Simple Enantioselective Syntheses of (2R,6R)-Hydroxynorketamine

Sep 11, 2017 - A novel strategy for accessing cyclic α-amino ketones enantioselectively has opened a simple synthetic route to the antidepressant (2R...
2 downloads 12 Views 437KB Size
Letter pubs.acs.org/OrgLett

Simple Enantioselective Syntheses of (2R,6R)‑Hydroxynorketamine and Related Potential Rapid-Onset Antidepressants Yixin Han,‡ Karla Mahender Reddy,‡ and E. J. Corey* Department of Chemistry and Chemical Biology, Harvard University, Cambridge, Massachusetts 02138, United States S Supporting Information *

ABSTRACT: A novel strategy for accessing cyclic α-amino ketones enantioselectively has opened a simple synthetic route to the antidepressant (2R,6R)-hydroxynorketamine and numerous analogues. Mechanistically guided catalyst selection was essential in an initial olefin epoxidation step. In a second crucial step, the epoxide was subjected to a novel O → N displacement that occurred with retention of configuration through the use of Al- or Ti-based azides, which promote epoxide activation and internal cis delivery of N3 to carbon.

T

his paper describes a useful enantioselective synthesis of a metabolite of the useful drug ketamine (1) which may provide a key to finding a valuable new antidepressant. The biomedical background of this research topic is unusually interesting, and so we open with a concise introduction. Mental illness causes a profound reduction in the quality of life worldwide as the source of much human suffering, conflict, crime, and social disruption (affecting about one person in six at some point in life). Chemical synthesis has contributed substantially to the development of useful treatments of depressive disorders by means of serotonin reuptake inhibitors (SRIs), e.g., fluoxetine and sertraline during the 1980−2000 period. Unfortunately, the benefits of SRIs appear only after several weeks of treatment, and many types of depression are totally resistant to this therapy.1 Further advances have been minimal because of the endless complexity and poor understanding of brain science, the lack of animal models, and the reluctance of pharmaceutical discovery companies to invest in this high-risk area. The research described herein developed as a result of recent findings with the long-known drug ketamine (1)2 which suggest the possibility of a new therapeutic approach to the treatment of drug-resistant depression (Figure 1). Injected ketamine (as the racemate) has been used as a shortacting (30−60 min) anesthetic since the 1960s. Its use is particularly advantageous in emergency situations, since it does not depress respiration or require intubation, and for shortduration procedures. However, ketamine causes many other

effects, e.g., euphoria, hallucination, muscle spasms, and agitation, and has become a drug of abuse. Commercial (S)-1 (prepared by resolution) was also marketed by Johnson and Johnson as a nasal spray in 2013 for anesthetic use. In 20063 a surprising report appeared that a single injected subanesthetic dose of (±)-ketamine can act as an antidepressant, even in treatment-refractory unipolar or bipolar depression. Remarkably, the antidepressant effect was rapidonset, often in a few hours, and then persisted for up to one week. It has been suggested that the effect of ketamine may be a result of it is ability to block the NMDA (N-methyl-D-aspartic acid) receptor, an important glutamate-activated postsynaptic ion (initially Na, then Ca) channel.4 These results were followed by the even more astonishing finding in 2016 that just one of numerous metabolites of ketamine, (2R,6R)-hydroxynorketamine (2), is responsible for the antidepressant activity that was originally ascribed to ketamine.5 The metabolite 2 is reported to be more potent than ketamine and also devoid of its side effects. It is possible that the metabolite 2 increases synaptic signaling in mice by binding to AMPA (α-amino-3hydroxy-5-methyl-4-isoxazole propionic acid) receptors, which are also glutamate-activated ion channels.5 NMDA and AMPA receptors act in partnership and are critical to synaptic plasticity, cognition, and memory. Deficits in synaptic signaling are associated with various psychological and neurological conditions, including depressive disorders.6−8 The research that led to ketamine grew out of results that had been obtained with the earlier drug phencyclidine (3), which was removed from the US market in 1965 because of very dangerous side effects. It seems a remarkable coincidence that ketamine (an anesthetic which works by blocking synaptic transmission via NMDAR antagonism) is converted by enzymic demethylation

Figure 1. Bioactive modulators of glutamate signaling in brain.

