the presence
of amide rotamers and diastereomeric mixtures.
Triflic-acid-mediated hydrolysis of diphenylsilane 24b also
proceeded rapidly at 0 °C. After 10 min, ammonium hydroxide
was added, and the products were extracted into organic
solvents. This gave a product that was, in comparison to 25,
(56) Corriu, R. J. P.; Guerin, C. J. Organomet. Chem. 1982, 225, 141-150.
(57) Corriu, R. J. P.; Gue´rin, C.; Moreau, J. J. E. Top. Stereochem. 1984, 15,
43-198. Holmes, R. R. Chem. ReV. 1990, 90, 17-31. Holmes, R. R. Chem.
ReV. 1996, 96, 927-950.
(58) Bassindale, A. R.; Taylor, P. G. In The Chemistry of Organic Silicon
Compounds; Rappoport, Z., Apeloig, Y., Eds.; John Wiley & Sons: New
York, 1998; Vol. 2, pp 495-511.
(59) Daniels, R. G.; Paquette, L. A. Organometallics 1982, 1, 1449-1453.
Hathaway, S. J.; Paquette, L. A. J. Org. Chem. 1983, 48, 3351-3353.
(60) Tacke, R.; Linoh, H.; Ernst, L.; Moser, U.; Mutschler, E.; Sarge, S.;
Cammenga, H. K.; Lambrecht, G. Chem. Ber. 1987, 120, 1229-1237.
(61) Tacke, R.; Pikies, J.; Wiesenberger, F.; Ernst, L.; Schomburg, D.;
Waelbroeck, M.; Christophe, J.; Lambrecht, G.; Gross, J.; Mutschler, E. J.
Organomet. Chem. 1994, 466, 15-27.
The diastereomeric silanediols 31 (two isomers) and 35 (four
isomers) could not be adequately separated into individual
components, and so they were collected and tested as mixtures.
Analysis by MALDI-TOF MS gave no indication of oligomer
formation, with the exception of 35 in which a trace of dimer
9
7368 J. AM. CHEM. SOC. VOL. 124, NO. 25, 2002

Silanol Protease Inhibitors
A R T I C L E S
The IC50 values for 31 and 35 (different by a factor of 4)
roughly parallel the data for 36 and 10 (different by a factor of
3). These pairs of inhibitors, 31/35 and 36/10, differ by the
absence or presence of a methyl group. The parallel of these
inhibitory responses suggests that the binding of the silanediols
to ACE is similar to that of the ketones.
Conclusions
These results are consistent with silanediols that chelate the
active site zinc ion of ACE, leading to tight binding of these
structures, whereas the methylsilanols contribute little to the
active site binding and may even sterically inhibit it. Silanediols
have been found here to be an effective functional group with
which to design metalloprotease inhibitors. Early synthetic
difficulties led to the synthesis of silanols with an isobutyl group
where Almquist’s inhibitors have a benzyl group. The enantio-
selective synthesis of analogues of 35, with substitution that
will allow a direct comparison with Almquist’s inhibitors, will
be reported in due course. Ongoing investigations are also
anticipated to define the details of the binding mode of
silanediols to metalloprotease enzymes.
Silanediols can be potent inhibitors of aspartic proteases (the
HIV protease64) and metalloproteases (this work), two of the
four classes of protease enzymes. Ongoing applications of this
new technology to other enzymes and other protease classes
are expected to shed additional light on the breadth of its
applicability.
Figure 6. Enzyme inhibition of ACE by ketones 10 and 36 and four silicon
analogues, silanols 30 and 34, and silanediols 31 and 35.
could be detected. The more complex silanols 30 (four diaster-
eomers) and 34 (eight diastereomers) were separable by HPLC,
and were therefore collected and tested individually. MALDI-
TOF analysis of the methylsilanols 30 and 34 showed only
monomeric species.
Enzyme Assay. The inhibition of ACE was evaluated using
Holmquist’s modification62 of the spectrophotometric method
of Cushman and Cheung,63 employing commercially available
angiotensin-converting enzyme (Sigma) and N-[3-(2-furyl)-
acryloyl]-L-phenylalanylglycylglycine (FAPGG, Sigma) as a
profluorescent substrate. The assays were conducted in duplicate
for those inhibitors with IC50 values above 1000 nM, and in
triplicate for all others. A summary of the resulting data is shown
in Figure 6.
The methylsilanols 30 and 34 proved to be modest inhibitors
at best, and data reported in Figure 6 are for the most active
diastereomers. With IC50 values of 3-4 µM, these silanols are
rather far from useful levels of activity. This is not terribly
surprising, as carbinols have not been found to be good inhibitors
of metalloproteases.18 The methyl group on silicon may be a
steric impediment to binding, and the lone silanol oxygen, as a
single point of ligation for the zinc ion, would not be expected
to contribute significantly to the binding of the inhibitor.
Silanediols 31 and 35, capable of chelating a metal ion, were
substantially better inhibitors of ACE, with IC50 values of 57
and 14 nM, respectively. These two silanediols were mixtures
of two and four diastereomers, respectively. Almquist found
that high affinity binding of ketone 10 was critically dependent
on the configuration of the steregenic centers. The same was
found for the diastereomers of 30 and 34, with IC50 values for
the least active diastereomers above 10 µM. It is likely,
therefore, that most of the enzyme inhibition of the silanediols
31 and 35 is because of a single diastereomer. If one assumes
that the observed inhibition can be attributed to a single
diastereomer, the concentration required for 50% inhibition of
the enyzme could be as low as 29 nM for 31 and 3.5 nM for
35.
Experimental Section
Difluoromethylphenylsilane (17a).65 To a solution of dichlorom-
ethylphenylsilane 16a (11.8 g, 61.5 mmol) in ethanol (200 mL) at 0
°C was added dropwise over 10 min hydrofluoric acid (48 wt % in
water, 10 mL), and the mixture was allowed to warm to room
temperature. After being stirred overnight, the reaction mixture was
poured into water (500 mL), and the resulting mixture was extracted
twice with 100 mL portions of hexane. The combined organic extracts
were washed with saturated aqueous NaCl, dried over Na2SO4, and
concentrated. The colorless oil of crude 17a was dissolved in toluene
and concentrated under reduced pressure to remove any remaining
moisture. This product was used in the next reaction without further
1
purification. H NMR (CDCl3): δ 7.68 (d, J ) 7.5 Hz, 2H), 7.60-
7.55 (m, 1H), 7.48-7.44 (m, 2H), 0.62 (t, JHF ) 6.0 Hz, 3H).
3-Butenyl(fluoro)methylphenylsilane (18a). A solution of 1-bro-
momagnesium-3-butene in ether (100 mL, prepared from 9.13 g (67.7
mmol) of 1-bromo-3-butene) was added to 17a (9.72 g, 61.5 mmol) in
toluene (100 mL) at room temperature via cannula over 30 min. After
being stirred overnight at room temperature, the reaction mixture was
quenched with water (20 mL). The aqueous layer was extracted twice
with 30 mL portions of ethyl acetate. The combined organics were
washed with saturated NaCl solution and dried over Na2SO4. Concen-
tration in vacuo provided crude title compound as a colorless oil in
quantitative yield. This product was used in the next reaction without
1
further purification. Rf ) 0.75 (hexane). H NMR (CDCl3): δ 7.60
(m, 2H), 7.40 (m, 3H), 5.88 (m, 1H), 5.04 (d, J ) 17.0 Hz, 1H), 4.95
(d, J ) 10.0 Hz, 1H), 2.19 (m, 2H), 1.05 (m, 2H), 0.52 (d, JHF ) 6.0
Hz, 3H).
3-Butenyl(1,3-dithian-2-yl)methylphenylsilane (19a). To a solution
of 1,3-dithiane (11.1 g, 91.9 mmol) in THF (150 mL) at -78 °C was
added dropwise over 10 min n-butyllithium (48 mL of a 1.6 M solution
(64) Chen, C.-A.; Sieburth, S. McN.; Glekas, A.; Hewitt, G. W.; Trainor, G.
L.; Erickson-Viitanen, S.; Garber, S. S.; Cordova, B.; Jeffrey, S.; Klabe,
R. M. Chem. Biol. 2001, 8, 1161-1166.
(65) Marans, N. S.; Sommer, L. H.; Whitmore, F. C. J. Am. Chem. Soc. 1951,
73, 5127-5130.
(62) Holmquist, B.; Bu¨nning, P.; Riordan, J. F. Anal. Biochem. 1979, 95, 540-
548.
(63) Cushman, D. W.; Cheung, H. S. Biochem. Pharmacol. 1971, 20, 1637-
1648.
