Scalable Methods for the Removal of Ruthenium ... - ACS Publications

Subscriber access provided by UNIV OF NEBRASKA - LINCOLN

Review

Scalable Methods for the Removal of Ruthenium Impurities from Metathesis Reaction Mixtures Philip Wheeler, John Hudson Phillips, and Richard L. Pederson Org. Process Res. Dev., Just Accepted Manuscript • DOI: 10.1021/acs.oprd.6b00138 • Publication Date (Web): 07 Jun 2016 Downloaded from http://pubs.acs.org on June 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

Scalable Methods for the Removal of Ruthenium Impurities from Metathesis Reaction Mixtures Philip Wheeler,* John H. Phillips, Richard L. Pederson AUTHOR ADDRESSES. Materia, Inc. 60 N San Gabriel Blvd. Pasadena, CA 91107, 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 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 27

TABLE OF CONTENTS GRAPHIC

Summary of methods for ruthenium removal used in clinical and commercial API syntheses.


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


KEYWORDS. Olefin metathesis, ruthenium metal removal, process chemistry.

ABSTRACT. To support the application of ruthenium-catalyzed olefin metathesis in drug development, we present a summary of the metal removal techniques employed in metathesis reactions intended for clinical or commercial API synthesis. Both extractive and adsorptive methods are discussed in the context of chemical process development leading to successful drug substance deliveries with sufficiently low residual ruthenium to meet regulatory specifications.


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 Since the first reports by Schrock and Grubbs of well-defined active catalysts,1 homogeneously-catalyzed olefin metathesis has found countless applications from polymer chemistry to pharmaceutical synthesis.2 Ruthenium-based catalysts (Chart 1) have emerged as the catalysts of choice for most small molecule synthesis applications, owing to their superior functional group tolerance and stability to handling.3 In the pharmaceutical industry in particular, ruthenium metathesis catalysts are prevalent in both medicinal and process chemistry applications.4 Chart 1. Homogeneous Ru Catalysts Discussed in this Review

PCy3 Cl Ru

PCy3 Cl Ru

Cl

N Ru Cl Cy3P

Cl Cy3P

O

1

N

2

N

N Cl

O

3

N

Ru

Cl

Cl NO2

N Cl

N Ru

Ph

Ru Cl

N Cl

Cl O

Cy3P 4

5

6

The posited improvement in clinical success rate for molecules with higher fraction-sp3, at least one chiral center, and greater overall 3D-character has increased demand for synthetic techniques capable of forming such structures.5 Ring-closing metathesis (RCM) is one such technique that has attracted considerable attention for generating sp3-rich scaffolds.6 Although


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


the reactive centers are sp2-hybridized, RCM can be used to close highly saturated and densely functionalized rings. The double bond in the product can also serve as an entry point to new chiral sp3 centers.7 These features make it a particularly good fit for the efficient synthesis of several key classes of so-called “escape from flatland” structures. Macrocyclic peptides, particularly HCV NS3/4A serine protease inhibitors, have served as a proving ground for olefin metathesis in chemical process development (Chart 2, 7-11). The scale-up of these challenging active pharmaceutical ingredients (APIs) was pioneered by Boehringer-Ingelheim (7) and further explored by several other process chemistry teams, including those at Janssen (8), Abbvie (9), and Merck (10, 11).8 Other highly substituted, saturated heterocycles produced in multi-kilogram quantities by olefin metathesis include an azapane (12) and spirocyclic amine (13). In each of these syntheses, olefin metathesis serves a crucial role in the overall efficiency of the process.


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

Chart 2. Examples of APIs and Drug Candidates Prepared by Olefin Metathesis.


