Practical Synthesis of MDM2 Antagonist RG7388. Part 1: A Cu(II

Subscriber access provided by University of Otago Library

Full Paper

Practical Synthesis of MDM2 Antagonist RG7388. Part 1: A Cu(II)-Catalyzed Asymmetric [3+2] Cycloaddition Lianhe Shu, Chen Gu, Dan Fishlock, and Zizhong Li Org. Process Res. Dev., Just Accepted Manuscript • DOI: 10.1021/acs.oprd.6b00320 • Publication Date (Web): 31 Oct 2016 Downloaded from http://pubs.acs.org on November 11, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Organic Process Research & Development is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Organic Process Research & Development

Practical Synthesis of MDM2 Antagonist RG7388. Part 1: A Cu(II)-Catalyzed Asymmetric [3+2] Cycloaddition Lianhe Shu,*† Chen Gu,‡ Dan Fishlock,§ and Zizhong Li‖

Process Research and Synthesis, Hoffmann-La Roche Inc., 340 Kingsland Street, Nutley, New Jersey 07110, United States

ACS Paragon Plus Environment

1


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48

Page 2 of 27

Graphic for TOC

Cu(OAc)2 3

(R)-BINAP DIPEA MeTHF

15, exo

10 M NaOH THF-MeOH

99.6% ee RG7388 (1) 8 16, endo


2

Page 3 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


ABSTRACT An efficient asymmetric synthesis of MDM2 antagonist RG7388 is reported. The highly functionalized chiral pyrrolidine carboxamide was assembled via a Cu(OAc)2/(R)-BINAP catalyzed asymmetric [3+2] cycloaddition, which gave the exo and endo adducts in a ratio of 10:1, with high enantiomeric excess for the exo isomer. A one pot hydrolysis and retro-Mannich/Mannich isomerization of the cycloaddition adducts in the presence of aqueous sodium hydroxide afforded RG7388 in high chemical and enantiomeric purities and 69% overall yield. KEYWORDS MDM2 Antagonist, Pyrrolidine carboxamide, Asymmetric [3+2] Cycloaddition, Copper


3


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 27

INTRODUCTION RG7388 (1, Chart 1),1 a chiral pyrrolidine carboxamide, is a small molecule MDM2 antagonist being investigated as a potential treatment for a variety of solid tumors and hematologic malignancies. Compared to compounds from the imidazoline series, e.g., RG7122 (2),2 the first MDM2 antagonist that has demonstrated proof of mechanism in the clinic, this molecule showed superior efficacy in xenograft models and pharmacological properties in preclinical studies. The Phase 3 trial of this compound is currently ongoing. Chart 1

1, RG7388

2, RG7112

In the original synthesis, 1 was prepared in five steps and ~10% overall yield from 3 (Scheme 1). This process was deemed not suitable for large scale preparation due to several issues: 1) the low overall yield, 2) the use of stoichiometric amount of silver fluoride, which caused the formation of silver mirror on the wall of the glassware, 3) the use of a Class I solvent,3 and 4) the chromatographic purification, especially the chiral supercritical fluid chromatography (SFC) for the isolation of the desired enantiomer 6.


4

Page 5 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Scheme 1 Original synthesis of 1

a

3

b

rac-5

4

c

d

6

rac-6

e

7

1

Reagents and Conditions: a) AgF (1.2 equiv), Et3N, ClCH2CH2Cl, 31%. b) 2 M NaOH, MeOH, 97%. c) chiral SFC separation, 49%. d) methyl 4-amino-3-methoxybenzoate, Ph2POCl, DIPEA, CH2Cl2, chromatography, 81%. e) aq. NaOH, MeOH, THF, chromatography, 80%. In order to fulfill the immediate demand of API for the toxicology studies in animals, a convergent catalytic asymmetric synthesis was developed, which provided the desired product 1 in 2 steps from 3 without chromatography (Scheme 2). Although this synthesis has been scaled up in the pilot plant and supplied the API required for initial clinical studies in Phase 1, it became evident that the process may not be practical for commercial scale production. Besides the concern of the availability of the chiral ligand 114 (Chart 2), the isomerization of 9 to 10, using milled powder of anhydrous lithium hydroxide, was inconsistent during the scale-up campaigns. In addition, the complex composition of