9
J. AM. CHEM. SOC. VOL. 124, NO. 25, 2002 7369

A R T I C L E S
Mutahi et al.
in hexane, 76.6 mmol), and the solution was stirred for 2 h. A solution
of 18a (11.9 g, 61.3 mmol) in THF (120 mL) was added dropwise
over 30 min, and the mixture was stirred for 3 h at -78 °C and
overnight at room temperature. The reaction mixture was diluted with
water (100 mL) and extracted with three 100 mL portions of ethyl
acetate. The combined organics were washed with saturated aqueous
NaCl, dried over Na2SO4, and concentrated. Flash chromatography over
silica gel (1:9 ethyl acetate/hexane) gave 19a mixed with residual 1,3-
dithiane. The latter was removed by sublimation (54 °C, 8.0 mmHg)
to provide pure 19a as a yellow oil (9.0 g, 50% for three steps from
(m, 2H), 7.39-7.34 (m, 3H), 5.95-5.82 (m, 1H), 5.03-4.89 (m, 2H),
3.70 (dd, J ) 12.0, 2.3 Hz, 1H), 2.15-2.07 (m, 2H), 1.86-1.76 (m,
1H), 1.59-1.49 (m, 2H), 1.23-1.12 (m, 1H), 1.03-0.84 (m, 7H), 0.33
and 0.34 (two singlets due to diastereomers, 3H). 13C NMR (CDCl3):
δ 141.5, 135.8, 135.6, 134.7, 134.6, 129.6, 129.5, 128.2, 128.1, 113.2,
62.5, 62.1, 42.3, 42.2, 27.5, 24.2, 24.1, 23.6, 20.7, 20.8, 10.7, 10.5,
-7.9, -8.0. IR (neat): 3434, 2954, 2913, 1111, 790, 736 cm-1. MS
(CI/CH4): m/e 263 (M+ + 1, 2), 175 (69), 137 (100), 115 (41). Anal.
Calcd for C16H26OSi: C, 73.22; H, 9.98. Found: C, 72.85; H, 10.24.
3-Butenyl(1-azido-3-methylbutyl)methylphenylsilane. To a solu-
tion of 21a (3.4 g, 13 mmol) in CH2Cl2 (150 mL) and triethylamine (9
mL, 65 mmol) at 0 °C was added dropwise over 5 min methanesulfonyl
chloride (7.4 g, 65 mmol), and the mixture was allowed to warm to
room temperature over 1 h. After being stirred overnight, the mixture
was cooled to 0 °C and diluted with water (50 mL). The aqueous layer
was extracted twice with 50 mL portions of CH2Cl2. The combined
organics were concentrated in vacuo, and the crude mesylate was
dissolved in DMF (150 mL). To this solution was added sodium azide
(4.2 g, 64.8 mmol), and, after being stirred overnight at room
temperature, the mixture was partitioned between water (200 mL) and
ethyl acetate (200 mL). The aqueous layer was extracted twice with
50 mL portions of ethyl acetate. The combined organics were washed
with saturated aqueous NaCl, dried over Na2SO4, and concentrated.
Flash chromatography over silica gel (1:19 ethyl acetate/hexane) gave
3-butenyl(1-azido-3-methylbutyl)methylphenylsilane as a colorless oil,
as a mixture of diastereomers (3.11 g, 84%). Rf ) 0.80 (1:9 ethyl
acetate/hexane). 1H NMR (CDCl3): δ 7.53-7.47 (m, 2H), 7.38-7.32
(m, 3H), 5.91-5.78 (m, 1H), 5.0-4.86 (m, 2H), 3.03 and 2.99 (two
triplets due to diastereomers, J ) 2.7 Hz, 1H), 2.08 (q, J ) 7.8 Hz,
2H), 1.82-1.70 (m, 1H), 1.58-1.47 (m, 1H), 1.19-1.10 (m, 1H), 1.06-
0.94 (m, 2H), 0.89-0.83 (m, 6H), 0.38 and 0.36 (two singlets due to
diastereomers, 3H). 13C NMR (CDCl3): δ 140.8, 134.2, 134.1, 129.6,
128.0, 113.3, 50.4, 50.3, 38.6, 38.5, 27.5, 25.8, 23.3, 20.7, 11.4, 11.0,
-6.8, -7.3. IR (neat): 2956, 2099, 1259 cm-1. MS (FAB): m/e 288
(MH+, 19), 260 (100), 258 (39), 244 (64), 204 (54). Exact mass (FAB)
calcd for C16H26N3Si, 288.1896; found, 288.1895.
1
dichlorosilane 16a). Rf ) 0.55 (1:9 ethyl acetate/hexane). H NMR
(CDCl3): δ 7.63-7.61 (m, 2H), 7.42-7.38 (m, 3H), 5.97-5.84 (m,
1H), 5.06-4.90 (m, 2H), 3.95 (s, 1H), 2.93-2.83 (m, 2H), 2.72-2.67
(m, 2H), 2.21-1.97 (m, 4H), 1.18-1.09 (m, 2H), 0.49 (s, 3H). 13C
NMR (CDCl3): δ 141.0, 134.5, 134.1, 129.9, 127.9, 113.2, 32.9, 31.1,
31.0, 27.2, 25.8, 10.8, -7.0. MS (FAB): m/e 293 (MH+, 23), 149 (58),
121 (100), 119 (37), 105 (34). Exact mass (FAB) calcd for C15H23S2-
Si, 293.0854; found, 293.0861. Anal. Calcd for C15H22S2Si: C, 61.16;
H, 7.53. Found: C, 61.31; H, 7.74.
3-Butenyl[2-(2-methylpropyl)-1,3-dithian-2-yl]methylphenylsi-
lane (20a). To a solution of 19a (6.81 g, 23.1 mmol) in THF (150
mL) at -78 °C was added dropwise over 10 min n-butyllithium (14.4
mL of a 1.6 M solution in hexanes, 23.1 mmol). After 3 h, 1-bromo-
2-methylpropane (3.52 mL, 32.3 mmol) was added dropwise, and the
mixture was stirred for 2 h at -78 °C and then overnight at room
temperature. The reaction mixture was quenched with water (10 mL)
and extracted with three 100 mL portions of ethyl acetate. The combined
organics were washed with saturated aqueous NaCl and dried over Na2-
SO4. Concentration gave 20a as a yellow oil in quantitative yield and
was used in the next reaction without purification. Rf ) 0.75 (1:19
ethyl acetate/hexane). 1H NMR (CDCl3): δ 7.69-7.66 (m, 2H), 7.40-
7.33 (m, 3H), 5.93-5.82 (m, 1H), 4.99 (dd, J ) 17.0, 1.7 Hz, 1H),
4.88 (dd, J ) 10.1, 1.7 Hz, 1H), 3.0-2.91 (m, 2H), 2.50-2.40 (m,
2H), 2.13-1.90 (m, 6H), 1.78-1.70 (m, 1H), 1.31-1.24 (m, 2H), 0.94
(d, J ) 6.6 Hz, 3H), 0.84 (d, J ) 6.6 Hz, 3H), 0.55 (s, 3H). 13C NMR
(CDCl3): δ 141.2, 135.1, 134.5, 129.5, 127.5, 112.9, 45.6, 39.5, 27.9,
27.0, 24.7, 24.6, 24.2, 24.1, 11.5, -6.1. IR (neat): 2896, 1430, 1254,
1114, 910, 798, 739, 700 cm-1. MS (CI/CH4): m/e 351 (MH+, 19),
295 (100), 273 (71), 175 (36), 149 (74). Exact mass (FAB) calcd for
C19H30S2Si, 350.1558 (MH+ - 1); found, 350.1561.
3-Butenyl-(3-methyl-1-oxobutyl)methylphenylsilane. To a solution
of 20a (6.19 g, 17.7 mmol) in acetonitrile (200 mL) were added water
(10 mL) and HgCl2 (24 g, 88 mmol). After being stirred overnight at
room temperature, the mixture was concentrated in vacuo and then
partitioned between water (100 mL) and hexane (200 mL). The aqueous
layer was extracted twice with 50 mL portions of hexane, and the
combined organic extracts were washed with saturated aqueous NaCl
and dried over Na2SO4. Concentration in vacuo at room temperature
gave the title compound as a yellow oil in quantitative yield. Rf )
0.75 (1:19 ethyl acetate/hexane). 1H NMR (CDCl3): δ 7.54-7.51 (m,
2H), 7.41-7.35 (m, 3H), 5.91-5.78 (m, 1H), 4.99 (dd, J ) 17.0, 1.7
Hz, 1H), 4.90 (dd, J ) 10.0, 1.7 Hz, 1H), 2.44 (d, J ) 6.7 Hz, 2H),
2.16-2.06 (m, 3H), 1.16-1.07 (m, 2H), 0.78 (pd, J ) 6.7 Hz, 6H),
0.51 (s, 3H).
N-[1-(3-Butenyl(methyl)phenylsilyl)-3-methylbutyl] Benzamide
(22a). A solution of 3-butenyl(1-azido-3-methylbutyl)methylphenylsi-
lane (2.75 g, 9.57 mmol) in ether (100 mL) at 0 °C was added dropwise
to lithium aluminum hydride (47.8 mL of a 1 M solution in ether, 47.8
mmol), and the mixture was then allowed to warm to room temperature
over 10 min. After being stirred for an additional 30 min, the mixture
was cooled to 0 °C and diluted with ether (100 mL). A saturated
aqueous Na2SO4 solution was added until evolution of hydrogen ceased,
the mixture was dried with anhydrous Na2SO4, filtered, and the residue
was extracted with ether (50 mL). The combined organics were
concentrated to give the amine as a colorless oil. This amine was taken
up in CH2Cl2 (60 mL) and triethylamine (5 mL), and the solution was
cooled to 0 °C. To this solution was added dropwise benzoyl chloride
(1.34 g, 9.57 mmol), and the mixture was allowed to warm to room
temperature. After being stirred overnight, the reaction mixture was
diluted with saturated aqueous NaHCO3 (20 mL). The aqueous layer
was extracted twice with 50 mL portions of CH2Cl2, and the combined
organics were washed with saturated aqueous NaCl, dried over Na2-
SO4, and concentrated. Flash chromatography over silica gel (1:9 ethyl
acetate/hexane) gave 22a, a mixture of diastereomers, as a sticky
3-Butenyl-(1-hydroxy-3-methylbutyl)methylphenylsilane (21a).
