Synthesis of Chiral Building Blocks for Use in Drug Discovery

In the past decade there has been a significant growth in the sales of pharmaceutical drugs worldwide, but more importantly there has been a dramatic growth in the sales of single enantiomer drugs. The pharmaceutical industry has a rising demand for chiral intermediates and research reagents because of the continuing imperative to improve drug efficacy. This in turn impacts on researchers involved in preclinical discovery work. Besides traditional chiral pool and resolution of racemates as sources of chiral building blocks, many new synthetic methods including a great variety of catalytic reactions have been developed which facilitate the production of complex chiral drug candidates for clinical trials. The most ambitious technique is to synthesise homochiral compounds from non-chiral starting materials using chiral metal catalysts and related chemistry. Examples of the synthesis of chiral building blocks from achiral materials utilizing asymmetric hydrogenation and asymmetric epoxidation are presented. Molecules 2004, 9 406


Introduction
Chiral chemical synthesis plays an important and growing role in the development of new drugs, agrochemicals and building blocks.This paper focusses on asymmetric synthesis with chiral catalysts to provide novel chiral amino acids and amino alcohols.Asymmetric hydrogenation with chiral DuPHOS Rh-catalyst to provide novel chiral α-amino acids and β-amino alcohols 1 and Noyori's chiral Ru-catalyst for synthesis of chiral amines 2 via chiral alcohols is discussed.Asymmetric epoxidation employing Jacobsen's Mn catalyst to afford α-amino alcohols 3 is also discussed.

Results and Discussion
A generic synthesis to access large numbers of diverse chiral α-amino acids 4 and β-amino alcohols 1 has been developed.There are many known methods of preparing the dehydrophenylalanine intermediate 5 from commercial starting materials (Scheme 1).

Scheme 1
This, coupled with selective reduction of the dehydrophenylalanine intermediate 5 using an asymmetric hydrogenation catalyst, such as DuPHOS, provides a very powerful route.This reaction synthesis is high yielding overall and provides the final chiral 2-amino-3-phenylpropanol building blocks 1 in high enantiomeric excess.
The preparation of the dehydro amino acid derivatives was accessed by several routes depending on the analogue and available starting materials.Dehydrophenylalanine intermediates 5a and 5b (Figure 2) were prepared from either an Erlenmeyer reaction with benzaldehydes and N-acetyl glycine or a Horner-Emmons reaction with Cbz-glycine phosphonate respectively.Alternatively, a Heck reaction between an aryl iodide (where the aldehyde is unavailable) and an acrylate also provides amidoacrylate 5a.The dehydrophenylalanine intermediate 5 was subjected to selective reduction with a chiral DuPHOS Rh-catalyst commercially available from Strem.The advantage of this catalyst is that it is predictable.The relative stereochemistry of the product is determined by the chiral catalyst ie [(COD)Rh((S,S)-Et-DuPHOS)] + will always give the (S)-enantiomer and [(COD)Rh((R,R)-Et-DuPHOS)] + gives the corresponding (R)-enantiomer [1,2].In addition, the double bond geometry of the starting alkene, whether it be E or Z gives the same absolute configuration in high enantiomeric excess.
Chiral 2-amino-3-phenylpropanol building blocks 1 were then prepared following several functional group manipulations.Confirmation of optical purity (stereochemistry obligated by the catalyst employed) was determined by preparation of either the Mosher amide or Mosher ester derivative.The diastereomeric excess (de) of the Mosher amide is representative of the enantiomeric excess (ee) of the precusor.The epimeric compounds were prepared in a similar manner in most cases.In general, 19 F-NMR integrals indicated >98% ee.In some cases the minor isomer was not observed at all.The absolute configuration of 2-substituted-1-propanols can be assigned by the 1 H-NMR splitting patterns of the Mosher's ester [3].Reverse phase HPLC was also used to distinguish between Mosher diastereoisomers.
In the first example, the 2,4-difluoro analogue was best accessed via the Erlenmeyer reaction of benzaldehyde 6 with N-acetyl glycine.The intermediate azlactone 7 was subsequently ring opened by reflux in acetone.Asymmetric hydrogenation of acrylic acid 8 with (S,S)-DuPHOS catalyst at 30 psi then installed the desired S stereochemistry in acetylamino acid 9.The next step in the synthesis was esterification of acid 9 with methanolic hydrogen chloride.It was observed here, that with longer reaction times the acetyl group was also cleaved.Thus, the installation and removal of protecting groups was achieved in a one pot synthesis in high yield.Reduction of (S)-amino ester 10 to (S)-amino alcohol 11 was achieved with lithium aluminium hydride.Preparation of the Mosher amide of 11 and analysis by 19 F-NMR (ratio 97:0.065) and HPLC indicated >98% ee.

