Abstract
We report a model study towards the enantioselective synthesis of the dibenzopyrrocoline alkaloid (−)-cryptowolinol. The key step involves a challenging enantioselective Pd0-catalyzed C(sp3)–H arylation performed with a chiral NHC ligand, which proceeds via parallel kinetic resolution (PKR). A very efficient PKR process was achieved on a deoxygenated model substrate and was successfully transposed to a potential intermediate en route to (−)-cryptowolinol.
Among the abundant isoquinoline alkaloids,1 very few, such as cryptowoline (1) and cryptaustoline, possess the dibenzopyrrocoline framework (Scheme 1a).2,3 (−)-Cryptowolinol 2, isolated together with 1 from the New Caledonian lauraceae Cryptocarya phyllostemon,4 displays an additional hydroxy group at C12, which makes its stereoselective synthesis challenging. In addition, doubts can be raised regarding the absolute configuration of both (−)-1 and (−)-2, as the absolute configuration of (−)-cryptausoline, which differs from 1 only by its four free phenols, was reassigned to its enantiomer via total synthesis by Meyers and co-workers.5 Motivated by these problems and our interest in this field, we embarked on the enantioselective synthesis of cryptowolinol using Pd-catalyzed C–H activation as the key step.6 Our retrosynthetic analysis is depicted in Scheme 1a. We hypothesized that 2 could arise from protected arylindoline 3 (PG = protecting group) via Friedel–Crafts-type construction of the six-membered ring and methylation of the tertiary amine. In turn, enantioenriched 3 could arise from racemic aryl bromide 4 via enantioselective Pd0-catalyzed C(sp3)–H arylation occurring via parallel kinetic resolution (PKR).7,8 Indeed, Kündig and co-workers reported that racemic aryl bromides 7 containing two different alkyl groups undergo efficient PKR in the presence of a Pd/chiral NHC catalyst to generate an equimolar mixture of enantioenriched regioisomers 8 and 9.9 We wondered if compound 10, which models intermediate 4 and possesses a phenyl substituent instead of one of the alkyl groups in 7, could undergo such a PKR process to furnish scalemic indoline 11 and dibenzophenanthridine 12. An efficient PKR requires that the two substituents react at comparable rates with the catalyst. In the case of 10, this is very challenging as C(sp2)–H bonds are usually much more reactive than secondary C(sp3)–H bonds,10,11 and therefore, 10 might rather undergo exclusive C(sp2)–H arylation to give racemic 12.12 This work reports our model studies of the PKR of aryl bromides 10 and its application to the more substituted intermediate 4 en route to (−)-cryptowolinol.
Scheme 1. Retrosynthetic Analysis of (−)-Cryptowolinol.
First, we investigated the effect of protecting groups on the primary alcohol (Y) and aniline (Z) on the PKR of racemic model substrates 10 (Table 1). Substrates bearing various silyl groups on the primary alcohol were first considered (entries 1–4). We employed conditions that were recently found to be optimal for the enantioselective C(sp3)–H arylation of secondary C–H bonds on linear alkyl chains,13 using IBioxtBu14 as the chiral ligand and [Pd(π-allyl)Cl]2 as the Pd source. Unfortunately, this resulted in the formation of C(sp2)–H arylation products 12a–d in low to moderate yields and enantioselectivities, along with variable amounts of indole 13, which was speculated to arise from elimination of the desired product. Given that 12 possesses a benzylic stereocenter, we probed its propensity to racemize by heating it under the reaction conditions but did not observe any significant racemization. Substrate 10e bearing an ethyl instead of methyl carbamate was tested but did not furnish better results (entry 5). A pivaloyl protecting group was also introduced on the primary alcohol but did not yield the desired indoline product 11f either (entry 6). In contrast, 11 was formed in low yield alongside 12 when ether protecting groups tBu, Bn, PMB, and MOM were employed (entries 7–10, respectively). In particular, the Bn, PMB, and MOM groups provided indolines 11h–j, respectively, in excellent enantioselectivities (entries 8–10, respectively), which indicates that C(sp3)–H arylation is