Synlett
DOI: 10.1055/s-0043-1775430
letter
Small Molecules in Medicinal Chemistry

A Concise Synthesis of a Key Azabicyclo[2.1.1]hexane Building Block for N-Heteroaryl Indazole LRRK2 Kinase Inhibitors

a   Discovery Chemistry, Merck & Co., Inc., Boston, MA 02115, USA
,
Vladimir Simov
a   Discovery Chemistry, Merck & Co., Inc., Boston, MA 02115, USA
,
Ping Liu
a   Discovery Chemistry, Merck & Co., Inc., Boston, MA 02115, USA
,
Michael J. Ardolino
b   Process Research and Development, Merck & Co., Inc., Boston, MA 02115, USA
,
Theodore Martinot
b   Process Research and Development, Merck & Co., Inc., Boston, MA 02115, USA
,
Thomas Lyons
b   Process Research and Development, Merck & Co., Inc., Boston, MA 02115, USA
,
Ji Qi
c   Process Research and Development, Merck & Co., Inc., Rahway, NJ 07065, USA
,
Jingjun Yin
c   Process Research and Development, Merck & Co., Inc., Rahway, NJ 07065, USA
,
Jinglai Hao
d   WuXi AppTec (Shanghai), No. 1 Building, 288 Fute Zhong Road, WaiGaoQiao Free Trade Zone, Shanghai 200131, P. R. of China
,
Bin Hu
d   WuXi AppTec (Shanghai), No. 1 Building, 288 Fute Zhong Road, WaiGaoQiao Free Trade Zone, Shanghai 200131, P. R. of China
,
Meng Chen
d   WuXi AppTec (Shanghai), No. 1 Building, 288 Fute Zhong Road, WaiGaoQiao Free Trade Zone, Shanghai 200131, P. R. of China
,
Jun Zhang
d   WuXi AppTec (Shanghai), No. 1 Building, 288 Fute Zhong Road, WaiGaoQiao Free Trade Zone, Shanghai 200131, P. R. of China
,
Yu-hong Lam
e   Computational and Structural Chemistry, Merck & Co., Inc., Rahway, NJ 07065, USA
,
Peter H. Fuller
a   Discovery Chemistry, Merck & Co., Inc., Boston, MA 02115, USA
,
J. Michael Ellis
a   Discovery Chemistry, Merck & Co., Inc., Boston, MA 02115, USA
,
Matthew Maddess
b   Process Research and Development, Merck & Co., Inc., Boston, MA 02115, USA
› Author Affiliations
 


Abstract

Increasing the fraction of sp3 character in lead compounds has been shown to increase their drug-likeness by improving their potency, selectivity, and physicochemical properties. Azabicyclo[2.1.1]hexanes have recently garnered much interest in the synthetic community as pyrrolidine analogues for their interesting biological properties and stereoelectronic effects. During the course of our optimization of N-heteroaryl LRRK2 inhibitors, we discovered that this unique bicyclic system leads to improvements in solubility and metabolic clearance. Herein, we outline a match-pair analysis showcasing the broad impact of this unique azabicyclo[2.1.1]hexane system on key drug-like properties, as well as a concise and scalable synthesis of this building block, featuring an intramolecular cyclization to forge a strained amide bond.


