Quantitative Structure-Selectivity Relationships in Enantioselective Catalysis: Past, Present, and Future

Abstract

The dawn of the 21st century has brought with it a surge of research related to computer-guided approaches to catalyst design. In the past two decades, chemoinformatics, the application of informatics to solve problems in chemistry, has increasingly influenced prediction of activity and mechanistic investigations of organic reactions. The advent of advanced statistical and machine learning methods, as well as dramatic increases in computational speed and memory, has contributed to this emerging field of study. This review summarizes strategies to employ quantitative structure-selectivity relationships (QSSR) in asymmetric catalytic reactions. The coverage is structured by initially introducing the basic features of these methods. Subsequent topics are discussed according to increasing complexity of molecular representations. As the most applied subfield of QSSR in enantioselective catalysis, the application of local parametrization approaches and linear free energy relationships (LFERs) along with multivariate modeling techniques is described first. This section is followed by a description of global parametrization methods, the first of which is continuous chirality measures (CCM) because it is a single parameter derived from the global structure of a molecule. Chirality codes, global, multivariate descriptors, are then introduced followed by molecular interaction fields (MIFs), a global descriptor class that typically has the highest dimensionality. To highlight the current reach of QSSR in enantioselective transformations, a comprehensive collection of examples is presented. When combined with traditional experimental approaches, chemoinformatics holds great promise to predict new catalyst structures, rationalize mechanistic behavior, and profoundly change the way chemists discover and optimize reactions.

Figure 1.

References obtained from a Scifinder search including the words or the concept “asymmetric catalysis” since 1990.

Figure 2.

An overview of chemoinformatics guided catalyst optimization.

Figure 3.

Charton and Sterimol steric parameters. Reproduced from Brethomé, A. V.; Fletcher, S. P.; Paton, R. S. Conformational Effects on Physical-Organic Descriptors: The Sterimol Steric Parameters. ACS Catal. 2019, 9, 2313–2323. Copyright 2019 American Chemical Society.

Figure 4.

Catalysts and substrates used in enantioselective NHK reaction and a univariate correlation of enantioselectivity with substituent Charton values. Adapted with permission from reference 63. Copyright 2008 John Wiley and Sons.

Figure 5.

Top: LFER for enantioselective, palladium catalyzed alkylation of allyl acetates. Center: LFER for enantioselective cyclopropanation of allylic alcohols. Bottom: LFER for enantioselective aziridination of styrene. Adapted with permission from reference 63. Copyright 2008 John Wiley and Sons.

Figure 6.

Catalysts with accompanying LFER plots for benzaldehyde and acetophenone allylation. Reproduced from Sigman, M. S.; Miller, J. J. Examination of the Role of Taft-Type Steric Parameters in Asymmetric Catalysis. J. Org. Chem. 2009, 74, 7633–7643. Copyright 2009 American Chemical Society.

Figure 7.

LFERs for desymmetrization of bisphenol substrates with a peptide catalyst (top). Reproduced from Sigman, M. S.; Miller, J. J. Examination of the Role of Taft-Type Steric Parameters in Asymmetric Catalysis. J. Org. Chem. 2009, 74, 7633–7643. Copyright 2009 American Chemical Society.

Figure 8.

LFERs between enantioselectivity and polarizability (top) and quadrupole moment (bottom) of the aromatic substituent on thiourea catalysts. Adapted with permission from reference 77. Copyright 2010 National Academy of Sciences.

Figure 9.

Interatomic distances (top left) in transition structures leading to R and S stereoisomers (right) and their correlation with enantioselectivity (bottom left). Reproduced from Zuend, S. J.; Jacobsen, E. N. Mechanism of Amido-Thiourea Catalyzed Enantioselective Imine Hydrocyanation: Transition State Stabilization via Multiple Non-Covalent Interactions. J. Am. Chem. Soc. 2009, 131, 15358–15374. Copyright 2009 American Chemical Society.

Figure 10.