favored by a lower steric hindrance on the protected primary alcohol. It is noteworthy that the achiral IBioxMe4 ligand15 provided a higher proportion of dibenzophenanthridine product 12h, which further illustrates the substrate innate preference for C(sp2)–H arylation with this family of ligands (entry 11; compare it with entry 8). Recrystallization of co-product 12j afforded a highly enantiomerically enriched material (er 98.5:1.5), which could be analyzed by X-ray crystallography. This allowed determination of the absolute configuration of 12j (as R) and by extension of indoline 11j (S at the same carbon atom) and other products in this study. Surprisingly, using the aryl iodide (10k) instead of the bromide (10h) led to the exclusive formation of racemic C(sp2)–H arylation product 12k (entry 12), which might reveal a change in the reaction mechanism. When COCF3 was used as an N-protecting group, multiple compounds were observed, with an only 10% yield of product 12l observed by 1H NMR (entry 13), hence giving limited options to protect the nitrogen atom in the synthesis of cryptowolinol. As shown in entry 8, the C(sp2)–H arylation product is favored over the C(sp3)–H arylation product with IBioxtBu as the ligand, which leads to a low er for 12 and a high er for 11. This corresponds to a high selectivity factor (s) for 11 (275) and a low s for 12 (6.7). To achieve a more efficient PKR, a slower C(sp2)–H arylation/faster C(sp3)–H arylation must be achieved so that both reactions occur at comparable rates. Tuning the substituents of the IBiox ligand did not lead to any improvement (entry 14). In contrast, chiral NHC ligands L1 and L2 employed by Kündig and co-workers for substrates 7 (see Scheme 1)9,16 provided a much more efficient PKR process (entries 15 and 16, respectively). In particular, (R,R)-L2 containing o-tolyl substituents provided a more equal rate of C(sp3)–H and C(sp2)–H arylation products 11h and 12h, thus translating into better er and s values for both products (entry 16). As these two products can be separated by standard column chromatography, the desired indoline 11h was isolated in close-to-ideal yield (47%) and enantioselectivity (>99% ee). It is noteworthy that indoline 11h was isolated as a single trans diastereoisomer in line with previous studies,9 which presumably results from minimization of repulsion between substituents at the C(sp3)–H activation transition state.
Table 1. Assessment of Various Substrates in the Parallel Kinetic Resolutiona.
| entry | substrate | X | Y | Z | ligand | yield of 11 (%)b | er of 11c | s of 11d | yield of 12 (%)b | er of 12c | s of 12d |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | 10a | Br | TBS | CO2Me | (S,S)-IBioxtBu | – | 22 | 78:22 | |||
| 2 | 10b | Br | TES | CO2Me | (S,S)-IBioxtBu | – | 14 | 68:32 | |||
| 3 | 10c | Br | TBDPS | CO2Me | (S,S)-IBioxtBu | – | 0 | ||||
| 4 | 10d | Br | TIPS | CO2Me | (S,S)-IBioxtBu | – | 43 | 84:16 | |||
| 5 | 10e | Br | TIPS | CO2Et | (S,S)-IBioxtBu | – | 32 | 72:28 | |||
| 6 | 10f | Br | Piv | CO2Me | (S,S)-IBioxtBu | – | 60 | 77:23 | |||
| 7 | 10g | Br | tBu | CO2Me | (S,S)-IBioxtBu | trace | 69 | 59:41 | |||
| 8 | 10h | Br | Bn | CO2Me | (S,S)-IBioxtBu | 22 | 99.5:0.5 | 275 | 62 | 69:31 | 6.7 |
| 9 | 10i | Br | PMB | CO2Me | (S,S)-IBioxtBu | 14 | 99.4:0.6 | 59 | 64:36 | ||
| 10 | 10j | Br | MOM | CO2Me | (S,S)-IBioxtBu | 27 | 99.6:0.4 | 38 | 69:31 | ||
| 11 | 10h | Br | Bn | CO2Me | IBioxMe4 | 12e | 50:50 | 88e | 50:50 | ||
| 12 | 10k | I | Bn | CO2Me | (S,S)-IBioxtBu | – | 89 | 50:50 | |||
| 13 | 10l | Br | Bn | COCF3 | (S,S)-IBioxtBu | – | 10e | ||||
| 14 | 10h | Br | Bn | CO2Me | (S,S)-IBioxAd | 21 | 99:1 | 136 | 60 | 69:31 | 5.9 |
| 15 | 10h | Br | Bn | CO2Me | (S,S)-L1 | 35 | 0.2:99.8 | 1083 | 47 | 84:16 | 14.3 |
| 16 | 10h | Br | Bn | CO2Me | (R,R)-L2 | 47 | 99.8:0.2 | 1268 | 52 | 8:92 | 33 |
Reaction conditions: 10 (0.1 mmol), [Pd(π-allyl)Cl]2 (5 mol %), ligand (10 mol %), CsOPiv (1 equiv), Cs2CO3 (1.5 equiv), 4 Å MS, toluene, 140 °C, 15 h. TBS = tert-butyldimethylsilyl. TES = triethylsilyl. TBDPS = tert-butyldiphenylsilyl. TIPS = triisopropylsilyl. Piv = pivaloyl. Bn = benzyl. PMB = p-methoxybenzyl. MOM = methoxymethyl.