To a solution of 3-butenyl-(3-methyl-1-oxobutyl)methylphenylsilane
(4.60 g, 17.7 mmol) in ethyl ether (150 mL) at 0 °C was added dropwise
lithium aluminum hydride (88.3 mL of a 1 M solution in ether, 88.3
mmol). After being stirred for 15 min at 0 °C, the reaction mixture
was diluted with ethyl ether (200 mL), and saturated Na2SO4 solution
was added until evolution of hydrogen had ceased. The mixture was
dried with anhydrous Na2SO4, filtered, the residue was extracted with
ether (50 mL), and the organics were combined. Concentration provided
21a as a yellow oil that was a mixture of diastereomers (3.6 g, 78%).
This product was used in the next reaction without further purification.
Rf ) 0.50 (1:9 ethyl acetate/hexane). 1H NMR (CDCl3): δ 7.58-7.53
1
colorless solid (2.6 g, 74%). Rf ) 0.20 (1:9 ethyl acetate/hexane). H
NMR (CDCl3): δ 7.69-7.60 (m, 2H), 7.56-7.54 (m, 2H), 7.48-7.37
(m, 6H), 5.95-5.80 (m, 1H), 5.57 (d, J ) 9.9 Hz, 1H), 5.0 (d, J )
17.0 Hz, 1H), 4.90 (d, J ) 8.6 Hz, 1H), 4.24-4.13 (m, 1H), 2.17-
2.07 (m, 2H), 1.68-1.55 (m, 1H), 1.44-1.24 (m, 2H), 1.07-0.99 (m,
2H), 0.93 (d, J ) 6.4 Hz, 3H), 0.85 and 0.84 (two doublets due to
diastereomers, J ) 6.6 Hz, 3H), 0.40 and 0.39 (two singlets due to
diastereomers, 3H). 13C NMR (CDCl3): δ 166.7, 140.9, 135.0, 134.7,
134.5, 134.34, 134.31, 131.1, 131.0, 130.5, 129.64, 129.61, 128.8,
128.54, 128.50, 128.0, 126.6, 113.3, 113.2, 40.5, 40.4, 37.2, 36.9, 27.5,
9
7370 J. AM. CHEM. SOC. VOL. 124, NO. 25, 2002

Silanol Protease Inhibitors
A R T I C L E S
25.1, 23.6, 21.2, 11.3, 10.7, -6.7, -7.4. IR (neat): 3279, 2954, 1628,
1536, 699 cm-1. MS (FAB): m/e 366 (MH+, 11), 310 (34), 288 (100).
Exact mass (FAB) calcd for C23H31NOSi, 365.2175 (MH+ - 1); found,
365.2172. Anal. Calcd for C23H31NOSi: C, 75.56; H, 8.55; N; 3.83.
Found: C, 75.18; H, 8.41; N, 3.72.
C29H37N2O2Si, 473.2624; found, 473.2621. Anal. Calcd for C29H36N2O2-
Si‚0.2H2O: C, 73.13; H, 7.70; N, 5.88. Found: C, 73.09; H, 7.49; N,
5.96.
N-[1-[Hydroxy(methyl)[3-oxo-3-[(phenylmethyl)amino]propyl]si-
lyl]-3-methylbutyl Benzamide (25). To a solution of 24a-1 (88 mg,
0.18 mmol) in CH2Cl2 (30 mL) at 0 °C was added trifluoromethane-
sulfonic acid (3 mL, 34 mmol). After the solution was stirred for 10
min, saturated aqueous NH4OH (30 mL) was added. The aqueous layer
was extracted twice with 5 mL portions of CH2Cl2. The combined
organics were washed with saturated aqueous NaCl, dried over Na2-
SO4, and concentrated at room temperature. The crude product was
passed through a pad of silica gel using ethyl acetate as eluant.
Concentration provided 25 as a colorless solid (61 mg, 81%). The
3-[[1-(Benzoylamino)-3-methylbutyl]methylphenylsilyl] Propionic
Acid (23a). To a solution of 22a (0.7 g, 1.91 mmol) in acetone (22
mL) were added 0.23 mL of a 4 wt % solution of OsO4 in water (2
mol %) and Jones reagent (2.43 mL, 6.49 mmol).54 After the mixture
was stirred for 24 h at room temperature, 2-propanol (0.5 mL) was
added followed by NaHSO3 (0.2 g). The mixture was diluted with water
(50 mL) and stirred until a dark-green, homogeneous solution was
produced. This solution was diluted further with water (90 mL) and
extracted with six 50 mL portions of ethyl acetate. The combined
organics were washed with saturated aqueous NaCl and dried over Na2-
SO4. Concentration gave a colorless solid of crude 23a as a mixture of
diastereomers. This product was used in the next reaction without further
1
diastereomeric ratio was determined by H NMR to be 1:1.6. Rf )
1
0.10 (1:1 ethyl acetate/hexane). mp 35-36 °C. H NMR (CDCl3): δ
7.72-7.67 (m, 2H), 7.45-7.13 (m, 8H), 6.73 and 6.60 (two doublets
due to diastereomers, J ) 8.3 Hz, 1NH), 6.35-6.24 (m, 1NH), 4.39-
4.24 (m, 3H), 3.67-3.55 (m, 1H), 2.42-2.30 (m, 2H), 1.73-1.60 (m,
2H), 1.38-1.17 (m, 2H), 0.89 (d, J ) 6.4 Hz, 6H), 0.14 and 0.05 (two
singlets due to diastereomers, 3H). 13C NMR (CDCl3): δ 174.7, 167.8,
167.6, 164.2, 138.1, 138.0, 134.7, 134.6, 131.3, 131.2, 128.7, 128.6,
128.5, 127.8, 127.72, 127.69, 127.50, 127.47, 127.4, 126.9, 43.7, 43.6,
40.0, 39.7, 39.0, 38.9, 30.3, 30.1, 25.4, 25.3, 23.6, 21.3, 10.9, 10.0,
1.0, -3.1, -3.3. IR (film): 3288, 2954, 2926, 1636, 1543, 1262, 888,
879, 804, 704 cm-1. MS (FAB): m/e 435 (MNa+, 100), 101 (92). Exact
mass (FAB) calcd for C23H32N2O3Si‚Na, 435.2080; found, 435.2071.
N-[1-[Trimethylsilyloxy(methyl)[3-oxo-3-[(phenylmethyl)amino]-
propyl]silyl]-3-methylbutyl Benzamide (26). To a solution of 33 (75
mg, 0.18 mmol) in THF (10 mL) at room temperature were added
triethylamine (1 mL) and chlorotrimethylsilane (1 mL), and the mixture
was stirred for 20 min. This mixture was concentrated in vacuo and
then partitioned between water (5 mL) and CH2Cl2 (5 mL). The aqueous
layer was extracted twice with 5 mL portions of CH2Cl2. The combined
organics were washed with saturated aqueous NaCl, dried over Na2-
SO4, and concentrated. Purification by preparative thin-layer chroma-
tography (1:3 ethyl acetate/hexane) gave two diastereomers as viscous
semisolids, in a ratio of 4:6.
1
purification. Rf ) 0.20 (1:9 ethyl acetate/hexane). H NMR (CDCl3):
δ 7.67-7.62 (m, 2H), 7.54-7.52 (m, 2H), 7.48-7.37 (m, 6H), 5.79
and 5.73 (two doublets due to diastereomers, J ) 10.0 Hz, 1H), 4.25-
4.13 (m, 1H), 2.44-2.34 (m, 2H), 1.64-1.57 (m, 1H), 1.49-1.38 (m,
1H), 1.34-1.21 (m, 3H), 1.03, 0.98, 0.90 and 0.84 (four doublets due
to diastereomers, J ) 6.6 Hz, 6H), 0.40 (s, 3H). MS (FAB): m/e 384
(MH+, 62), 310 (100), 306 (88). Exact mass (FAB) calcd for C22H30-
NO3Si, 384.1995; found, 384.1995.
N-[3-Methyl-1-[[3-oxo-3-[(phenylmethyl)amino]propyl]meth-
ylphenylsilyl]butyl] Benzamide (24a). To a solution of benzylamine
(0.25 g, 2.29 mmol) in DMF (25 mL) at 0 °C were added 4-methyl-
morpholine (0.21 mL, 1.91 mmol), 1-(3-dimethylaminopropyl)-3-
ethylcarbodiimide hydrochloride (0.55 g, 2.87 mmol), 1-hydroxyben-
zotriazole (0.26 g, 1.91 mmol), and crude carboxylic acid 23a (1.91
mmol). After being stirred for 30 min at 0 °C under argon, the mixture
was allowed to warm to room temperature and was stirred overnight.
This mixture was partitioned between water (30 mL) and ethyl acetate
(30 mL). The aqueous layer was extracted twice with 30 mL portions
of ethyl acetate. The combined organics were washed with saturated
aqueous NaHCO3 (20 mL) and saturated aqueous NaCl, dried over Na2-
SO4, and concentrated. Flash chromatography over silica gel (1:1:3 ethyl
acetate/hexane/CH2Cl2) gave two diastereomers, 24a-1 and 24a-2, in
a ratio of 3:7 (69% from 22a).
