B1OORGANIC & MEDICINAL CHEMISTRY LETTERS B ioorganic & Medicinal Chemistry Letters 9 (1999) 1625-1630 Pergamon AMINO-SUBSTITUTED THALIDOMIDE ANALOGS: POTENT INHIBITORS OF TNF-ct PRODUCTION George W. Muller, *a Roger Chen, a Shaei-Yun Huang, a Laura G. Corral, a Lu Min Wong, a Rebecca T. Patterson, a Yuxi Chen, b Gilla Kaplan, b and David I. Stirling. a aCelgene Corporation, 7Powder Horn Drive, Warren, NJ07059, U.S.A. bRockefeller University, 1230 York Avenue, New York, NY10021, U.S.A. Received 31 March 1999; accepted 30 April 1999 Abstract: Thalidomide, (1), is a known inhibitor of TNF-c~ release in LPS stimulated human PBMC. Herein we describe the TNF-c~ inhibitory activity of amino substituted analogs of thalidomide (1) and its isoindolin-1- one analog, EM-12 (2). The 4-amino substituted analogs were found to be potent inhibitors of TNF-c~ release in LPS stimulated human PBMC. © 1999 Elsevier Science Ltd. All rights reserved. Introduction: Thalidomide (2-(2,6-dioxo-3-piperidyl)isoindoline-l,3-dione), (1) was developed as a sedative without the side effects of barbiturates in the 1950's by Chemie GmnenthalJ Thalidomide quickly became a popular sedative in Europe and Australia and was subsequently used for the treatment of morning sickness in pregnant women. However, thalidomide was removed from the marketplace when its use was linked to birth defects. Thalidomide's teratogenic properties made the drug infamous and catalyzed the development of the current drug approval regulations. A serendipitous discovery in 1965 by Sheskin while treating erythema nodosum leprosum (ENL), an acute inflammatory condition associated with lepromatous leprosy led to the discovery that thalidomide possesses immunomodulatory properties} Since this initial discovery, thalidomide has been found to afford clinical benefit in a variety of autoimmune and inflammatory disease states, s oo. o Thalidomide (1) EM-12 (2) In 1991 it was reported that thalidomide was a selective inhibitor of tumor necrosis factor-c~ (TNF-cg over production in stimulated human monocytes. 4 TNF-c~ is a key cytokine in the inflammatory cascade and elevated TNF-c~ levels are associated with inflammatory diseases. 5 Recent successful clinical trials in rheumatoid arthritis and inflammatory bowel disease with TNF-c~ antibodies and soluble TNF-c~ receptors have validated the inhibition of TNF-c~ as a clinical treatment. 6 The clinical activity of thalidomide and the importance of TNF-c~ inhibition led us to initiate a program to improve the TNF-cc inhibitory activity of thalidomide by structural modification. We have previously 0960-894X/99/$ - see front matter © 1999 Elsevier Science Ltd. All rights reserved. PIh S0960-894X(99)00250-4
`
`ALVOGEN, Exh. 1008, p. 0001
`
`

`

`1626 G. W. Muller et al. / Bioorg. Med. Chem. Lett. 9 (1999) 1625-1630 reported a series of thalidomide analogs derived from 13-amino-13-arylpropanoic acid derivatives that are potent inhibitors of TNF-~. 7 Further studies revealed these compounds to be potent inhibitors of phosphodiesterase type 4 (PDE4). 8 The PDE4 inhibitory potency for most of these compounds has correlated with their TNF-ct inhibitory activity. PDE4 is the major PDE isoenzyme present in monocytes and macrophages, key producers of TNF-ct. PDE enzymes control the levels of cyclic adenosine monophosphate (cAMP) by hydrolysis of cAMP to 5'-AMP. Inhibition of PDE4 in stimulated monocytes has been demonstrated to elevate levels of cAMP and inhibit of TNF-ct production. 9 In further studies to improve the TNF-ct inhibitory activity of thalidomide, we prepared a series of amino-phthaloyl substituted analogs of thalidomide (1) and its isoindoline-l-one analog, EM-12 (2). Some amino substituted thalidomide analogs have previously been reported but were not assayed for their TNF-ct inhibitory activity, l° EM-12 (2) has been reported to be a more potent teratogen than thalidomide in rabbits, rats, and monkeys. 11 When 2 was evaluated for TNF-ct inhibitory activity in LPS stimulated human PBMC it was found to have similar activity to thalidomide. The isoindolinone replacement of the phthaloyl ring increases the stability of the molecule and may lead to increased bioavailability. Herein, we report the structure-activity relationships of amino substitution of the phthaloyl ring of thalidomide and isoindolinone ring of EM-12 on the TNF-ct inhibitory activity in LPS stimulated human PBMC. Scheme 1 %H Cbz-L-glutamine a, b }. CI H3N~O 3 O O O H O2N O X O 4a X = 4-NO 2 5a X = 4-NH 2 4b X = 5-NO 2 5b X = 5-NH 2 Reagents: (a) CDI, THF, reflux; (b) H 2, 10% PdlC, EtOAcI4N HCI; (c) 3, AcOH, reflux; (d) 10% PdlC, acetone. Chemistry. The amino substituted analogs of thalidomide were prepared as illustrated in Scheme 1.12 The amino thalidomide analogs were prepared via the condensation of 3-aminopiperidine-2,6-dione hydrochloride, (3). Compound 3 was prepared in two steps from commercially available Cbz-L-glutamine. Treatment of Cbz- L-glutamine with carbonyl diimidazole (CDI) in refluxing THF afforded Cbz-aminoglutarimide. The Cbz protecting group was readily removed by hydrogenolysis under 50-60 psi of hydrogen in the presence of 10% Pd/C in a mixture of ethyl acetate and 4 N HC1. The hydrochloride (3) was used directly in the anhydride
`
`ALVOGEN, Exh. 1008, p. 0002
`
`

