Towards Chemoenzymatic Syntheses of Both Enantiomers of Phosphoemeriamine

An enzyme-promoted addition of nitromethane to the appropriate phosphorylated imine (aza-Henry reaction) intended to be used in the synthesis of the title phosphoemeriamine, a phospha-analog of emeriamine (aminocarnitine), failed due to the tautomerization of the imine to the corresponding enamine. Nevertheless, both enantiomers of phosphoemeriamine were synthesized in high yield and enantiomeric purity using another chemoenzymatic approach, starting with a crucial step involving a CAL-B-mediated acetylation of the appropriate racemic precursor—diethyl 2-amino-3-dimethylaminopropylphosphonate—under kinetic resolution conditions. The enzymatic reaction was very efficient and provided each enantiomeric product in acceptable yield and with enantiomeric excess of 91 and 92%. The following appropriate chemical transformations led to the desired enantiomers of phosphoemeriamine in the form of phosphoemeriamine sesquichloride with enantiomeric excess up to 90%.


Introduction
Continuing our interest in the enzyme catalytic promiscuity [1], i.e., the ability of a single active site of the enzyme to catalyze more than one reaction, particularly those which are different from that designed in studies published in Nature [2][3][4][5][6], we have recently reported on our investigations of the enzyme-promoted addition of nitromethane to aldimines (i.e., the aza-Henry reaction).We succeeded in obtaining, for the first time, the desired products of this type of addition 3 in the yields of 14-81%.The most efficient enzymes turned out to be lipase TL from Pseudomonas stutzeri and oxynitrilase from Arabidopsis thaliana.We have expected that the compounds 3 will be produced with high enantiomeric excess.However, much to our disappointment, all the reactions investigated turned out to be non-stereoselective (Scheme 1) [7].

Introduction
Continuing our interest in the enzyme catalytic promiscuity [1], i.e., the ability of a single active site of the enzyme to catalyze more than one reaction, particularly those which are different from that designed in studies published in Nature [2][3][4][5][6], we have recently reported on our investigations of the enzyme-promoted addition of nitromethane to aldimines (i.e., the aza-Henry reaction).We succeeded in obtaining, for the first time, the desired products of this type of addition 3 in the yields of 14-81%.The most efficient enzymes turned out to be lipase TL from Pseudomonas stutzeri and oxynitrilase from Arabidopsis thaliana.We have expected that the compounds 3 will be produced with high enantiomeric excess.However, much to our disappointment, all the reactions investigated turned out to be non-stereoselective (Scheme 1) [7].Although the above results have not been fully satisfactory, we have decided to apply the newly elaborated procedure in the synthesis of a particular diaminophosphonic acid, namely phosphoemeriamine 8, a phospha-analog of emeriamine (aminocarnitine).The latter is a natural amino acid showing interesting biological properties.It behaves as an inhibitor of fatty acid oxidation and acts as a hypoglycemic and antiketogenic agent [8][9][10][11].It should be stressed that both enantiomers of phosphoemeriamine have so far been only synthesized ones, using an enantiomeric sulfinyl group as a chiral auxiliary.This methodology was developed in our department and has proven to be very efficient [8][9][10][11].The synthetic strategy was based on a highly diastereoselective addition of the O,O-diethyl methylphosphonate carbanion to (S)-N-(p-toluenesulfinyl)cinnamaldimine, followed by isolation of the major diastereoisomeric β-amino adduct and its further conversion to the desired target through proper transformations.However, to obtain the opposite enantiomer, it would be necessary to apply the (R) enantiomer of the starting aldimine and to repeat the whole troublesome transformation procedures.To overcome these obstacles, we have decided to develop an environmentally friendly, chemoenzymatic methodology, which would use the approach shown above.

Results and Discussion
Our plan to accomplish the synthesis of phosphoemeriamine based on the aza-Henry reaction is shown in Scheme 2. Thus, N-p-toluenesulfonyl-2-diethoxyphosphorylethanimine 4 was planned to be treated with nitromethane in the presence of various enzymes (which were reported efficient in the cited publication [7]).
inhibitor of fatty acid oxidation and acts as a hypoglycemic and antiketogenic agent [8][9][10][11].It should be stressed that both enantiomers of phosphoemeriamine have so far been only synthesized ones, using an enantiomeric sulfinyl group as a chiral auxiliary.This methodology was developed in our department and has proven to be very efficient [8][9][10][11].The synthetic strategy was based on a highly diastereoselective addition of the O,O-diethyl methylphosphonate carbanion to (S)-N-(p-toluenesulfinyl)cinnamaldimine, followed by isolation of the major diastereoisomeric β-amino adduct and its further conversion to the desired target through proper transformations.However, to obtain the opposite enantiomer, it would be necessary to apply the (R) enantiomer of the starting aldimine and to repeat the whole troublesome transformation procedures.To overcome these obstacles, we have decided to develop an environmentally friendly, chemoenzymatic methodology, which would use the approach shown above.

