J. Med. Chem. 1990,33, 21-31
`21
`( f )-cis - [ 3- (5,7-Diamino-3H- 1,2,3-triazolo[ 4,5- d Ipyrimi-
`(B+); IR (KBr) 3600-3000 (NH2, OH), 1750,1600 cm-' (C=C,
`din-3-yl)cyclopentyl]carbinol (1 lb). Compound 9b (268 mg,
`253 nm in 0.1 N HCl; NMR (dimethyl-d8 sulf-
`C=N); UV A,,
`oxide) 6 11.05-10.95 (s, 1 H, 7-OH, DzO exchangeable), 7.10-6.90
`1 mmol) was processed as described for 9a to yield 220 mg of 1 lb
`(br, 2 H, NHz, DzO exchangeable), 4.95-4.80 (m, 1 H, H-l'),
`(88%), which was recrystallized from ethanol-water (1:2) to afford
`4.70-4.50 (br, 1 H, CHzOH, DzO exchangeable), 3.50-3.40 (d, 2
`pink-white crystals: mp 223-225 "C; MS (30 eV, 250 "C) m/e
`249 (M'), 218 (M+ - 31), 151 (B'); IR (KBr) 3600-3100 (NHz,
`H, CHzOH), 2.32-1.55 (m, 7 H, H-4', CHzCHz, CHH'). Anal.
`OH), 1700,1600 cm-' (C=C, C=N); UV A,
`253,283 nm in 0.1
`(CloH1~N602*1.25H20) C, H, N.
`(*)-cis -[ 4 4 5,7-Diamino-3H- 1 ,2,3-triazolo[ 4,5-d]pyrimi-
`N HCl; NMR (dimethyl-d, sulfoxide) 6 7.85-7.25 (br, 2 H, NHz,
`din-3-yl)-2-cyclopentenyl]carbinol(l la). Compound 9a (267
`D20 exchangeable), 6.50-6.30 (s, 2 H, NHz, D20 exchangeable),
`4.95-4.85 (m, 1 H, H-l'), 4.65-4.60 (t, 1 H, CHzOH, DzO ex-
`mg, 1 mmol) was processed as described for compound 6a with
`changeable), 3.50-3.40 (d, 2 H, CH20H), 2.35-1.60 (m, 7 H, H-4',
`a reaction time of 20 h at 60 "C. The residual mixture was
`absorbed onto silica gel (2 g); it was packed into a column (2.0
`CHzCHz, CHH'). Anal. (CloHl,N70) C, H, N.
`X 10 cm) and eluted by CHC13-MeOH (151) to yield lla as white
`Acknowledgment. This work was supported by Public
`crystals, 204 mg (83%). The crude product was recrystallized
`Health Service Grant CA23263 from the National Cancer
`from ethanol-water (2:l) to yield lla: mp 240-242 "C dec; MS
`(30 eV, 240 "C) mle 247 (M+), 229 (M+ - 18), 217 (M+ - 30), 151
`Institute. We gratefully acknowledge the valuable assis-
`(B+); IR (KBr) 3600-3100 (NHz, OH), 1700, 1650, 1600 cm-'
`tance of J a y Brownell.
`(C=O, C=C, C=N); UV A,,
`253,283 nm in 0.1 N HCl; NMR
`(dimethyl-d6 sulfoxide) 6 7.80-7.20 (br, 2 H, NHz, D20 ex-
`Registry No. la, 61865-50-7; lb, 65898-98-8; 2a, 122624-72-0;
`changeable), 6.50-6.30 (s,2 H, NH2, D20 exchangeable), 6.15-6.10
`2b, 78795-20-7; 3a, 122624-73-1; 3b, 122624-74-2; 4a, 122624-75-3;
`and 5.95-5.90 (dd, 2 H, CH=CH vinyl, J = 5.0 Hz), 5.65-5.55
`4b, 122624-76-4; 5a, 122624-77-5; 5b, 12262478-6; 6a, 118237-87-9;
`6b, 118237-86-8; 7a, 118353-05-2; 7b, 11291500-1; 8a, 118237-88-0;
`(m, 1 H, H-l'), 4.75-4.65 (t, 1 H, CHzOH, D20 exchangeable),
`8b, 120330-36-1; 9a, 122624-79-7; 9b, 12262480-0; loa, 12262481-1;
`3.55-3.40 (m, 2 H, CH20H), 2.95-2.85 (m, 1 H, H-49, 2.65-2.55
`lob, 122624-82-2; 1 la, 122624-83-3; 1 lb, 122624-71-9; 2-amino-
`(m, 1 H, CH"), 1.90-1.80 (m, 1 H, CHH?. Anal. (CloH13N7-
`O-HzO) C, H, N.
`4,6-dichloropyrimidine, 56-05-3; p-chloroaniline, 106-47-8.
`
`Inhibitors of Cholesterol Biosynthesis. 1.