1
26a. 21 mg, 29%, Rf ) 0.25 (1:3 ethyl acetate/hexane). H NMR
(CDCl3): δ 7.72 (d, J ) 17.2 Hz, 2H), 7.49-7.22 (m, 8H), 6.58 (d, J
) 9.7 Hz, 1H), 6.37 (t, J ) 4.8 Hz, 1H), 4.38 (d, J ) 5.6 Hz, 2H),
3.86 (m, 1H), 2.45-2.34 (m, 1H), 2.31-2.20 (m, 1H), 1.75-1.66 (m,
1H), 1.55 (dt, J ) 14.1, 3.8 Hz, 1H), 1.38-1.29 (m, 1H), 1.26-0.86
(m, 8H), 0.14 (s, 3H), 0.10 (s, 9H). 13C NMR (CDCl3): δ 174.4, 166.8,
138.2, 134.7, 131.1, 128.6, 128.5, 127.7, 127.3, 126.7, 43.6, 39.3, 38.3,
30.1, 25.2, 23.6, 21.3, 11.0, 1.9, -2.9. IR (neat): 3277, 2955, 1635,
1544, 1254, 1066, 848, 703 cm-1. MS (FAB): m/e 507 (MNa+, 100),
395 (55), 294 (50). Exact mass (FAB) calcd for C26H40N2O3Si2‚Na,
507.2475; found, 507.2473.
24a-1. 0.18 g, 20%, colorless crystals, Rf ) 0.50 (1:1 ethyl acetate/
1
hexane). mp 42-44 °C. H NMR (CDCl3): δ 7.60 (d, J ) 7.3 Hz,
2H), 7.53-7.25 (m, 13H), 6.56 (t, J ) 5.3 Hz, 1H), 6.13 (d, J ) 10.1
Hz, 1H), 4.40 (d, J ) 5.7 Hz, 2H), 4.41 (m, 1H), 2.63-2.52 (m, 1H),
2.37-2.27 (m, 1H), 1.65-1.55 (m, 1H), 1.45 (dt, J ) 14.1, 3.6 Hz,
1H), 1.35-1.21 (m, 3H), 0.88 (d, J ) 6.4 Hz, 3H), 0.85 (d, J ) 6.6
Hz, 3H), 0.37 (s, 3H). 13C NMR (CDCl3): δ 174.9, 167.1, 138.5, 134.8,
134.6, 134.4, 131.4, 129.9, 128.8, 128.7, 128.3, 128.0, 127.4, 126.8,
43.5, 39.4, 36.6, 30.6, 25.0, 23.3, 20.9, 7.6, -7.1. IR (film): 3284,
1636, 1539, 698 cm-1. MS (FAB): m/e 473 (MH+, 58), 395 (72), 282
(100), 157 (72), 91 (68). Exact mass (FAB) calcd for C29H37N2O2Si,
473.2624; found, 473.2625. Anal. Calcd for C29H36N2O2Si‚H2O: C,
70.98; H, 7.81; N, 5.71. Found: C, 71.20; H, 7.49; N, 5.76.
1
26b. 33 mg, 46%, Rf ) 0.20 (1:3 ethyl acetate/hexane). H NMR
(CDCl3): δ 7.62 (d, J ) 7.1 Hz, 2H), 7.44-7.18 (m, 8H), 6.38 (t, J )
5.5 Hz, 1H), 6.27 (d, J ) 9.2 Hz, 1H), 4.34 (d, J ) 5.4 Hz, 2H), 3.88-
3.80 (m, 1H), 2.47-2.37 (m, 1H), 2.26-2.15 (m, 1H), 1.73-1.6 (m,
1H), 1.49 (dt, J ) 14.2, 4.0 Hz, 1H), 1.35-1.26 (m, 1H), 1.02-0.76
(m, 8H), 0.09 (s, 3H), 0.04 (s, 9H). 13C NMR (CDCl3): δ 174.5, 167.0,
138.3, 134.7, 131.1, 128.6, 128.5, 127.8, 127.3, 126.6, 43.5, 39.4, 38.2,
30.2, 25.3, 23.6, 21.3, 11.2, 1.9, -3.1. IR (film): 3278, 2954, 1633,
1547, 1259, 1074, 846, 704 cm-1. MS (FAB): m/e 507 (MNa+, 31),
395 (100), 294 (35). Exact mass (FAB) calcd for C26H40N2O3Si2‚Na,
507.2475; found, 507.2478.
3-Butenyl(fluoro)diphenylsilane (18b). In a two-neck flask equipped
with a condenser were placed magnesium (3.46 g, 142 mmol) and a
crystal of iodine. The flask was warmed with a heat gun until iodine
had sublimed. A solution of 4-bromo-1-butene (9.62 g, 71.3 mmol) in
ether (100 mL) was added dropwise in 30 min, and the resulting mixture
was refluxed for 2 h. This Grignard solution was cooled to room
temperature and added over 30 min via cannula to a second flask
24a-2. 0.44 g, 49%, colorless powder, Rf ) 0.45 (1:1 ethyl acetate/
1
hexane). mp 190-191 °C. H NMR (CDCl3): δ 7.66 (d, J ) 7.4 Hz,
2H), 7.54-7.21 (m, 13H), 6.10-6.06 (m, 2H), 4.37 (d, J ) 5.6 Hz,
2H), 4.32-4.24 (m, 1H), 2.45-2.35 (m, 1H), 2.24-2.13 (m, 1H), 1.68-
1.59 (m, 1H), 1.48 (dt, J ) 14.4, 3.5 Hz, 1H), 1.33-1.15 (m, 3H),
0.91 (d, J ) 6.5 Hz, 3H), 0.85 (d, J ) 6.5 Hz, 3H), 0.37 (s, 3H). 13C
NMR (CDCl3): δ 174.5, 167.4, 138.6, 135.0, 134.6, 134.5, 131.5, 130.0,
128.9, 128.8, 128.4, 128.1, 127.6, 126.9, 43.6, 40.0, 36.6, 30.7, 25.1,
23.5, 21.0, 8.7, -7.4. IR (Nujol): 3261, 2949, 2922, 2854, 1652, 1628,
1559, 699 cm-1. MS (FAB): m/e 473 (MH+, 32), 395 (47), 282 (58),
157 (100), 137 (48), 93 (42), 91 (46). Exact mass (FAB) calcd for
9
J. AM. CHEM. SOC. VOL. 124, NO. 25, 2002 7371

A R T I C L E S
Mutahi et al.
containing 17b (15.7 g, 71.3 mmol) in ether (100 mL) at room
temperature. After being stirred overnight at room temperature under
argon, the reaction mixture was quenched with water (20 mL), and the
organic layer was isolated. The aqueous layer was extracted twice with
50 mL portions of ether. The combined organic extracts were washed
with saturated aqueous NaCl, dried over Na2SO4, and concentrated.
Distillation (134 °C, 1.5 mmHg) provided pure 18b as a colorless oil
The combined organic extracts were washed with saturated aqueous
NaCl and dried over Na2SO4. Concentration to dryness in vacuo at
room temperature gave the crude title compound as a yellow oil (3.2
1
g, 88%). Rf ) 0.80 (1:9 ethyl acetate/hexane). H NMR (CDCl3): δ
7.58 (d, J ) 6.4 Hz, 4H), 7.44 (m, 6H), 5.88 (m, 1H), 4.99 (dd, J )
17.1, 1.6 Hz, 1H), 4.91 (dd, J ) 10.2, 1.6 Hz, 1H), 2.50 (d, J ) 6.6
Hz, 2H), 2.14 (m, 2H), 1.36 (m, 2H), 0.95 (m, 1H), 0.76 (d, J ) 6.7
Hz, 6H). IR (neat): 3068, 3046, 2956, 1641, 1431, 1113, 743, 702
cm-1. MS (FAB): m/e (rel intensity) 323 (MH+, 15), 321 (89), 236
(36), 182 (63), 169 (100). Exact mass (FAB) calcd for C21H25OSi,
321.1675 (MH+ - 2); found, 321.1670.
1
(15.2 g, 83%). H NMR (CDCl3): δ 7.62 (d, J ) 7.8 Hz, 4H), 7.50-
7.39 (m, 6H), 5.97-5.83 (m, 1H), 5.02 (dd, J ) 17.1, 1.7 Hz, 1H),
4.94 (dd, J ) 10.2, 1.7 Hz, 1H), 2.28-2.20 (m, 2H), 1.38-1.30 (m,
2H). 13C NMR (CDCl3): δ 140.9, 135.9, 134.1, 129.7, 127.8, 113.1,
26.9, 14.0.
3-Butenyl(1-hydroxy-3-methylbutyl)diphenylsilane (21b). To a
solution of the 3-butenyl(3-methyl-1-oxobutyl)diphenylsilane (3.0 g,
9.3 mmol) in ethyl ether (100 mL) at 0 °C was added lithium aluminum
hydride (1.0 M in ethyl ether, 47 mmol). After being stirred for 15
min at 0 °C under argon, the reaction mixture was diluted with ethyl
ether (300 mL) and quenched with saturated aqueous Na2SO4 until
evolution of hydrogen had ceased. The mixture was dried with solid
Na2SO4 and filtered, and the residue was extracted with ether (50 mL).
The organic extracts were combined and concentrated in vacuo. Flash
chromatography over silica gel (1:9 ethyl acetate/hexane) gave pure
21b as a colorless oil (2.1 g, 69%). Rf ) 0.60 (1:9 ethyl acetate/hexane).
