Enantioselective synthesis of α-phosphono sulfonate squalene synthase inhibitors: Chiral recognition in the interactions of an α-phosphono sulfonate inhibitor with squalene synthase

Full text of DOI:10.1021/ja962505o

11668
J. Am. Chem. Soc. 1996, 118, 11668-11669
Enantioselective Synthesis of r-Phosphono Sulfonate
Squalene Synthase Inhibitors: Chiral Recognition
in the Interactions of an r-Phosphono Sulfonate
Inhibitor with Squalene Synthase
R. Michael Lawrence,† Scott A. Biller,*,†
John K. Dickson, Jr.,† Janette V. H. Logan,†
David R. Magnin,† Richard B. Sulsky,† John D. DiMarco,‡
Jack Z. Gougoutas,‡ Barbara D. Beyer,§ Susan C. Taylor,§
Shih-Jung Lan,| Carl P. Ciosek, Jr., Thomas W. Harrity,
Kern G. Jolibois, Lori K. Kunselman, and
Dorothy A. Slusarchyk
Figure 1. Newman projections of transition states (A and B) and
product (C) of the reaction of chiral phosphonic diamide anions with
electrophiles.
Chart 1
DiVision of DiscoVery Chemistry, Department of Solid
State Chemistry, Analytical Research and DeVelopment
Department of Pharmacokinetics and Metabolism
Department of Metabolic Diseases
Bristol-Myers Squibb Pharmaceutical Research Institute
P.O. Box 4000, Princeton, New Jersey 08543
ReceiVed July 19, 1996
Inhibitors of squalene synthase1 have great potential as
selective cholesterol lowering agents due to the strategic location
of the enzyme at the first committed step in the cholesterol
biosynthetic pathway. We recently reported the rational design
and discovery of a potent squalene synthase inhibitor, R-phospho-
no sulfonate triacid 1,2 and the orally bioavailable prodrug ester
23 (Chart 1). R-Phosphono sulfonates 1 and 2 exist as pairs of
enantiomers. An important question remained unresolved:
Could squalene synthase discriminate between the two enanti-
omers of 1? This is a critical issue, both with respect to
understanding the molecular interactions of R-phosphono sul-
fonates with squalene synthase and to selecting a single
enantiomer for further pharmaceutical development.
Scheme 1a
a (a) CH3P(O)Cl2, Et3N, toluene, 81%; (b) n-BuLi, THF, -78 °C;
(c) 3-(3′-phenoxyphenyl)propyl iodide, -78 °C to rt, 89%; (d)
[(CH3)2NCS2]2, -90 to -70 °C, 84%; (e) aqueous HCl, CH3CN, 99%;
(f) 30% H2O2, HOAc/formic acid (9:1) ratio; (g) aqueous KOH, 82%.
1-Substituted phosphono methylsulfonates were virtually
unknown in the chemical literature prior to our studies.4
A
major concern in our synthetic planning was the configurational
stability of the chiral center, which is geminally substituted with
two electron-withdrawing groups. Prior to attempting the
enantioselective syntheses, deuterium incorporation studies were
2 was subjected to these same solvolytic conditions, no
D-incorporation was detected. We concluded that both the
triacid 1 and diester 2 were isolable as single enantiomers,
whereas triester 3 was very sensitive to racemization.
1
performed by H NMR to assess the configurational stability
of the chiral centers of 1-3. While triacid 1 was resistant to
D-incorporation over a wide range of pD values, the racemic
Our strategy for the introduction of the asymmetric center
relied upon the sulfuration or alkylation of a chiral phosphorus
carbanion (Figure 1).5,6 The chiral auxiliary that appeared most
promising was the C2-symmetric phosphonic diamide derived
from trans-1,2-cyclohexanediamine.5 The sulfuration of anion
A was predicted to provide C having the (S)-configuration, on
the basis of the sterically favored approach to the carbanion as
depicted in the Hanessian transition state model. Alkylation
of the dianion B was predicted to provide C with the same
configuration, on the basis of the sterically favored approach
to the chelated dianion.5c Thus, the stereochemical outcome
for both routes should be identical, since a change in the
geometry of the anion (A versus B, Figure 1) is accompanied
by a change in the order of group introduction.
prodrug 2 exhibited a slow H/D exchange at basic pD (t1/2
)
110 h at pD 9.3, room temperature (rt)). Under the mild
solvolytic conditions utilized to deprotect the cyclohexyl sul-
fonate ester of 3 (10:1 CF3CH2OD/D2O, 3 equiv of KOAc, 40
°C, 20 h), 2 was obtained with complete D-incorporation. When
* Corresponding Author: Bristol-Myers Squibb Pharmaceutical Research
Institute, P.O. Box 4000, Princeton, New Jersey 08543; tel. 609-252-5585;
fax 609-252-6601; email biller@bms.com.