`trans -6-(2-Pyrrol- l-ylethyl)-4-hydroxypyran-2-ones, a Novel Series of HMG-CoA
`Reductase Inhibitors. 1. Effects of Structural Modifications at the 2- and
`5-Positions of the Pyrrole Nucleus
`
`B. D. Roth,* D. F. Ortwine,* M. L. Hoefle, C. D. Stratton, D. R. Sliskovic, M. W. Wilson, and R. S. Newton
`Parke-Davis Pharmaceutical Research Division, Warner-Lambert Company, 2800 Plymouth Road,
`Ann Arbor, Michigan 48105. Received January 25, 1989
`A novel series of trans-6-(2-pyrrol-l-ylethyl)-4-hydroxypyran-2-ones and their dihydroxy acid derivatives were prepared
`and evaluated for their ability to inhibit the enzyme HMG-CoA reductase in vitro. A systematic study of substitution
`at the 2- and 5-positions of the pyrrole ring revealed that optimum potency was realized with the 2-(4-fluoro-
`phenyl)-&isopropyl derivative 8x (Table In), which possessed 30% of the in vitro activity of the potent fungal metabolite
`compactin (I). A molecular modeling analysis led to the description of a pharmacophore model characterized by
`(A) length limits of 5.9 and 3.3 8, for the 2- and 5-substituenta, respectively, as well as an overall width limit of 10.6
`8, across the pyrrole ring from the 2- to the 5-substituent and (B) an orientation of the ethyl(ene) bridge to the
`4-hydroxypyran-2-one ring nearly perpendicular to the planes of the parent pyrrole, hexahydronaphthalene, and
`phenyl rings of the structures examined (Figure 3,O = 80-110"). Attempts to more closely mimic compactin's polar
`isobutyric ester side chain with the synthesis of 2-phenylpyrroles containing polar phenyl substituents resulted in
`analogues (Table 111, 8m-p) with equal or slightly reduced potencies when compared to the 2-[(unsubstituted or
`4-fluoro)phenyl]pyrroles, supporting the hypothesis that inhibitory potency is relatively insensitive to side-chain
`polarity or charge distribution in this area.
`The discovery that the fungal metabolites compactin (I)'
`and mevinolin (11)2 are not only potent inhibitors of the
`enzyme HMG-CoA reductase (HMGR), the rate-limiting
`enzyme in cholesterol biosynthesis, b u t are also effective
`hypocholesterolemic agents in man3 has led t o a plethora
`
`of publications describing synthetic and biological studies
`of close structural analogue^.^
`
`(1) (a) Endo, A.; Kuroda, M.; Tsujita, Y. J. Antibiot. 1976, 1346-8.
`(b) Endo, A,; Kuroda, Y.; Tanzawa, K. FEBS Lett. 1976, 72(2),
`323-6. (c) Brown, A. G.: Smale. T. C.: King. T. J.: Hassen-
`kamp, R.; Thompson, R.' H. J. Chem. Soc.,?'erkin' Trans. 1
`1976, 1165-9.
`(a) Endo, A. J. Antibiot. 1979, 32, 852. (b) Alberts, A.; Chen,
`J.; Kuron, G.; Hunt, V.; Huff, J.; Hoffman, C.; Rothrock, J.;
`Lopez, M.; Joshua, H.; Harris, E.; Pachett, A.; Monaghan, R.;
`Currie, S.; Stapley, E.; Albers-Schonberg, G.; Hensens, 0.;
`Hirshfield, J.; Hoogsteen, K.; Liesch, J.; Springer, J. R o c .
`Natl. Acad. Sci. U.S.A. 1980, 77(7), 3957-61.
`(a) Therapeutic response to Lovastatin (Mevinolin) in Non-
`Familial Hypercholesterolemia. J. Am. Med. Assoc. 1986,256,
`2829. (b) Vega, L.; Grundy, S. J. Am. Med. Assoc. 1987,
`257(1), 33-38 and references contained therein.
`
`I: R = H (compactin)
`CHJ
`I11
`11: R = CH3 (mevinolin)
`T h e disclosure of a series of very potent 6-(o-bi-
`phenyly1)-substituted 4-hydroxypyran-2-ones (111) by
`Willard et al.5 led us to hypothesize that the key structural
`
`(4) For a review, see: Rosen, T.; Heathcock, C. Tetrahedron 1986,
`42 (18), 4909-51.
`
`0022-2623/90/1833-0021$02.50/0
`0 1989 American Chemical Society
`
`NCI Exhibit 2030
`Page 1 of 11
`
`

`
`22 Journal of Medicinal Chemistry, 1990, Vol. 33, No. 1
`
`Both et al.
`
`Scheme Io
`Method A
`RICOCH=CHz + RzCHO a R1COCHZCH2CORz
`
`Scheme 11"
`
`1
`
`2
`
`3
`
`-
`
`Rlw 2 R
`
`3
`
`a(method A) L
`i - k (mhod C)
`
`CN
`I
`X
`I
`
`5
`
`CHO
`I
`X
`I
`J
`6
`f
`
`e
`
`
`
`z
`
`
`
`1
`\
`
`Method B
`
`b c
`3
`R ~ C O C H ~ C O Z C H ~
`a (a) 3-Benzyl-5-(2-hydroxyethyl)-4-methylthiazolium chloride,
`Et3N, 70 OC. (b) NaH, RICOCHzBr. (c) NaOH, CH30H.
`
`d, c (method 6 )
`
`feature possessed by all of these agents was a large lipo-
`philic group held in a particular spatial relationship with
`respect to the 4-hydroxypyran-2-one moiety. Indeed, ex-
`amination of CPK models of these inhibitors suggested
`that the ortho phenyl ring might occupy the same space
`as the isobutyric ester moiety of compactin and mevinolin.
`This hypothesis is supported by the 100-fold loss in po-
`tency found on hydrolysis of the isobutyric ester group?
`as well as the suggestion by Nakamura and Abeles that
`this portion of mevinolin fits into a lipophilic pocket in
`the active site of HMGR normally occupied by coenzyme
`A.'
`If this were true, then any connecting group that
`served to hold the lactone and the lipophilic moiety in the
`correct spatial relationship might be sufficient for potent
`inhibition. To investigate this, we selected the pyrrole ring
`as the anchor for various connecting groups, since there
`appeared to be sufficient synthetic methodology to allow
`for the simultaneous introduction of a variety of 2- and
`5-substituents. By varying the steric and electronic
`properties of these substituents, modifying the connecting
`group, and employing a molecular modeling analysis, we
`hoped to discern, at least in part, the optimal spatial re-
`lationship between the lipophilic group and the 4-
`hydroxypyran-2-one moiety and use this information in
`the design of potent HMGR inhibitors.