Pairwise combinations of X and Y substituents on the common core give 25 unique catalysts (top right). Experimentally determined enantioselectivities are used to construct a multivariate relationship between catalyst descriptors and selectivity (bottom). Predictive models are constructed by using a 9-member training set evenly covering the descriptor space. Adapted with permission from reference 81. Copyright 2011 National Academy of Sciences

Figure 11.

Scaffold comparison for the enantioselective propargylation of ketones.

Figure 12.

Palladium catalyzed enantioselective Heck arylation.

Figure 13.

Enantioselective alkylation of aryl aldehydes (top). LFERs with Charton parameters (center), Sterimol parameters (bottom). Reproduced from Huang, H.; Zong, H.; Bian, G.; Song, L. Constructing a Quantitative Correlation between N-Substituent Sizes of Chiral Ligands and Enantioselectivities in Asymmetric Addition Reactions of Diethylzinc with Benzaldehyde. J. Org. Chem. 2012, 77, 10427–10434. Copyright 2012 American Chemical Society.

Figure 14.

Enantioselective propargylation reaction (top) and test catalyst / substrate combinations with their predicted and observed values in kcal/mol (bottom). Reproduced from Harper, K. C.; Vilardi, S. C.; Sigman, M. S. Prediction of Catalyst and Substrate Performance in the Enantioselective Propargylation of Aliphatic Ketones by a Multidimensional Model of Steric Effects. J. Am. Chem. Soc. 2013, 135, 2482–2485. Copyright 2013 American Chemical Society.

Figure 15.

Enantioselective Henry reaction catalyzed by 1-amino-2-phosphinamido ligands. Reproduced from Huang, H.; Zong, H.; Bian, G.; Yue, J.; Song, L. Correlating the Effects of the N-Substituent Sizes of Chiral 1,2-Amino Phosphinamide Ligands on Enantioselectivities in Catalytic Asymmetric Henry Reaction Using Physical Steric Parameters. J. Org. Chem. 2014, 79, 9455–9464. Copyright 2014 American Chemical Society.

Figure 16.

Models with and without molecular vibrations as descriptors.

Figure 17.

Dehydrogenative Heck reaction optimized using a predictive model. Reproduced from Zhang, C.; Santiago, C. B.; Crawford, J. M.; Sigman, M. S. Enantioselective Dehydrogenative Heck Arylations of Trisubstituted Alkenes with Indoles to Construct Quaternary Stereocenters. J. Am. Chem. Soc. 2015, 137, 15668–15671. Copyright 2015 American Chemical Society.

Figure 18.

(a) Phosphoric acid catalyzed dehydrogenative C-N coupling indicating substituent variability in substrates and catalyst members. (b) Multivariate LFER with indicated descriptors, for the triazole substituted catalyst and the resulting correlation. Adapted with permission from reference 92. Copyright 2015 American Association for the Advancement of Science.

Figure 19.

Predictive models for the kinetic resolution of secondary alcohols (top) and the enantioselective fluorination of allylic alcohols (bottom). Reproduced from Orlandi, M.; Coelho, J. A. S.; Hilton, M. J.; Toste, F. D.; Sigman, M. S. Parameterization of Non-covalent Interactions for Transition State Interrogation Applied to Asymmetric Catalysis. J. Am. Chem. Soc. 2017, 139, 6803–6806. Copyright 2017 American Chemical Society.

Figure 20.

Four case studies to identify structural effects of amino acid ligands in palladium catalyzed C-H activation reactions and the resulting suggested ligand set. Reproduced from Park, Y.; Niemeyer, Z. L.; Yu, J.-Q.; Sigman, M. S. Quantifying Structural Effects of Amino Acid Ligands in Pd(II)-Catalyzed Enantioselective C-H Functionalization. Organometallics. 2018, 37, 203–210. Copyright 2018 American Chemical Society.

Figure 21.

A multivariate LFER for an enantioselective Pummerer reaction. Adapted with permission from reference 97. Copyright John Wiley and Sons.