† Division of Discovery Chemistry.
‡ Department of Solid State Chemistry.
§ Analytical Research and Development.
| Department of Pharmacokinetics and Metabolism.
Department of Metabolic Diseases.
(1) For reviews on squalene synthase inhibitors, see: Biller, S. A.;
Neuenschwander, K.; Ponpipom, M. M.; Poulter, C. D. Curr. Pharm. Des.
1996, 2, 1-40. Abe, A.; Tomesch, J. C.; Wattanasain, S.; Prestwich, G.
D. Nat. Prod. Rep. 1994, 279-302.
In the sulfuration approach (Scheme 1), the diamine 75a,d was
converted to phosphonic diamide 8, which was deprotonated
and alkylated with 3-(3′-phenoxyphenyl)propyl iodide to yield
9. Anion A was formed by deprotonation of 9 with n-BuLi,
(2) Magnin, D. R.; Biller, S. A.; Chen, Y.; Dickson, J. K., Jr.; Fryszman,
O. M.; Lawrence, R. M.; Logan, J. V. H.; Sieber-McMaster, E. S.; Sulsky,
R. B.; Traeger, S. C.; Hsieh, D. C.; Lan, S.-J.; Rinehart, J. K.; Harrity, T.
W.; Jolibois, K. G.; Kunselman, L. K.; Rich, L. C.; Slusarchyk, D. A.;
Ciosek, C. P., Jr. J. Med. Chem. 1996, 39, 657-660.
(5) (a) Hanessian, S.; Delorme, D.; Beaudoin, S.; Leblanc, Y. J. Am.
Chem. Soc. 1984, 106, 5754-5756. (b) Hanessian, S.; Bennani, Y. L.;
Delorme, D. Tetrahedron Lett. 1990, 31, 6461-6464. (c) Hanessian, S.;
Bennani, Y. L. Tetrahedron Lett. 1990, 31, 6465-6468. (d) Bennani, Y.
L. Synthese Asymmetrique D’Acides Phosphoniques Alpha-Substitutes; Ph.D.
Thesis, University of Montreal, 1991.
(6) Denmark, S. E.; Dorow, R. L. J. Am. Chem. Soc. 1990, 112, 864-
866. Denmark, S. E.; Miller, P. C.; Wilson, S. R. J. Am. Chem. Soc. 1991,
113, 1468-1470.
(7) Control studies indicate that the reduced diastereoselectivity in the
sulfuration is not the result of epimerization of sulfurated phosphonate 10.
(3) Dickson, J. K., Jr.; Biller, S. A.; Magnin, D. R.; Petrillo, E. W., Jr.;
Hillyer, J. W.; Hsieh, D. C.; Lan, S.-J.; Rinehart, J. K.; Gregg, R. E.; Harrity,
T. W.; Jolibois, K. G.; Kalinowski, S. S.; Kunselman, L. K.; Mookhtiar,
K. A.; Ciosek, C. P., Jr. J. Med. Chem. 1996, 39, 661-664.
(4) To the best of our knowledge, there is only one other report of carbon
substituted R-phosphono sulfonates prior to our own work:2,3 Christensen,
B. G.; Beattie, T. R.; Graham, D. W. U.S. Patent 3 657 282, April 18,
1972. An unsubstituted R-phosphono methylsulfonate ester was reported
as an adenosine phospho sulfate analog: Callahan, L.; Ng, K.; Geller, D.
H.; Agarwal, K.; Schwartz, N. B. Anal. Biochem. 1989, 177, 67-71.
S0002-7863(96)02505-X CCC: $12.00 © 1996 American Chemical Society

Communications to the Editor
J. Am. Chem. Soc., Vol. 118, No. 46, 1996 11669
followed by sulfuration with [(CH3)2NCS2]2 (dithiuram) at -90
°C to provide 10(S)/10(R) in 84% yield as a 3:1 mixture of
diastereomers, which were readily separated by silica gel
chromatography. The lower diastereoselectivity observed for
the sulfuration as compared to that reported for alkylation5 of
anions such as A is presumably due to the long C-S bond length
in the transition state, which would serve to attenuate the steric
hindrance sensed by dithiuram relative to an alkyl halide.7
Indeed, alkylation of A with benzyl bromide at -90 °C led to
a 10:1 mixture of diastereomers. The separated diastereomers
10(S) and 10(R), respectively, were hydrolyzed with mild acid
to remove the chiral auxiliary, and the resulting dithiocarbamates
were oxidized with H2O2 in HCO2H/CH3CO2H to provide
triacids 1(S) [99.8% ee,8 [R]D -10.7° (c ) 0.88, H2O)] and
1(R) [(95.5% ee,8 [R]D +9.5° (c ) 0.89, H2O)].