`We herein present our initial investigations into this
`series of inhibitors that define the structure-activity re-
`lationships at the 2- and 5-positions of the pyrrole nucleus
`and in the connecting group to the lactone ring. Also
`reported is the molecular modeling study and associated
`pharmacophore model, which describe conformational
`requirements of the side chain and steric requirements at
`the 2- and 5-positions of the pyrrole ring.
`Chemistry
`Our general synthetic strategy entailed the preparation
`of a suitable 1,4-diketone (3, Table I), either by the thia-
`zolium salt chemistry developed by Stetter (Scheme I,
`method A)8 or by alkylation of a @-keto ester with an a-halo
`ketone followed by hydrolysis and decarboxylation (me-
`thod B). The Stetter reaction proved to be the more
`versatile and generally higher yielding of the two. Paal-
`Knorr cyclization with 3-aminopropionitrile or an w-amino
`acetal provided the pyrroles in good yield (Scheme 11).
`The one exception was 1-(4-fluorophenyl)-5,5-dimethyl-
`
`(a) Willard. A.: Novello. F.: Hoffman. W.: Craeoe. E. USP
`4459422. (b) Stokker, G.; Hoffman, W.; Aibert; A:; Cragoe,
`E.; Deana, A.; Gilfillan, J.; Huff, J.; Novello, F.; Prugh, J.;
`Smith, R.; Willard, A. J. Med. Chem. 1985,28, 347-358. (c)
`Stokker, G. E.; Alberts, A. W.; Anderson, P. S.; Cragoe, E. J.;
`Deana, A. A.; Gilfillan, J. L.; Hirshfield, J.; Holtz, W. J.;
`Hoffman, W. F.; Huff, J. W.; Lee, T. J.; Novello, F. C.; Prugh,
`J. D.; Rooney, C. S.; Smith, R. L.; Willard, A. K. J. Med. Chem.
`1986,29, 17C-181.
`Endo, A. J. Med. Chem. 1985, 28, 401-5.
`Nakamura, C.; Abeles, R. Biochemistry 1985, 24, 1364-76.
`(a) Stetter, H. Angew. Chem., Int. Ed. Engl. 1976,15,639. (b)
`Stetter, H.; Kuhlmann, H. Chem. Ber. 1976, 109, 2890. (c)
`Stetter, H.; Schreckenberg, M. Chem. Ber. 1974,107,2453. (d)
`Stetter, H.; Kuhlmann, H. Synthesis 1975, 379.
`
`HO-0
`
`HO,,,-O
`
`9
`8
`7
`"(a) H,N-X-CN, HOAc, reflux. (b) DIBAL-H, toluene, -78 "C.
`(c) aqueous HCl. (d) HzN-X-CH(OEt)2, toluene, cat. p-TSA, re-
`flux. (e) -CH2CO-CHCH3CH3, THF, -78 OC. (0 n-Bu3B, NaBH,,
`-78 "C.
`(9) HzOZ, OH-.
`(h) Toluene, reflux.
`(i) HzN-X-OH,
`HOAc. 6 ) CH3S02C1, pyr. (k) KCN, DMF-H20, 100 "C.
`
`Scheme I11
`
`HO-0
`
`8U
`
`Scheme IV
`
`HO-0
`
`8v
`
`8k n, P
`81, rn, o
`hexane-1,4-dione (3q), which was extremely resistant to
`cyclization. After considerable experimentation, it was
`found that treatment with ethanolamine in acetic acid
`resulted in an exothermic reaction from which the pyrrole
`was isolated in 84% yield. Mesylation and displacement
`with potassium cyanide in DMF/H20 afforded the re-
`quisite nitrile. Reduction of the nitriles 5 with DIBAL-H
`produced the desired aldehydes 6 in good yields (Table 11).
`Condensation of 6 with the dianion of methyl or ethyl
`acetoacetate under the conditions of Weilerg afforded the
`corresponding alcohols 7. Sih et a1.I0 reported the re-
`duction of a related 6-hydroxy-@-keto ester in their syn-
`thesis of compactin in which little stereoselectivity (2:l
`erythro:threo) was found employing either sodium or zinc
`borohydride. We, and others,5b have found excellent se-
`lectivity (>10:1 erythro:threo) employing the procedure
`of Narasaka and Pai," in which 7 was complexed with a
`trialkylborane prior to treatment with borohydride at low
`temperature. The resultant boronate was hydrolyzed with
`
`(9) Huckin. S. N.: Weiler. L. J. Am. Chem. SOC. 1974. 96.
`1082-1087.
`(10) Wang. N. Y.: Hsu. C. T.: Sih. C. J. J. Am. Chem. SOC. 1981.
`' 103, z538-6539. '
`(11) (a) Narasaka, K.; Pai, H. C. Chem. Lett. 1980, 1415-1418. (b)
`Ibid. Tetrahedron 1984, 40, 2233-2238.
`
`I
`
`,
`
`
`
`I
`
`,
`
`
`
`NCI Exhibit 2030
`Page 2 of 11
`
`

`
`Inhibitors of Cholesterol Biosynthesis. 1
`- 4 1
`
`J o u r n a l of Medicinal Chemistry, 1990, Vol. 33, No. 1 23
`
`n
`
`Table 11. 2,fi-Disubstituted Pyrrol-1-yl Carbox- or Benzaldehydes
`
` %
`
`O n
`
`-8
`
`-8
`
`- 7
`
`-6
`
`-5
`
`I
`-1
`
`Figure 1. Correlation between CSI a n d COR ICm's.
`
`LOG COR IC50
`
`Table I. Substituted 1,4-Diketones
`RICOCH2CH2COR2
`3
`
`R,
`
`no.