`

`G. W. Muller et al. / Bioorg. Med. Chem. Lett. 9 (1999) 1625-1630 1627 condensation reaction without purification. Treatment of 3 with 3- or 4-nitrophthalic anhydrides in refluxing acetic acid afforded the 4- and 5-nitro substituted thalidomide analogs 4a and 4b, respectively, in good yields. The nitro groups of 4a and 4b were reduced by hydrogenation in a Parr shaker under 50-60 psi of hydrogen in the present of 10% Pd/C to afford the desired 4- and 5-amino substituted thalidomide analog 5a and 5b, respectively. 13 The amino substituted isoindolinone analogs were prepared as illustrated Scheme 2.12 Treatment of 3 with the appropriately substituted nitro substituted methyl 2-(bromomethyl)benzoates, 6a-d yielded the four isomeric nitro EM-12 analogs 7a-d. The nitro groups were hydrogenated to the desired amino compound as described above to afford 8a-d. The four isomeric nitro substituted methyl 2-(bromomethyl)benzoates (6a-d) were prepared by benzylic bromination of the corresponding commercially available nitro substituted methyl 2- methylbenzoates. Scheme 2 0 0 0 / ~. 0 / a b,c~_ N O O2N 02 Br X 4 6a 3-NO 2 7a X = 4-NO 2 8a X= 4-NH 2 6b 4-NO 2 7b X = 5-NO 2 8b X= 5-NH 2 6c 5-NO 2 7c X = 6-NO 2 8c X= 6-NH 2 6d 6-NO 2 7d X = 7-NO 2 8d X= 7-NH 2 Reagents: (a) light, NBS, CCI 4, reflux; (b) 3, EtaN, DMF, 80 °C; (d) H2, 10% Pd/C, MeOH The R and S isomers of 5a were prepared starting from the S- and R-isomers of glutamine t-butyl ester (Scheme 3). The nitro substituted Nefs reagent analog, 10 was prepared by treatment of 3-nitrophthalimide with ethyl chloroformate. 14 Nef's reagent is a reagent commonly used in the preparation of chiral N-phthaloyl protected amino acids. Treatment of 10 with the single isomers of t-butyl glutamine afforded the phthaloyl glutamine derivatives, (S)- and (R)- 11. The t-butyl group was removed using standard acidic conditions to afford (S)- and (R)- 12. To avoid racemization, the ring closure was accomplished using the method reported by Casini and Ferappi 15 for the synthesis of the single isomers of thalidomide to afford (R)- and (S)- of 4a. 13 The nitro groups were reduced as described earlier in acetone to afford the single isomers of 5a. Scheme 3 ONco t • , ~ccd ~ N~)--O 0 2 X O O (S)- and (R)- 11 R = t-Butyl (S)- and (R)- 12 R = H NO 2 10 (S)- and (R)- 4a X = NO 2 (S)- and (R)- 5a X = NH 2 Reagents: (a) EtaN, (R) or (S) t-butyl glutamine HCI; (b) HCI, CH2CI2; (c) SOCI 2, pyr/Et3N; (d) H 2, 10% Pd/C, acetone.
`
`ALVOGEN, Exh. 1008, p. 0003
`
`

