Tenofovir Disoproxil Fumarate: New Chemical Developments and Encouraging in vitro Biological Results for SARS-CoV-2

The recent emergence of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has led society to live with a serious public health problem. In this sense, repositioning of antiretrovirals has captured the attention of the scientific community. Tenofovir disoproxil fumarate (TDF) is an antiretroviral compound that is used to treat acquired immune deficiency syndrome (AIDS) and hepatitis B. In this short report, we present a scale-up investigation of TDF by in situ infrared spectroscopy monitoring and a forced degradation study to describe a new degradation product. Finally, we have evaluated TDF in vitro for SARS-CoV-2 for the first time foreseeing the using of this medicine in pre-clinical and clinical investigations for the COVID-19 (coronavirus disease 2019) treatment.

vessels, which helped to achieve both the heating and the cooling effect around the reactor in the temperature range from −50 to 220 °C. The central nozzle was employed for the mechanical stirrer, driven by a sealed electric motor (0.18 kW, EX II 2G EEx) and equipped with a baffle and a Pt100 temperature sensor. The heating and cooling operation was performed with a UNISTAT 510 thermal control system, which circulated thermal fluid in the reactor jacket; the capacity was 3 kW, the temperature ranged between −50 and 250 °C, and temperature stability was ± 0.01 °C. A thermal sensor of the Pt100 type was used to control and to correct temperature variation in the reaction medium (variation ± 0.5 °C). The in situ Fourier transform infrared spectroscopy with attenuated total reflectance (FTIR-ATR) analyses were carried out on a ReactIR 15® (Mettler Toledo) spectrophotometer equipped with an AgX 9.5 mm × 1.5 m fiber (Silver Halide) / DiComp (Diamond) tipped probe. The spectral data were collected with the iC IR 7.0/7.1 software. Before the reactions, the FTIR-ATIR spectra of all the reagents and products had been acquired in the solvent that would be employed in the reaction, at the temperature at which the experiment would be accomplished.
______________________________________________________________ *e-mail: gclososki@fcfrp.usp.br; eaneto@fmrp.usp.br; npelopes@gmail.com The spectra of neat samples were also recorded with the same FTIR-ATR apparatus. The 1 H nuclear magnetic resonance (NMR) spectra were obtained on a Bruker DPX-400 spectrometer (400 MHz) or on a Bruker DRX-500 spectrometer (500 MHz). The chemical shifts are expressed in ppm from tetramethylsilane (TMS). The data are reported as follows: chemical shifts, coupling constants (Hz), and integration. The 13 C NMR spectra were acquired on a Bruker DPX-400 spectrometer (100 MHz) or on a Bruker DPX-500 spectrometer (125 MHz) with complete proton decoupling. The chemical shifts are reported in ppm from tetramethylsilane, with the solvent resonance as the internal standard (CDCl3 77.26 ppm, CD3OD 49.86 ppm). The HRMS spectra were measured on a Bruker Daltonics micrOTOF II -ESI-TOF Mass Spectrometer. The high-performance liquid chromatography analyses were performed on HPLC system (Shimadzu), comprising LC-20 AD solvent pumps, a CBM-20A system controller, a CTO-20A column oven, an SIL-20A injector, and an SPD-M20A diode array detector (DAD) operating at 270 nm. A Luna C18 column Gemini ® (5 µm, 4.6 × 250 mm, Phenomenex) was used for the chromatographic analyses.