Yield of the isolated product.
er values were determined by HPLC using a chiral stationary phase.
Selectivity factors calculated from the equation s = ln[1 – c(1 + eePr)]/ln[1 – c(1 – eePr)] = krel = kfast/kslow, where c is the conversion determined by 1H NMR and Pr is the product.
Determined by 1H NMR using trichloroethylene as the internal standard.
Thermal ellipsoids shown at 50% probability. The absolute configurations of the other products were assigned in analogy to 12j.
Following up on this model study, we embarked on the synthesis of advanced tetraoxygenated intermediate 3 toward (−)-cryptowolinol (Scheme 2). Commercially available benzaldehyde 14 first underwent Strecker and Pinner reactions to generate aminoester 6. Reduction of the ester and protection as TBS ether were followed by Buchwald–Hartwig N-arylation with dibromide 5, leading to intermediate 16 in excellent yield. Protecting group exchange to install the required benzyl group, followed by aniline protection with methyl chloroformate, furnished compound 4 in 22.7% overall yield over eight steps.
Scheme 2. Enantioselective Synthesis of Advanced Intermediate 3.
Reaction conditions: (a) LiHMDS (1.1 equiv), THF, −40 to 25 °C, then acetone cyanohydrin (2 equiv), 25 °C; (b) 3 M HCl/MeOH, 25 °C; (c) NaBH4 (3 equiv), EtOH, 0 °C; (d) TBSCl (1.1 equiv), Et3N (2.2 equiv), DMAP (10 mol %), CH2Cl2, 0 °C; (e) 15 (1.0 equiv), 5,6-dibromo-1,3-benzodioxole 5 (1.05 equiv), Pd2dba3 (4 mol %), (±)-BINAP (8 mol %), NaOtBu (1.5 equiv), toluene, 80 °C; (f) TBAF (2 equiv), THF, 0 to 25 °C; (g) NaH (2 equiv), BnBr (1.5 equiv), THF, 0 to 25 °C; (h) ClCO2Me, neat, reflux; (i) see Table 2.
Application of the conditions optimized for substrate 10 using (R,R)-L2 as the ligand (see Table 1, entry 16) to the more oxygenated substrate 4 provided lower yields of both C(sp3)–H and C(sp2)–H arylation products 3 and 17 (Table 2, entry 1). Indeed, the corresponding protodebrominated product was formed in 22% yield, which affected the efficiency of the PKR process. Analogous ligands L1 and L3 were tested on this substrate but did not furnish better results (entries 2 and 3, respectively). Nevertheless, scaling up the reaction with enantiomeric ligand (S,S)-L2 (entry 4) allowed the isolation of indoline 3 in satisfying yield (107 mg, 40%) and enantioselectivity (er 98:2). Unfortunately, despite significant experimentation, all attempts to cleave the methyl carbamate in 3 led to the elimination of the benzyloxy group and the formation of the corresponding indole. Future investigations will be devoted to the introduction of a more labile N-protecting group to complete the synthesis of (−)-cryptowolinol and confirm or reassign its absolute configuration.