Received: August 11, 2017 Published: September 11, 2017

© 2017 American Chemical Society

5224

DOI: 10.1021/acs.orglett.7b02498 Org. Lett. 2017, 19, 5224−5227

Letter

Organic Letters and hydroxylation to an AMPA-activating compound that increases synaptic transmission. It has also been proposed by a different group that the antidepressant activity is a result of its effect on eEF2, increased protein translation and stimulation of synthesis of brain-derived neurotrophic factor (BDNF), downstream of NMDA inhibition.9,10 These significant developments on ketamine-related fastacting antidepressants prompted us to apply synthetic chemistry to the development of a new generation of therapeutic molecules which might lead to better understanding in an area of great medical need. Access to synthetic molecules related to 2 could allow probing to determine its biological target. Also, it seems possible that (2R,6R)-hydroxynorketamine (2, (R,R)-HNK) may not emerge as the ideal medicine and that a range of other molecules will have to be synthesized and tested in order to find the most effective drug and pathways of action. In this paper, we outline a simple, practical enantioselective route to 2 (or ent-2) which can allow access to many other structures that could potentially be useful in this field. We hope that this work will facilitate additional research. Previous literature has described other methods for the synthesis of ketamine and also the conversion of ketamine to 2.5,11,12 The pathway that was developed for the enantioselective synthesis of (2R,6R)-HNK (2), which is outlined in Scheme 1,

analysis using the mechanistic model of the Jacobsen epoxidation that we advanced several years ago14 which suggested the use of the less bulky catalyst B (5 mol %) instead of A. In the mechanistic model,14 the oxirane ring is formed asynchronously with a styrene-type substrate such as 4, and the benzylic carbon develops radical-cation character. The π-face of the double bond aligns with the Mn-oxo group and one of the phenoxy oxygens nearby, so that the benzylic positive charge is stabilized by interaction with a phenoxide lone pair, as shown in Figure 2. Fortunately, both the rate of

Scheme 1. Enantioselective Synthesis of (2R,6R)Hydroxynorketamine Hydrochloride

epoxidation and the yield of the required epoxide 5 (60%, 93:7 enantioselection) were considerably improved with catalyst B. Reaction of the epoxide 5 with a reagent made from Ti(O-iPr)4 and Me3SiN315 in toluene at 70 °C afforded a cishydroxybenzylic azide by displacement of the benzylic oxygen by N3 with retention, as shown in detail in Scheme 2. The

Figure 2. Pre-transition-state assembly for the enantioselective conversion of 4 to 5.

Scheme 2. Ring Opening of Epoxide 5 and Internal Azide Transfer in an Intermediate Carbocation To Form a cisAzido Alcohol with Overall Retention of Configuration at the Benzylic Center

resulting azido alcohol was directly oxidized to the corresponding ketone 6 (66% from 5) using Dess−Martin periodinane. Upon storage of the oily product for several hours, crystals of enantiomerically pure azide separated (>99% ee by HPLC analysis using a Chiralcel OD-H column). Reduction of the azido ketone to the corresponding amino ketone and tertbutoxy carbonyl protection provided (R)-N-Boc-norketamine (7) in high yield.16 The transformation of 7 to 2 was effected efficiently as shown in Scheme 1. The synthetic (2R,6R)hydroxynorketamine so produced was identical with the natural metabolite spectroscopically and by optical rotation (see the SI). Radiolabeled (tritium) 2 has also been generated from synthetic 2 for biological studies, which are ongoing. The transformation of the chiral epoxide 5 to the corresponding β-azido alcohol also presented a challenge because of the tendency of 5 to undergo ring contraction to 1-o-chlorophenylcyclopentane carboxaldehyde. The only satisfactory reagents that we found for the conversion of 5 to 6 were Ti(O-i-Pr)2(N3)2 and i-Bu2AlN3 (generated from iBu2AlCl and NaN3). It is important to note that the

was based on the enantioselective epoxidation of 1-ochlorophenylcyclohexene 4 using a Jacobsen’s salen-type catalyst and NaOCl as oxidant.13 When the reaction was attempted with the standard Jacobsen catalyst (A in Scheme 1) the reaction results were disappointing not only because the rate was very slow but also because the major product was 2-ochlorophenyl-2-cyclohexenone, with the epoxide 5 being a minor product (30% max. yield). It seemed likely that the steric bulk of the o-chlorophenyl group and its nonplanarity with C C are detrimental to the formation of 5. This led us to an 5225

DOI: 10.1021/acs.orglett.7b02498 Org. Lett. 2017, 19, 5224−5227

Letter

Organic Letters replacement of the epoxide oxygen by N3 is unusual since it occurred specifically at the benzylic position of 5 and with selective retention of configuration. This observation points to the reaction pathway shown in Scheme 2, and the possibility that stereochemistry is determined by a cis-transfer of azide from Ti to the benzylic carbon. We are currently investigating the scope of this novel process. The enantioselective synthesis of (2S,6S)-hydroxynorketamine (ent-2) was also successfully accomplished using the method outlined in Scheme 1. Further, we applied a parallel sequence of reactions to the synthesis of the o-fluoro analogue of (2S,6S)-hydroxynorketamine (13), as shown in Scheme 3.