Synthetic procedures
Preparation of 1-(6-amino-9H-purin-9-yl)propan-2-ol (4) In a 15-L jacketed reactor vessel equipped with a mechanical stirring and a reflux condenser, adenine (2) (8.03 mol, 1085 g), anhydrous potassium carbonate (0.27 mol, 37.5 g), and 5 L of dimethylformamide (DMF) were added. The flask containing the reactional medium was heated to 120 °C, and then propylene carbonate (3) (10.0 mol, 1020 g) was added dropwise to the stirred solution, under nitrogen atmosphere. The reaction was monitored by in situ FTIR-ATR spectroscopy. After 5 h, the mixture was cooled to 70 °C, and isopropanol (3 L) and methanol (  Synthesis of diethyl (((1-(6-amino-9H-purin-9-yl)propan-2-yl)oxy)methyl)phosphonate (6) In a 15-L jacketed reactor vessel, alcohol 4 (2.59 mol, 500 g), magnesium di-tert-butoxide (2.84 mol, 485 g), and DMF (4.5 L) were added. This suspension was stirred (120 rpm) at 70 °C for 1 h, under nitrogen atmosphere. Then, under stirring, a solution of tosylphosphonate 5 (3.10 mol, 1000 g) in DMF (500 mL) was added dropwise (the temperature of the reaction medium was kept under control and did not vary more than ± 0.5 °C). The reaction was monitored by in situ FTIR-ATR spectroscopy. The reaction was stirred at 70 °C for 8 h. After that, DMF was removed by distillation under reduced pressure (85-90 °C). The mixture was cooled to room temperature; water (3 L) and saturated NaCl solution (2 L) were added; and the pH was adjusted to 7 with HCl solution (50% v ̸ v). The reaction medium was extracted with ethyl acetate (4 × 1000 mL); the organic phases were combined and distilled under low pressure. The crude product was purified by crystallization from acetonitrile; 1 L of acetonitrile was used for every Synthesis of (((1-(6-amino-9H-purin-9-yl)propan-2-yl)oxy)methyl)phosphonic acid (7) In a 10-L jacketed reactor under nitrogen atmosphere, a suspension of diethylphosphonate 6 (1.45 mol, 500 g) in NMP (1500 mL) and sodium bromide (3.03 mol, 312.5 g), previously dried under reduced pressure at 120 °C for 24 h, were added, and the reaction medium was cooled to 0 °C. Trimethylsilyl chloride (5.91 mol, 750 mL) was then added dropwise to the reaction mixture, under nitrogen atmosphere. The reaction was kept under stirring (120 rpm) at 75 °C for 1.5 h. The reaction was monitored by in situ FTIR-ATR spectroscopy. Excess TMSCl and other volatile impurities were removed under reduced pressure at 50 °C. The system was cooled to 5 °C; distilled water (2 L) was added; the pH was adjusted to 3 with 40% (m/v) aqueous NaOH solution; and a bulky white solid precipitated. After 48 h at 0 °C, the white solid was filtered under reduced pressure and washed with a cold mixture of water/acetone (1:1).

Synthesis of tenofovir disoproxil fumarate (TDF) (1)
In a 10-L jacketed reactor vessel, tenofovir (7) (1.74 mol, 500 g) previously dried under vacuum at 120 °C for 36 h (analysis showed 0.8% of water), NMP (2600 mL), and cyclohexane (1500 mL) were added. A Dean-Stark apparatus was connected, and the solution was refluxed for 24 h under nitrogen atmosphere and mechanical stirring (200 rpm). After that, under stirring (400 rpm), a new portion of cyclohexane (500 mL) and triethylamine (200 mL) was added to the suspension, and then the cyclohexane was distilled off (60-65 °C) under reduced pressure. The reaction temperature was set to 60 °C. Next, triethylamine (6.86 mol; 955 mL; 4 equiv.) followed by chloromethylisopropyl carbonate (7.54 mol; 1150 g; 5 equiv.) was added dropwise to the suspension under stirring. The conversions were determined by analyzing the formed products by HPLC. Care was taken to control the formation of impurities, which should not exceed 10% yield. After 3 h, the remaining triethylamine and chloromethylisopropyl carbonate were distilled off at reduced pressure (60-65 °C). A saturated NaCl solution (200 mL) was added to the reaction mixture, and the aqueous layer was extracted with ethyl acetate (2 × 2000 mL). The organic phases were combined, dried over anhydrous magnesium sulfate, and concentrated in vacuo. The resulting yellowish oil (822.45 g) was dissolved in isopropanol (1600 mL); fumaric acid (1.29 mol; 150 g) was added; and the mixture was heated to 70 °C for 3 h under stirring and nitrogen atmosphere. After that, the reaction medium was cooled and kept at 0 °C for 24 h. Then, the product was filtered under vacuum, the solid was washed with cold isopropanol and dried under vacuum for 24 h, to