#

During the course of a drug-discovery program, the material requirements of lead compounds for profiling and safety studies increase exponentially as the program progresses. This, in turn, places a greater demand on the practicality of the chemistry used to supply that material. We recently disclosed a number of our efforts toward the identification of low-dose, kinome-selective, CNS-penetrant LRRK2 kinase inhibitors for the treatment of Parkinson’s disease.[2] The dimethylmorpholine moiety in compounds of type 1 (Figure [1a]) was found to impart good potency and kinome selectivity; however, it suffered from significant oxidative metabolism, and generated reactive metabolites.[2a] [b] Through optimization efforts, we identified the lead compound 3a (Figure [1a]), where a key innovation was the unique bicyclic 2-(2-azabicyclo[2.1.1]hexan-4-yl)propan-2-ol solvent-front moiety, which led to a significant improvement in the in vivo rat clearance, mean residence time (MRT), and solubility relative to the baseline pyrrolidine 2a. Spirocycles and bridged bicycles have been used to great effect in medicinal chemistry to improve compound selectivity, potency, and pharmacokinetics, and these changes can be ascribed to an increased sp3 fraction.[3] [4] When we generated multiple match-pair analogues with the monocyclic pyrrolidine in 2a and the bicyclic pyrrolidine in 3a, we found that this was the case, with molecules containing the bicyclic pyrrolidine showing a consistently improved intrinsic clearance and fasted simulated intestinal fluid (FaSSIF) solubility relative to compounds containing the pyrrolidine group, as illustrated by the matched-molecular-pair analysis in Figure [1b].[5] Given these benefits, the azabicyclic pyrrolidine became a privileged motif for our LRRK2 inhibitors and an essential synthetic intermediate.

Zoom Image
Figure 1 (a) Impact of the 2-(2-azabicyclo[2.1.1]hexan-4-yl)propan-2-ol moiety. (b) Matched-pair analysis, showing a broad impact on intrinsic clearance and solubility.

Our initial route to access 2-(2-azabicyclo[2.1.1]hexan-4-yl)propan-2-ol 9 began from glycerol (Scheme [1]). The average yield per step was 73%, and the lowest-yielding step was the intramolecular cyclization to form the azabicyclo[2.1.1]hexane core. Although the starting materials were inexpensive and >100 g of intermediate 8 were generated by this route, the overall process was lengthy and inefficient, requiring 22 steps and giving a 0.6% overall yield. It became clear that this first-generation route would be insufficient to support the increasing needs of the program, as we hoped to advance 3a to in vivo studies.

The azabicyclo[2.1.1]hexane motif has garnered a lot of interest from the synthetic community, and a number of photochemical and intramolecular SN2 cyclization strategies have been developed.[6] We considered whether an amide cyclization of intermediate 11 to give 10 would be a viable disconnection (Scheme [2]). In this approach, a reaction would occur between the amine and the syn-ester group to yield a product with differentiated amide and ester functional groups. Initial concerns regarding this approach were that ring strain could make for inefficient cyclization or that the resulting product would be unstable. Despite these concerns, we were encouraged by several reports describing the generation of strained and bridged amides.[7]

Zoom Image
Scheme 1 First-generation route
Zoom Image
Scheme 2 Amide cyclization strategy

Our synthesis began with the commercially available ketone 12 (Scheme [3]). Oxime formation was facile and the crude product 13 was isolated in 87% yield as a white solid and used without further purification. A screen of reduction conditions revealed Raney nickel to be the best reagent for the reduction of oxime 13 to the cyclobutylamine 11, although substantial oxime hydrolysis was observed with catalytic amounts of the metal. This problem was solved by increasing the Raney nickel loading to three equivalents, providing 11 in an excellent 96% yield. This material could be taken forward after filtration and concentration. When exploring the key cyclization reaction toward intermediate 14, it was found that Lewis acids such as AlMe3, La(OTf)3, or Mg(OMe)2 failed to convert the starting material 11, and multiple unidentified byproducts were observed upon heating. Attempts to use the activated acyl chloride also failed to provide any reaction. After extensive optimization, we began to gain some traction by employing basic conditions. Whereas LiHMDS provided multiple byproducts, LDA gave the desired product 14, with product decomposition observed upon prolonged reaction times. We hypothesized that the decomposition was likely due to the basic amine reacting with the strained amide of the product.