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


To date, a total of four compounds for which olefin metathesis was investigated as a synthetic approach during development have reached the market as approved therapeutics,9 with more in various stages of preclinical and clinical development. As the use of olefin metathesis on scale has developed, so have the methods for the removal of ruthenium catalyst byproducts. Like other transition metal impurities, ruthenium levels in a final drug product must be carefully controlled according to regulatory guidelines. As such, specific purification steps may be necessary to eliminate residual ruthenium from intermediates and/or drug substances to meet the criteria specified by regulatory bodies, typically to less than ten parts-per-million (100 kg to deliver material for clinical studies, as described in a previous article.22 Scheme 2. Pilot Plant Scale RCM to Deliver BILN-2061a

a

Conditions: (i) 1 (3 mol %), PhMe (0.014 M), 80 °C, then sequential MNA/NaHCO3 washes

and charcoal filtration; (ii) 21, Cs2CO3, NMP, 50 °C, then charcoal treatment and crystallization (90%); (iii) LiOH (2 equiv), THF, H2O, 40-45 °C (90%). A more detailed study focused on the removal of residual ruthenium from the RCM product of diene acid 25 was reported the following year.23 Several scavengers were screened to remove ruthenium from crude 26, which was prepared with varying levels of residual Ru by changing catalyst and loading (Table 1). In this screen, it appeared that THMP was the most effective scavenger.24 When combined with a charcoal treatment, Ru levels as low as 32 ppm could be achieved in macrocycle 26, which could be taken on to BILN-2061 with acceptable purity.


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 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 27

Entry

Catalyst (loading)

Equivalents of THMP (17)

Ru level in 26 (ppm)

1

2 (5 mol%)

25

623

2

3 (3.5 mol%)

60

260; 32a 64a

3 (3.5 mol%) 40 3 a After an additional charcoal treatment and filtration. Table 1. RCM of 25 and Screened Ru Scavengers

However, the long processing times and large excess of scavenger needed for these workup procedures led the team at Boehringer Ingelheim to develop a semi-continous supercritical fluid extraction using supercritical CO2. It was found that the presence of an organic solvent improved the solubility of the product in the sCO2 layer. In the event, the crude reaction mixture (0.01M in toluene) containing 5 mol % of 1 (50,000 ppm Ru) was injected directly into a pressure vessel with a continuous stream of CO2 at 40 °C and 138 bar and extracted without equilibration. After the extraction was complete, the mixture was stirred with charcoal and filtered, resulting in 90% recovery of 26 with 100-115 ppm of residual ruthenium (Table 2).

Run

Ru before charcoal treatment (ppm)

Ru after charcoal filtration (ppm)

1

708

100

839 115 2 Table 2. Ruthenium Removal from 26 by Semi-continuous sCO2 Extraction (40 °C, 138 bar)


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 49 50 51 52 53 54 55 56 57 58 59 60


As the development of BILN-2061 progressed, it was found that substitution of the secondary amide present in the ring helped favor cyclization, and a more suitable N-Boc-protected RCM precursor 27 was identified.25 The combination of a more optimal substrate and secondgeneration catalyst 4 allowed for reduced ruthenium catalyst loading (0.05-0.1 mol%) while maintaining greater than 90% yield in the macrocyclization.26 With lower catalyst loading, significantly less MNA was needed to reduce ruthenium levels to ≤50 ppm in 28. Aside from standard crystallization, no additional treatment downstream was necessary to achieve 5 ppm or less in final API (Scheme 3). Scheme 3. Optimized Route to BILN-2061a

a

Conditions: (i) 4 (0.1 mol%), PhMe (0.2M), 110 °C, then sequential MNA/NaHCO3 washes

(95%); (ii) C6H5SO3H (2 equiv), PhMe, 70 °C (95%); (iii) LiOH (1.3 equiv), THF, 0-5 °C


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 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 27