16-fold more potent than 1(R) as an inhibitor of squalene
synthase (IC50 values of 68 and 1120 nM, respectively).
Furthermore, 1(S) was found to be considerably more potent
as an inhibitor of the biosynthesis of cholesterol in rats.11 On
intravenous dosing, 1(S) had an ED50 of 0.16 mg/kg, while 1-
(R) was inactive at 1 mg/kg.
Alkylation of the trisilver salt of 1(S) with iodomethyl
pivalate12 (anisole, CH2Cl2, 4 Å sieves), followed by solvolysis
of the labile triester 6 (CH3CN, H2O) and careful neutralization
of the sulfonic acid with potassium phosphate provided 2(S) in
98% ee8 [57% yield, [R]D -6.6° (c ) 1.0, CH3OH)], indicating
minimal racemization. Following oral dosing of 2(S) to rats,
1(S) was recovered from plasma and bile without significant
racemization, proving that 2(S) possessed suitable chiral stability
to deliver 1(S) in ViVo.13
The asymmetric routes to R-phosphono sulfonates described
herein efficiently provide the potent squalene synthase inhibitor
1 and prodrug 2 in nonracemic form and are generalizable to
other members of this class of molecules. Of particular interest
is the enantioselectivity observed for the inhibitory potency of
1(S) and 1(R), indicating that squalene synthase is able to
discriminate between the phosphonate and sulfonate moieties
in distinct binding sites. The dibasic phosphonate group and
the monobasic sulfonate group have considerable structural
similarity. Both are tetrahedral, second-row functions with a
C3V display of negatively charged oxygen atoms. In addition,
they show significant similarity, but not identity, in their
interaction with Lewis acids.14 Certain proteins can discriminate
between binding phosphonyl and sulfonyl groups,15 whereas
others bind both functions comparably well.16 Bond angle and
bond length data from the X-ray crystal structure of 5(S) and a
related R-phosphono sulfonate10,17 support the close isosteric
relationship between the phosphonate and sulfonate groups.
Further structural studies will be required to understand the basis
for this selective binding. The enantioselectivity displayed by
squalene synthase reinforces the importance of single enantiomer
drug development,18 where the “inactive” enantiomer can
possess unwanted pharmacologic and toxicologic properties.
Acknowledgment. We acknowledge helpful discussions with Dr.
Youssef L. Bennani, currently of Abbott Laboratories. In addition,
the assistance of Bristol-Myers Squibb Analytical Research and
Development and the Discovery Chemistry NMR Group is appreciated
in the determination of analytical and spectral data.
Scheme 2a
a (a) P(O)Cl3, Et3N, 85%; (b) CH3SO3Et, n-BuLi, THF, -78 °C to
rt, 85%; (c) Bu4NI, THF, 100%; (d) n-BuLi, THF, -90 °C; (e) 3-(3′-
phenoxyphenyl)propyl iodide, THF, -90 to -78 °C; (f) HCl (aqueous);
(g) AG-50 ×8 (K+ form), 57% from 14; (h) AG-50 ×8 (H+ form); (i)
R1NH2 (2 equiv); for 4, R1 ) 1-adamantyl (99.5% ee); for 5, R1 )
(S)-R-methylbenzyl.
For the chelation-controlled alkylation approach (Scheme 2),
the precursor to anion B was synthesized by converting the
diamine 7 to the Bu4N+ salt 14.9 Treatment of 14 with n-BuLi
followed by alkylation with 3-(3′-phenoxyphenyl)propyl iodide
at -90 °C provided 15 as a >20:1 ratio of diastereomers by
31P NMR. Removal of the chiral auxiliary afforded 1(S) (92%
ee8) in 57% overall yield from 14. The enantiomeric purity
could be increased to 99.5%8 by recrystallization of the bis-
adamantylamine salt 4(S). The major enantiomers from the
sulfuration and alkylation routes proved to be identical, which
is consistent with the proposed transition state models (Figure
1). The absolute configuration of the major enantiomer was
confirmed by single crystal X-ray analysis of the bis-(S)-R-
methylbenzylamine salt 5(S).10
Supporting Information Available: Details of the X-ray structural
analysis of compound 5(S), as well as the related R-phosphono sulfonate
i of ref 2 (18 pages). See any current masthead page for ordering and
Internet access instructions.