Scheme 2
In a second example, Horner-Emmons reaction of 2,4-dimethylbenzaldehyde 12 with N-Cbzglycine phosphonate gave amidoacrylate 13.Catalytic hydrogenation with (S,S)-DuPHOS provided the (S)-amido ester 14.Hydrogenation of Cbz-amide 14 with Pd/C followed by reduction of ester 15 with lithium aluminium hydride afforded the desired (S)-amino alcohol 16.LCMS and reverse phase HPLC of the (R)-Mosher amide derivative estimate the enantiomeric excess to be >98%, as the alternate (R,R)-diastereoisomer was undetectable.In a third example, 2-fluoro-4-methylaniline (17) was used as a starting reagent where 2-fluoro-4methylbenzaldehyde was not commercially available.The Heck reaction with the diazotized aniline and methyl acrylate works well, but with methyl 2-acetamidoacrylate this reaction fails completely.Aniline 17 was thus converted to aryl iodide 18 via the Sandmeyer reaction followed by Heck reaction with methyl-2-acetamidoacrylate and Pd-acetate to form amidoacrylate 19 [4].The reaction was done neat and gave the acrylic ester as a black tar which was purified by chromatography.Catalytic hydrogenation with (S,S)-DuPHOS provided (S)-amido ester 20.Reduction of ester 20 with lithium aluminium hydride gave an acetamido alcohol intermediate.Derivatisation of this intermediate as the (R)-Mosher ester indicated, from the 1 H-NMR, that we had obtained 2S stereochemistry [2,3] Next we examined the synthesis of chiral amines via chiral alcohols using Noyori's chiral Rucatalyst.The Noyori asymmetric hydrogenation catalyst is one of a family of catalysts consisting of a ruthenium based system with a phosphine and 1,2-diamine ligand.They provide highly efficient enantio-and diastereoselective hydrogenation of simple ketones.Asymmetric hydrogenation of 2substituted cyclohexanone with the Noyori catalyst favors the cis isomer and provides the cyclohexanol in high ee.This is attributed to the fact that from a racemic mixture, the 2R substituted cyclohexanone undergoes hydrogenation faster and the slower 2S substrate undergoes epimerisation under basic conditions faster than it is hydrogenated [5].

80%
The hydrogenation of 2-methylcyclohexanone 22 with Noyori's ruthenium catalyst, (S)-Binap-(+)-(1R, 2R)-diphenylethylenediamine, in the presence of potassium hydroxide at 65 psi affords (1R, 2S)-2-methyl-1-cyclohexanol 23 as a 40:1 mixture of cis and trans isomers in high yield.Analysis of the Mosher ester of 23 by 19 F-NMR indicated 70% ee.Importantly, the C1 stereochemistry was inverted by Mitsonobu reaction to afford the corresponding azide 24 in 100% trans form.Reduction of azide 24 with PPh 3 gave the desired (1S, 2S)-methyl cyclohexylamine 25 in 70% ee as confirmed by the Mosher amide of 25.Thus, the enantiomeric excess was maintained throughout the synthesis.This was then further enhanced by resolution with tartaric acid.Amine 25 was resolved by recrystallisation with (+)tartaric acid from methanol to give the hydrogen tartrate salt in 92% ee.The low yields in the synthesis can be attributed to the volatility of the products compounded by the poor UV absorption characteristics of these molecules which made the synthesis very difficult.Due to the high volatility of the free-base 25, the compound was prepared as the hydrogen tartrate salt.When evaporating solutions of the free-base 25, solvent should be removed at pressures greater than 600 mbar at 60 °C in order to minimise losses of the free-base through evaporation.

