Enantioselectivity factors

Unprotected racemic amines can be resolved by enantioselective acylations with activated esters (110,111). This approach is based on the discovery that enantioselectivity of some enzymes strongly depends on the nature of the reaction medium. For example, the enantioselectivity factor (defined as the ratio of the initial rates for (3)- and (R)-isomers) of subtiHsin in the acylation of CX-methyl-ben zyl amine with tritiuoroethyl butyrate varies from 0.95 in toluene to 7.7 in 3-methyl-3-pentanol (110). The latter solvent has been used for enantioselective resolutions of a number of racemic amines (110). 

The enantioselectivity factors in Table 10 indicate that amino acid samples with different enantiomeric excess (ee) should show differences in the R = (1 + I5)/Iref term. Indeed, In Rchkai is linearly related to optical purity of the specimen and the relevant calibration curve can be established. Accordingly, it is possible to rapidly determine ee of an amino acid sample by a single measurement of Rchirai in a tandem... 

It becomes obvious from the clustered data points that the binding constant for the 7 -enantiomer is too small to be accurately determined by this method. Hence, indirect affinity CE (resolution method) was utilized to determine the binding constant for the 7 -enantiomer. Indirect affinity CE makes use of the knowledge of the constant for one enantiomer (here, A s) and in addition of experimental separation data as obtained with the racemate of DNB-Leu in presence of the tBuCQN selector as BGE additive. By use of Equation 1.10, an enantioselectivity factor may be defined as the ratio of the binding constants of S- and 7 -enantiomers yielding the following equation ... 

In simple experiments, particulate silica-supported CSPs having various cin-chonan carbamate selectors immobilized to the surface were employed in an enantioselective liquid-solid batch extraction process for the enantioselective enrichment of the weak binding enantiomer of amino acid derivatives in the liquid phase (methanol-0.1M ammonium acetate buffer pH 6) and the stronger binding enantiomer in the solid phase [64]. For example, when a CSP with the 6>-9-(tcrt-butylcarbamoyl)-6 -neopentoxy-cinchonidine selector was employed at an about 10-fold molar excess as related to the DNB-Leu selectand which was dissolved as a racemate in the liquid phase specified earlier, an enantiomeric excess of 89% could be measured in the supernatant after a single extraction step (i.e., a single equilibration step). This corresponds to an enantioselectivity factor of 17.7 (a-value in HPLC amounted to 31.7). Such a batch extraction method could serve as enrichment technique in hybrid processes such as in combination with, for example, crystallization. In the presented study, it was however used for screening of the enantiomer separation power of a series of CSPs. 

Moderate enantioselectivity factors have also been found for electron transfer reactions between HRP or GO and resolved octahedral ruthenium or osmium complexes, respectively. In particular, the rate constants for the oxidation of GO(red) by electrochemically generated and enantiomers of [Os(4,4 - 2 ) ]3 + equal 1.68 x 106 and 2.34 x 106 M-1 s-1, respectively (25 °C, pH 7) (41). The spectral kinetic study of the HRP-catalyzed oxidation of and A isomers of the cyclo-ruthenated complex [Ru(phpy)(phen)2]PF6 (Pig. 21) by hydrogen peroxide has revealed similarities with the oxidation of planar chiral 2-methylferrocene carboxlic acid (211). In both cases the stereoseleci-vity factor is pH dependent and the highest factors are not observed at the highest rates. The kA/kA ratio for [Ru(phpy)(phen)2]PF6 is close to 1 at pH 5-6.5 but increases to 2.5 at pH around 8 (211). 

The calibration curves for the pure R- and S-enantiomers of halodiether B with SPR are shown in Fig. 10. A significant and reproducible difference upon exposure to the optical antipodes was observed as the response of the chiral sensors to the chiral compounds. Enantioselectivity factors a can be determined by dividing the respective signal heights. For a concentration of 20 xgl 1, the a value for SPR is 9.6 ( 0.7). The a values decrease with increasing concentrations, because fewer complexation sites are available for guest molecules. 