`3aEb Ph
`3b
`4-FCeH4
`3~
`4-PhCeH4
`3dac 4-ClCeH4
`4-CH30C6H4
`3eh
`3f
`3-F3CCeH4
`3g
`3-CH3OCeH4
`3h
`2-CH3OCeH4
`3i
`2-naphthyl
`3j
`1-naphthyl
`3k
`
`A &-
`A
`
`3'
`3m"
`3~
`30
`3p
`3q
`3r
`38
`3t
`3~
`
`cyclohexyl
`PhzCH
`4-FCeHL
`4-FCiH;
`4-FCeH4
`4-FCBH4
`4-FCeH4
`4-FCeH4
`4-FCeH4
`
`% yield"
`(Drocedure)
`
`R,
`
`bp
`(mmHd. "C
`100 (0.1)
`46-8
`109-112
`116-8 (1.0)
`b
`b
`143-5 (0.2)
`133-5 (1.0)
`87-8
`105 (0.1)
`114-6 (1.0)
`
`b
`
`110 (4)
`b
`b
`133-5 (1.0)
`108-9 (0.2)
`132-3 (0.2)
`b
`132-5 (1.0)
`150-5 (0.1)
`b
`
`X
`THO
`
`Rl&* \ I
`
`6
`
`R1
`
`I-FCeH,
`
`4-FCeH4
`
`4-FCeH4
`
`R2
`
`% yieldnrb
`(method)
`
`4-FCeH4
`4-FCeH4
`4-FCeHd
`Ph
`4-PhCeH4
`4-CH3OCeH4
`4-C1C&
`3-F&H4
`3-CH3OCeH4
`2-CH3OCeH4
`2-naphthyl
`1-naphthyl
`
`cyclohexyl LB
`4
`
`PhzCH
`4-FCsH4
`4-FCeH4
`4-FCeH4
`4-FCeH4
`4-FCeH4
`4-FCaHA
`
`no.
`
`X
`
`6d
`-CH2CHzCHz-
`6e -CH(CH3)CHz-
`6f
`-CHZCHz-
`6g
`-CHZCHz-
`6h
`-CHpCHz-
`6i
`-CHzCHz-
`6j
`-CHzCH2-
`6k
`-CH2CH2-
`61
`-CHPCHp-
`6m
`-CHzCHz-
`6n
`-CHzCH2-
`60
`-CHzCHz-
`-CH&HZ-
`6p
`6q
`-CHzCHp-
`
`6r
`
`-CHzCHz-
`
`6s
`-CHzCHz-
`6t
`-CHzCHz-
`6~
`-CHZCHz-
`6~
`-CHzCHz-
`6~
`-CHzCHz-
`-CHZCHz-
`6 s
`6y
`-CHZCHz-
`62
`-CHzCHz-
`6aa
`-CHzCH2-
`6bb
`-CHzCH2-
`~ C C -CHzCHz-
`6dd
`-CH,CH,-
`6ee
`-CH;CHi-
`6ff
`-CH,CHI-
`6gg
`-CH,CH;-
`6hh -CHzCH2-
`6ii
`-CH2CHz-
`
`32
`b
`CHiCH;);
`2,4-F&H3
`3aa
`138-141 (0.2)
`CH(CH&
`2-CH30C6H4
`68 (B)
`3bb
`160-2 (2)
`2,6-(CH3O)zCeH3 CH(CH3)Z
`All spectral data were consistent with assigned structures.
`Purified by silica gel chromatography.
`
`aqueous peroxide and base.'* The dihydroxy acids were
`then lactonized by refluxing in toluene with azeotropic
`removal of water. Generally, the lactones were crystalline,
`such that the small amounts of the cis ladone stereoisomer
`9 present were easily removed by recrystallization, pro-
`viding >95% of the racemic trans stereoisomer (8). The
`conversion of 8u to 8v was accomplished by hydrogenation
`over Pd-C at 1 atm (Scheme 111). Finally, the phenol
`analogues 8k, 8h, and 8p were prepared from the corre-
`
`~~~~~
`
`Isolated yields after chromatography on silica gel. bAll compounds
`possessed 'H NMR spectra in accord with assigned structure (aldehydic
`proton, singlet, 6 8.95-9.65). eMp 70-3 "C. dMp 104-6 "C. Anal. C,
`H, N. 'Mp 105-7 "C. Anal. C, H, N.
`sponding methyl ethers Si, Sm, and 80 by BBr,-mediated
`demethylation (Scheme IV).13
`Biological Results
`The target lactones (8, Table 111) were saponified and
`tested for their ability to inhibit HMGR employing two
`protocols. Method 114 (cholesterol synthesis inhibition
`screen, or CSI) measured the rate of conversion of [14C]-
`
`(12) A detailed examination of this reaction has appeared: Ka-
`thawala, F.; Prager, B.; Prasad, K.; Repic, 0.; Shapiro, M.;
`Stabler, R.; Widler, L. Helu. Chirn. Acta 1986, 69, 803-5.
`
`(13) McOmie, J.; Watts, M.; West, D. Tetrahedron 1968,24, 2289.
`(14) Dugan, R.; Slakey, L.; Briedis, A.; Porter, J. Arch. Biochim.
`Biophys. 1972, 152, 21-7.