Table 2. PKR of Aryl Bromide 4a.
| entry | ligand | yield of 3 (%)b | er of 3c | yield of 17 (%)b | er of 17c |
|---|---|---|---|---|---|
| 1 | (R,R)-L2 | 34 | 0.5:99.5 | 35 | 40:60 |
| 2 | (S,S)-L1 | 15 | 98.5:1.5 | 17 | 73:27 |
| 3 | (R,R)-L3 | 16 | 4:96 | 6 | 34:66 |
| 4d | (S,S)-L2 | 40 | 98:2 | –e |
Reaction conditions: 4 (0.1 mmol), [Pd(π-allyl)Cl]2 (5 mol %), ligand (10 mol %), CsOPiv (1 equiv), Cs2CO3 (1.5 equiv), 4 Å MS, toluene, 140 °C, 15 h.
Yield of the isolated product.
er values were determined by HPLC using a chiral stationary phase.
Run on a 0.5 mmol scale.
This product could not be isolated in pure form on this scale. Absolute configurations were assigned on the basis of the X-ray crystal structure of 12j.
In conclusion, we performed a model study of the enantioselective synthesis of the rare dibenzopyrrocoline alkaloid (−)-cryptowolinol. The key step involves a challenging enantioselective Pd0-catalyzed C(sp3)–H arylation proceeding via parallel kinetic resolution. A very efficient PKR was achieved on a deoxygenated model substrate and was successfully transposed to a potential intermediate en route to (−)-cryptowolinol.
Acknowledgments
This work was financially supported by the State Secretariat for Education, Research and Innovation (2019.0454), Osaka University Engineering Science Student Dispatch Program 2022, and the University of Basel. The authors thank Dr. D. Häussinger for NMR experiments, Dr. M. Pfeffer for MS analyses, and Dr. A. Prescimone for X-ray diffraction analysis.
Data Availability Statement
The data underlying this study are available in the published article and its Supporting Information.
Supporting Information Available
The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.orglett.4c00386.
Procedural and spectral data (PDF)
The authors declare no competing financial interest.
Supplementary Material
References
- Bentley K. W. β-Phenylethylamines and the Isoquinoline Alkaloids. Nat. Prod. Rep. 2006, 23, 444–463. 10.1039/B509523A. [DOI] [PubMed] [Google Scholar]
- Elliott I. W. Dibenzopyrrocoline Alkaloids. Alkaloids 1987, 31, 101–116. 10.1016/S0099-9598(08)60259-X. [DOI] [Google Scholar]
- a Ewing J.; Hughes G. K.; Ritchie E.; Taylor W. C. An Alkaloid Related to Dehydrolaudanosoline. Nature 1952, 169, 618–619. 10.1038/169618b0. [DOI] [Google Scholar]; b Ewing J.; Hughes G. K.; Ritchie E.; Taylor W. C. The Alkaloids of Cryptocasrya bowiei (Hook.) Druce. Aust. J. Chem. 1953, 6, 78–85. 10.1071/CH9530078. [DOI] [Google Scholar]; c Takano S.; Satoh S.; Oshima Y.; Ogasawara K. Stereochemistry of the Dibenzopyrrocoline Alkaloids Cryptaustoline and Cryptowoline. Heterocycles 1987, 26, 1487–1489. 10.3987/R-1987-06-1487. [DOI] [Google Scholar]; d Lee S.-S.; Chen C.-K.; Huang F.-M.; Chen C.-H. Two Dibenzopyrrocoline Alkaloids from Litsea cubeba. J. Nat. Prod. 1996, 59, 80–82. 10.1021/np960014b. [DOI] [Google Scholar]