Imprinted Ti02-gel films were prepared in a similar way by using L- or D-amino acid derivatives (Cbz-Ala, Cbz-Leu, Cbz-Phe) as templates. The template molecule is removed by treating with 0.1 wt % aqueous ammonia. The rebinding data for the l- and D-enantiomers in these imprinted films are shown in Table 6.5. The enantioselectivity factor (a) in each imprintedfilm is a ratio of... 

In the Cbz-Ala imprinted films, the frequency change between the template and its enantiomer was 5 or less, and the enantioselectivity factor was about 1.1 in either of the l- and D-imprinted films. In contrast, the template molecule was bound much better than its enantiomer in the Cbz-Leu- or Cbz-Phe-imprinted... 

The first separations of enantiomers in GC on cyclodextrin modified column were carried out by Sybilska et al. in 1983 [5], They applied a formamide solution of a-cyclodextrin as a stationary phase in the classical packed column. The column allowed an efficient separation of chiral monoterpenes - a- and P-pinenes into enantiomers. This system of using CDs in GC is characterised by obtaining high enantioselectivity factors, so enantioseparation is still possible for receiving not very efficient packed columns. Unfortunately, the columns appeared to be not very stable at higher temperatures. 

Because the molecular basis of enantioselectivity is poorly understood, directed evolution seems to be an excellent choice for engineering enantioselective biocatalysts. Several impressive examples have been documented. In a classical study, Reetz and coworkers used error-prone PCR coupled with a 96-well plate based colorimetric screening method to increase the enantioselectivity of a Pseudomonas aeruginosa lipase toward 2-methyldecanoate. After several rounds of directed evolution, the enantioselectivity of the lipase increased from E = 1.04 (2% enantiomeric excess) to E = 25 (90-93% enantiomeric excess, ee) (E is the enantioselectivity factor). Using a similar approach. 

The ratio of Eq. (22) to Eq. (23) provides a direct differential equation between epi-chlorohydrin enantiomer concentrations and the enantioselectivity factor, which is a function of temperature and catalyst concentration as variables (Eq. 25). This equation is readily solved to give the mathematical relation between enantiomeric excess and yield (R) as a function of a (Eq. 26). The maximum obtainable yield for a given ee specification (and vice versa) is provided straightforwardly in Eq. (26). Moreover, this equation shows that the yield-ee pair is independent of the amount of water used (provided the amount is sufficient to reach the desired ee) and of the water addition rate. Thus, the water addition mode will only impact the reaction time necessary to reach the desired yield and ee target. 

A more thorough analysis of the enantioselectivity of several commercially available alkaline proteases was performed by determining the enantioselectivity fador E of these enzymes in the hydrolysis of (R,S)-4. In order to determine the enantioselectivity factor E accurately (the kinetic ratios for the conversion of the two enantiomers in a first-order kinetic resolution) [11], accurate measurement of the optical purity of the product acid and unreacted ester is necessary. Separation of the carboxylic acid enantiomers by gas chromatography was not possible due to decomposition of the acid on the column, so a derivatization method was developed to convert the acid into the corresponding methyl ester. Treatment of the extracted acid and ester with trimethylsilyldiazomethane resulted in the conversion of the acid into the methyl... 

The low efficiency and short migration distances typical of thin-layer chromatography limit useful separations to those with relatively large enantioselectivity factors. Absorption by the chiral selector can cause baseline instability and reduced sample detectability for quantitative measurements by scanning densitometry. The chiral sepa-... 

Moderate enantioselectivity factors have also been found for electron transfer reactions between HRP or GO and resolved octahedral ruthenium or osmium complexes, respectively. In particular, the rate constants for the oxidation of GO(red) by electrochemically generated A and A enantiomers of [Os(4,4 -Me2bpy)3] equal 1.68 x 10 and 2.34 X 10 respectively (25 °C, pH 7) 41). The spectral kinetic... 

The advantage of using van t Hoff plot of the logarithm of the enantioselectivity factor (In a) vs. reciprocal absolute temperature (1/7) is that it does not require the knowledge of phase ratio for determination A AH) and A(A5) values. In turn, it does not bring a solution of determination of the entropy of the solute transfer from the mobile phase to stationary phase (A5i). 