`

`1628 G. W. Muller et al. /Bioorg. Med. Chem. Lett. 9 (1999) 1625-1630 The 4-amino-ct-methyl analog (14) of thalidomide was prepared from ct-methylglutamic acid (Scheme 4). By standard chemistry ct-methylglutamic acid was converted to Cbz-c~-methylglutamic acid anhydride (15). Treatment of the anhydride with ammonia afforded a mixture of ct- and ,/-amides, 16. This mixture was cyclized with CDI to the Cbz-protected aminoglutarimide 17. The Cbz-group was removed by hydrogenation under acidic conditions to afford aminoglutarimide hydrochloride 18. Condensation with 3-nitrophthalic anhydride followed by reduction of the nitro group afforded 14. Scheme 4 O O H ZHN~-O a, b, c:,.. X~-~O 15 17 X = Z-NH 18 X = CI H3N d, e X O 19 X = NO= 14 X = NH 2 Reagents: (a) NH3, CH2CI2; (b) CDI, THF; (c) H2, 10% Pd/C, EtOH/4N HCI; (d) 3-NO2-phthalic anydride, AcOH, reflux; (e) H2, Pd/C, acetone. Biological Assays. TNF-~ inhibitory activity was measured in lipopolysacharide (LPS) stimulated PBMC as previously reported. 7 The human whole blood TNF-cc inhibition assay was run in a similar fashion to the PBMC assay except heparinized fresh human whole blood was plated directly into microtiter plates. The assay was then continued as previously reported for the PBMC assay. The assay for PDE4 enzyme inhibition was run as previously described. 8 Results and Discussion. Thalidomide has been reported to be a selective inhibitor of TNF-c~ in LPS stimulated human monocytes. 4 Thalidomide has a TNF-ct ICso 0f-200 p.M in LPS stimulated PBMC. 8 Previous research with thalidomide analogs suggested that phthaloyl substitution could lead to increases in activity. 8 Although the amino substitution had been previously described, these analogs had not been tested for their ability to inhibit TNF-ct production (Table 1). The 5-amino analog, 5b, was found to have a TNF-c~ IC50 of-100 p.M. No inhibitory activity was observed at the lower concentrations tested (less than or equal to 10 p.M). The 4-amino analog, 5a, was significantly more potent with an IC50 of 13 nM. Thus, this compound was -15,000 times more potent than thalidomide as a TNF-ct inhibitor in vitro. The novel isomeric amino-substituted EM-12 analogs were then prepared and tested. Unlike thalidomide where there are only two regio isomers, there are four possible regio isomers, 8a-d. Only the 4-amino analog 8a potently inhibited TNF-et production (IC50 less than 100 p.M). Compound 8a was found to have an IC50 of 100 nM (Table 1). This substitution correlates with the amino substitution on 5a and demonstrated that the amino group needed to be opposite to the carbonyl of the isoindolinone for optimal activity. The S- and R-isomers of 5a were prepared and evaluated. The S-isomer of 5a was found to be the more active isomer with a TNF-ct IC50 of 3.9 nM. The R-isomer was -20-fold less active with a TNF IC50 of 94 nM. Although (R)-5a's optical purity was greater than 95% ee, some activity was probably due to residual (S)-isomer in the sample.
`
`ALVOGEN, Exh. 1008, p. 0004
`
`

`

`G. W. Muller et al. / Bioorg. Med. Chem. Lett. 9 (1999) 1625-1630 1629 The a-methyl analog of thalidomide, 13, has also been reported to demonstrate similar TNF-a inhibitory activity to thalidomide.16 This compound does not contain the racemizable chiral center found in thalidomide. The 4-amino analog 14 was a potent inhibitor of TNF-a with an IC50 of 44 nM. Work is in progress to prepare the single isomers of 14 and will be reported on in the future. Compounds 5a, 8a, and 14 were evaluated for PDE4 inhibitory activity using PDE4 enzyme isolated from U937 cells. 8 All three compounds were inactive (<50% inhibition) at 100 p.M, the highest concentration assayed. These results strongly suggested that these compounds do not act by PDE4 inhibition. The three active analogs, 5a, 8a, and 14 were evaluated for their ability to inhibit TNF-a levels LPS stimulated human whole blood to mimic their activity in vivo. The compounds had only modest declines in activity in this assay. (Table 1). Table 1 TNF-a Inhibition in LPS Stimulated Human PBMC and Whole Blood Compd TNF-a Inhibit. TNF-cz Whole Blood At 100 jxM ICs0 TNF-a ICs0 5a 95% 13 nM 25 riM 5b 55% -100,000 nM ND 8a 74% 100 nM 480 nM 8b 15% ND ND 8e 12% ND ND 8d 18% ND ND 14 98% 44 nM 216 nM (S)-5a 99% 3.9 nM 14 nM (R)-5a 85% 93 nM 73 nM In summary, we have discovered three high potency inhibitors of TNF-a by 4-amino substitution of thalidomide, EM-12, and a-methylthalidomide. The (S)-4-amino substituted analog of 5a was found to be -50,000 times more potent than thalidomide at inhibiting TNF-c~ levels in LPS stimulated human PBMC. None the three compounds showed significant activity as a PDE4 inhibitor. A recent publication reported 14 to enhance TNF-a production in 12-O-tetradecanoyl-phorbol 13-acetate stimulated human leukemia HL-60 cells.17 These discordant results are possibly related to our use of primary human cells stimulated with LPS in contrast to the other investigators use of the HL-60 cell line stimulated with TPA. Further, we have demonstrated that these compounds retain high activity in the milieu of whole human blood. We are presently investigating the structure-activity relationships of other substituted phthaloyl and isoindolinone analogs of thalidomide and EM- 12 and will be publishing on the biological profiles of 5a, 8a, and 14.18
`
`ALVOGEN, Exh. 1008, p. 0005
`
`