5


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 27

isomers/intermediates in the key cycloaddition step indicated that multiple transition states were involved, which was the major challenge for further process improvement. Scheme 2 Silver-catalyzed asymmetric synthesis of 1

a 3

Intermediates

2 3 4

5

8 9 mixture of isomers

c

b 2 3 4

5

10

1

Reagents and Conditions: a) AgOAc (1.0 mol%), 11 (1.1 mol%), 4 °C, MeTHF, not isolated. b) anhydrous LiOH (milled), 60 °C, then n-heptane, 92-96% yield, up to 68% ee. c) i. LiOH, iPrOH−H2O, 70 °C; ii. filter off lithium salt of rac-1; iii. AcOH-H2O, 70 °C, 42% yield, >99% ee.


6

Page 7 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Chart 2

11, (R)-MeOBIPHEP

In the present paper, we describe the development of a copper/(R)-BINAP catalyzed asymmetric synthesis, which gives 1 with high chemical and chiral purities in 69% overall yield from 3, without the issues encountered with the use of silver catalyst. RESULTS AND DISCUSSION Silver and copper are the most commonly used metal catalysts for the [3+2] cycloadditions of azomethine ylides to electron-deficient olefins. While silver-catalyzed reactions normally proceed via the endo transition states, copper catalyzed reactions selectively give the exo adducts as the major products.5 Previously during the synthesis of a spiroindolinone analogue of 1, it has been discovered that the stereochemistry of the product from the cycloaddition step is dictated by the configuration of C-3.6 Once the desired chirality at C-3 is established, the chiral centers at C-2, C-4 and C-5 could be inverted to give the target product, which is the thermodynamically most stable isomer. The conversion of 9 to 10 upon treatment with lithium hydroxide suggests that compound 9 behaves in the same manner as the spiroindolinone analogue. Therefore the isomerization process used for the previous compound6 could potentially be applied to this synthesis.


7


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 27

The cycloaddition reaction was initially examined using (R)-BINAP as chiral ligand in combination with four different copper salts, and methyl ester 126 was used instead of 8. Cu(CH3CN)4PF6 has very good solubility in the organic media, but its reaction was poorly reproducible. The complex of CuOAc with (R)-BINAP is sparingly soluble in THF and the reactions were also inconsistent. For copper(II) salts, the complex of Cu(OAc)2 and (R)-BINAP showed better solubility than that of Cu(OTf)2 in THF, and thus was selected for investigation. The reaction gave a mixture of 13 (exo) and 14 (endo) in a ratio of ~10:1 (Scheme 3), indicating that it proceeded primarily via concerted transition states as expected.5 A number of unidentified minor isomers (~20 area% in total) were detected by LC-MS as they all exhibit the same molecular weight and mass fragment patterns with 13. Upon addition of methanol, the major isomer, 13, precipitated and was collected by filtration in 42% yield and 99.7% ee. Scheme 3 Cu(OAc)2-catalyzed reaction of 3 with 12

Cu(OAc)2 (R)-BINAP 3 NEt 3 THF

12

13

exo

14

endo

When the crude reaction mixture containing 13, 14 and the copper salt was treated with 1 equiv of DBU at reflux, the isomerization indeed occurred to afford 7 as the major product, however the yield was low due to significant degradation. On the other hand, when pure 13 was exposed to DBU in refluxing THF for 5 h, no sign of degradation or isomerization was observed and 13 was cleanly


8

Page 9 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


recovered. Pure 13 was also very stable when treated with fine powder of anhydrous lithium hydroxide in THF. After the mixture was stirred at 60 °C for 5 h, only trace amount of 7 was detected by HPLC. Scheme 4 Conversion of 13 to 1

aq. NaOH THF

1

13

Subsequently, it was discovered that 13 could be directly converted to the final product when treated with aqueous sodium hydroxide (Scheme 4). And 1 was obtained in >99% ee and up to 40% yield from 3. The hydrolysis/isomerization of the endo isomer 14 could also be performed under the same conditions.7 The inversion of C-5 in 13 likely proceeds via a retro-Mannich/Mannich isomerization cascade.6,8 Interestingly, compound 13 and its isomers differ from the spiroindolinone analogues6 in terms of the base needed for such transformation. The isomerization of the latter occurred in the presence of DBU. With aqueous sodium hydroxide, no apparent isomerization was observed.9 Optimization of the [3+2] Cycloaddition Step Since the exo product 13 was obtained in high enantiopurity and successfully converted to 1, the goal of the optimization was to maximize the formation of this isomer in the [3+2] cycloaddition. The methyl ester 12 was initially used, but later replaced by the ethyl ester 8 as the Kilo Lab was using the latter for large-scale campaigns. This change had no impact on the selectivities. The reactions were