- a Leboeuf M.; Cavé A.; Ranaivo A.; Moskowitz H. Cryptowolinol et Cryptowolidine, Nouveaux Alcaloïdes de Type Dibenzopyrrocoline. Can. J. Chem. 1989, 67, 947–952. 10.1139/v89-145. [DOI] [Google Scholar]; b Cavé A.; Leboeuf M.; Moskowitz H.; Ranaivo A.; Bick I. R. C.; Sinchai W.; Nieto M.; Sévenet T.; Cabalion P. Alkaloids of Cryptocarya phyllostemon. Aust. J. Chem. 1989, 42, 2243–2263. 10.1071/CH9892243. [DOI] [Google Scholar]
- Meyers A. I.; Sielecki T. M. Total Synthesis of the Dibenzopyrrocoline Alkaloid (S)-(+)-Cryptaustoline. Revision of Absolute Configuration Due to an Unusual Inversion in Stereochemistry. J. Am. Chem. Soc. 1991, 113, 2789–2790. 10.1021/ja00007a084. [DOI] [Google Scholar]; b Meyers A. I.; Sielecki T. M.; Crans D. C.; Marshman R. W.; Nguyen T. H. (−)-Cryptaustoline: Its Synthesis, Revision of Absolute Stereochemistry, and Mechanism of Inversion of Stereochemistry. J. Am. Chem. Soc. 1992, 114, 8483–8489. 10.1021/ja00048a020. [DOI] [Google Scholar]
- Selected reviews:; a Yamaguchi J.; Yamaguchi A. D.; Itami K. C–H Bond Functionalization: Emerging Synthetic Tools for Natural Products and Pharmaceuticals. Angew. Chem., Int. Ed. 2012, 51, 8960–9009. 10.1002/anie.201201666. [DOI] [PubMed] [Google Scholar]; b Chen D. Y. K.; Youn S. W. C–H Activation: A Complementary Tool in the Total Synthesis of Complex Natural Products. Chem. - Eur. J. 2012, 18, 9452–9474. 10.1002/chem.201201329. [DOI] [PubMed] [Google Scholar]; c Karimov R. R.; Hartwig J. F. Transition-Metal-Catalyzed Selective Functionalization of C(sp3)–H Bonds in Natural Products. Angew. Chem., Int. Ed. 2018, 57, 4234–4241. 10.1002/anie.201710330. [DOI] [PMC free article] [PubMed] [Google Scholar]; d Abrams D. J.; Provencher P. A.; Sorensen E. J. Recent Applications of C–H Functionalization in Complex Natural Product Synthesis. Chem. Soc. Rev. 2018, 47, 8925–8967. 10.1039/C8CS00716K. [DOI] [PubMed] [Google Scholar]; e Baudoin O. Multiple Catalytic C–H Bond Functionalization for Natural Product Synthesis. Angew. Chem., Int. Ed. 2020, 59, 17798–17809. 10.1002/anie.202001224. [DOI] [PubMed] [Google Scholar]; f Lam N. Y. S.; Wu K.; Yu J.-Q. Advancing the Logic of Chemical Synthesis: C–H Activation as Strategic and Tactical Disconnections for C–C Bond Construction. Angew. Chem., Int. Ed. 2021, 60, 15767–15790. 10.1002/anie.202011901. [DOI] [PMC free article] [PubMed] [Google Scholar]; g Sinha S. K.; Ghosh P.; Jain S.; Maiti S.; Al-Thabati S. A.; Alshehri A. A.; Mokhtar M.; Maiti D. Transition-Metal Catalyzed C–H Activation as a Means of Synthesizing Complex Natural Products. Chem. Soc. Rev. 2023, 52, 7461–7503. 10.1039/D3CS00282A. [DOI] [PubMed] [Google Scholar]
- a Eames J. Parallel Kinetic Resolutions. Angew. Chem., Int. Ed. 2000, 39, 885–888. . [DOI] [PubMed] [Google Scholar]; b Dehli J. R.; Gotor V. Parallel Kinetic Resolution of Racemic Mixtures: A New Strategy for the Preparation of Enantiopure Compounds?. Chem. Soc. Rev. 2002, 31, 365–370. 10.1039/b205280f. [DOI] [PubMed] [Google Scholar]