1H NMR (CDCl3): δ 7.57 (m, 4H), 7.37 (m, 6H), 5.87 (m, 1H), 4.97
(dd, J ) 17.1, 1.7 Hz, 1H), 4.88 (dd, J ) 10.2, 1.7 Hz, 1H), 4.08 (d,
J ) 12.1 Hz, 1H), 2.12 (m, 2H), 1.85 (m, 1H), 1.58 (m, 1H), 1.26 (m,
4H), 0.88 (d, J ) 6.2 Hz, 6H). 13C NMR (CDCl3): δ 141.4, 135.7,
133.8, 129.8, 128.2, 113.3, 61.5, 42.4, 27.5, 24.2, 23.7, 20.7, 9.8. IR
3-Butenyl(1,3-dithian-2-yl)diphenylsilane (19b). To a solution of
1,3-dithiane (6.77 g, 56.3 mmol) in THF (120 mL) at -78 °C was
added dropwise over 10 min n-butyllithium (1.6 M in hexane, 50 mmol),
and the solution was stirred for 2 h under argon. A solution of 18b
(11.1 g, 43.3 mmol) in THF (100 mL) was added, and the mixture
was stirred for 3 h at -78 °C and overnight at room temperature. The
reaction mixture was quenched with water (100 mL), and the organic
layer was isolated. The aqueous layer was extracted with two 100 mL
portions of ethyl ether. The combined organic extracts were washed
with saturated aqueous NaCl, dried over Na2SO4, and concentrated.
Flash chromatography over silica gel (1:9 ethyl acetate/hexane) gave
19b contaminated with 1,3-dithiane. The latter was removed by
sublimation (54 °C, 8.0 mmHg). Recrystallization from Et2O provided
pure 19b as a colorless solid (14.8 g, 96%). Rf ) 0.40 (1:49 ethyl
acetate/hexane). mp 47-49 °C. 1H NMR (CDCl3): δ 7.66 (d, J ) 7.8
Hz, 4H), 7.43 (m, 6H), 5.90 (m, 1H), 5.02 (dd, J ) 17.1, 1.7 Hz, 1H),
4.91 (dd, J ) 10.2, 1.7 Hz, 1H), 4.26 (s, 1H), 2.92 (t, J ) 11.8 Hz,
2H), 2.71 (m, 2H), 2.19 (m, 2H), 2.07 (m, 2H), 1.37 (m, 2H). 13C NMR
(CDCl3): δ 141.1, 135.9, 132.4, 130.2, 128.0, 113.2, 32.3, 31.5, 27.5,
25.8, 10.5. IR (film): 3068, 3047, 2897, 1639, 1429, 1112, 911, 744,
703 cm-1. MS (FAB): m/e (rel intensity) 357 (MH+, 11), 237 (36),
(neat): 3565, 3452, 2955, 2910, 1642, 1434, 1116, 743, 705 cm-1
.
MS (FAB): m/e (rel intensity) 323 (MH+ - 2, 5), 269 (21), 198 (57),
183 (100), 159 (56). Exact mass (FAB) calcd for C21H27OSi, 323.1831
(MH+ - 2); found, 323.1824. Anal. Calcd for C21H28SiO: C, 77.72;
H, 8.70. Found: C, 77.60; H, 8.67.
3-Butenyl(1-azido-3-methylbutyl)diphenylsilane. To a solution of
21b (1.68 g, 5.16 mmol) in methylene chloride (100 mL) and
triethylamine (3.6 mL) at 0 °C was added dropwise over 5 min
methanesulfonyl chloride (2.96 g, 25.8 mmol), and the solution was
allowed to warm to room temperature over 1 h. After being stirred
overnight under argon, the mixture was cooled to 0 °C and quenched
with water (50 mL). The organic layer was isolated, and the aqueous
layer was extracted twice with 20 mL portions of methylene chloride.
The combined organic extracts were concentrated in vacuo at room
temperature. The crude mesylate was dissolved in DMF (100 mL), and
to this solution was added sodium azide (1.68 g, 25.8 mmol). After
being stirred for 8 h at room temperature, the mixture was partitioned
between water (200 mL) and ethyl acetate (200 mL). The organic layer
was isolated, and the aqueous layer was extracted twice with 50 mL
portions of ethyl acetate. The combined organic extracts were washed
with saturated aqueous NaCl, dried over Na2SO4, and concentrated.
Flash chromatography over silica gel (1:9 ethyl acetate/hexane) gave
the pure title compound as a colorless oil (1.57 g, 87%). Rf ) 0.90
211 (100). Exact mass (FAB) calcd for C20H23S2Si, 355.1010 (MH+
-
2); found, 355.1020. Anal. Calcd for C20H24S2Si: C, 67.36; H, 6.78.
Found: C, 67.08; H, 6.78.
3-Butenyl[2-(2-methylpropyl)-1,3-dithan-2-yl]diphenylsilane (20b).
To a solution of 19b (4.68 g, 13.1 mmol) in THF (100 mL) at -78 °C
was added dropwise over 10 min n-butyllithium (1.6 M in hexanes,
18.4 mmol). After the mixture was stirred for 3 h under argon, 1-bromo-
2-methylpropane (2.14 mL, 19.7 mmol) was added dropwise over 5
min, and the mixture was stirred for 2 h at -78 °C and overnight at
room temperature. The reaction mixture was quenched with water (10
mL), and the excess organic solvent was removed under reduced
pressure. The crude mixture was extracted with three 100 mL portions
of ethyl acetate. The combined organic extracts were washed with
saturated aqueous NaCl, dried over Na2SO4, and concentrated. Flash
chromatography over silica gel (1:9 ethyl acetate/hexane) gave pure
20b as a colorless solid (5.0 g, 92%). Rf ) 0.75 (1:9 ethyl acetate/
1
hexane). mp 67-69 °C. H NMR (CDCl3): δ 7.83 (d, J ) 6.0 Hz,
1
4H), 7.41 (m, 6H), 5.88 (m, 1H), 4.97 (dd, J ) 17.1, 1.7 Hz, 1H), 4.88
(dd, J ) 10.1, 1.7 Hz, 1H), 3.04 (m, 2H), 2.49 (t, J ) 4.3 Hz, 1H),
2.44 (t, J ) 4.1 Hz, 1H), 2.08 (d, J ) 5.1 Hz, 2H), 1.98 (m, 4H), 1.79
(m, 1H), 1.54 (m, 2H), 0.82 (d, J ) 6.7 Hz, 6H). 13C NMR (CDCl3):
δ 141.5, 136.7, 132.9, 129.9, 127.7, 113.0, 45.7, 39.9, 28.3, 27.2, 24.6,
24.3, 24.2, 11.0. IR (film): 3068, 3043, 2951, 2912, 1432, 1113, 738,
705 cm-1. MS (EI): m/e (rel intensity) 289 (M+ - 123, 3), 183 (72),
175 (100), 159 (32), 105 (35). Exact mass (FAB) calcd for C24H32S2-
Si, 412.1715 (MH+ - 1); found, 412.1727. Anal. Calcd for C24H32S2-
Si: C, 69.84; H, 7.82. Found: C, 70.08; H, 8.07.
(1:9 ethyl acetate/hexane). H NMR (CDCl3): δ 7.57 (d, J ) 6.7 Hz,
4H), 7.41 (m, 6H), 5.88 (m, 1H), 5.00 (dd, J ) 17.0, 1.5 Hz, 1H), 4.92
(dd, J ) 10.2, 1.5 Hz, 1H), 3.44 (d, J ) 12.4 Hz, 1H), 2.12 (m, 2H),
1.84 (m, 1H), 1.60 (t, J ) 12.5 Hz, 1H), 1.30 (m, 3H), 0.92 (d, J ) 6.7
Hz, 3H), 0.89 (d, J ) 6.7 Hz, 3H). 13C NMR (CDCl3): δ 140.7, 135.3,
132.6, 129.9, 128.0, 113.3, 49.2, 38.8, 27.5, 25.9, 23.3, 20.7, 10.5. IR
(neat): 2957, 2102, 1432, 1116, 743, 701 cm-1. MS (FAB): m/e (rel
intensity) 350 (MH+, 18), 322 (100), 315 (42), 306 (42), 236 (73).
Exact mass (FAB) calcd for C21H28N3Si, 350.2053; found, 350.2040.
Anal. Calcd for C21H27N3Si: C, 72.16; H, 7.79; N, 12.02. Found: C,
71.99; H, 7.91; N, 11.60.
3-Butenyl(3-methyl-1-oxobutyl)diphenylsilane. To a solution of
20b (4.64 g, 11.2 mmol) in CH3CN (300 mL) were added water (10
mL) and HgCl2 (15.26 g, 56.21 mmol). After being stirred overnight
at room temperature, the mixture was concentrated and partitioned
between water (100 mL) and hexane (200 mL). The organic layer was
isolated, and the aqueous layer was extracted with hexane (50 mL).
N-[1-(3-Butenyldiphenylsilyl)-3-methylbutyl] Benzamide (22b).
To a solution of 3-butenyl(1-azido-3-methylbutyl)diphenylsilane (1.3
g, 3.72 mmol) in ethyl ether (50 mL) at 0 °C was added dropwise over
5 min lithium aluminum hydride (1 M in ether, 18.6 mmol), and the
mixture was allowed to warm to room temperature over 10 min. After
9
7372 J. AM. CHEM. SOC. VOL. 124, NO. 25, 2002

Silanol Protease Inhibitors
A R T I C L E S
1.45 (m, 4H), 0.98 (d, J ) 6.5 Hz, 3H), 0.82 (d, J ) 6.5 Hz, 3H). 13
being stirredfor 30 min under argon, the reaction mixture was cooled
to 0 °C and quenched successively with water (0.7 mL), 15% NaOH
in water (0.7 mL), and water (2.1 mL). The mixture was filtered, and
the residue was extracted twice with 20 mL portions of ethyl ether.