9


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 27

carried out at room temperature with 1 or 2 mol% of Cu(OAc)2 and (R)-BINAP in 6 volumes of solvents. Lowering the reaction temperature to 0 °C resulted in significantly slower reaction, while the effect on reaction selectivities was negligible. Dichloromethane, THF, MeTHF, CPME, and toluene were examined as the solvent. Poor selectivity (~45 area%) for the exo isomer was observed in dichloromethane. In CPME, MeTHF and toluene, the reactions gave the exo adducts in ~80 area% as compared to ~75 area% in THF. MeTHF was selected for further investigation as the reaction was faster in this solvent than in CPME and toluene. In the absence of base, the reaction of 3 with 8 or 12 proceeds slower and results in a higher level of impurities.10 Three bases, triethylamine, DIPEA, and DABCO, were tested and found to work equally well on the suppression of impurities arising from the competing stepwise processes. One equivalent of the base was sufficient for the reaction to complete in 24 h, and no further improvement was observed when an excess amount of base was used. The ligand screening was conducted with Cu(OAc)2 (1.0 mol%), phosphine ligand (Chart 3) (1.1 mol%) and DIPEA (1 equiv) in MeTHF at room temperature for 2 days to ensure complete conversion achieved. All reactions gave a mixture of 15 and 16 (Scheme 5) as the major products (Table 1). At the time of this study, it was not possible to evaluate the performance of the ligand at this stage, due to the lack of a chiral HPLC method that could separate all the enantiomers. Therefore the reaction mixtures were treated with aqueous sodium hydroxide to convert both 15 and 16 to 1, and the resulting mixtures were analyzed by chiral HPLC. As summarized in Table 1, reactions with better selectivity for the exo isomer 15 (entries 1, 3, 5−7, 14−18) generally led to higher ee for 1. The best ee obtained was 90.7% (entry 3) with ligand 19, as compared to 89.0% ee (entry 18) with (R)-BINAP. The modest gain in enantioselectivity, however,


10

Page 11 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


did not justify a switch from a more readily available and inexpensive ligand, thus the studies continued with (R)-BINAP.


11


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48

Page 12 of 27

Scheme 5 Cu(OAc)2-catalyzed reaction of 3 with 8

Cu(OAc)2 Chiral Ligand 3

NaOH

8

1 DIPEA MeTHF

MeTHF-MeOH

15

exo

16

endo


12

Page 13 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48


Table 1. HPLC result of the ligand screening entry

ligand

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

17 18 19 20 21 22 23 24 25 26 27 28 29 30 11 31 32 (R)-BINAP

HPLC area11 15 : 16 91 : 9 29 : 71 94 : 6 9 : 91 90 : 10 94 : 6 94 : 6 86 : 14 10 : 90 47 : 53 56 : 44 68 : 32 86 : 14 90 : 10 93 : 7 91 : 9 92 : 8 91 : 9

ee of 1 after hydrolysis/isomerization 80.8 25.4 90.7 39.8 83.2 83.8 86.3 66.4 77.9 -20.0 -58.0 60.0 71.3 83.0 84.0 84.8 86.0 89.0


13


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48

Page 14 of 27

Chart 3

Ar = 3,5-di-Me-Ph 17

22

28

18

23

29

Ar = p-Tolyl 19

20

24

30

25

31

21

26

27

32


14

Page 15 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Since copper could catalyze the [3+2] cycloaddition in the absence of a ligand, it is critical to pre-form the metal/ligand complex to minimize the background reactions. When Cu(OAc)2 and (R)-BINAP were mixed in MeTHF and stirred for 2-3 h before the addition of the substrates and DIPEA, the reaction was complete after stirring at room temperature overnight, and the ratio of exo to endo was ~10:1. Short catalyst aging (i.e.,

Practical Synthesis of MDM2 Antagonist RG7388. Part 1: A Cu(II

Recommend Documents