Figure 3. 2,5-Difluorophenyl and phenyl analogues of 2 and (R,R)HNK.

Scheme 4. Derivatives of 2 or 7 for Use as Biological Probes

Scheme 3. Enantioselective Synthesis of o-Fluoro Analogue of (2S,6S)-Hydroxynorketamine Hydrochloride

resolution brings new tools and possibilities for research on these potential future medicines. The methodology described above for the synthesis of the enantiomers of 6-hydroxynorketamine and various aryl-differentiated analogues can be applied more generally for the enantioselective synthesis of a range of α-amino cyclic ketones via cis-1,2-amino alcohols. The use of mechanistically guided catalyst selection for enantioselective epoxidation and internally directed diastereoselective delivery of azide from Ti or Al to carbon are key features of the synthetic approach. The serendipitous discovery of the antidepressant properties of ketamine together with experimental indications that this activity is probably due to the human metabolite 2 open new avenues for research in an area of great medical need. Despite the fact that 2 has probably been in the bodies of many thousands of humans, current regulations for clinical development prevent its study in humans until extensive preclinical investigations have been performed, since 2 is considered legally a new chemical entityan especially clear example of the tension between efficient drug discovery and ensuring drug safety. Although it may be some years before the utility of 2 (or a synthetic analogue) will be determined, in the meantime medical research in this area combining molecular synthesis with chemical biology can advance the neuroscience of depression, e.g., in exploring signaling pathways or BDNF elevation.

The epoxidation reaction of the o-fluoro olefin 8 was faster than that with the o-chloro olefin 4 and proceeded well even with the standard Jacobsen catalyst, R,R-catalyst A, to give epoxide 5 with 32:1 enantioselectivity. Treatment of the epoxide 9 with iBu2AlN3 (from i-Bu2AlCl and NaN3) produced the corresponding cis-azido alcohol, again via a benzylic cation and cis-internal delivery of N3 from Al to C, in a process paralleling that outlined in Scheme 2. One recrystallization of the penultimate intermediate 12 afforded that substance in 99.5:0.5% enantiomeric purity. The approach shown in Scheme 3 also provides access to the (2R,6R)-fluoro analogue of 2, which is of interest not only in terms of antidepressive activity but also as a tool compound for applying magnetic resonance imaging to determining in vivo the distribution of ent-13 or 13 in the brain of an experimental animal, e.g., mouse. In addition, we have applied the methodology outlined above to the synthesis of the 2,5-difluorophenyl (14) and phenyl (15) analogues of 2 (Figure 3), as detailed in the Supporting Information. Many interesting new molecules that could be useful in ongoing research to find novel antidepressants are readily available synthetically from 2 or 7 as exemplified in Scheme 4. For instance, aminal 18 and imine 19 may be more bioavailable as prodrugs for intranasal administration. The facile synthesis of (2R,6R)-hydroxynorketamine, numerous analogues, and various derivatives without the need for 5226

DOI: 10.1021/acs.orglett.7b02498 Org. Lett. 2017, 19, 5224−5227

Letter

Organic Letters



(13) For reviews on the Jacobsen epoxidation, see: (a) Katsuki, T. Synlett 2003, 281−297. (b) Larrow, J. F.; Jacobsen, E. N. Top. Organomet. Chem. 2004, 6 (Organometallics in Process Chemistry), 123−152. (c) Muniz-Fernandez, K.; Bolm, C. Manganese-catalyzed epoxidations. In Transition Metals for Organic Synthesis, 2nd ed.; Wiley, 2004; Vol. 2, pp 344−356. (d) McGarrigle, E. M.; Gilheany, D. G. Chem. Rev. 2005, 105, 1563−1602. (e) Senanayake, C. H.; Smith, G. B.; Ryan, K. M.; Fredenburgh, L. E.; Liu, J.; Roberts, F. E.; Hughes, D. L.; Larsen, R. D.; Verhoeven, T. R.; Reider, P. J. Tetrahedron Lett. 1996, 37, 3271−3274. (14) Kürti, L.; Blewett, M.; Corey, E. J. Org. Lett. 2009, 11, 4592− 4595. (15) Choukroun, R.; Gervais, D. J. Chem. Soc., Dalton Trans. 1980, 1800. (16) The absolute configurations of 5−7 follow from analogous Jacobsen oxidations and the mechanistic model.14