The combined organic extracts were washed with saturated aqueous
NaCl and dried over Na2SO4. Concentration in vacuo gave quantitatively
the crude amine as a colorless oil. This amine was dissolved in
methylene chloride (30 mL) and triethylamine (5.0 mL), and the solution
was cooled to 0 °C. To this solution was added dropwise over 5 min
benzoyl chloride (0.52 g, 3.72 mmol), and the mixture was allowed to
warm to room temperature. After being stirred overnight under argon,
the reaction mixture was quenched with 10% aqueous K2CO3 (20 mL).
The organic layer was isolated, and the aqueous layer was extracted
twice with 50 mL portions of CH2Cl2. The combined organic extracts
were washed with saturated aqueous NaCl, dried over Na2SO4, and
concentrated. Column chromatography over silica gel (1:9 ethyl acetate/
hexane) gave pure amide 22b as a colorless crystalline solid (1.40 g,
88%). Rf ) 0.25 (1:9 ethyl acetate/hexane). mp 99-101 °C. 1H NMR
(CDCl3): δ 7.61 (m, 6H), 7.42 (m, 9H), 5.83 (m, 1H), 5.55 (d, J )
10.3 Hz, 1H), 4.95 (dd, J ) 17.1, 1.6 Hz, 1H), 4.87 (dd, J ) 10.2, 1.6
Hz, 1H), 4.70 (dt, J ) 11.3, 3.1 Hz, 1H), 2.08 (m, 2H), 1.64 (m, 1H),
1.39 (m, 2H), 1.30 (br t, J )7.3 Hz, 2H), 1.03 (d, J ) 6.5 Hz, 3H),
0.74 (t, J ) 6.5 Hz, 3H). 13C NMR (CDCl3): δ 167.1, 141.1, 135.8,
135.7, 135.2, 133.3, 132.5, 131.3, 130.1, 128.7, 128.4, 128.4, 126.8,
113.4, 40.9, 34.8, 27.2, 24.9, 23.6, 21.2, 10.6. IR (film): 3302, 2955,
2923, 1640, 1519, 1487, 1429, 1322, 1114, 744, 706 cm-1. MS
(FAB): m/e (rel intensity) 428 (MH+, 8), 372 (22), 351 (34), 350 (100).
Exact mass (FAB) calcd for C28H34NOSi, 428.2410; found, 428.2411.
Anal. Calcd for C28H33NSiO: C, 78.64; H, 7.77; N, 3.28. Found: C,
78.28; H, 7.96; N, 3.22.
C
NMR (CDCl3): δ 174.8, 167.3, 138.4, 135.6, 135.4, 134.6, 132.4, 131.4,
130.3, 130.2, 128.7, 128.4, 127.9, 127.4, 126.8, 43.4, 40.0, 34.9, 30.5,
25.0, 23.3, 20.9, 7.6. IR (film): 3280, 2951, 2925, 1636, 1541, 1114,
700 cm-1. MS (FAB): m/e (rel intensity) 535 (MH+, 30), 457 (87),
344 (100). Exact mass (FAB) calcd for C34H38N2O2Si, 534.7724 (MH+
- 1); found, 534.7719. Anal. Calcd for C34H38N2SiO2: C, 76.36; H,
7.16; N, 5.24. Found: C, 76.21; H, 7.19; N, 5.23.
2-[1-(Benzoylamino)-3-methylbutyl]-1,1,1,5,5,5-hexamethyl-2-[3-
oxo-3-[(phenylmethyl)amino]propyl]silyl]trisiloxane (28). To a solu-
tion of 24b (20 mg, 0.037 mmol) in methylene chloride (4.6 mL) at 0
°C was added triflic acid (0.44 mL, 5.0 mmol). After being stirred for
10 min at 0 °C under nitrogen, the reaction mixture was diluted with
methylene chloride (20 mL) and transferred via cannula to a second
flask containing saturated aqueous NaCl (20 mL) and NaHCO3 (0.82
g, 10.0 mmol) at 0 °C. This mixture was stirred for 15 min at 0 °C,
and the organic layer was isolated and dried with Na2SO4. The organic
solution was cooled to 0 °C under nitrogen, and treated successively
with TMSCl (3 mL) and triethylamine (2 mL). After being stirred for
30 min at °C, the mixture was quenched with water (10 mL). The
organic layer was isolated and washed with saturated aqueous NaCl,
dried over Na2SO4, and concentrated. Flash chromatography over silica
gel (1:2 ethyl acetate/hexane) gave trisiloxane 28 as a thick colorless
oil (12 mg, 58%). Rf ) 0.50 (1:3 ethyl acetate/hexane). 1H NMR
(CDCl3): δ 7.65 (d, J ) 7.1 Hz, 2H), 7.45-7.19 (m, 8H), 6.36-6.33
(m, 2H), 4.35 (d, J ) 5.7 Hz, 2H), 3.84-3.76 (m, 1H), 2.45-2.33 (m,
1H), 2.25-2.15 (m, 1H), 1.70-1.62 (m, 1H), 1.51-1.41 (m, 1H), 1.37-
1.28 (m, 1H), 0.96-0.79 (m, 8H), 0.09 (s, 9H), 0.07 (s, 9H). 13C NMR
(CDCl3): δ 174.4, 166.8, 138.4, 134.8, 131.1, 128.6, 128.5, 127.8,
127.3, 126.7, 43.6, 39.4, 37.7, 30.1, 25.2, 23.7, 21.5, 10.5, 2.0, 1.9. IR
(neat): 3272, 2956, 1660, 1638, 1539, 1255, 1080, 845 cm-1. MS
(FAB): m/e (rel intensity) 581 (MNa+, 48), 469 (99), 368 (100), 207
(51). Exact mass (FAB) calcd for C28H46N2O4Si3‚Na, 581.2662; found,
581.2662.
3-[[1-(Benzoylamino)-1-(3-methylbutyl)]diphenylsilyl] Propanoic
Acid (23b). To a solution of olefin 22b (0.63 g, 1.46 mmol) in acetone
(23 mL) were added 0.18 mL (2 mol %) of a 4 wt % solution of OsO4
in water and Jones reagent (1.89 mL, 5.05 mmol). After the mixture
was stirred for 24 h at room temperature, 2-propanol (0.73 mL) was
added followed by NaHSO3 (0.22 g). The mixture was diluted with
water (45 mL) and stirred until a dark-green, homogeneous solution
was produced. This solution was diluted with water (90 mL) and
extracted with six 50 mL portions of ethyl acetate. The combined
organic extracts were dried over MgSO4. Concentration in vacuo gave
crude carboxylic acid 23b as a colorless solid (0.62 g, 96%). Rf )
0.40 (ethyl acetate). 1H NMR (CDCl3): δ 7.50 (m, 6H), 7.35 (m, 9H),
5.61 (d, J ) 10.2 Hz, 1H), 4.63 (dt, J ) 10.8, 3.5 Hz, 1H), 2.40 (m,
1H), 2.21 (m, 1H), 1.55 (m, 1H), 1.36 (m, 4H), 0.94 (d, J ) 6.4 Hz,
3H), 0.76 (d, J ) 6.5 Hz, 3H). IR (film): 3320, 2957, 2925, 1713,
1630, 1536, 1430, 1114, 707 cm-1. MS (FAB): m/e (rel intensity) 446
(MH+, 13), 372 (100), 368 (30), 199 (20). Exact mass (FAB) calcd for
C27H32NO3Si, 446.2151; found, 446.2159.
1-[3-[[1-(Benzoylamino)-3-methylbutyl]methylphenylsilyl]-1-oxo-
propyl] L-Proline 1,1-Dimethylethyl Ester (29a). To a solution of
23a (52 mg, 0.136 mmol) in methylene chloride (1.5 mL) at 0 °C were
added DCC (42 mg, 0.204 mmol) and HOBt (28 mg, 0.204 mmol).
