Enantioselectivity chemical optimization

Chem. Soc., 126, 14411-14418 Skander, M., Malan, C., Ivanova, A. and Ward, TR. (2005) Chemical optimization of artificial metaUoenzymes based on the biotin-avidin technology (S)-selective and solvent-tolerant hydrogenation catalysts via the introduction of chiral amino acid spacers. Chem. Commun., 4815-4817 Ward, TR. (2005) Artificial metallo-enzymes for enantioselective catalysis based on the noncovalent incorporation of organometallic moieties in a host protein. Chem.-Eur. J., 11, 3798-3804 Letondor, C. and Ward, TR. (2006) Artificial metaUoenzymes for enantioselective catalysis Recent advances. Chem. Bio. Chem., 7, 1845-1852. 

Fig. 2 Biotin-avidin technology Artificial metalloenzymes [M(L )(biotin-ligand)]c(strept)avidin for enantioselective catalysis are based on the anchoring of a catalyticaUy active metal fragment within a host protein via a hgand, a spacer, and biotin. Chemical optimization can be achieved either by varying the spacer or the metal chelate moiety ML ). Saturation mutagenesis at a position close to the metal moiety ( ) can be used for genetic optimization...

As shown above, it was not so easy to optimize the Michael addition reactions of l-crotonoyl-3,5-dimethylpyrazole in the presence of the l ,J -DBFOX/ Ph-Ni(C104)2 3H20 catalyst because a simple tendency of influence to enantio-selectivity is lacking. Therefore, we changed the acceptor to 3-crotonoyl-2-oxazolidi-none in the reactions of malononitrile in dichloromethane in the presence of the nickel(II) aqua complex (10 mol%) (Scheme 7.49). For the Michael additions using the oxazolidinone acceptor, dichloromethane was better solvent than THF and the enantioselectivities were rather independent upon the reaction temperatures and Lewis base catalysts. Chemical yields were also satisfactory. 

Here, a few comments should be made with regards to Table 1.1. It is clear that for aU three substrate classes, enantioselectivities can range from good to very high, whereas the TON-and especially TOF-values are less impressive (and in most cases have not been optimized). Most catalyst systems require pressures of 20-100 bar (2-10 x lO" hPa) to achieve realistic reaction times but, as a rule, the chemical yields are very high. In the presence of Ti(OiPr)4, Ir/f-binaphane... 

After completing his initial intramolecular cycloaddition, Hodgson utilized conditions that had been optimized for the intermolecular cycloaddition of DMAD with simple cyclic carbonyl ylides used by Hashimoto and co-workers (139). Hodgson et al. (140) found that the reaction indeed gave excellent overall chemical yield, but the enantioselectivity dropped to 1%, giving essentially a racemic mixture. It appeared that ee ratios were sensitive to the electronic nature of the dipole. Hodgson chose to screen several binaphthol derived rhodium catalysts of the type developed by McKervey and Pirrung, due in part to the reports of... 

Widenhoefer and co-workers have developed an effective Pd-catalyzed protocol for the asymmetric cyclization/ hydrosilylation of functionalized 1,6-dienes that employed chiral, non-racemic pyridine-oxazoline ligands." " " Optimization studies probed the effect of both the G(4) substituent of the pyridine-oxazoline ligand (Table 7, entries 1-6) and the nature of the silane (Table 7, entries 6-15) on the yield and enantioselectivity of the cyclization/ hydrosilylation of dimethyl diallylmalonate. These studies revealed that employment of isopropyl-substituted catalyst (N-N)Pd(Me)Gl [N-N = (i )-( )-4-isopropyl-2-(2-pyridinyl)-2-oxazoline] [(i )-43f and a stoichiometric amount of benzhydryldimethylsilane provided the best combination of asymmetric induction and chemical yield, giving the corresponding silylated cyclopentane in 98% yield as a single diastereomer with 93% ee (Table 7, entry 15). 

From the results reviewed above, one might get the impression that the choice of an organic solvent that optimizes the enantioselectivity of the enzyme in a given resolution reaction is a matter of tedious trial and error, with little guidance from established rules or insights. In practice, however, one has to consider several mitigating circumstances. In many cases of interest, the choice will be limited to a relatively small number of solvents that are either industrially approved or readily available in the laboratory. Since most practical resolutions start from a racemic mixture obtained by chemical synthesis, batch-mode enrichment requiring relatively modest 5-values will be an attractive method. In that case, solubility and easy... 

Major advances have been made in recent years in the development and optimization of chiral resolutions of derivatized CD-based CSPs [68]. It has been reported that CSPs based on CD derivatives were more enantioselective than CSPs obtained from native CDs [68]. An acetyl /CCD column exhibited enhanced separation for scopolamine in comparison to the native / -CD CSP in the reversed-phase mode [69]. The enantiomeric resolution of some drugs was compared on the native [l-CD and on CSPs based on (S)- and (i )-2-hydroxy-propyl /I-CD, and the best resolution was reported on the derivatized CSPs [44]. Five types of natural and chemically modified [>- or y-C D stationary phases were developed and used for the chiral resolution of dansyl amino acids. The best resolution of dansyl amino acids was provided by y-CD CSPs [70]. 