TDF forced degradation studies
The purpose of forced degradation studies is to investigate the drug intrinsic stability, as required by regulatory agencies such as the National Health Surveillance Agency (ANVISA, Agência Nacional de Vigilância Sanitária) in Brazil, the European Medicines Agency (EMA) in Europe, and the Food and Drug Administration (FDA) in the United States, to protect the population from damage caused by the presence of impurities in medicines. To date, forced degradation studies of tenofovir, either alone or in association with other antiretrovirals, have demonstrated their stability when they are exposed to dry heat, humidity, and photolysis. Variable degradation has been observed under acid and alkaline hydrolysis and oxidation conditions, according to the exposure conditions, in which impurities always remain within acceptable limits for commercialization. [1][2][3][4][5][6][7][8][9][10][11] To reinforce the safety of using tenofovir disoproxil fumarate, a new forced degradation study was carried out under more extreme conditions to obtain information on not yet known degradation pathways. The stress conditions employed in this study (concentration and type of stressing agent, temperature, and incubation time) were standardized, whenever possible, to compare the results obtained in the various experiments performed herein (Table S1). When significant degradation was not achieved, the condition endpoint was used according to the specialized literature.  Three main degradation products originated during the forced degradation assays, as represented by the total ion chromatogram of tenofovir disoproxil oxidation by peroxide ( Figure S20). The peak at 25 min refers to tenofovir disoproxil, resulting from the dissociation of the fumarate salt, with a mass/charge ratio (m/z) ion of 520 ( Figure S21a). The degradation product with retention time 8.37 min presented a high-intensity ion with m/z 288 ( Figure S21b). This degradation product was more evident in acid and alkaline hydrolysis conditions and has a molecular mass that is equivalent to the hydrolysis of the prodrug tenofovir disoproxil to tenofovir mediated by esterases, as already described in the literature. 12 The degradation product with retention time 16.95 min presented a high-intensity ion with m/z 404 ( Figure S21c). This degradation product has a molecular mass that is equivalent to tenofovir monosoproxil, also called impurity A by the International Pharmacopoeia of the World Health Organization. 13 The dry thermal and photolysis condition also provided this degradation product in minor proportions.
The degradation product with retention time of 23.7 min presented a spectrum with a greater intensity ion with m/z 536 ( Figure S21d). The molecular mass of this degradation product corresponds to an increase of 16 Da in the chemical structure of tenofovir disoproxil, due to introduction of an oxygen atom. This degradation product has never been reported in the literature and was observed only for the oxidation by peroxide. There was no significant degradation of tenofovir disoproxil in the oxidation by transition metal. To define the structure of the new degradation product, a detailed analysis of the TDF ESI-MS/MS spectra was conducted ( Figure S22). The discussion of the mechanism is included in the Results and Discussion section of the paper. A detailed analysis of the TDF-oxide (degradative product) revealed the same MS/MS profile as TDF ( Figure S23). Figure S24 shows a comparative fragmentation of TDF and TDF-oxide.  Finally, Figure S24 shows a comparison of the fragmentation pathways of TDF and TDF-oxide. The mechanisms are described in the paper file, and the data suggest oxidation at the purine ring. One possible metabolite is an N-oxide metabolite that can emerge in forced studies, but the correct position cannot be defined by MS/MS analysis only. Figure S24. Fragmentation pathway proposed on the basis of the MS/MS spectra of TDF and TDF-oxide. With the MS/MS data, it is not possible to define on which N atom the reaction occurred. One of the atoms was arbitrarily used just to demonstrate that oxidation does not occur in the side chain.

Docking study
The cryo-EM structure of SARS-CoV-2 RNA-dependent RNA polymerase (PDB id = 7BV2, 2.50 Å) 14 was prepared via HERMES (version 2020.1 CSD release) 15 with the removal of water molecules and the complex remdesivir-primer, and addition of hydrogens. Then, the binding site was defined by the region occupied by remdesivir with the inclusion of all residues within 18 Å via GOLD (version 2020.1 CSD release). 16 CHEMPLP was chosen as score function, and 50 runs per ligand were performed ( Figure S25).