JA962505O
(12) Srivastva, D. N.; Farquhar, D. Bioorg. Chem. 1984, 12, 118-129.
(13) Compound 1(S) was isolated from the 3 and 6 h plasma samples
and the 0-12 h bile samples of rats that had received single 40 µmol/kg
oral doses of 2(S) and analyzed for 1(R) and 1(S) by HPLC using the assay
reported in ref 8. The results showed that the isolated samples contained
greater than 99.5% of 1(S).
(14) Kanyo, F. K.; Christianson, D. W. J. Biol. Chem. 1991, 266, 4264-
4268.
(15) Luecke, H.; Quiocho, F. A. Nature 1991, 347, 402-406.
(16) Anderson, C. J.; Lucas, L. J. H.; Widlanski, T. S. J. Am. Chem.
Soc. 1995, 117, 3889-3890.
Both enantiomers of 1 were tested for their ability to inhibit
the conversion of farnesyl diphosphate to squalene by rat
microsomal squalene synthase in Vitro.11 Enantiomer 1(S) was
(8) Chiral analytical HPLC was utilized to determine the enantiomeric
enrichment in 1 and 2. Both sets of enantiomers (1 and 2) were separated
using the R1-acid glycoprotein column (Chiral-AGP) available from
ChromTech (Sweden), using an 100 × 4.0 mm i.d. column (10 µm particle
size) at 1 mL/min flow rate. For the separation of 1(S) and 1(R), the mobile
phase was 85% 0.1 M KH2PO4 (pH ) 4.6) and 15% acetonitrile. The
retention times were 10.3 and 18.8 min, respectively. For the separation
of 2(S) and 2(R), the mobile phase was 78% 0.1 M KH2PO4 (pH ) 4.6),
16% 2-propanol, and 6% methanol. The retention times were 24.5 and
15.7 min, respectively.
(17) Crystallographic assignment of the stereochemical configuration of
5(S)10 required a distinction between two strictly isosteric groups at the
chiral center: the PO3 and SO3 groups of the R-phosphono sulfonate dianion
(the remaining acidic OH was not observed experimentally). The C-S
(1.78 Å) and O-S (av 1.44 Å) bonds are only several hundredths of an
angstrom shorter than the corresponding bonds to phosphorus, although
angles C-S-O (av ) 107°) and O-S-O (105°, 109°, 121°) are not
significantly different for the phosphonate group. The phosphonate is
intermolecularly H-bonded to itself and therefore bears the only non-ionized
acidic hydrogen: O3-P-O4-H‚‚‚O3, where the O4‚‚‚O3 distance is 2.61
Å. Phosphonate oxygen O4 is assigned as the hydroxyl on the basis of its
longer bond length (P-O4 ) 1.56 Å versus P-O3 ) 1.48 Å). The other
phosphonate oxygen and the three sulfonate oxygens are involved in
H-bonds only as acceptors with the protonated amine counterions. Similar
isosteric geometries have been observed in the crystal structure10 of the
related R-phosphono sulfonate (structure i of ref 2).
(9) Carretero, J. C.; Demillequand, M.; Ghosez, L. Tetrahedron 1987,
43, 5125-5134.
(10) Details of the X-ray structural analysis for 5(S) and the related
compound i of ref 2 are included as Supporting Information. Atomic
coordinates for both compounds have been deposited in the Cambridge
Crystallographic Database and can be obtained upon request to the Director,
Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge, CB2
1EZ, U.K.
(11) Ciosek, C. P., Jr.; Magnin, D. R.; Harrity, T. W.; Logan, J. V. H.;
Dickson, J. K., Jr.; Gordon, E. M.; Hamilton, K. A.; Jolibois, K. G.;
Kunselman, L. K.; Lawrence, R. M.; Mookhtiar, K. A.; Rich, L. C.;
Slusarchyk, D. A.; Sulsky, R. B.; Biller, S. A. J. Biol. Chem. 1993, 268,
24832-24837.
(18) Crossely, R. Chirality and Biological ActiVity of Drugs; CRC
Press: Boca Raton, FL, 1995.