The development of an effective synthetic route of rilpivirine

Background Rilpivirine (RPV) was approved by the U.S. FDA (Food and Drug Administration) in 2011 to treat individuals infected with human immunodeficiency virus 1 (HIV-1). Significantly, rilpivirine is three fold more potent than etravirine. Once-daily, it is used with a low oral dose (25 mg/tablet), decreasing the drug administration and bringing a better choice to the patients. However, there are many shortcomings in the existing synthesis route of RPV, such as the high cost, prolonged reaction time and low yield (18.5%). Results This article describes our efforts to develop an efficient and practical microwave-promoted method to synthesize rilpivirine using less toxic organic reagents and low boiling solvents. The last step's reaction time decreased from 69 h to 90 min through this optimized synthetic procedure, and the overall yield improved from 18.5 to 21%. In addition, the yield of intermediate 3 increased from 52 to 62% compared to the original patent. Conclusion Overall, through a series of process optimization, we have developed a practical synthesis method of rilpivirine, which is easy to scale with higher yield and shorter reaction time.


Synthesis of 4-[(4-chloropyrimidin-2-yl)amino]benzonitrile (intermediate 3)
Scheme 3 represents the chemical pathway for the preparation of intermediate 3 (4-[(4-chloropyrimidin-2-yl) amino]benzonitrile). The starting materials 2-thioxo-2,3-dihydropyrimidin-4(1H)-one (10) and iodomethane were blended in the presence of NaOH base to attain compound 11. Further, the nucleophilic substitution of 11 with p-aminobenzonitrile (12) in DME afforded compound 13(iia). Finally, the target intermediate 3 was obtained by chlorinating the hydroxy group of 13 with POCl 3 under reflux conditions. This route's advantage is that the reaction step is simple and short, and the total yield is up to 47%, and the disadvantage is that the reaction produces methanethiol, which is highly toxic and smelly. Complete elimination of this odour is highly challenging, which may increase the cost of large-scale industrial production. In addition, the boiling point of DME used in the reaction process is too high [15]. So it has been reported that the step (ii) using the melting method also has a higher yield and shorter reaction time [16,17]. However, there is no report on the influence of temperature and reaction time on the yield, the optimal reaction temperature and reaction time cannot be determined.
Scheme 4 also illustrates the synthesis of intermediate 3, and the cyclization of starting material 1-(4-cyanophenyl)guanidine (14) with substituted diethyl malonate (15) in the presence of NMP and sodium acetate afforded compound 16 with pyrimidine moiety. The subsequent reaction of 16 with water, acetic acid, leads to 13, which further treated with POCl 3 afforded the desired intermediate 3 by "one-pot method". Although it is a simple reaction with an overall yield of 58.5%, the starting material (14) is expensive, limiting the large-scale industrial application of this route [18].
In Scheme 5, uracil's reaction with POCl 3 leads to the formation of 2,4-dichloropyrimidine (18), which was further treated with sodium methoxide in methanol afforded 19. The acid-catalyzed nucleophilic addition between 19 and 4-aminobenzpnitrile in the presence of 2-toluenesulphonic acid in dioxane gave 4-[4-methoxypyrimidin-2-yl) amino]benzonitrile (20). The target intermediate 3 was achieved by reacting 20 with pyridine, followed by POCl 3 under reflux condition. The reaction is relatively simple, but the process is too long and reducing the yields. Also, the highly toxic phosphorus oxychloride reagent is used repeatedly, which is harmful to the environment and public health. The entire route has a long reaction time and consumes a lot of energy, which is not suitable for industrial production [14].