Zoom Image
Scheme 3 Synthesis of 9 naproxen salt

We reasoned that the reaction could be improved by switching to an organometallic reagent with a chelating metal, such as an alkyl Grignard, which could potentially activate the nucleophile and electrophile through a preorganized transition state.[8] Furthermore, following deprotonation, no base would be present in the mixture to promote further reaction and product decomposition. This hypothesis was supported by a DFT exploration of the key reactive species I and II and transition state III (Figure [2]). Species I was used as a model of the complex formed from the Grignard reagent and 11 in an unbridged geometry. The dissociation of one ether molecule from magnesium (represented as Me2O) from I to form the bridged complex II was exergonic by 7.0 kcal/mol. The distance between the reactive N and C in I was well preserved in II. The four-membered ring moieties of I and II are also largely congruent as shown by an overlay of these two structures. The formation of a chelated complex was thus expected to be facile. The DFT calculations suggested that the addition step should also be facile, with a low activation barrier of 6.7 kcal/mol from II. In transition state III, the nucleophilic N atom forms an angle of 100° with the reactive C=O bond, which is within the range of attack angles proposed by Bürgi and Dunitz et al. for nucleophilic carbonyl additions.[9] Taken together, these computational modeling results supported our proposal that Mg provided a templating effect, pre-organizing the reactant into a transition-state-like geometry.

Zoom Image
Figure 2 DFT geometries of I, II, and transition state III leading to intramolecular Grignard addition promoted by Mg chelation (M06/6-31G**, SMD(THF)).

We were pleased to find that 15 could be formed directly by the addition of MeMgBr to 11. After optimization of the equivalents of MeMgBr and the order of addition, it was possible to obtain a 40% isolated yield of lactam 15; however, a side product resulting from the addition of the Grignard reagent to ester 11 without subsequent cyclization was observed. Unfortunately, further experimentation and purification could not eliminate this byproduct at a large scale. Ultimately, this problem was solved by the addition of the bulkier t-BuMgCl, which provided the intermediate lactam 14 in a moderate 64% yield after column chromatography. Lactam 14 is a versatile intermediate, as additional analogues can readily be prepared by manipulation of the ester functionality. Addition of MeMgBr to the ester cleanly afforded lactam 15 in 69% yield (Scheme [3]). Reduction of the lactam 15 was best accomplished using BH3·SMe2. Product 9 could be obtained in a yield of 67%; however, the amino diol resulting from ring opening of 15, was observed and proved challenging to remove. After screening acid salts and solvents, we found that the product could be cleanly isolated as its naproxen salt in 89% yield by precipitation from tert-amyl alcohol. Overall, this route provided the pure naproxen salt of 9 in five steps and 22% overall yield from the commercially available ketone 12. Importantly, the described route has been executed on a kilogram scale without the need for purification by column chromatography.

In conclusion, we have developed a robust and scalable synthesis of intermediate lactam 14, featuring a t-BuMgCl-promoted intramolecular cyclization. The use of the organomagnesium reagent was critical to the success of this transformation by avoiding product decomposition and by potentially pre-organizing the transition state to cyclization. Lactam 14 proved to be a useful intermediate en route to the privileged 2-(2-azabicyclo[2.1.1]hexan-4-yl)propan-2-ol solvent-front moiety 9,[10] which leads to consistent improvements in drug-like properties, such as solubility and metabolic clearance.


#

Conflict of Interest

The authors declare no conflict of interest.