MeOH
The absolute stereochemistry was confirmed by X-ray analysis of the L-(+)-tartaric acid salt (R, R-(+)-tartaric acid).Finally, the synthesis of (1R, 2S)-cis-amino alcohol 30 is discussed.α-Amino alcohols can be prepared via asymmetric epoxidation employing Jacobsen's catalyst.This chiral manganese epoxidation catalyst gives high ee's with unactivated alkenes.Single enantiomer amino alcohols are then efficiently produced from the corresponding chiral epoxide by acid or base catalysed epoxide ring openings.The synthesis of aminohydroxyindanes via epoxidation of indenes has been reported in the literature [7].Sodium borohydride reduction of indanone 26 to indanol 27 followed by acid catalysed dehydration gave indene 28 in 50% yield after distillation.Chiral epoxidation of indene 28 with (R,R)-Jacobsen's catalyst, 4-phenylpyridine N-oxide and sodium hypochlorite provides the (1R, 2S)-epoxide 29.Epoxide 29 then underwent a Ritter reaction with acetonitrile under acidic conditions via a cyclic intermediate which was then hydrolysed to afford the (1R, 2S)-cis-amino alcohol 30 and subsequently converted to the hydrochloride salt 31.The enantiomeric excess was determined to be >98% by LCMS and 19 F-NMR analysis of the (S)-Mosher amide derivative of 30.

Conclusions
The demand for chiral intermediates and chiral active ingredients is predicted to grow.Chiral catalysis has proven an effective means of preparing key chiral intermediates in high ee.Ready access to these and more advanced intermediates and chiral drugs targets accelerates our collaborative drug discovery programs.

General
Reverse phase HPLC analysis was carried out using a Column Engineering Monitor C18 (5 µm), 50 x 4.6 mm.LCMS were run on a Perkin-Elmer Sciex API-100 instrument.ESMS were run on an API III LC/MS/MS from Perkin Elmer/Sciex using an electrospray inlet. 1 H-, 19 F-and 13 C-NMR spectra were acquired on a 400 MHz Varian UNITY INOVA spectrometer and were recorded at 400 MHz, 377 MHz and 100 MHz respectively.X-ray crystallography was carried out on a Bruker Smart Apex X-ray diffractometer.The images were generated using Mercury 1.1 from .pdbfiles.Chiral catalysts (DuPHOS, Noyori and Jacobsen's) were purchased from Strem.Flash chromatography was conducted on Merck Silica gel type 9385.

Synthesis of 2-acetylamino-3-(2,4-difluorophenyl)acrylic acid (8).
A solution of azlactone 7 (21.8g, 97.68 mmol) in acetone (210 mL) and water (82 mL) was refluxed for 4 h.The reaction mixture was then cooled to room temperature and solvent removed under reduced pressure to yield a yellow solid.The crude product was dissolved in water (100 mL) and adjusted to pH 10 with potassium carbonate then the aqueous layer was extracted with ethyl acetate (3 x 100 mL).The organic layer was discarded and the aqueous layer adjusted to pH 3-4 with citric acid then extracted with ethyl acetate (5 x 100 mL).The organic layer was collected then dried over sodium sulfate and the solvent removed under reduced pressure to afford as a yellow powder (23.5 g, 88% yield).This crude material was then purified by recrystalization using boiled water to give acrylic acid 8 as a yellow solid (17.79 g, 75.6% yield).HPLC (214 nm), t R 4.76 min.(92.7%); 1 H-NMR (CDCl 3 ) δ: 2.41 (s, 3H), 3.49 (s, 1H), 6.85-6.89(m,1H),