Figure 22.

Enantioselective, palladium catalyzed substitution of allylic alcohols. Predicted vs observed selectivities for various substrates for two different catalyst classes. Reproduced from Wang, Y.; Zhou, H.; Yang, K.; You, C.; Zhang, L.; Luo, S. Steric Effect of Protonated Tertiary Amine in Primary-Tertiary Diamine Catalysis: A Double-Layered Sterimol Model. Org. Lett. 2019, 21, 407–411. Copyright American Chemical Society.

Figure 23.

Enantioselective benzoin reaction.

Figure 24.

Copper catalyzed, enantioselective cyclopropanation with a representative oxazoline ligand set (top) and generation of distance-weighted volume and Charton-Taft values (bottom).

Figure 25.

Chiral Lewis acid catalyzed enantioselective Diels-Alder reaction. (a) CCM vs. dihedral angle for 2,2’-biaryldiols. Sets a, b and c refer to the different ester residues in the starting material (b) Biphenyl dihedral angle and CCM as they relate to enantioselectivity. Reproduced from Gao, D.; Schefzick, S.; Lipkowitz, K. Relationship between Chirality Content and Stereoinduction: Identification of a Chiraphore. J. Am. Chem. Soc. 1999, 121, 9481–9482. Copyright 1999 American Chemical Society.

Figure 26.

Illustration of twist, bite, and pucker distortions and their relation to CCM. Reproduced from Lipkowitz, K.; Schefziek, S.; Avnir, D. Enhancement of Enantiomeric Excess by Ligand Distortion. J. Am. Chem. Soc. 2001, 123, 6710–6711. Copyright 2001 American Chemical Society.

Figure 27.

The four distortions studied in reference 128 and their relation to CCM. The numbering system described in the original work has been included for reference.

Figure 28.

Ruthenium-catalyzed, enantioselective, transfer hydrogenation reaction.

Figure 29.

Selectivity of phosphoric acid catalysts in the synthesis of chiral, N,S- acetals.

Figure 29.

Selectivity of phosphoric acid catalysts in the synthesis of chiral, N,S- acetals.

Figure 30.

External test sets for models generated with CCM parameters, Sterimol parameters, and both set representing catalysts. Adapted with permission from reference 137. Copyright 2019 Elsevier.

Figure 31.

CICC calculations. Reproduced from Aires-de-Sousa, J.; Gasteiger, J. New Description of Molecular Chirality and Its Application to the Prediction of the Preferred Enantiomer in Stereoselective Reactions. J. Chem. Inf. Comput. Sci. 2001, 41, 369–375. Copyright American Chemical Society.

Figure 32.

Enantioselective addition of diethylzinc to benzaldehyde and 45 training catalysts. The major isomer formed when the catalyst is used in the reaction is designated by the (+) or (−) under the catalyst structure.

Figure 33.

Test catalysts for the enantioselective addition of diethylzinc into benzaldehyde. The major isomer formed when the catalyst is used in the reaction is designated by the (+) or (−) under the catalyst structure.

Figure 34.

Secondary alcohols synthesized by the reduction of the corresponding ketone with (−)-DIP-chloride. The major isomer of the product is depicted in each case.

Figure 35.

Enantioselective diethylzinc alkylation of benzaldehyde with the predicted vs. observed plot. Reproduced from Aires-de-Sousa, J.; Gasteiger, J. Prediction of Enantiomeric Excess in a Combinatorial Library of Catalytic Enantioselective Reactions. J. Comb. Chem. 2005, 7, 298–301. Copyright American Chemical Society.

Figure 36.

Enantioselective transfer hydrogenation for the reduction of acetophenone.

Figure 37.

Designation of “left” and “right” substituents used in reference 155.

Figure 38.

Alignment dependent MIF workflow to represent a molecule with a grid based descriptor.

Figure 39.