`
`NCI Exhibit 2030
`Page 3 of 11
`
`

`
`24 Journal of Medicinal Chemistry, 1990, Vol. 33, No. 1
`
`Roth et ai
`
`I
`
`60 00 100 120 140 100 180 200 220 240 200 280 300 320 340 300
`
`0
`
`20
`
`40
`
`COMPO
`
`- I
`----. 0cc
`
`-
`
`ANGLE
`
`-
`
`I11
`t ~ h a 0e (R
`t*-+ ann
`?+?Ex 0 x
`
`isomeP1
`
`Obb
`-*-
`0e IS isomer1
`0t lend01
`+*-c 0"
`
`tt-• 0f
`cf+ 0t lex01
`i i 4 i t a z
`Figure 2. cAMsE9-n energies calculated for comparahle orientations of the lactone side chain. Dashed lines represent less potent analogues
`(8j, 82,8bb, 8cc, and 8nn; CSI ICm > 5 rM).
`acetate to cholesterol employing a crude liver homogenate
`derived from rats fed a chow diet containing 5% chole-
`styramine. Method II', (CoA reductase inhibition screen,
`or COR) was a more specific screen employing a partially
`purified microsomal enzyme preparation to measure the
`direct conversion of o,b[W]HMG-CoA to mevalonic acid.
`The biological activities are reported as IC, values and
`as a ratio to compactin, which was employed as the internal
`standard in each testing protocol. Compactin consistently
`displayed an IC,, between 0.02 and 0.03 pM. The IC,
`values from the two assays were moderately correlated (eq
`1,'6 Figure 1).
`log (IC50, COR) = 0.81 (+0.09) log (ICM, CSI) - 1.32
`(1)
`
`aside from a length limitation of the 2-substituent (see the
`molecular modeling section below), no obvious structure-
`activity relationships could be discerned. Optimum po-
`tency resided in the 4-fluorophenyl analogue, 8f. With
`2-subtitution held constant as the optimal 4-fluorophenyl,
`potency increased with increasing length of the 5-sub-
`stituent from methyl (80 through cyclopentyl(8aa) to a
`maximum with isopropyl (8x) (length = 2.5 A; see mod-
`eling section below). Potency decreased thereafter to a low
`of >lo0 pM with 5-cyclohexyl substitution (8cc).
`With 5-substitution held constant as the optimal iso-
`propyl, additional variation of the 2-phenyl substituents,
`now keeping within the length limit of 5.9 A suggested by
`the modeling analysis (h-mrn), failed to improve the
`potency over the 2-(4-fluorophenyl)-5-isopropyl derivative,
`8s. Indeed, an additional "front-to-back" width limitation
`(Figure 3) may be apparent with 8ii and 8mm, which
`project significantly greater bulk in these directions than
`the other analogs. Finally, of interest is the 2-(4-flUOrO-
`phenyl)-5-trifluoromethyl analogue add, whose high po-
`tency may be due in part to stabilization of the pyrrole ring
`by the electron-withdrawing trifluoromethyl group, an
`aspect to be addressed in future communications.
`These results, combined with results from the molecular
`modeling study, confirmed our belief that 8x possessed the
`optimum substitution pattern, since structural modifica-
`tions at the 2- and 5-positions, as well as variation of the
`bridge to the lactone ring, led to decreased potency. A
`similar conclusion can be inferred from the examination
`of other 5-membered ring heterocycles reported in the
`patent 1iterature.l'
`(16) Compounds Be and 800 were assigned IC, values of 100 PM
`so they could be included in the correlation.
`(17) Kathawala, F. G. WIPO Patent WO 84/02131, 1984.
`
`n = 36, 9 = 0.70, F = 81, s = 0.39
`Structure-Activity Relationships
`As very little was known about heterocycle-containing
`inhibitors at the outset of this study, our strategy was to
`systematically examine each portion of the structure,
`keeping the 4-hydroxypyran-2-one ring intact. Initially,
`the optimum chain length between the lactone and the
`pyrrole ring was determined. A two-carbon bridge (80 was
`superior to either a three-carbon (8d) or aryl spacer (8a-c)
`(Table In). This is consistent with the fmdmgs of Stokker
`et ai."
`Holding the bridge constant as ethyl, the structure-ac-
`tivity relationships of the 2 and 5 pyrrole substituents were
`explored. With &methyl substitution (8f-w), high potency
`was conferred by bulky cycloalkyl 2-substitutents (8s-v).
`Among Z-(substituted-phenyI)-&methyl derivatives (8f-r),
`
`(15) Kita. T.; Brown. M.; Goldstein, J. J. Clin. Invest. 1980, 66,
`1094-1100.
`
`NCI Exhibit 2030
`Page 4 of 11
`
`

`
`Inhibitors of Cholesterol Biosynthesis. 1
`
`Journal of Medicinal Chemistry, 1990, Vol. 33, NO. 1 25
`
`Table 111. truns-6-(2-Pyrrol-l-ylalkyl or -aryl)-4-hydroxypyran-2-ones
`
`X
`
`Ri
`
`Rz
`
`d-
`
`4-FCeH4
`
`no.