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.orglett.7b02498. Experimental procedures and characterization data for novel reactions and products including 1H and 13C NMR spectra and chiral HPLC traces (PDF)



AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Yixin Han: 0000-0002-4941-092X Karla Mahender Reddy: 0000-0002-9601-2967 Author Contributions ‡

Y.H. and K.M.R. contributed equally.

Notes

The authors declare no competing financial interest.

■ ■

ACKNOWLEDGMENTS We are grateful for financial assistance from Pfizer, Inc. and Peking University (Prof. Zhen Yang and Jiahua Chen). REFERENCES

(1) Rush, A. J.; Trivedi, M. H.; Wisniewski, S. R.; Nierenberg, A. A.; Stewart, J. W.; Warden, D.; Niederehe, G.; Thase, M. E.; Lavori, P. W.; Lebowitz, B. D.; McGrath, P. J.; Rosenbaum, J. F.; Sackeim, H. A.; Kupfer, D. J.; Luther, J.; Fava, M. Am. J. Psychiatry 2006, 163, 1905− 1917. (2) Elia, N.; Tramer, M. R. Pain 2005, 113, 61−70. (3) Zarate, C. A., Jr.; Singh, J. B.; Carlson, P. J.; Brutsche, N. E.; Ameli, R.; Luckenbaugh, D. A.; Charney, D. S.; Manji, H. K. Arch. Gen. Psychiatry 2006, 63, 856−864. (4) Duman, R. S.; Aghajanian, G. K.; Sanacora, G.; Krystal, J. Nat. Med. 2016, 22, 238−249. (5) (a) Zanos, P.; Moaddel, R.; Morris, P. J.; Georgiou, P.; Fischell, J.; Elmer, G. I.; Alkondon, M.; Yuan, P.; Pribut, H. J.; Singh, N. S.; Dossou, K. S. S.; Fang, Y.; Huang, X.-P.; Mayo, C. L.; Wainer, I. W.; Albuquerque, E. X.; Thompson, S. M.; Thomas, C. J.; Zarate, C. A., Jr.; Gould, T. D. Nature 2016, 533, 481−486. (b) The absolute configuration and structure of 2 were established by X-ray diffraction analysis of the hydrochloride salt.. (6) (a) Fleming, J. J.; England, P. M. Nat. Chem. Biol. 2010, 6, 89−97. (b) Clem, R.; Huganir, R. L. Science 2010, 330, 1108−1112. (7) Zhang, Y.; Cudmore, R.; Lin, D.-T.; Linden, D.; Huganir, R. L. Nat. Neurosci. 2015, 18, 402−407. (8) For background information on AMPA and NMDA receptors see (a) Neuroscience - Long-Term Potentiation. https://www.youtube. com/watch?v=vso9jgfpI_c (accessed Nov 6, 2013). Long Term Potentiation/Depression. https://www.youtube.com/watch?v=OBif222B6Q (accessed May 13, 2014). (b) Tyler, M. W.; Yourish, H. B.; Ionescu, D. F.; Haggarty, S. J. ACS Chem. Neurosci. 2017, 8, 1122− 1134. (9) Monteggia, L. M.; Gideons, E.; Kavalali, E. T. Biol. Psychiatry 2013, 73, 1199−1203. (10) Suzuki, K.; Nosyreva, E.; Hunt, K. W.; Kavalali, E. T.; Monteggia, L. M. Nature 2017, 546, E1−E3. (11) (a) Stevens, C. L.; Elliot, R. D. J. Am. Chem. Soc. 1963, 85, 1464−1470. (b) Stevens, C. L.; Thuillier, A.; Daniher, F. A. J. Org. Chem. 1965, 30, 2962−2966. (c) Stevens, C. L. U.S. Patent 3,254,124, 1966. (12) Yang, X.; Toste, F. D. J. Am. Chem. Soc. 2015, 137, 3205−3208. 5227

DOI: 10.1021/acs.orglett.7b02498 Org. Lett. 2017, 19, 5224−5227