CoMFA for enantioselective, Diels-Alder reaction. Most selective catalyst in reference 158 with areas where high occupancy corresponds to selectivity (green) and where low occupancy corresponds to selectivity (yellow). Reproduced from Lipkowitz, K.; Pradhan, M. Computational Studies of Chiral Catalysts: A Comparative Molecular Field Analysis of an Asymmetric Diels-Alder Reaction with Catalysts Containing Bisoxazoline or Phosphinooxazoline Ligands. J. Org. Chem. 2003, 68, 4648–4656. Copyright 2003 American Chemical Society.

Figure 40.

Sparteine and sparteine analogs employed as ligands in enantioselective lithiation of N-Bocpyrrolidine.

Figure 41.

Regions with grid points (red and blue) correlated with catalyst enantioselectivity.

Figure 42.

Sparteine and sparteine analogs employed as ligands in enantioselective lithiation of N-Bocpyrrolidine.

Figure 43.

Temperature dependent selectivity predictions in the enantioselective diethylzinc addition to benzaldehyde.

Figure 44.

Catalysts for the enantioselective addition of diethylzinc into aryl aldehydes.

Figure 45.

Cinchonidinium alkaloid catalyzed, enantioselective phase transfer alkylations along with Predicted vs. Observed plot for training (diamonds) and test (circles) catalyst sets. Adapted with permission from reference 172. Copyright Royal Society of Chemistry.

Figure 46.

Enantioselective alkylation with biaryl-derived phase transfer catalysis.

Figure 47.

Different possible conformations of the catalyst scaffold. ^aaryl = Ph, 1-naphthyl, mesityl. ^bLibrary containing different conformer combinations, ^cConformation of scaffold. Reproduced from Denmark, S. E.; Gould, N. D.; Wolf, L. M. A Systematic Investigation of Quaternary Ammonium Ions as Asymmetric Phase-Transfer Catalysts. Application of Quantitative Structure Activity/Selectivity Relationships. J. Org. Chem. 2011, 76, 4337–4357. Copyright 2011 American Chemical Society.

Figure 48.

Top: steric contour maps from two different perspectives. Green contours indicate regions where steric bulk leads to increased enantioselectivity, whereas yellow regions indicate regions where less steric bulk leads to increased enantioselectivity. Bottom: Electrostatic contour maps from two different perspectives. Blue contours indicate regions where increased positive charge leads to greater enantioselectivity, whereas red contours indicate regions where decreased positive charge (or increased negative charge) leads to increased enantioselectivity. Reproduced from Denmark, S. E.; Gould, N. D.; Wolf, L. M. A Systematic Investigation of Quaternary Ammonium Ions as Asymmetric Phase-Transfer Catalysts. Application of Quantitative Structure Activity/Selectivity Relationships. J. Org. Chem. 2011, 76, 4337–4357. Copyright 2011 American Chemical Society.

Figure 49.

Top left: steric contour map for catalysts in enantioselective ketone hydrogenation reactions. Green contours indicate regions where steric bulk leads to increased enantioselectivity, whereas yellow regions indicate regions devoid of steric bulk which lead to increased enantioselectivity. Top right: Electrostatic contour map. Blue contours indicate regions where increased positive charge leads to greater enantioselectivity whereas red contours indicate regions where decreased positive charge (or increased negative charge) lead to increased enantioselectivity. Experimental catalyst (A1) and theoretically improved catalyst (C1). Adapted with permission from reference 180. Copyright Royal Society of Chemistry.

Figure 50:

Rhodium catalyzed enantioselective addition of phenylboronic acid to 1-naphthaldehyde with depiction of ligand library. (a)–(d) Space filling models and digitized structures of two catalysts. The red areas designate regions where steric bulk is associated with diminished selectivity, and the blue regions designate areas where steric bulk is associated with high selectivity. The circled region in (b) and (d) contains the methyl unit that was removed (it is absent is (a) and(c)). Reprinted from with permission from reference 186. Copyright 2017 John Wiley and Sons.

Figure 51.