Synthesis of 2-benzyloxycarbonylamino-3-(2,4-dimethylphenyl)acrylic acid methyl ester (13).
To a solution of N-(benzyloxycarbonyl)-α-phosphonoglycine trimethyl ester (14.574 g, 44 mmol, 1.1 eq) in dichloromethane (80 mL) was added DBU (6.28 mL, 42 mmol, 1.05 eq).The reaction mixture was stirred for 0.5 h then 2,4-dimethylbenzaldehyde (12, 6.2 mL, 40 mmol, 1 eq) was added.The reaction was exothermic after addition of aldehyde 12.The reaction mixture was stirred at room temperature over the weekend.To this mixture was added aqueous 1N H 2 SO 4 followed by separation of the dichloromethane layer.The aqueous phase was further extracted with an aliquot of ethyl acetate and the combined organic layers were dried over sodium sulfate, filtered and concentrated to give an oil.The crude product was purified by flash chromatography over silica gel with 20% ethyl acetate/ petroleum ether as eluent followed by 50% dichloromethane/petroleum ether.The fractions containing the major band were combined and evaporated to dryness to provide amidoacrylate 13 as a colourless solid (12.19 g, 89% yield).HPLC (214nm), t R = 9. 95
In a 500 mL 3 necked round bottom flask fitted with a thermometer and flushed with nitrogen, lithium aluminium hydride (1.71 g, 45.1 mmol, 4eq) was suspended in anhydrous tetrahydrofuran (90 mL) and cooled in an ice bath to 0 °C.(S)-Amino ester 15 (2.75 g, 11.28 mmol) was solubilised in anhydrous tetrahydrofuran (140 mL) then slowly cannulated into the reaction mixture whilst maintaining temperature < 3°C.The reaction mixture was allowed to warm to room temperature, stirred for 1 h, then recooled to 0 °C and very cautiously diluted with water (15 mL), 15% sodium hydroxide (15 mL) and water (15 mL).The reaction mixture was allowed to stir for a further 0.5 h at 0°C then the aluminium salts were filtered off and the filtrate concentrated.The residue was taken up in ethyl acetate, washed with brine, extracted with ethyl acetate (x 3) and the combined organic extracts dried over sodium sulfate, filtered and concentrated.The crude material was diluted with dichloromethane, washed with 1N hydrochloric acid (x 5) and the combined aqueous layers were reextracted with dichloromethane then basified and re-extracted with dichloromethane (x 5).The organic layer was dried over sodium sulfate, filtered and concentrated to afford a white solid (2.09 g) which was then taken up in 3 N hydrochloric acid and lyophilised to provide (S)-amino alcohol 16 as a pale yellow solid (1.99 g, 81% yield).Amine 15 (42 mg, 0.2 mmol) was dissoved in tetrahydrofuran (2 mL) under an atmosphere of N 2 , DIEA (104 mL, 0.6 mmol, 3 eq), (R)-(+)-MTPA (47 mg, 0.2 mmol), EDC (38 mg, 0.2 mmol) and HOBt.H 2 O (31 mg, 0.2 mmol) were added and the reaction stirred overnight.The reaction mixture was diluted with water and extracted with Et 2 O (2 x 25 mL).The combined organic layers were washed with brine (10 mL) then dried over anhydrous sodium sulfate, filtered and concentrated to afford the (R)-Moshers amide ester of 15 (26 mg, 30% yield) as a colourless oil. 1 H-NMR, LCMS and reverse phase HPLC estimate the enantiomeric excess to be > 98% as the alternate (R, R)diastereoisomer was undetectable.HPLC (214nm), t R = 5.76 (15, 13.8%), 10.66 (

Synthesis of 2-fluoro-1-iodo-4-methyl benzene (18) via the Sandmeyer reaction.
In a round bottom flask, 2-fluoro-4-methylaniline (17, 1 g, 7.99 mmol) was suspended in water (2 mL) and concentrated hydrochloric acid (2 mL).This solution was then cooled in an ice bath with vigorous stirring.To this stirring solution was added sodium nitrite (662 mg, 9.58 mmol) dissolved in water (2 mL) dropwise over 0.5 h, keeping the temperature below 10 °C.The reaction was then stirred for a further 0.5 h.The resulting solution was then added dropwise to a solution of potassium iodide (1.99 g, 11.98 mmol) dissolved in water (2 mL) stirring in an ice bath.The reaction was then refluxed for 2 h before being allowed to stir at room temperature over night.The reaction mixture was then taken up in ethyl acetate and washed with 3N hydrochloric acid and 1M sodium hydroxide containing a small portion of sodium metabisulfite.The organic layer was then dried over sodium sulfate and the solvent removed under reduced pressure to afford a dark brown oil (1.46 g, 77% yield).On scale-up of this reaction the crude yield increased to 93%.This material was then purified via flash chromatography using a hexane running solvent and washed with 1M hydrochloric acid (x 2), 2M sodium hydroxide (x 1), brine (x 1) and dried over sodium sulfate to recover aryl iodide 18 (829 mg, 44% yield) as a colourless oil.HPLC (214 nm), t R =10.13 min.(76%); 1 H-NMR (CDCl 3 ) δ: 2.29 (s, 3H), 6.68 -7.55 (m, 3H).
Determination of the enantiomeric excess of 25: preparation of the Mosher amide.
. Deacetylation of the acetamido alcohol intermediate gave the resulting (S)-amino alcohol 21.Analysis of the derived (R)-Mosher amide of 21 by LC/LCMS confirmed >98% ee.