Results and Discussion
Our plan to accomplish the synthesis of phosphoemeriamine based on the aza-Henry reaction is shown in Scheme 2. Thus, N-p-toluenesulfonyl-2-diethoxyphosphorylethanimine 4 was planned to be treated with nitromethane in the presence of various enzymes (which were reported efficient in the cited publication [7]).
( First, we attempted to synthesize substrate imine 4 via condensation of α-diethoxy phosphorylacetaldehyde with p-toluenesulfonamide, according to a procedure in the literature [12].However, this derivative turned out to be useless in the envisaged transformation since it easily underwent tautomerization to the corresponding enamine 4a, thus making the appropriate addition impossible (Scheme 3).The same transformation happened with the N-toluenesulfinyl analog of 4 since it behaved in a similar way.In continuing our efforts, we decided to employ another chemoenzymatic approach.The latter is a recently and widely used application in asymmetric synthesis [13,14].One of the crucial parts of this approach would comprise a kinetic resolution of an appropriate Scheme 2. Envisaged synthesis of phosphoemeriamine 8 via the addition of nitromethane 2 to N-p-toluenesulfonyl-2-diethoxyphosphorylethanimine 4. * denotes chirality centre.
First, we attempted to synthesize substrate imine 4 via condensation of α-diethoxy phosphorylacetaldehyde with p-toluenesulfonamide, according to a procedure in the literature [12].However, this derivative turned out to be useless in the envisaged transformation since it easily underwent tautomerization to the corresponding enamine 4a, thus making the appropriate addition impossible (Scheme 3).The same transformation happened with the N-toluenesulfinyl analog of 4 since it behaved in a similar way.inhibitor of fatty acid oxidation and acts as a hypoglycemic and antiketogenic agent [8][9][10][11].It should be stressed that both enantiomers of phosphoemeriamine have so far been only synthesized ones, using an enantiomeric sulfinyl group as a chiral auxiliary.This methodology was developed in our department and has proven to be very efficient [8][9][10][11].
The synthetic strategy was based on a highly diastereoselective addition of the O,O-diethyl methylphosphonate carbanion to (S)-N-(p-toluenesulfinyl)cinnamaldimine, followed by isolation of the major diastereoisomeric β-amino adduct and its further conversion to the desired target through proper transformations.However, to obtain the opposite enantiomer, it would be necessary to apply the (R) enantiomer of the starting aldimine and to repeat the whole troublesome transformation procedures.To overcome these obstacles, we have decided to develop an environmentally friendly, chemoenzymatic methodology, which would use the approach shown above.

Results and Discussion
Our plan to accomplish the synthesis of phosphoemeriamine based on the aza-Henry reaction is shown in Scheme 2. Thus, N-p-toluenesulfonyl-2-diethoxyphosphorylethanimine 4 was planned to be treated with nitromethane in the presence of various enzymes (which were reported efficient in the cited publication [7]).First, we attempted to synthesize substrate imine 4 via condensation of α-diethoxy phosphorylacetaldehyde with p-toluenesulfonamide, according to a procedure in the literature [12].However, this derivative turned out to be useless in the envisaged transformation since it easily underwent tautomerization to the corresponding enamine 4a, thus making the appropriate addition impossible (Scheme 3).The same transformation happened with the N-toluenesulfinyl analog of 4 since it behaved in a similar way.In continuing our efforts, we decided to employ another chemoenzymatic approach.The latter is a recently and widely used application in asymmetric synthesis [13,14].One of the crucial parts of this approach would comprise a kinetic resolution of an appropriate In continuing our efforts, we decided to employ another chemoenzymatic approach.The latter is a recently and widely used application in asymmetric synthesis [13,14].One of the crucial parts of this approach would comprise a kinetic resolution of an appropriate racemic precursor [15,16].Noteworthy, a similar approach was successfully applied in our laboratory earlier in the synthesis of enantiomers of phosphocarnitine [17].Thus, we chose racemic diethyl 2-amino-3-N,N-dimethylaminopropylphosphonate 12 as a substrate in the lipase-promoted acetylation under kinetic resolution conditions.It was synthesized by us starting from diethyl 2-hydroxy-3-N,N-dimethylaminopropylphosphonate 9 [18].The ensuing mesylation resulted in 10 and followed by its reaction with sodium azide gave diethyl 2-azido-3-N,N-dimethylaminopropylphosphonate 11.The latter was subjected to the Staudinger reaction with triphenylphosphine to furnish the desired racemic substrate 12 (Scheme 4).
The ensuing chemical transformations of each enantiomer of the acetamide 13 were as follows (Scheme 6).The subsequent methylation of the terminal tertiary amino group furnished the final phosphoemeriamine precursor, trimethyl (2-acetamido)-(3-diethoxyphosphoryl)-propylammonium iodide 14.It was subjected to hydrolysis, using the procedure described in a previous publication [11] to give the desired products 8.It must be stressed that the enantiomers of 8 were obtained in the form of the complexes in which two molecules of phosphoemeriamine were complexed with one additional chloride anion, indicated here as 8a, which is in accordance with the result described earlier [11].Hence, in one molecule of the product, prepared as above, the ratio between the phosphoemeriamine cation and the chloride anion is 1:1.5;thus, the final structure may be described as phosphoemeriamine sesquichloride 8a.In this way, each enantiomer of phosphoemeriamine sesquichloride 8a was obtained and characterized and their absolute configurations were ascribed by comparing the sign of their optical rotation with that described previously (for the (−)-(R) enantiomer of >99% ee, [α] D = − 9.2) [11], although a partial racemization was observed in our work, most probably at the stage of the acidic hydrolysis of 14.The determination of the absolute configuration of the final products enabled us to also determine absolute configurations of the intermediates 12, 13 and 14 since their transformations into phosphoemeriamine sesquichloride proceeded without involvement of the stereogenic center (Schemes 5 and 6).
(though obviously of the opposite sign) of optical rotation as (−)-13, which could be taken as proof of the efficiency of the kinetic resolution (Scheme 5).The enantiomeric excess of the recovered amine (+)-12 was determined by 31 P NMR using optically pure (+)-(R)-tbutylphenylphosphinothioic acid as a chiral solvating agent [22,23].In turn, the enantiomeric excess of the acetylation product, acetamide 13, was determined on the basis of its optical rotation and via a chemical correlation shown in Scheme 5.
The ensuing chemical transformations of each enantiomer of the acetamide 13 were as follows (Scheme 6).The subsequent methylation of the terminal tertiary amino group furnished the final phosphoemeriamine precursor, trimethyl (2-acetamido)-(3-diethoxyphosphoryl)-propylammonium iodide 14.It was subjected to hydrolysis, using the procedure described in a previous publication [11] to give the desired products 8.It must be stressed that the enantiomers of 8 were obtained in the form of the complexes in which two molecules of phosphoemeriamine were complexed with one additional chloride anion, indicated here as 8a, which is in accordance with the result described earlier [11].Hence, in one molecule of the product, prepared as above, the ratio between the phosphoemeriamine cation and the chloride anion is 1:1.5;thus, the final structure may be described as phosphoemeriamine sesquichloride 8a.In this way, each enantiomer of phosphoemeriamine sesquichloride 8a was obtained and characterized and their absolute configurations were ascribed by comparing the sign of their optical rotation with that described previously (for the (−)-(R) enantiomer of >99% ee, [α]D = − 9.2) [11], although a partial racemization was observed in our work, most probably at the stage of the acidic hydrolysis of 14.The determination of the absolute configuration of the final products enabled us to also determine absolute configurations of the intermediates 12, 13 and 14 since their transformations into phosphoemeriamine sesquichloride proceeded without involvement of the stereogenic center (Schemes 5 and 6).Scheme 6. Final synthesis of each enantiomer of phosphoemeriamine sesquichloride 8a.