(A) Formation of chiral N,S-acetals with train and test substrate combinations. (B) Catalyst and substrate combinations forming different train and test sets. (C) External test catalysts. Adapted with permission from reference 138. Copyright 2019 American Association for the Advancement of Science.

Figure 52.

A) Predicted vs. observed free energies (kcal / mol) of the train and test sets overlaid for a support vector machine using a second order polynomial kernel. Accuracy metrics are listed in the table below. B) The selectivity space as represented by the first three principal components of the full feature-space. Adapted with permission from reference 138. Copyright 2019 American Association for the Advancement of Science.

Figure 53.

(A) Predicted vs. Observed plot for simulated reaction optimization. (B) Average predicted and observed selectivity data for all catalysts wit average selectivity over 80 % ee. Adapted with permission from reference 138. Copyright 2019 American Association for the Advancement of Science.

Figure 54.

Graphical representation of the process for the calculation of GRIND.

Figure 55.

Predicted and observed selectivities for enantioselective addition of diethylzinc to benzaldehyde. GSI ratios for different structural motifs are depicted in color.

Figure 56.

Rhodium-catalyzed hydroformylation of terminal alkenes.

Figure 57.

Predicted and observed selectivity data for selected catalysts.

Figure 58.

Construction of matrices Z₁, Z₂, and Z₃.

Figure 59.

Enantioselective cyclopropanation and predictive model of enantioselectivity. Adapted from Jiang, C.; Li, Y.; Tian, Q.; You, T. QSAR Study of Catalytic Asymmetric Reactions with Topological Indices. J. Chem. Inf. Comput. Sci. 2003, 43, 1876–1881. Copyright 2003 American Chemical Society.

Figure 60.

Enantioselective cyclopropanation and enantioselectivity predictive model. Reproduced from Jiang, C.; Li, Y.; Tian, Q.; You, T. QSAR Study of Catalytic Asymmetric Reactions with Topological Indices. J. Chem. Inf. Comput. Sci. 2003, 43, 1876–1881. Copyright 2003 American Chemical Society.

Figure 61.

Diastereoselective pinacol coupling and predictive model of diastereoselectivity. Reproduced from Jiang, C.; Li, Y.; Tian, Q.; You, T. QSAR Study of Catalytic Asymmetric Reactions with Topological Indices. J. Chem. Inf. Comput. Sci. 2003, 43, 1876–1881. Copyright 2003 American Chemical Society.

Figure 62.

Enantioselective Kumada coupling reaction. Reproduced from Jiang, C.; Li, Y.; Tian, Q.; You, T. QSAR Study of Catalytic Asymmetric Reactions with Topological Indices. J. Chem. Inf. Comput. Sci. 2003, 43, 1876–1881. Copyright 2003 American Chemical Society.

Figure 63.

Catalyzed addition of diethylzinc to benzaldehyde.

Figure 64.

Enantioselective addition of diethylzinc to aldehydes using amino thiol ligands. Adapted with permission from reference 206. Copyright 2006 Elsevier.

Figure 65.

Enantioselective addition of diethylzinc to aldehydes using an amino pyridine ligand. Adapted with permission from reference 206. Copyright 2006 Elsevier.

Figure 66.

Enantioselective diethylzinc alkylation, case study 4 from reference 206. Adapted with permission from reference 206. Copyright 2006 Elsevier.

Figure 67.

Reaction and catalysts studies in enantioselective acetophenone reduction. Adapted with permission from reference 192. Copyright 2004 Elsevier.

Figure 68.

Enantioselective hydrogenation reaction studied in reference 218.

Figure 69.

Enantioselective β-arylation reaction. Adapted with permission from reference 220. Copyright 2012 Royal Society of Chemistry.

Chart 1.

UTS for chiral, phosphoric acid catalysts.

Scheme 1.

Enantioselective Heck-Heck cascade reaction.

Scheme 2.

Chiral Bronsted acid catalyzed addition of enecarbamates to acyliminium ions.