(89%); (iv) p-BrC6H4SO2Cl, Et3N, cat. DMAP, CH2Cl2 23 °C (93%); (v) 21, Cs2CO3, NMP, 50 °C, then crystallization (90%); (vi) LiOH (2 equiv), THF, H2O, 40-45 °C (90%). In closely related work, scientists at Janssen developed conditions to close the 14-membered macrocycle present in TMC-435 (simeprevir), a macrocyclic NS3 protease inhibitor, using second generation catalyst 5 (Scheme 4).27 Taking a cue from the protection strategy reported for BILN-2061, the team at Janssen added a chlorodifluoroacetyl group to the amide to help favor cyclization, and found it to be much easier to remove than a Boc group. They also found that the addition of an iodide source such as tetraethylammonium iodide improved the rate and yield of the reaction. In the patent disclosure, catalyst loading of 6 mol% in refluxing CH2Cl2 is described; however in a more recent publication, 0.3 mol % of 5 in refluxing toluene is reported for a similar RCM.28 Upon complete reaction of diene 30, an aqueous solution of MNA and dimethylamine is added to quench the catalyst and cleave the acyl group, respectively. After extractive workups, the desired macrocyclic product 31 is isolated by crystallization from 2butanone in 70% yield. This intermediate is then converted to the API by saponification and amide bond formation. Unfortunately, the level of residual ruthenium in neither 31 nor the final API was disclosed.


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


Scheme 4. Reported RCM Conditions enroute to TMC-435a

a

Conditions: (i) 5 (6 mol %), Et4NI, chlorodifluoroacetic anhydride, CH2Cl2 (30 vol), 40 °C;

then aqueous MNA and Me2NH; then crystallization; (ii) NaOH, EtOH/H2O, reflux (90%) (iii) EDCI, CH2Cl2, 20 °C, then cyclopropanesulfonamide, DBU (89%). In the large scale synthesis of SB-462795, a clinical candidate developed by GlaxoSmithKline, RCM was used to access the azapane ring present in the API. Both aminoalcohol diastereomers of 32 were explored as potential intermediates, each of which could be elaborated to the desired product after hydrogenation of the olefin, amide coupling with valine fragment 33, and oxidation to the ketone (Scheme 5).17


15


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 16 of 27

Scheme 5. Potential Approaches to SB-462795 via RCM.

To access cis-32, the RCM of diene 34 is carried out in EtOAc at 60 °C using 1 mol% of catalyst 6 (Scheme 6). The RCM is followed by extraction with a basic aqueous solution of cysteine, and after crystallization the product was isolated in 89% yield with 148 ppm residual ruthenium. After two more steps, including a hydrogenation using Pd/C, cis-32 was isolated with 14 ppm of residual ruthenium.


16

Page 17 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 6. RCM of 34 and Subsequent Stepsa

a

Conditions: (i) 6 (1 mol%), EtOAc (15 vol), 60 °C, then cysteine/NaOH, then crystallization

(90%); (ii) H2, Pd/C, DME; (iii) aq. NaOH, reflux (91% over 2 steps). In the synthesis of the diastereomeric azapane trans-32, phthalyl-protected amino alcohol 35 is used as the RCM precursor (Scheme 7). This substrate was found to be exceptionally suitable for the RCM, with complete conversion achievable with only 0.1-0.2 mol % catalyst loading, depending on substrate purity. Initially, the RCM was run with 0.5 mol % 6 in toluene at 110 °C , from which the product readily crystallized. The crude product slurry was treated with THMP (17), formed in situ by treating tetrakis(hydroxymethyl)phosphonium chloride with sodium bicarbonate.15 After filtration and washes, the RCM product 37 was isolated in 96% yield with 359 ppm of residual ruthenium. Azapane trans-32 is isolated after deprotection of the phthalimide and hydrogenation with Pd/C. The authors do not report the specific ruthenium content in trans-32, but note that the downstream steps did lower the ruthenium content further. In a separate communication, a further optimized version of this route was reportedly used for multi-kilogram deliveries of SB-462795 for clinical studies, which will be discussed later in this review.