After 5 min, a solution of L-proline tert-butyl ester (L-proline tert-butyl
ester hydrochloride (42 mg, 0.204 mmol) and triethylamine (28 µL,
0.204 mmol) in 1.5 mL of methylene chloride) was added, and the
resulting mixture was stirred at 0 °C for 1 h and then 23 h at room
temperature. The mixture was filtered, and the filtrate was washed with
0.2 N NaOH (3 mL). The aqueous phase was extracted with methylene
chloride (3 × 2 mL), and the combined organics were washed with
water (3 mL), the water wash was extracted with methylene chloride
(3 × 2 mL), and the combined extracts were concentrated in vacuo
and purified by flash chromatography (1:4 ethyl acetate/methylene
chloride) to give 57 mg of 29a as a clear oil (78%, a mixture of four
N-[3-Methyl-1-[[3-oxo-3[(phenylmethyl)amino]propyl]diphenylsilyl]-
butyl] Benzamide (24b). To a solution of benzylamine (28 mg, 0.26
mmol) in DMF (5 mL) at 0 °C were added 4-methylmorpholine (0.025
mL, 0.22 mmol), DEC (65 mg, 0.34 mmol), HOBT (30 mg, 0.22 mmol),
and crude carboxylic acid 23b (100 mg, 0.22 mmol). After being stirred
for 30 min at 0 °C under argon, the mixture was allowed to warm to
room temperature and stirred overnight. This mixture was concentrated
in vacuo and partitioned between water (8 mL) and ethyl acetate (8
mL). The organic layer was isolated, and the aqueous layer was
extracted twice with 8 mL portions of ethyl acetate. The combined
organic extracts were washed successively with saturated aqueous
NaHCO3 (8 mL) and saturated aqueous NaCl, dried over Na2SO4, and
concentrated. Flash chromatography over silica gel (2:3 ethyl acetate/
hexane) and recrystallization from ethyl acetate gave pure diamide 24b
as a colorless crystalline solid (96 mg, 80%). Rf ) 0.50 (2:3 ethyl
acetate/hexane). mp 131-132 °C. 1H NMR (CDCl3): δ 7.41 (m, 20H),
6.42 (t, J ) 5.2 Hz, 1H), 5.98 (d, J ) 10.1 Hz, 1H), 4.72 (m, 1H),
4.33 (d, J ) 5.8 Hz, 2H), 2.54 (m, 1H), 2.10 (m, 1H), 1.65 (m, 1H),
1
diastereomers). Rf ) 0.40 (1:9 ethyl acetate/methylene chloride). H
NMR (CDCl3) (complex mixture of four diastereomers): δ 7.91-7.73
(m), 7.55 (m), 7.45-7.34 (m), 7.07 (d, J ) 10.4), 6.88 (d, J ) 9.7),
6.71 (d, J ) 11.0), 6.48 (d, J ) 11.0), 6.37 (d, J ) 9.8), 4.41-4.02
(m), 3.63-3.24 (m), 2.59-2.46 (m), 2.40-2.35 (m), 2.24-1.81 (m),
1.68-1.55 (m), 1.44 (m), 1.41 (m), 1.39 (s), 1.32-1.23 (m), 0.95-
0.79 (m), 0.38 (s), 0.36 (s), 0.33 (s), 0.28 (s), 0.26 (s), 0.25 (s). Exact
mass (FAB) calcd for C31H45N2O4Si (MH+), 537.3149; found, 537.3146.
1-[3-[[1-(Benzoylamino)-3-methylbutyl]diphenylsilyl]-1-oxopro-
pyl] L-Proline 1,1-Dimethylethyl Ester (29b). To a 0 °C solution of
23b (29 mg, 0.065 mmol) in methylene chloride (1 mL) were added
DCC (20 mg, 0.098 mmol) and HOBt (13 mg, 0.098 mmol). After 5
min, a solution of L-proline tert-butyl ester (L-proline tert-butyl ester
hydrochloride (20 mg, 0.098 mmol) and triethylamine (14 µL, 0.098
mmol) in 0.5 mL of methylene chloride) was added, and the resulting
mixture was stirred at 0 °C for 1 h and then 10 h at room temperature.
The mixture was filtered and washed with 0.2 N NaOH. The aqueous
9
J. AM. CHEM. SOC. VOL. 124, NO. 25, 2002 7373

A R T I C L E S
Mutahi et al.
phase was extracted with methylene chloride (3 × 2 mL), and the
combined organics were washed with water (3 mL). The water wash
was extracted with methylene chloride (3 × 2 mL), and the combined
organics were concentrated in vacuo and purified by flash chromatog-
raphy (1:9 ethyl acetate/methylene chloride) to give 22.5 mg of 29b as
a clear oil (58%, a mixture of two diastereomers). Rf ) 0.40 (1:9 ethyl
(0.5 mL) was added a solution of LDA, prepared from diisopropylamine
(40 mL, 0.28 mmol), THF (1 mL), and n-butyllithium (0.12 mL, 2.3
M in hexanes, 0.28 mmol). The resulting solution was stirred at 0 °C
for 1 h and then cooled to -78 °C. Iodomethane (18 µL, 0.28 mmol)
was added, and stirring continued for 2 h. After the solution was warmed
to room temperature, water (2 mL) was added, and the aqueous phase
was acidified to pH 1 with 6 N HCl, and extracted with 1:1 ethyl acetate:
hexanes (3 × 2 mL). The combined organics were concentrated and
purified by flash chromatography (1:9 methanol/methylene chloride)
to give 36 mg of 32b as a tan oil (89%). Rf ) 0.47 (1:9 methanol/
methylene chloride). 1H NMR (CDCl3): δ 7.61-7.54 (m), 7.47-7.32
(m), 7.23 (m), 5.66 (d, J ) 10.3), 4.70 (m), 2.54-2.47 (m), 2.31-2.23
(m), 1.65-1.55 (m), 1.48-1.36 (m), 1.26 (m), 1.05 (d, J ) 6.4), 1.02
(s, J ) 6.8), 0.93 (m), 0.83 (m). Exact mass (FAB) calcd for C28H34-
NO3Si (MH+), 460.2308; found, 460.2303.
1
acetate/methylene chloride). H NMR (CDCl3) (complex mixture of
two diastereomers with proline rotamers): δ 7.80 (m), 7.70 (m), 7.64-
7.49 (m), 7.45-7.31 (m), 7.05 (d, J ) 9.8), 6.86 (d, J ) 9.8), 6.47 (d,
J ) 10.4), 4.68 (m), 4.40-3.83 (m), 3.57-3.15 (m), 2.49-2.39 (m),
2.35-2.23 (m), 2.09-1.79 (m), 1.67-1.50 (m), 1.42 (m), 1.35 (s), 1.26
(m), 1.00-0.93 (m), 0.83-0.79. Exact mass (FAB) calcd for C36H47N2O4-
Si (MH+), 599.3305; found, 599.3313.
1-[3-[[1-(Benzoylamino)-3-methylbutyl]hydroxymethylsilyl]-1-
oxopropyl] L-Proline (30). To a 0 °C solution of 29a (52 mg, 0.097
mmol) in methylene chloride (12 mL) was added trifluoromethane-
sulfonic acid (1.63 mL), and the resulting solution was stirred for 25
min. Concentrated NH4OH was added to give an aqueous phase (pH
5-6), and the pH was adjusted to 1 with 6 N HCl. The aqueous phase
was extracted with methylene chloride (3 × 2 mL), and the combined
extracts were concentrated in vacuo. Purification by RP-HPLC (250
× 22 mm Vydac Peptide and Protein C18, with a linear gradient of
40-50% aqueous acetonitrile containing 1% trifluoroacetic acid,
monitoring at 215 nM) and lyophilyzing the collected fractions gave
15.5 mg of 30 as a colorless solid (38%). 1H NMR (acetone-d6) (mixture
of four diastereomers): δ 10.67 (m), 10.44 (m), 8.08 (m), 7.92 (m),
7.73 (m), 7.58 (m), 7.43 (m), 4.90 (m), 4.77 (m), 4.71 (m), 4.62 (m),
3.83-3.74 (m), 3.52 (m), 3.28-3.18 (m), 2.99 (m), 2.77 (m), 2.60-
2.53 (m), 2.42 (m), 2.16 (m), 1.91 (m), 1.77-1.62 (m), 1.28 (m), 0.98-
0.88 (m), 0.43 (s), 0.42 (s), 0.38 (s), 0.32 (s), 0.31 (s), 0.28 (s), 0.14
(m). MS (FAB): m/e (rel intensity) 403 (M - OH, 100), 230 (10),
109 (14). Exact mass (FAB) calcd for C21H31N2O4Si, 403.2053; found,
403.2048.
1-[3-[[1-(Benzoylamino)-3-methylbutyl]methylphenylsilyl]-2-methyl-
1-oxopropyl] L-Proline 1,1-Dimethylethyl Ester (33a). To a 0 °C
solution of 32a (94.4 mg, 0.238 mmol) in methylene chloride (1 mL)
were added DIC (56 µL, 0.357 mmol) and HOBt (48 mg, 0.357 mmol).
After 5 min, a solution of L-proline tert-butyl ester (L-proline tert-butyl
ester hydrochloride (74 mg, 0.357 mmol) and triethylamine (49 µL,
0.357 mmol) in 1.5 mL of methylene chloride) was added, and the
resulting mixture was stirred at 0 °C for 1 h and then 11 h at room
temperature. The mixture was filtered and washed with 0.2 N NaOH
(3 mL). The aqueous phase was extracted with methylene chloride (3
× 2 mL), and the combined organics were washed with water (10 mL).
The water wash was extracted with methylene chloride (3 × 2 mL),
and the combined organics were concentrated in vacuo and purified
by flash chromatography (1:4 ethyl acetate/methylene chloride) to give
131.1 mg of 33a as a clear oil (58%). Rf ) 0.36 (1:4 ethyl acetate/
1
methylene chloride). H NMR (CDCl3): δ 7.78 (m), 7.57 (m), 7.37
(m), 6.91 (m), 6.60-6.49 (m), 6.09-5.95 (m), 4.36-4.26 (m), 4.20-
4.08 (m), 3.56-3.26 (m), 2.70 (m), 2.59 (m), 2.56 (m), 1.82 (m), 1.60
(m), 1.47-1.35 (m), 1.25 (m), 1.13 (m), 1.06-0.81 (m), 0.47-0.36
(m), 0.22 (s). Exact mass (FAB) calcd for C32H47N2O4Si (MH+),
551.3305; found, 551.3329.
1-[3-[[1-(Benzoylamino)-3-methylbutyl]dihydroxysilyl]-1-oxopro-
pyl] L-Proline (31). To a 0 °C solution of 29b (20 mg, 0.033 mmol)
in methylene chloride (2.5 mL) was added trifluoromethanesulfonic
acid (0.55 mL), and the resulting solution was stirred for 25 min.