`8a
`
`8b
`
`8C
`
`8d
`Be
`8f
`8g
`8h
`8i
`8j
`8k
`81
`8m
`8n
`80
`8P
`8q
`8r
`8s
`8t
`
`8U
`
`8 V
`
`8
`
`mp, OC
`155-7
`
`%
`formula'
`yield
`32 C22H2oFNOS
`
`relative
`ICm,b,r pM, log ICw, potency,d ICw>' pM, log ICw,
`CSI
`CSI
`CSI
`COR
`COR
`-
`-
`20
`-4.7
`0.10
`
`54-7
`
`29 C22H~$N03
`
`24
`
`-4.6
`
`0.01
`
`63
`
`-4.2
`
`142-5
`
`21 C22HaN03
`
`>lo0
`
`-4.0
`
`<0.01
`
`>lo0
`
`oil
`167-9
`oil
`89-91
`104-7
`95-96
`118-1 2 1
`161-2
`oil
`106-9
`144-5
`112-3
`140-2
`foam
`137-8
`129-130
`125-6
`
`41
`30
`32
`29
`35
`50
`28
`-
`65
`21
`-
`38
`-
`30
`21
`25
`20
`
`C19H25N03
`
`53
`5.0
`0.51
`1.4
`23
`12
`10
`2.6
`1.5
`2.5
`1.9
`2.1
`2.5
`16
`1.8
`0.69
`1.4
`
`-4.3
`-5.3
`-6.3
`-5.9
`-4.6
`-4.9
`-5.0
`-5.6
`-5.8
`-5.6
`-5.7
`-5.7
`-5.6
`-4.8
`-5.8
`-6.2
`-5.8
`
`0.02
`0.50
`0.90
`0.40
`0.10
`0.10
`0.20
`1.0
`0.30
`0.80
`1.40
`0.90
`1.10
`0.10
`0.70
`0.50
`1.10
`
`-
`40
`2.8
`13
`23
`28
`3.2
`6.3
`5.4
`11
`12
`25
`30
`3.6
`4.0
`2.2
`5.8
`
`-4.0
`
`-
`-4.4
`-5.6
`-4.9
`-4.6
`-4.6
`-5.5
`-5.2
`-5.3
`-5.0
`-5.0
`-4.6
`-4.5
`-5.4
`-5.4
`-5.6
`-5.2
`
`135-8
`
`13 C20H27N03'
`
`1.3
`
`-5.9
`
`1.60
`
`3.2
`
`-5.5
`
`135-9
`
`68 CzoHaNOs
`
`2.3
`
`-5.6
`
`1.10
`
`2.3
`
`-5.6
`
`8W
`8 X
`8Y
`82
`8aa
`8bb
`8cc
`8dd
`8ee
`8ff
`8gg
`8hh
`8ii
`8jj
`8kk
`811
`8mm -CH2CH2-
`
`129-132
`105-6
`117-8
`107-8
`CH(C;H&
`cyclopropyl foam
`cyclobutyl
`88-9
`cyclohexyl 64-6
`CF,
`oil
`87-9
`CH(CHJ2
`CH(CH&
`oil
`2,4-F&Hs
`CH(CH&
`75-7
`2-MeOC6H4
`CH(CH3):,
`oil
`2,6-(Me0)2C6H3 CH(CHd2
`foam
`2,5-Me2CBH3 CH(CH3)2 oil
`2-iPrOC&
`CH(CH3)2 oil
`2-ClC&
`CH(CH3)2
`foam
`CH(CH3)2 oil
`
`%LH,
`
`13
`33 C25H27N03
`0.40
`34 CZoH24FN03
`24 C,rH,cFNOq
`1.6
`20
`36 C,;HzFNO,
`2.2
`22 CzoHzzFN03
`17
`5 C21H24FN03
`30 C23H28FN03 >lo0
`58 CrsHr7FANOq
`0.25
`40 CiHGFNO3-
`1.3
`9 C2oH2rFNO:
`3.2
`8 CmH23FzN03
`1.6
`2.2
`l6 C21H27N04
`36 Cz2HzsNO5
`19
`25 CZ2HzsN0J,
`12
`12 CsH31N04'
`3.2
`25 CmHZ4ClNO3
`3.2
`34 CZ5HzsNOt
`9.6
`
`-4.9
`-6.4
`-5.8
`-4.7
`-5.7
`-4.8
`-4.0
`-6.6
`-5.9
`-5.5
`-5.8
`-5.6
`-4.7
`-4.9
`-5.5
`-5.5
`-5.0
`
`0.10
`30.2
`1.70
`0.10
`1.30
`0.20
`<0.01
`8.0
`1.8
`0.9
`1.5
`1.0
`0.2
`0.2
`0.9
`0.5
`0.2
`
`8.9
`0.23
`1.8
`32
`2.6
`-
`>lo0
`0.63
`2.6
`1.8
`2.6
`5.6
`87
`16
`-
`9.1
`25
`
`-5.4
`-6.6
`-5.7
`-4.5
`-5.6
`-
`-4.0
`-6.2
`-5.6
`-5.8
`-5.2
`5.2
`-4.1
`-4.8
`-
`-5.0
`-4.6
`
`CH(C2H5)2
`
`CH(C2H5)2 oil
`
`-
`-
`<0.01
`-4.0
`>IO0
`-CH2CH2-
`8nn
`I
`-7.6
`0.025
`-7.6
`0.026
`compactin
`100
`"Analytical results are within f0.4% of theoretical values unless otherwise noted. bCholesterol synthesis inhibition screen; a measure of the rate
`of conversion of [Wlacetate to cholesterol employing a crude liver homogenate. 'ICw values were determined with four dose levels of each inhibitor
`in the assay systems described in ref 14 (CSI) and 15 (COR). dCalculated as follows: (ICw of test compound)/(ICw of compactin determined
`simultaneously) X 100. 'CoA reductase inhibition screen; a measure of the direct conversion of D,L-['~C]HMG-COA to mevalonic acid employing a
`partially purified microsomal enzyme preparation. 'C: calcd, 75.62; found, 75.12. 'C: calcd, 72.92; found, 72.50. hC: calcd, 69.54; found, 71.37; H:
`calcd, 7.01; found, 7.54. IC: calcd, 74.33; found, 74.78. IC: calcd, 71.66; found, 72.09. *C: calcd, 73.69; found, 72.09.
`
`20 C21H35N03
`
`NCI Exhibit 2030
`Page 5 of 11
`
`

`
`y
`
`26 Journal of Medicinal Chemistry, 1990, Vol. 33, No. 1
`
`
`
`
`pyrrole Out Ethyl(ene) o f 0' t h e the bridge
`plane
`f a = 90-i:oo)
`
`0
`----__4_L ?
`
`OH
`
`/
`
`\
`
`Y
`
`Roth et al.