17


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 18 of 27

Scheme 7. RCM of 35 and Subsequent Stepsa

a

Conditions: (i) 6 (0.5 mol%), PhMe (11 vol), 110 °C, then crystallization and wash with aq.

THMP (96%) (ii) NH2OH, MeOH, 30 °C (73%); (iii) H2, Pd/C, EtOH/MeOH/H2O, 33–37 °C (98%). Though each of the above methods comprises simple Lewis base coordination of the Ru(II), chemical reduction of the ruthenium has been reported to give a water-soluble Ru(0) species. In the patent disclosure describing the preparation of rolapitant, a diazaspirocyclic NK-1 receptor antagonist, researchers at Schering-Plough employed an RCM of 38 followed by hydrogenation to form the piperidine ring present in the API (Scheme 8).29 A specific example describes the treatment of 38 with 1 mol % of catalyst 6 in toluene at 60-80 °C in the presence of TsOH to effect complete conversion to 39. The acidic conditions are likely necessary due to the presence of a secondary amine in the substrate and product.30 After completion, tetrabutylammonium chloride (TBACl) and aqueous sodium metabisulfite are added to the crude reaction and stirred for 1h. The inventors suggest that the metabisulfite acts to reduce the ruthenium to an unidentified, water soluble ML2 species, though this species is not characterized. The RCM product is isolated by extraction and then crystallization as the hydrochloride salt. The salt is


18

Page 19 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


then freebased and hydrogenated in the presence of Pd/C, isolated as the hydrochloride salt, and recrystallized to give 13 as the hydrochloride hydrate. Scheme 8. RCM and Subsequent Hydrogenation to Prepare Rolapitant

a

Conditions: (i) 6 (1 mol%), TsOH (1.5 equiv), PhMe (14 vol), 60-80 °C, then Bu4NCl aq.

Na2S2O5/NaHCO3, then 12N HCl and crystallization (85%); (ii) aq. NaOH, then Pd/C, Nuchar® Aquaguard®, PhMe, then HCl precipitation (95%) (iii) Recrystallization from EtOH/H2O (89%). It should be noted that oxidation of the ruthenium center can also facilitate extraction into aqueous solutions.31 Though effective on lab scale, these methods have yet to be reported in applications at multi-kilogram scale. Adsorption Methods Aided by Hydrogenation While charcoal treatments of mixtures containing the direct ruthenium byproducts of the olefin metathesis reaction have been employed successfully (vide supra), adsorption of ruthenium onto activated charcoal can be facilitated by hydrogenation. First and second generation catalysts such as 2 and 3 are converted to ruthenium hydride species under hydrogen pressures as low as 1


19


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 20 of 27

atm.32 These species seem to adsorb quite efficiently onto activated charcoal, and with sufficiently low metathesis catalyst loading, no additional treatment is necessary beyond the filtration to remove the heterogeneous Pd catalyst. This is particularly advantageous for synthetic sequences intended to deliver saturated rings.33 Ruthenium adsorption facilitated by hydrogenation is illustrated in the previously described multi-kilo scale synthesis of SB-462795. As discussed above, the optimal RCM substrate explored by scientists at GSK is the phthalimide-containing diene 35 (Scheme 9). By identifying and carefully controlling the impurities in 35 that negatively impacted the rate of the RCM, the process team at GSK achieved complete conversion in 2 hours using 0.25 mol% of 6 at 80 °C in only 5 volumes of toluene.34 The product 37 was precipitated directly from the crude reaction mixture in 90% yield with only 121 ppm of residual ruthenium. The RCM product was then subjected to hydrogenation with Pd/C, after which the crude mixture was filtered over Celite to remove the charcoal-supported catalyst. The saturated azapane product 40 is then isolated by crystallization. The authors note that after the remaining three chemical steps (deprotection, amide bond formation, and oxidation to the ketone) and two crystallizations, SB-462795 was isolated with

Scalable Methods for the Removal of Ruthenium ... - ACS Publications

Recommend Documents