A statistical thermodynamic study of CSP-enantiomer interaction demonstrated that the possible enantioselectivity factor a was not significantly different when an interaction dominated the two others or when the three interactions were of comparable strength. However, in the former case. In a should be a linear function of 1/r, with r, the absolute temperature and a departure from this Van t Hoff behavior would suggest that multiple retention modes compete [17]. 

Table 4 allows for the computation of the theoretical enantioselectivity factor for any of the possible 1.6 million enantiomeric pairs built with four of the substituents. However, this value is for the -NEC-P-CD CSP with a hexane-IPA mobile phase only. For example, the dinitrobenzoyl derivative of propranolol has a chiral center bearing substituents 3 (H-), 36 (-0-C0-DNB), 77 (-CH2-0-(l-naphthyl), and 80 (-CH2-N(t-Bu)-CO-DNB. These four substituents contribute, respectively, for 0, 146, 20, and 101 cal/mol making a A(AG) of 267 cal/mol that would produce an enantioselectivity factor a of 1.58 [Eq. (3), with 586 cal/mol for the/ rproduct at 22 C or 295 K and exp(—267/586) = 1.58]. The experimental a was only 1.06. Table 4 shows at the bottom that the four substituents do not interact fiilly independently. Of course, two identical substituents make the chirality disappears, so the A(AG) energy is necessarily nil a = 0). It was found that two DNB derivatives were detrimental for the enantioselectivity factor decreasing the A(AG) energy by 230 cal/mol (Table 4). Both the amine and alcohol group of propranolol were DNB derivatized, so the calculated 267 cal/mol for the four substituents must be decreased by 230 cal/mol giving a final A(AG) value of only 37 cal/mol and the corresponding enantioselectivity factor a = 1.065 very close to the experimental value.

For the five enantiomers smdied in the reversed-phase mode, two terms dominated the AeE and the AvV terms, the first being positive and the second being negative (Table 5, bottom). They almost cancel each other. The AeE term encodes interactions through polarizable n and jt electrons. The e coefficient has a minor importance in overall solute retention [47]. It has a major effect on enantioselectivity. The negative enantioselective contribution of the AvV term is likely an indication of steric repulsion. Since these two terms almost cancel each other for our five test solutes, it means that the dipolar, AsS, and especially H-bonding, AbB, terms will be mainly responsible for the experimentally observed enantioselectivity factor. 

Column 25 cm x 4.6 mm i.d. retention factor of the first enantiomer, a enantioselectivity factor, Rs resolution factor between enantiomers. Mobile phases RP = reversed phase, methanol/buffer pH 4.1 40/60 v/v PIM = polar ionic mode, methanol/acetonitrile 45/55 v/v with 0.1% acetic acid and 0.1% triethylamine. Data from [1, 2, 12-16]. 

Structurally to vancomycin (Fig. 1), shows promising efficacy toward native amino acid enantiomer separation (Table 2) [8], The eremomycin enantioselectivity factors of amino acids are comparable if not higher than those of teicoplanin (Table 2). 

Table 2 shows that the derivation of the amine group of amino acids can improve chiral recognition, e.g., the enantioselectivity factors of alanine are 1.8 and 2.7 on teicoplanin and TAG CSPs they become 13 and 3.6, respectively, upon derivati-zation of the amine group forming A-benzoyl alanine (Table 2). Enantioselectivity enhancement is very often obtained by N-derivatization of amino acids [14, 16]. Such enantiorecognition enhancement is not an absolute rule as shown by iV-benzoyl phenylalanine in Table 2. The phenylalanine enantioselectivity factors jumps from 1.5 (native form, Rs = 3.1) to 4.4 (N-benzoylated form, Rs = 11.4) on teicoplanin CSP. It decreases from 3.7 (native form, Rs = 13.7) down to 1.5 Rs = 2.6) after N-benzoylation on the TAG CSP. There is a clear steric effect due to the attached benzoyl group very beneficial for enantiorecognition on teicoplanin CSP and detrimental when the TAG CSP is used.

---