`
`S > z e c o n s t r a i n t s
`
`dl - less than 5 9 8.
`d2 - l e s s thali 3 3 ;i
`d j - l e s s than 10 6 A
`Figure 3. Summary of conclusions from the molecular modeling
`study.
`
`Molecular Modeling
`In order to identify the required spatial relationship
`between the lipophilic group (represented by the substi-
`tuted pyrrole, phenyl, and hexahydronaphthalene ring
`systems) and the 4-hydroxypyran-2-one moiety, quantify
`steric tolerances across the pyrrole ring, and evaluate the
`relationship between potency and the polarity (charge
`distribution) of the side chains, selected analogues from
`Table 111, compactin (I), and the potent biphenyl inhibitor
`111 were modeled by using the CAMSEQ-11 program pack-
`age18J9 (Table IV; see the Experimental Section). Con-
`formational preferences of the ethyl (or ethylene) bridge
`to the lactone ring, size of the R1 and R2 substituents
`(Table IV), and charge distribution were compared to
`potency in the CSI screen (at the outset of this study,
`affinities in the COR screen were unavailable for the
`majority of the analogues studied) in order to develop a
`pharmacophore model for HMGR inhibition.
`Lactone Side Chain Conformations. For reference
`purposes, calculated energies for the 0", 90°, MOO, and
`lowest energy conformations of 0 are summarized in Table
`IV. Figure 2 depicts the calculated energies for individual
`conformations. From Figure 2, all of the modeled com-
`pounds, including compactin (I), the biphenyl analogue 111,
`and the less potent analogues 8z,8bb, 8cc, and 8nn, can
`adopt an eneretically favorable conformation where the
`ethyl(ene) bridge is nearly perpendicular to the parent
`pyrrole, benzene, or hexahydronaphthalene ring systems.
`Indeed, for the potent derivatives 8t and 111, the calcula-
`tions show that the out of plane (0 = 80-110") orientation
`is the only one allowed. In addition, the reduced potency
`of the tert-butyl(8y) over the isopropyl (8x1 analogue may
`be explained by the fact that the out of plane conformation
`(0 = 110") of 8y is calculated to be energetically disfavored
`over the in-plane (0 = 0-70") orientations.
`Thus, it is concluded that a conformation of the
`ethyl(ene) bridge to the 4-hydroxypyran-2-one ring out of
`the plane (90-120") of the parent ring systems is consistent
`with increased potency as a HMGR inhibitor. Interest-
`ingly, this corresponds to the calculated minimum energy
`and not the X-ray conformationlb of compactin. The X-ray
`conformation represents a secondary minimum at 0 =
`
`(18) (a) Potenzone, R., Jr.; Cavicchi, E.; Weintraub, H. J. R.; Hop-
`finger, A. J. Comput. Chem. 1977,1, 187. (b) Potenzone, R.,
`Jr.; Hopfinger, A. J. A Demonstration of the CAMSEQ-II
`Software System In DHEW Publ. (FDA) (U.S.), Issue FDA
`78-1046, Structural Correlations of Carcinogenesis and Muta-
`genesis, 1978, pp 102-103.
`(19) In-house conversion of the program to run on an IBM 3033
`under MVS/TSO (J. W. Vinson, unpublished work).
`
`i
`a t
`Figure 4. Charge distribution of compactin and selected ana-
`logues. Hatched and open spheres represent positive and negative
`charges, respectively. Sphere size is proportional to the magnitude
`of the atomic charge.
`
`24.6", 1.2 kcal/mol higher in energy, probably due to
`packing interactions.
`Steric Tolerances. In determining steric tolerances,
`the substituents were somewhat arbitrarily assigned.
`Larger substituents such as substituted phenyl, nor-
`bornenyl, and the isobutyric ester on compactin were
`placed at R1 (Table IV); small alkyl groups were assigned
`to R,. Changing the assignment would affect the conclu-
`sions regarding these tolerances. Low-energy, extended
`conformations of the substituents were used in the distance
`calculations; other orientations of flexible groups such as
`CH(C,H,), could produce different distances.
`The maximum lengths of R1 and Rz and the overall
`width of the molecule across the parent ring system from
`R1 to R2 are given in Table IV. The calculations show a
`clear dependence of CSI potency on all three distances
`summarized in Figure 3. High potency (ICw < 1.6 pM)
`is observed only for those analogues whose (a) maximum
`length of R1 (Figure 3, d,) is <5.9 8, (Table IV: compare
`8f and Sj), (b) maximum length of R2 (Figure 3, d,) is <3.3
`8, (compare 8x and 82 or 8114, and (c) overall width
`(Figure 3, d3) is C10.6 A (compare 8y and 8bb). Other
`analogues not included in Table IV reinforce the length
`constraints at R,: the 2-naphthyl analogue 8q (d, = 6.40
`A) is less potent than the l-naphthyl (d, = 4.20 A), and
`the para-substituted derivatives 8h and 8i possess reduced
`potency.
`Charge Distribution. Initially, it was hypothesized
`that the spatial orientation of polar regions with relatively
`large partial charges within the molecule might be con-
`nected to CSI potency. Compactin contains two distinct
`regions of relatively large partial charges corresponding
`to the 4-hydroxypyran-2-one ring and the isobutyric ester
`side chain (Figure 4). The potent inhibitors 8f and 8x
`also present relatively large partial charges, albeit weaker
`in strength, in roughly the same region as this side chain.