- a Newton C. G.; Wang S.-G.; Oliveira C. C.; Cramer N. Catalytic Enantioselective Transformations Involving C–H Bond Cleavage by Transition-Metal Complexes. Chem. Rev. 2017, 117, 8908–8976. 10.1021/acs.chemrev.6b00692. [DOI] [PubMed] [Google Scholar]; b Sumit; Chandra D.; Sharma U. Merging Kinetic Resolution with C–H Activation: An Efficient Approach for Enantioselective Synthesis. Org. Biomol. Chem. 2021, 19, 4014–4026. 10.1039/D1OB00232E. [DOI] [PubMed] [Google Scholar]
- a Katayev D.; Nakanishi M.; Bürgi T.; Kündig E. P. Asymmetric C(sp3)–H/C(Ar) Coupling Reactions. Highly Enantio-enriched Indolines via Regiodivergent Reaction of a Racemic Mixture. Chem. Sci. 2012, 3, 1422–1425. 10.1039/c2sc20111a. [DOI] [Google Scholar]; b Katayev D.; Larionov E.; Nakanishi M.; Besnard C.; Kündig E. P. Palladium–N-Heterocyclic Carbene (NHC)-Catalyzed Asymmetric Synthesis of Indolines through Regiodivergent C(sp3)–H Activation: Scope and DFT Study. Chem. - Eur. J. 2014, 20, 15021–15030. 10.1002/chem.201403985. [DOI] [PubMed] [Google Scholar]
- Baudoin O. Ring Construction by Palladium(0)-Catalyzed C(sp3)–H Activation. Acc. Chem. Res. 2017, 50, 1114–1123. 10.1021/acs.accounts.7b00099. [DOI] [PubMed] [Google Scholar]
- Wheatley M.; Zuccarello M.; Tsitopoulou M.; Macgregor S. A.; Baudoin O. Effect of α-Substitution on the Reactivity of C(sp3)–H Bonds in Pd0-Catalyzed C–H Arylation. ACS Catal. 2023, 13, 12563–12570. 10.1021/acscatal.3c03806. [DOI] [PMC free article] [PubMed] [Google Scholar]
- a Yang S.-Y.; Han W.-Y.; Zhang D.-L.; Zhou X.-J.; Bai M.; Cui B.-D.; Wan N.-W.; Yuan W.-C.; Chen Y.-Z. Synthesis of Phenanthridines through Palladium-Catalyzed Cascade Reaction of 2-Halo-N-Ms-arylamines with Benzyl Halides/Sulfonates. Eur. J. Org. Chem. 2017, 2017, 996–1003. 10.1002/ejoc.201601608. [DOI] [Google Scholar]; b Miyakoshi T.; Niggli N. E.; Baudoin O. Remote Construction of N-Heterocycles via 1,4-Palladium Shift-Mediated Double C–H Activation. Angew. Chem., Int. Ed. 2022, 61, e202116101. 10.1002/anie.202116101. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Melot R.; Zuccarello M.; Cavalli D.; Niggli N.; Devereux M.; Bürgi T.; Baudoin O. Palladium(0)-Catalyzed Enantioselective Intramolecular Arylation of Enantiotopic Secondary C–H Bonds. Angew. Chem., Int. Ed. 2021, 60, 7245–7250. 10.1002/anie.202014605. [DOI] [PubMed] [Google Scholar]
- Glorius F.; Altenhoff G.; Goddard R.; Lehmann C. Oxazolines as Chiral Building Blocks for Imidazolium Salts and N-Heterocyclic Carbene Ligands. Chem. Commun. 2002, 2704–2705. 10.1039/b208045a. [DOI] [PubMed] [Google Scholar]
- Altenhoff G.; Goddard R.; Lehmann C. W.; Glorius F. Sterically Demanding, Bioxazoline-Derived N-Heterocyclic Carbene Ligands with Restricted Flexibility for Catalysis. J. Am. Chem. Soc. 2004, 126, 15195–15201. 10.1021/ja045349r. [DOI] [PubMed] [Google Scholar]
- Kündig E. P.; Seidel T. M.; Jia Y.-x.; Bernardinelli G. Bulky Chiral Carbene Ligands and Their Application in the Palladium-Catalyzed Asymmetric Intramolecular α-Arylation of Amides. Angew. Chem., Int. Ed. 2007, 46, 8484–8487. 10.1002/anie.200703408. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data underlying this study are available in the published article and its Supporting Information.