Supporting Information

  • References and Notes

  • 1 Theodore Martinot is now affiliated to Parabalis Medicines (formerly FogPharma). Thomas Lyons is now affiliated to Takeda Pharmaceutical Company. J. Michael Ellis is now affiliated to Bristol Myers Squibb. Michael J. Ardolino is now affiliated with LifeMine Therapeutics.
    • 2a Candito DA, Simov V, Gulati A, Kattar S, Chau RW, Lapointe BT, Methot JL, DeMong DE, Graham TH, Kurukulasuriya R, Keylor MH, Tong L, Morriello GJ, Acton JJ, Pio B, Liu W, Scott JD, Ardolino MJ, Martinot TA, Maddess ML, Yan X, Gunaydin H, Palte RL, McMinn SE, Nogle L, Yu H, Minnihan EC, Lesburg CA, Liu P, Su J, Hegde LG, Moy LY, Woodhouse JD, Faltus R, Xiong T, Ciaccio P, Piesvaux JA, Otte KM, Kennedy ME, Bennett DJ, DiMauro EF, Fell MJ, Neelamkavil S, Wood HB, Fuller PH, Ellis JM. J. Med. Chem. 2022; 65: 16801
    • 2b Scott JD, DeMong DE, Greshock TJ, Basu K, Dai X, Harris J, Hruza A, Li SW, Lin SI, Liu H, Macala MK, Hu Z, Mei H, Zhang H, Walsh P, Poirier M, Shi ZC, Xiao L, Agnihotri G, Baptista MA, Columbus J, Fell MJ, Hyde LA, Kuvelkar R, Lin Y, Mirescu C, Morrow JA, Yin Z, Zhang X, Zhou X, Chang RK, Embrey MW, Sanders JM, Tiscia HE, Drolet RE, Kern JT, Sur SM, Renger JJ, Bilodeau MT, Kennedy ME, Parker EM, Stamford AW, Nargund R, McCauley JA, Miller MW.. J. Med. Chem. 2017; 60:  2983
    • 2c Gulati A, Yeung CS, Lapointe B, Kattar SD, Gunaydin H, Scott JD, Childers KK, Methot JL, Simov V, Kurukulasuriya R, Pio B, Morriello GJ, Liu P, Tang H, Neelamkavil S, Wood HB, Rada VL, Ardolino MJ, Yan XC, Palte R, Otte K, Faltus R, Woodhouse J, Hegde LG, Ciaccio P, Minnihan EC, DiMauro EF, Fell MJ, Fuller PH, Ellis JM. RSC Med. Chem. 2021; 12:  1164
    • 2d Keylor MH, Gulati A, Kattar SD, Johnson RE, Chau RW, Margrey KA, Ardolino MJ, Zarate C, Poremba KE, Simov V, Morriello GJ, Acton JJ, Pio B, Yan X, Palte RL, McMinn SE, Nogle L, Lesburg CA, Adpressa D, Lin S, Neelamkavil S, Liu P, Su J, Hegde LG, Woodhouse JD, Faltus R, Xiong T, Ciaccio PJ, Piesvaux J, Otte KM, Wood HB, Kennedy ME, Bennett DJ, DiMauro EF, Fell MJ, Fuller PH. J. Med. Chem. 2022; 65:  838
    • 2e Logan KM. Kaplan W, Simov V, Zhou H, Li D, Torres L, Morriello GJ, Acton JJ, Pio B, Chen Y.-H, Keylor MH, Johnson R, Kattar SD, Chau R, Yan X, Ardolino M, Zarate C, Otte KM, Palte RL, Xiong T, McMinn SE, Lin S, Neelamkavil SF, Liu P, Su J, Hegde LG, Woodhouse JD, Moy LY, Ciaccio PJ, Piesvaux J, Zebisch M, Henry C, Barker J, Wood HB, Kennedy ME, DiMauro EF, Fell MJ, Fuller PH. J. Med. Chem. 2024; 67: 16807
  • 4 Lovering F, Bikker J, Humblet C. J. Med. Chem. 2009; 52: 6752
  • 7 Szostak M, Aubé J. Chem. Rev. 2013; 113: 5701
  • 8 Bruckner R. Organic Mechanisms: Reactions, Stereochemistry and Synthesis. Harmata M. Springer; Berlin: 2010
  • 9 Bürgi HB, Dunitz JD, Lehn JM, Wipff G. Tetrahedron 1974; 30: 1563
  • 10 2-(2-Azabicyclo[2.1.1]hex-4-yl)propan-2-ol (9) A 500 mL three-necked round-bottomed flask was charged with 15 (4.75 g, 0.031 mol, 1 equiv) and THF (150 mL). The solution was cooled to 0 °C (ice–salt bath), and BH3·Me2S (16.1 mL, 0.161 mol, 5.3 equiv) was added dropwise over 0.5 h while the temperature was maintained between 0 and 5 °C. When the addition was complete, the mixture was warmed to 25 °C and stirred for 1 h then heated to 50 °C for 15 h. The mixture was cooled to 0 °C, and MeOH (50 mL) was added slowly over 2 h. The resulting mixture was heated to 50 °C for 2 h and then concentrated to give the crude intermediate. A 1 M solution of HCl in MeOH (200 mL) was added dropwise at 20 °C over 0.5 h, and the resulting mixture was stirred at 50 °C for 12 h. The mixture was then concentrated to provide a crude residue. 50 wt% aq K2CO3 (200 mL) was added, and the resultant mixture was extracted with 2-methyltetrahydrofuran (4 × 250 mL). The organic layer was dried (Na2SO4), filtered, and concentrated to give a colorless oil; yield: 3 g (69%). 1H NMR (400 MHz, MeOD): δ = 3.73–3.72 (m, 1 H), 3.53 (s, 1 H), 2.94 (s, 2 H), 1.84–1.83 (m, 2 H), 1.38–1.36 (m, 2 H), 1.21 (s, 6 H). 13C NMR (101 MHz, CD3CN): δ = 117.9, 70.09, 69.6, 61.0, 55.8, 47.7, 39.8, 26.0. LCMS: m/z [M + H]+ calcd: 142.1; found; 142.4.