Synthesis of rilpivirine
Synthesis of the target compound rilpivirine (1) was achieved by allowing the reaction between (2E)-3-(4amino-3,5-dimethylphenyl)prop-2-enenitrile hydrochloride (2) and 4-[(4-chloropyrimidin-2-yl)amino] benzonitrile (3) in acetonitrile under reflux condition for 69 h (Scheme 6, yield: 68.6%) [19]. This reaction's conditions are relatively simple, but the reaction time is longer, extending the industrial cycle production time and reducing the product's yield and quality. Also, it has been reported that NMP used as a solvent at 95 °C for 17 h to produce rilpivirine with 71.4% yield, wherein cis-isomer accounts for 0.7% [20]. However, it isn't easy to reclaim NMP in industrial production because of its higher boiling point.
There are many shortcomings in the synthetic strategy of rilpivirine. Notably, (a) intermediate 2 was mainly obtained by Heck reaction, and the catalyst palladium acetate and its ligands used in this reaction are expensive; (b) the starting material (14) used for the preparation of intermediate 3 is relatively expensive and the reaction temperature is too high; (c) when uracil is used as a starting material, the synthetic route in Scheme 5 is rather tedious, and the final reaction yield is too low; (d): the final step in the synthesis of rilpivirine, is too long (69 h) and causing energy consumption. Hence there is an urgent need to find more efficient and practical methods for synthesizing rilpivirine in the pharmaceutical industry. Herein, we represent our efforts to develop an efficient synthetic route with increased overall yield and reasonable reaction time.

Optimization of synthesis steps and process parameters of intermediate 13
The reaction for the synthesis of compound 13 was carried out through melting/fusion method with the temperature ranging from 160 °C to 200 °C. Herein, we investigated the effect of reaction temperature and reaction time on the yield of compound 13 (Table 1). During this process, the observations are: (1) when the reaction temperature was 160 °C, the reaction solution was completely solidified after 2 h, and the product yield was 64%; (2) when the temperature was 180 °C for 6 h, the product yield was 70%; (3) when the reaction temperature was 190 °C for 6 h, the product yield was 70.2%; (4) when the reaction temperature was 200 °C, no obvious increase in the yield was observed compared to the yield at 190 °C. To sum up, the reaction condition is selected to be 160 °C for 2 h, and then 180 °C for 4 h.

Synthesis optimization of rilpivirine (1)
Considering the longer reaction time of the final step in the previous reports, we attempt to apply the microwave-assisted synthesis method in the amination step to shorten the reaction time and to improve the yield of  rilpivirine. The orthogonal experiment investigated the effects of reaction solvent, temperature, pressure and reaction time on microwave-assisted amination yield.

Optimization of reaction solvent
Among the methods reported in the literature, the most commonly used solvents are acetonitrile and NMP. In the preliminary study, we conducted the reaction under atmospheric pressure (method A). When we led our first attempt, dioxane, acetonitrile, and NMP solvents were used. It was observed that the reaction with acetonitrile solvent provided good yields, NMP with moderate to good and with dioxane the reaction is not formed. However, when we tried to use these three kinds of solvents for microwave reaction, the reaction results were not satisfactory. Table 2 shows that although the microwave reaction can proceed smoothly, the microwave reaction yield is still far from the traditional heating conditions. Considering that different solvents are used as the reaction, none of the reactions was ideal. We speculated that the reaction temperature and time have a significant influence on the reaction yield. Since the acetonitrile yield as solvent was similar to NMP and considering industrial production cost, we choose low-boiling acetonitrile as the reaction solvent for further exploration. Therefore, a single variable method was applied to investigate the effect of reaction temperature and time on yield. The results are shown in Table 3.
We can conclude that the yield was improved with the increased reaction time and temperature. But there is a decreasing tendency of the yield when the temperature is 150 °C and reaction time is more than 120 min. Accordingly, the amination reaction's optimized conditions were determined as follows: microwave-irradiation reaction with acetonitrile solvent at 140 °C for 90 min produced the amination product with 71% yield, which was higher than that of the traditional method (69%).