`However, attempts to increase potency by more closely
`mimicking the polar regions associated with the isobutyric
`ester of compactin with the more polar 2- and 3-(methoxy
`and hydroxy)phenyl analogues 8m-p resulted in equipo-
`
`NCI Exhibit 2030
`Page 6 of 11
`
`

`
`Inhibitors of Cholesterol Biosynthesis. 1
`Table IV. Results of Modeling Studies on Compactin and Substituted Pyrroles
`
`Journal of Medicinal Chemistry, 1990, Vol. 33, No. 1 27
`
`5.0
`
`-37.10'
`
`-41.43e
`
`100'
`
`60°, -42.92'
`
`10.12
`
`5.0
`0.51
`
`-46.93'
`-40.92
`
`-27.09'a
`-39.27
`
`OD, -46.931'
`lW
`-10.03 Oo, -40.92
`
`10
`1.4
`
`67.11
`
`-44.98
`
`-16.40 90°, -44.98
`
`10.12
`7.66
`
`9.33
`7.22
`
`1.4
`
`19.63
`
`-43.65
`
`-15.01 70°, -44.65
`
`7.87
`
`0.40
`
`-46.64
`
`-45.06
`
`46.29 OD, -46.64
`
`1.6
`
`-47.77
`
`-24.10'
`
`100
`
`OD, -47.77
`
`10.12
`
`10.20
`
`20
`
`-52.35
`
`-50.97
`
`100
`
`Oo, -52.35
`
`10.99
`
`5.58 2.48 [
`
`@H,]
`
`,
`'
`*:&Ha
`>so bond from a-Me
`to lactone side chain
`from Oo to 60° by 20°
`5.58 2.48 as above
`5.58 1.50 methyl group (&) from
`0' to 60° by 10'
`5.89 1.50 as above
`3.64 1.50 bond from R1 to pyrrole
`from OD to 360° by 20°
`
`4.27 1.50 as above
`
`5.58 2.48 [ 4;]
`5.58 2.48 [
`5.58 3.74 [ +&&]
`
`H&H,]
`
`*&H3
`
`all bonds from Oo to
`60' by 20'
`
`"YY
`
`100
`0.026
`
`-47.28
`100
`10.17' -56.04'
`
`-46.46
`
`-44.82
`
`6.01 60°, -46.64
`
`17
`
`100
`
`-51.76
`
`-50.31
`
`100
`
`100
`100'
`
`OD, -51.76
`looo, -54.31
`120°, -61.74'
`
`H*?H,
`terminal methyls set
`to a staggered
`conformation
`5.58 3.35 bond from & to pyrrole
`from Oo to 360° by 20°
`5.58 4.33 bond from & to pyrrole
`from Oo to 360° by 20°
`3.74 3.74 see compound 82 above
`5.66 1.50
`
`10.62
`
`11.92
`
`9.41
`8.81
`
`H
`
`O
`
`W
`
` 0.01
`
`100
`
`-48.89
`
`100
`
`130°, -52.92
`
`8.74
`
`F$
`
`H3
`
`0
`
`CH,
`
`terminal alkyl groups
`set to a staggered
`conformation
`
`5.52 1.50 bond from & (Me) to
`phenyl from Oo to 60°
`by 20°; bond from R1
`(4-F,3-MeC6H3) to
`phenyl from Oo
`(biphenyl coplanar) to
`90' by 1 5 O
`
`8X
`
`4-FC6H4
`
`8y
`
`4-FCaH4
`
`82
`
`4-FCeHd
`
`I
`
`I11
`
`"CSI screen (see Table 111). *Counterclockwise rotation of 0 from 0 to 180° by loo, unless otherwise noted, starting from the in-plane conformation shown
`(atoms A, B, C, D in a cis orientation). Steric and electrostatic (using charges calculated via the CNDO/2 method) terms were used. Energies are in kilocalo-
`ries/mole. CAt each conformation of the lactone side chain, rotations were performed on the marked bonds from 0" to 180° by 20°, unless otherwise indicated.
`was scanned from OD to 250° by loD. f S stereoisomer. 8 8 = 110'
`Substituted phenyl rings at R1 were held perpendicular to the pyrrole. d R stereoisomer.
`conformer, -46.09 kcal/mol. hEndo isomer. 'Ex0 isomer. ' 8 = 70" conformer, -46.93 kcal/mol. kchair form; equatorial attachment to pyrrole. '-9 was scanned
`from 0' to 350" by loo.
`sensitive to the polarity of the group at R1.
`tent, not more potent, analogues. In addition, compounds
`containing bicyclo moieties at R1 (8t-v) demonstrated that
`Conclusions
`a polar substituent in this area (or an aryl ring, for that
`A series of 6-(2-pyrrol-l-ylethyl)-4-hydroxypyran-2-ones
`matter) was not required for CSI potency at the 1 FM level.
`(8) has been identified as inhibiting the enzyme HMG-CoA
`Thus, it is concluded that CSI potency is relatively in-
`
`NCI Exhibit 2030
`Page 7 of 11
`
`

`
`28 Journal of Medicinal Chemistry, 1990, Vol. 33, No. I
`reductase (HMGR). By measuring t h e inhibition of
`HMGR in vitro, the 2- and 5-substituents on the pyrrole
`ring have been optimized, thus obtaining a compound (8x1
`that possesses 30% of t h e in vitro potency of the potent
`fungal metabolite compactin.
`From a molecular modeling study, i t was determined
`that so long as the 2- and 5-substituents did not interfere
`with the ability of the ethyl bridge to t h e lactone ring t o
`attain an out-of-plane conformation (0 = 90-110°), and the
`substituents were within the distance contraints given in
`Figure 3, one could expect t o achieve potency at the 1 pm
`level in the CSI screen. Attempts to enhance potency by
`mimicking partial charges in the polar isobutyric ester side
`chain in compactin failed. It is concluded that there are
`no strong electronic requirements for binding in this area.
`In addition, the reduced potency of 8w, 8ii, and 8mm
`relative to other substituted phenyl derivatives suggests
`a steric intolerance off of one of the ortho phenyl positions
`of the