Corresponding Author

David A. Candito
Discovery Chemistry, Merck & Co., Inc.
Boston, MA 02115
USA   

Publication History

Received: 08 November 2024

Accepted after revision: 09 December 2024

Article published online:
13 January 2025

© 2025. Thieme. All rights reserved

Georg Thieme Verlag KG
Oswald-Hesse-Straße 50, 70469 Stuttgart, Germany

  • References and Notes

  • 1 Theodore Martinot is now affiliated to Parabalis Medicines (formerly FogPharma). Thomas Lyons is now affiliated to Takeda Pharmaceutical Company. J. Michael Ellis is now affiliated to Bristol Myers Squibb. Michael J. Ardolino is now affiliated with LifeMine Therapeutics.
    • 2a Candito DA, Simov V, Gulati A, Kattar S, Chau RW, Lapointe BT, Methot JL, DeMong DE, Graham TH, Kurukulasuriya R, Keylor MH, Tong L, Morriello GJ, Acton JJ, Pio B, Liu W, Scott JD, Ardolino MJ, Martinot TA, Maddess ML, Yan X, Gunaydin H, Palte RL, McMinn SE, Nogle L, Yu H, Minnihan EC, Lesburg CA, Liu P, Su J, Hegde LG, Moy LY, Woodhouse JD, Faltus R, Xiong T, Ciaccio P, Piesvaux JA, Otte KM, Kennedy ME, Bennett DJ, DiMauro EF, Fell MJ, Neelamkavil S, Wood HB, Fuller PH, Ellis JM. J. Med. Chem. 2022; 65: 16801
    • 2b Scott JD, DeMong DE, Greshock TJ, Basu K, Dai X, Harris J, Hruza A, Li SW, Lin SI, Liu H, Macala MK, Hu Z, Mei H, Zhang H, Walsh P, Poirier M, Shi ZC, Xiao L, Agnihotri G, Baptista MA, Columbus J, Fell MJ, Hyde LA, Kuvelkar R, Lin Y, Mirescu C, Morrow JA, Yin Z, Zhang X, Zhou X, Chang RK, Embrey MW, Sanders JM, Tiscia HE, Drolet RE, Kern JT, Sur SM, Renger JJ, Bilodeau MT, Kennedy ME, Parker EM, Stamford AW, Nargund R, McCauley JA, Miller MW.. J. Med. Chem. 2017; 60:  2983
    • 2c Gulati A, Yeung CS, Lapointe B, Kattar SD, Gunaydin H, Scott JD, Childers KK, Methot JL, Simov V, Kurukulasuriya R, Pio B, Morriello GJ, Liu P, Tang H, Neelamkavil S, Wood HB, Rada VL, Ardolino MJ, Yan XC, Palte R, Otte K, Faltus R, Woodhouse J, Hegde LG, Ciaccio P, Minnihan EC, DiMauro EF, Fell MJ, Fuller PH, Ellis JM. RSC Med. Chem. 2021; 12:  1164
    • 2d Keylor MH, Gulati A, Kattar SD, Johnson RE, Chau RW, Margrey KA, Ardolino MJ, Zarate C, Poremba KE, Simov V, Morriello GJ, Acton JJ, Pio B, Yan X, Palte RL, McMinn SE, Nogle L, Lesburg CA, Adpressa D, Lin S, Neelamkavil S, Liu P, Su J, Hegde LG, Woodhouse JD, Faltus R, Xiong T, Ciaccio PJ, Piesvaux J, Otte KM, Wood HB, Kennedy ME, Bennett DJ, DiMauro EF, Fell MJ, Fuller PH. J. Med. Chem. 2022; 65:  838