Concentrated NH4OH was added to give an aqueous phase (pH 5-6),
and then was adjusted to pH 1 with 6 N HCl. The aqueous phase was
extracted with methylene chloride (3 × 2 mL), and the combined
extracts were concentrated in vacuo. Purification by RP-HPLC (250
× 4.6 mm Vydac Peptide and Protein C18, with a linear gradient of
20-50% aqueous acetonitrile containing 1% trifluoroacetic acid,
monitoring at 215 nM) and lyophilyzing the collected fractions gave
1-[3-[[1-(Benzoylamino)-3-methylbutyl]diphenylsilyl]-2-methyl-
1-oxopropyl] L-Proline 1,1-Dimethylethyl Ester (33b). To a 0 °C
solution of 32b (33 mg, 0.072 mmol) in methylene chloride (0.5 mL)
were added DIC (17 µL, 0.108 mmol) and HOBt (15 mg, 0.108 mmol).
After 5 min, a solution of L-proline tert-butyl ester (L-proline tert-butyl
ester hydrochloride (22 mg, 0.108 mmol) and triethylamine (15 µL,
0.108 mmol) in 0.5 mL of methylene chloride) was added, and the
resulting mixture was stirred at 0 °C for 1 h and then 13 h at room
temperature. The mixture was washed with 0.2 N NaOH (3 mL). The
aqueous phase was extracted with methylene chloride (4 × 2 mL),
and the combined organics were washed with water (10 mL). The water
wash was extracted with methylene chloride (2 × 2 mL), and the
combined organics were concentrated in vacuo and purified by flash
chromatography (1:4 ethyl acetate/methylene chloride) to give 24 mg
of 33b as a tan oil (56%). Rf ) 0.49 (1:4 ethyl acetate/methylene
chloride). 1H NMR (CDCl3): δ 7.81-7.76 (m), 7.72-7.70 (m), 7.63-
7.57 (m), 7.55-7.50 (m), 7.45-7.29 (m), 7.06 (d, J ) 9.8), 6.86 (d, J
) 10.7), 6.73 (d, J ) 10.7), 6.47 (d, J ) 9.8), 4.68 (m), 4.55 (m), 4.31
(m), 4.20 (m), 4.05 (m), 3.95 (m), 3.60 (m), 3.52 (m), 3.44 (m), 3.30
(m), 3.20 (m), 2.43 (m), 2.07-1.80 (m), 1.65-1.45 (m), 1.42 (m), 1.35
(m), 1.26 (m), 0.96 (m), 0.81 (m). Exact mass (FAB) calcd for
C37H49N2O4Si (MH+), 613.3462; found, 613.3470.
1
5.5 mg of 31 as a colorless solid (39%). H NMR (D2O): δ 7.93 (s),
7.74 (s), 7.59 (s), 3.74 (s), 3.51 (s), 2.89 (m), 2.39 (m), 2.08 (m), 1.68
(m), 0.93 (m). MS (FAB): 405 (MH+ - H2O).
3-[[1-(Benzoylamino)-3-methylbutyl]methylphenylsilyl]-2-meth-
ylpropionic Acid (32a). To a solution of 23a (128 mg, 0.334 mmol)
in THF (1 mL) was added a solution of LDA, prepared from
diisopropylamine (164 µL, 1.17 mmol), THF (1 mL), and n-butyllithium
(0.5 mL, 2.3 M in hexanes, 1.16 mmol). The resulting solution was
stirred at 0 °C for 1 h and then cooled to -78 °C. Iodomethane (73
µL, 1.17 mmol) was added, and stirring continued for 2.5 h. After the
solution was warmed to room temperature, water (2 mL) was added,
and the aqueous phase was washed with 1:3 ethyl acetate/hexane,
acidified to pH 1 with 6 N HCl, and extracted with methylene chloride
(3 × 2 mL). The combined organics were concentrated and purified
by flash chromatography (1:9 methanol/methylene chloride) to give
94.4 mg of 32a as a tan oil (71%). Rf ) 0.54 (1:9 methanol/methylene
chloride). 1H NMR (CDCl3): δ 7.69-7.65 (m), 7.60-7.51 (m), 7.47-
7.33 (m), 7.28-7.22 (m), 6.98-6.94 (m), 5.92 (d, J ) 9.2), 5.76 (d, J
) 11), 4.43 (m), 4.19 (m), 1.63-0.83 (m), 0.53-0.40 (m). Exact mass
(FAB) calcd for C23H31NO3Si (MH+), 398.2151; found, 398.2156.
3-[[1-(Benzoylamino)-3-methylbutyl]diphenylsilyl]-2-methylpro-
pionic Acid (32b). To a solution of 23b (36 mg, 0.081 mmol) in THF
1-[3-[[1-(Benzoylamino)-3-methylbutyl]hydroxymethylsilyl]-2-
methyl-1-oxopropyl] l-Proline (34). To a 0 °C solution of 33a (58
mg, 0.105 mmol) in methylene chloride (9 mL) was added trifluo-
romethanesulfonic acid (1.77 mL), and the resulting solution was stirred
for 25 min. Concentrated NH4OH was added to give an aqueous phase
(pH 5-6), and then was adjusted to pH 1 with 6 N HCl. The aqueous
phase was extracted with methylene chloride (3 × 4 mL), and the
combined extracts were concentrated in vacuo. Purification by RP-
9
7374 J. AM. CHEM. SOC. VOL. 124, NO. 25, 2002

Silanol Protease Inhibitors
A R T I C L E S
HPLC (250 × 22 mm Vydac Peptide and Protein C18, with a linear
gradient of 35-50% aqueous acetonitrile containing 1% trifluoroacetic
acid, monitoring at 215 nM) and lyophilyzing the collected fractions
gave 21.1 mg of 34 as a colorless solid (46%). 1H NMR (CD3CN): δ
9.14 (m), 7.92 (d, J ) 6.9), 7.71 (t, J ) 6.9), 7.57 (t, J ) 7.8), 4.74
(m), 4.56 (m), 4.47 (m), 3.79 (m), 3.72 (m), 3.57 (m), 3.11 (m), 2.93
(m), 2.36-1.81 (m), 1.46 (m), 1.32-1.24 (m), 0.96 (m), 0.40-0.30
(m). MS (FAB): m/e (rel intensity) 417 (M - OH, 100), 403 (20).
Exact mass (FAB) calcd for C22H33N2O4Si, 417.2210; found, 417.2221.
1-[3-[[1-(Benzoylamino)-3-methylbutyl]dihydroxysilyl]-1-oxopro-
pyl] L-Proline (35). To a 0 °C solution of 33b (24 mg, 0.039 mmol)
in methylene chloride (3.7 mL) was added trifluoromethanesulfonic
acid (0.66 mL), and the resulting solution was stirred for 25 min.
Concentrated NH4OH was added to give an aqueous phase (pH 5-6),
and then was adjusted to pH 1 with 6 N HCl. Saturated NaCl (2 mL)
was added, the aqueous phase was extracted with methylene chloride
(3 × 2 mL), and the combined extracts were concentrated in vacuo.
Purification by RP-HPLC (250 × 22 mm Vydac Peptide and Protein
C18, with a linear gradient of 25-40% aqueous acetonitrile containing
1% trifluoroacetic acid, monitoring at 215 nM) and lyophilyzing the
dissolved in deionized water buffered with 0.3 M NaCl and 0.05 M
Trizma (Sigma), to give a pH of 8.3 at 37 °C. A solution of porcine
ACE (Sigma) in a buffered human serum base (0.10 mL) was added
to 1.0 mL of a 0.5 mM solution of N-[3-(2-furyl)acryloyl]-L-phenyl-
alanylglycylglycine (FAPGG, Sigma) in a 1 cm disposable cuvette.
The test compound solution was added, and the mixture was incubated
for 5 min. Enzymatic activity was followed by monitoring at 340 nm
over a period of 5 min, relative to an identical solution not containing
the inhibitor. The silanediols 31 and 35, the most active species, were
assayed in triplicate, and the less active methylsilanols 30 and 34 were
assayed in duplicate.
Acknowledgment. Acknowledgment is made to the Center
for Biotechnology, supported by the New York State Science
and Technology Foundation, the National Science Foundation
(CHE 9712772), the National Institutes of Health (HL62390),
and Wyeth-Ayerst Research for support of this research.
M.wa.M. gratefully acknowledges support from Schering-
Plough Research Institute and Dr. Robert Aslanian for helpful
discussions.
1
collected fractions gave 3.6 mg of 35 as a colorless solid (21%). H
NMR (D2O): δ 7.97 (m), 7.78-7.69 (m), 7.63 (m), 4.56 (m), 3.77
(m), 3.57-3.19 (m), 2.91 (m), 2.42 (m), 2.11 (m), 1.75 (m), 2.10 (m),
1.75 (m), 1.44-1.18 (m), 0.99 (m). MS (MALDI-TOF): 420 (M -
16).
General Assay Procedure. Following the procedure of Cushman
and Cheung,63 as modified by Holmquist,62 test compounds were
Supporting Information Available: Proton NMR spectra for
19-28 and plotted inhibition data for silane inhibitors (PDF).
This material is available free of charge via the Internet at
http://pubs.acs.org.
JA026158W
9
J. AM. CHEM. SOC. VOL. 124, NO. 25, 2002 7375