Conclusion
In conclusion, we provide a synthetic method for rilpivirine, an essential drug in the treatment of acquired immunodeficiency syndrome (AIDS). The experiment mainly optimized the route of rilpivirine and the key process parameters of each step by controlling the single variable method to determine the optimal experimental conditions. The total yield was improved from 18.5 to 21%, and the reaction time of the last step decreased from 69 h to 90 min. The details are as follows: (1) synthesis of 4-((4-hydroxypyrimidin-2-yl)amino)benzonitrile (13) the reaction was carried out at 160 °C for 2 h, and then the temperature was raised to 180 °C for 4 h. After the treatment, a small amount of N,N-dimethylformamide was added to dissolve the residues, subsequently, the mixture was filtered and washed with acetonitrile to give the product with 70% yield; (2) the total yield of the intermediate 3 increased from 52 to 62% compared to the original patent. The total synthesis yield of the intermediate 2 and rilpivirine is equivalent to original patents. (3) For the rilpivirine synthetic step, a microwave-irradiation method is selected instead of the traditional method in the original patent, and the yield of amination reaction is increased to 71%. Therefore, the total synthetic yield of rilpivirine risen from 18.5 to 21%. The method proceeded in six linear steps with a gram scale and did not use high toxicity starting materials and high boiling-point solvent. This efficient and environmental-friendly process and the optimum conditions for rilpivirine preparation may form the basis for future manufacturing route.

Experimental section
Thin-layer chromatography (TLC) was performed on Silica Gel GF254 for TLC (Merck), and spots were visualized by irradiation with UV light (λ = 254 nm). 1 H NMR and 13 C NMR spectra were recorded on a Bruker AV-400 NMR spectrometer or a Bruker Avance-600 NMR spectrometer using solvents as indicated. Chemical shifts were reported in δ values (ppm) with tetramethylsilane as the internal reference, and J values were reported in hertz (Hz). Melting points (m.p.) were determined on a micro melting point apparatus (TianJin Analytical Instrument Factory, Nankai, Tianjin, China). Mass spectra were recorded on an LC Autosampler device: Standard G1313A instrument. The microwave reaction was conducted on a CEM Discover (0-600 W, 2450 MHz) instrument and the conventional high-pressure reaction was performed on Parr 4590 device. Rotary evaporators were served in the concentration of the reaction solutions under reduced pressure.

(E)-3-(4-amino-3,5-dimethylphenyl)acrylonitrile hydrochloride (2)
To a round-bottomed flask consisting of 10 mL of phosphorus oxychloride was added compound 6 slowly at 0 °C and then refluxed for 8 h at 40 °C. After completion of the reaction was monitored by TLC, and the reaction mixture was slowly poured into the ice-water mixture and stirred for 1 h and then filtered to obtain the crude (E)-3-(4-amino-3,5-dimethylphenyl)acrylonitrile (7) as white crystal solid.

Preparation of ethanolic hydrochloric acid solution
The solvent of anhydrous ethanol was stirred under ice bath, and then 41 mL of acetyl chloride reagent added slowly dropwise and stirred for 30 min to obtain an ethanolic hydrochloric acid solution.

4-((4-chloropyrimidin-2-yl)amino)benzonitrile (3)
Compound 13 was dissolved in 20 mL of phosphorus oxychloride at 0 °C and then refluxed for 1h. After completion, the mixture was cooled to room temperature, slowly poured into 100 g of crushed ice and stirred for half an hour. After filtration, the obtained filter cake was dissolved in 50 mL of water, and the pH was adjusted to 7-8 with potassium carbonate and filtered. The filtrate was washed with acetonitrile and dried to obtain 4-((4-chloropyrimidin-2-yl)amino)benzonitrile (3) as a white solid (9.9 g). Yield: 89%. mp: 209-210 °C. 1

Rilpivirine (1)
A mixture of compound 2 (1 g, 4.8 mmol) and 3 (1.2 g, 5.2 mmol) along with 10 mL acetonitrile was transferred into 20 mL microwave vial and set the reaction temperature at 140 °C, 2 h. After completion, the reaction mixture was cooled to room temperature and then 10% potassium carbonate solution was added dropwise to the vial to adjust the pH to 8-9. The obtained filtrate was washed with acetonitrile to obtain the target molecule rilpivirine as a white solid (1.28 g). Yield 71%. mp: 138-140 °C. 1