`

`1630 G. W. Muller et al./ Bioorg. Med. Chem. Lett. 9 (1999) 1625-1630 References and Notes 1. Stifling, D. I. Pharmaceutical News 1996, 3, 17. 2. Sheskin, J. Clinical Pharmacol. Ther. 1965, 6, 303. 3. (a) Gutierrez-Rodriguez, O. Arth. And Rheum. 1984, 27, l118. (b) Peterson, D. L.; Georghiou, P. R.; Allworth, A. M.; Kemp, R. J. Infection. Clin. Infect. Dis. 1995, 20, 250. (c) Schuler, U.; Ehninger, G. Drug Safety, 1995, 12, 364. (d) Vogelsang, G. B.; Hess, A. D.; Gordon, G.; Brundrette, R.; Santos, G. W. Transplantation Proc. 1987, X/X, 2658. (e) Klausner, J.; Makonkawkeyoon, S.; Akarasewi, P.; Nakata, K.; Kasinrerk, W.; Corral, L.; Dewar, R.; Lane, C.; Freedman, V.; Kaplan, G. J Acq. Immun. Def. Synd 1996, 11,247. (f) Stifling, D.; Sherman, M.; Strauss, S. J. Amer. Pharm Assoc. 1997, NS37(3). 4. Sampaio, E. P.; Samo, E. N.; Galilly, R.; Cohn, Z. A.; Kaplan, G. J Exp. Meal 1991, 173, 699. 5. Eigler, A.; Sinha, B.; Hartmann, G.; Endres, S. Immunology Today 1997, 18, 487. 6. (a) van Dullemen, H; van Deventer, S; Hommes, D; Bijl, L; Jansen, J; Tytgat, G; Woody, J. Gastroenterology 1995, 109, 129. (b) Elliott, M.; Feldmann, M.; Maini, R. Int. J. Immunopharmac. 1995, 17, 141. 7. Muller, G. W.; Corral, L. G; Shire, M. G.; Wang, H.; Moreira, A.; Kaplan, G.; Stirling, D. I. J. Med Chem 1996, 39, 3238. 8. Muller, G.W.; Shire, M. G.; Wong, L. M.; Corral, L. G.; Patterson, R. T.; Chen, Y.; Stirling, D. I. Bioorg Med Chem. Lett. 1998, 8, 2669. 9. Torphy, T. J. Am. J. Resp. Crit. Care 1998, 157, 351. 10. He, W.; Engstrom, L.; Walker, S.; Vogelsang, G.; Leong, K. W. Abstracts of Papers, 206 th American Chemical Society, Chicago, IL, Fall 1993; Med. Chem., paper 216. l l. Heger, W.; Schmahl, H.; Klug, S.; Felies, A.; Nau, H.; Merker, H.; Neubert, D. Teratogenesis. Carcinogenesis, and Mutagenesis 1994, 14, 115. 12. Muller, G. W.; Stifling, D. I.; Chen, R. S. C. United States Patent No. 5,635,517, 1997. Chem Abstr. 1997, 127, 86110. 13. Chiral HPLC analyses were carried out at 240 nM using a Daicel chiralpak AD column (0.46 x 25 cm) eluting with 30/70 hexane/IPA at 0.75mL/min. 14. Siedlecka, R.; Skarzewski, J. Pol. Synth. Commun. 1997, 27, 2081. 15. Casini, G.; Ferappi, M. Farmaco, Ed. Sci. 1964, 563. 16. Wnedt, S.; Finkam, M.; Wniter, W.; Ossig, J.; Raabe, G; Zwingenberger, K. Chirality 1996, 8, 390. 17. Miyachi, H.; Koiso, Y.; Shirali, R; Niwayama, S.; Liu, J. O.; Hashimoto, Y. Chem. Pharm. Bull. 1998, 46, 1165. 18. Corral, L. G.; Haslett, P. A.; Muller, G. W.; Chen, R.; Wong, M. L.; Ocampo, C.; Stirling D. I.; Kaplan, G. J. lmmun, accepted.
`
`ALVOGEN, Exh. 1008, p. 0006
`
`