General Information
All solvents were dried and distilled prior to use.The starting materials and enzymes were purchased from Merck (Poznan, Poland), Sigma-Aldrich (Poznań Poland, TCI Chemicals (Trimen, Łódź, Poland) or Fluorochem (Hadfield, UK).The synthesized products were purified by column chromatography on a Merck 60 silica gel (0.063-0.200 mm) or preparative plate chromatography using a Merck 60 F254 silica gel plate (2.5 mm).TLC was performed on a Merck 60 F254 silica gel plate (0.25 mm).All the reactions were run in Scheme 6. Final synthesis of each enantiomer of phosphoemeriamine sesquichloride 8a.

General Information
All solvents were dried and distilled prior to use.The starting materials and enzymes were purchased from Merck (Poznan, Poland), Sigma-Aldrich (Pozna ń Poland, TCI Chemicals (Trimen, Łódź, Poland) or Fluorochem (Hadfield, UK).The synthesized products were purified by column chromatography on a Merck 60 silica gel (0.063-0.200 mm) or preparative plate chromatography using a Merck 60 F 254 silica gel plate (2.5 mm).TLC was performed on a Merck 60 F 254 silica gel plate (0.25 mm).All the reactions were run in duplicate.The NMR spectra were recorded in CDCl 3 or D 2 O using a Bruker AV III 500 spectrometer (Pozna ń, Poland) at 500 MHz ( 1 H), 126 MHz ( 13 C) and 202 MHz ( 31 P).Mass spectra, including HRMS, were measured on a Finnigan MAT instrument (Bremen, Ger-many) Optical rotations were measured on a Perkin-Elmer 241MC polarimeter (Uberlingen, Germany).

Conclusions
An attempt to synthesize enantiomeric phosphoemeriamine via the enzymepromoted addition of nitromethane to aldimines (aza-Henry reaction) failed due to the tautomerization of the appropriate phosphorylated imine to the corresponding enamine.However, another chemoenzymatic approach, in which the crucial step involved a CAL-B-mediated acetylation of the appropriate racemic precursor-diethyl 2-amino-3dimethylaminopropylphosphonate under kinetic resolution conditions-proved to be very efficient and provided each enantiomeric product (the starting amine and the acetamide formed) in acceptable yield and with enantiomeric excesses up to 92%.The following appropriate chemical transformations led to the desired enantiomers of phosphoemeriamine in the form of their sesquichlorides with ee up to 90%.
Copyright: © 2024 by the authors.Submitted for possible open access publication under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/license s/by/4.0/).