By screening solvent and inorganic bases to establish the optimal reaction conditions for dimeric chiral PTCs, a toluenexhloroform (7 3, v/v) solvent system and a 50% aqueous KOH base were found to afford the best enantioselectivity and chemical yield within a reasonable reaction time. As dimeric cinchona-PTCs are very poorly soluble in toluene (one of the popular solvents in asymmetric alkylation), this might act as an obstacle for the catalyst to show its maximum ability. However, the addition of chloroform to toluene provided better results due to an improved solubility of the dimeric PTC. This difference in ability to dissolve the dimeric PTC might be heavily associated not only with the reaction rate but also with the chemical/ optical yield. However, the use of chloroform alone proved to be inadequate as an optimal solvent [10]. 

Arai et al. also reported another BINOL-derived two-center phase-transfer catalyst 31 for an asymmetric Michael reaction (Scheme 6.11) [8b]. Based on the fact that BINOL and its derivatives are versatile chiral catalysts, and that bis-ammonium salts are expected to accelerate the reaction due to the two reaction sites - thus preventing an undesired reaction pathway - catalyst 31 was designed and synthesized from the di-MOM ether of (S)-BINOL in six steps. After optimization of the reaction conditions, the use of 1 mol% of catalyst 31a promoted the asymmetric Michael reaction of glycine Schiff base 8 to various Michael acceptors, with up to 75% ee. When catalyst 31b or 31c was used as a catalyst, a lower chemical yield and selectivity were obtained, indicating the importance of the interaction between tt-electrons of the aromatic rings in the catalyst and substrate. In addition, the amine moiety in catalyst 31 had an important role in enantioselectivity (34d and 34e lower yield and selectivity), while catalyst 31a gave the best results. 

An optimized version of the enantioselective SMB-GC unit was subsequently presented for enflurane enantiomers (chemical structure cf. insert in Figure 24) (Biressi et al., 2002b). It consisted of eight 80 cm x 15 mm (i.d.) stainless steel columns assembled in a home-made SMB-GC unit operated at 35°C (Scheme, cf. Figure 24). Each column with an adsorption bed volume of 140 ml each contained 20 % unpurified Lipodex E in the polysiloxane SE-54 and coated (17 %, w/w) on Chromosorb A (NAW, 20-30 mesh) 0.6 mm). This set-up represented the first gas-chromatographic SMB-GC unit for the preparative-scale separation of enantiomers. 

Quantum chemical DFT calculations at the B3LYP/6-31G(d) level have been used to study the enantioselective lithiation/deprotonation of O -alkyl and O-alk-2-enyl carbamates in the presence of (—)-sparteine and (—)-(f )-isosparteine.7 Complete geometry optimization of the precomplexes consisting of the carbamate, the chiral ligand, and the base (/-PrLi), for the transition states of the proton-transfer reaction, and for the resulting lithio carbamates have been performed in order to quantify activation barriers and reaction energies. 

The production of 10-methylcarbapenem (184), which has antibacterial activities and enhanced chemical and metabolic stability, has been reported by asymmetric hydroformylation of 4-vinyl-0-lactams 185 catalyzed by Rh-BINAPHOS complexes (Scheme 12.75). Under optimized conditions, the observed regioselectivity was 55/45 (b/1), enantioselectivity was 93/7 (1860 18600 at 95% conversion, and S/C = 1000.233... 

A similar system was studied a few years later by the S chaus group [89], who compared several binaphthol-derived chiral Bronsted acids such as 92a and 94a-d in the triethylphosphine-mediated MBH reaction between cyclohexenone and aldehydes. Optimized conditions were found with 2-20 mol% of chiral Bronsted acid and an excess of triethylphosphine (200 mol%) as the nucleophilic promoter at 0-10 °C in THF. Using PMe3 or P(n-Bu)3 in the reaction afforded 76 in yields similar to that of PEt3, but in lower enantioselectivity (50% and 64% ee, respectively). The use of only (R)-BINOL in the MBH reaction of dihydrocinnamaldehyde 74 and cyclohexenone 75 resulted in the formation of 76 in 16% ee. Partially saturated BINOL derivatives such as 94a-d offered high chemical yield and enantio selectivity (Scheme 5.19) [91]. Optimal results with the addition of aliphatic al-... 

The enantioselective syntheses of (R)-oc-cuparenone and (S)-a-cuparenone, both of which are natural products from different sources, were also completed using the solid-state photodecarbonylation of diasteromerically pure difluorodioxaborinane ketones 192 and 194 (Scheme 2.47). The latter were prepared in two steps from 191, and irradiated as nanocrystalline suspensions to optimize the chemical yields of the transformation. The photoreaction of the optically pure ketones was 100% stereoselective with an isolated yield of 80%. The two natural products were obtained by simple acid removal of the chiral auxiliary. 

The enantioselectivity of a biocatalytic resolution or asymmetrization is primarily dependent on the enzyme and the structure of the substrate. Both of these can be changed in order to optimize the selectivity. The enzyme can be optimized by molecular genetic methods, while the substrate can be modified by organic chemical synthesis. These ways of optimizing selectivity will not be discussed in this chapter. 

These enzymes must have a large activity spectrum, and ideally enantioselectivity for toxic stereoisomers. Their mass production under GMP conditions must be realizable at a reasonable cost. Long-term storage without activity loss (in solution, lyophilized, or adsorbed/bound on a matrix) must be possible under field conditions. Conformational stability can be optimized by chemical modification or addition of stabihzers such as polyols. Thermostable enzymes from thermophilic bacteria (Merone et al., 2005) or mutated/ evolved highly stable enzymes from mesophilic bacteria (Elias et al., 2008) are promising alternatives. 

---