    • 2e Logan KM. Kaplan W, Simov V, Zhou H, Li D, Torres L, Morriello GJ, Acton JJ, Pio B, Chen Y.-H, Keylor MH, Johnson R, Kattar SD, Chau R, Yan X, Ardolino M, Zarate C, Otte KM, Palte RL, Xiong T, McMinn SE, Lin S, Neelamkavil SF, Liu P, Su J, Hegde LG, Woodhouse JD, Moy LY, Ciaccio PJ, Piesvaux J, Zebisch M, Henry C, Barker J, Wood HB, Kennedy ME, DiMauro EF, Fell MJ, Fuller PH. J. Med. Chem. 2024; 67: 16807
  • 4 Lovering F, Bikker J, Humblet C. J. Med. Chem. 2009; 52: 6752
  • 7 Szostak M, Aubé J. Chem. Rev. 2013; 113: 5701
  • 8 Bruckner R. Organic Mechanisms: Reactions, Stereochemistry and Synthesis. Harmata M. Springer; Berlin: 2010
  • 9 Bürgi HB, Dunitz JD, Lehn JM, Wipff G. Tetrahedron 1974; 30: 1563
  • 10 2-(2-Azabicyclo[2.1.1]hex-4-yl)propan-2-ol (9) A 500 mL three-necked round-bottomed flask was charged with 15 (4.75 g, 0.031 mol, 1 equiv) and THF (150 mL). The solution was cooled to 0 °C (ice–salt bath), and BH3·Me2S (16.1 mL, 0.161 mol, 5.3 equiv) was added dropwise over 0.5 h while the temperature was maintained between 0 and 5 °C. When the addition was complete, the mixture was warmed to 25 °C and stirred for 1 h then heated to 50 °C for 15 h. The mixture was cooled to 0 °C, and MeOH (50 mL) was added slowly over 2 h. The resulting mixture was heated to 50 °C for 2 h and then concentrated to give the crude intermediate. A 1 M solution of HCl in MeOH (200 mL) was added dropwise at 20 °C over 0.5 h, and the resulting mixture was stirred at 50 °C for 12 h. The mixture was then concentrated to provide a crude residue. 50 wt% aq K2CO3 (200 mL) was added, and the resultant mixture was extracted with 2-methyltetrahydrofuran (4 × 250 mL). The organic layer was dried (Na2SO4), filtered, and concentrated to give a colorless oil; yield: 3 g (69%). 1H NMR (400 MHz, MeOD): δ = 3.73–3.72 (m, 1 H), 3.53 (s, 1 H), 2.94 (s, 2 H), 1.84–1.83 (m, 2 H), 1.38–1.36 (m, 2 H), 1.21 (s, 6 H). 13C NMR (101 MHz, CD3CN): δ = 117.9, 70.09, 69.6, 61.0, 55.8, 47.7, 39.8, 26.0. LCMS: m/z [M + H]+ calcd: 142.1; found; 142.4.

Zoom Image
Figure 1 (a) Impact of the 2-(2-azabicyclo[2.1.1]hexan-4-yl)propan-2-ol moiety. (b) Matched-pair analysis, showing a broad impact on intrinsic clearance and solubility.
Zoom Image
Scheme 1 First-generation route
Zoom Image
Scheme 2 Amide cyclization strategy
Zoom Image
Scheme 3 Synthesis of 9 naproxen salt
Zoom Image
Figure 2 DFT geometries of I, II, and transition state III leading to intramolecular Grignard addition promoted by Mg chelation (M06/6-31G**, SMD(THF)).