As a commonly used IV sedative hypnotic drug, propofol has rapid onset, short duration, and rapid recovery.1 Though widely used in induction and maintenance of general anesthesia, propofol still causes some adverse effects, for example, injection site pain,2 limitation of physical stability,3 hypertriglyceridemia,4,5 and potential embolism. Most of the adverse effects are caused by the application of lipid-based emulsion for the unsatisfactory water solubility of propofol. Therefore, the development of the water-soluble prodrugs was considered.6
Fospropofol (GPI15715), a water-soluble phosphate ester prodrug of propofol approved by the Food and Drug Administration on December 12, 2008, is a potent IV sedative hypnotic drug, which can be hydrolyzed by the endothelial cell alkaline phosphatases and then release propofol as an active metabolite.7 However, the metabolites of this N-phosphono-O-methyl prodrug include formaldehyde and phosphate8 (Fig. 1). Although formaldehyde is a normal metabolite in several catabolic pathways,9 it may lead to potential risks in the body, for instance, alteration of homeostasis in the cells.10 In addition, as one of the exogenous toxins, formaldehyde may be a main factor in nasal cancer occurrence or mutagenic effects that were observed in several previous studies.11,12
To solve this problem, we designed and synthesized the lead compound “HX0969,” which is a highly potent prodrug without the risk of producing formaldehyde and its water-soluble derivative “HX0969W.”13–15 However, the by-products of HX0969W still contained phosphate ester. In addition, paresthesia and pruritus were observed to be caused by fospropofol in >20% of patients. Perineum itching, burning, or tingling also occurred in up to 85% of patients at all dose ranges.7 According to clinical research on fosphenytoin and dexamethasone, these adverse reactions were considered to be caused by accumulation of a phosphate ester component.16 Phosphate was used in well-known water-soluble prodrug designs for IV use, but it had adverse side effects. Another adverse effect of phosphate accumulation is hyperphosphatemia, which could cause vascular calcification and cardiovascular disease.17
To avoid the accumulation of the unwanted by-products, we synthesized 2 propofol prodrugs HX0969-Ala-HCl and HX0969-Gly-F3 based on the lead compound HX0969 and introduced alanine and glycine individually. To investigate these novel prodrugs, we assessed their water solubility, their hydrolysis to propofol in different physiological media (rat plasma, rhesus monkey plasma, and rat hepatic microsomes), their onset time, and duration in rats. We expected that these novel propofol prodrugs could not only eliminate the adverse effects caused by the accumulation of formaldehyde and phosphate but also retain their rapid action onset and good water solubility.
The lead compound HX0969 was designed based on the cyclization mechanism, which rapidly releases propofol and γ-hydroxybutyric acid (Fig. 2). Though it is a highly potent intermediate, HX0969 still has limited solubility in water. To solve this problem, the phosphate group was introduced into HX0969. The water-soluble derivative HX0969W (2,6-diisopropylphenyl 4-hydroxybutanoate disodium phosphate, C16H23Na2O6P) was then designed, which was able to release HX0969 first and then release propofol (Fig. 3). To avoid the unwanted metabolite phosphate group and maintain favorable water solubility, alanine and glycine were considered for introduction into the present prodrug design (Figs. 4 and 5). Appendix A describes the compound synthesis procedures.
Experimental animal studies were approved by the Committee of Scientific Research and the Institutional Animal Experimental Ethics Committee of West China Hospital, Sichuan University (Chengdu, China). Animals were cared for according to the Guide for Care and Use of Laboratory Animals.
Adult Sprague-Dawley rats weighing 180 to 220 g were used in this study. Rats were housed at ambient temperatures of 25°C ± 1°C, with a relative humidity of 60% and a 12-hour light/12-hour dark cycle. All rats had free access to water and food. All experiments were performed during the light phase, and rats were acclimatized for 1 week.
Test of Solubility and Stability of Prodrugs
The solubility of these prodrugs was determined at room temperature in normal saline (0.9%). A fixed amount (20 mg) of the prodrugs was consecutively added to normal saline in a test tube with the consequent 5-minute vortex until the solution was saturated. The mixture was then centrifuged mechanically in a Biofuge stratus (Labofuge 400, Heraeus, Hanau, Germany) at 3500g for 10 minutes. The supernatant (10 μL) of the mixture was collected and diluted with methanol (fospropofol 1:1000, HX0969W 1:200, HX0969-Ala-HCl 1:1000, HX0969-Gly-F3 1:200). The diluted mixture was then withdrawn, filtered (0.45 μm Millipore), and analyzed by high-performance liquid chromatography (HPLC).
Before the study, the propofol concentration was measured in these prodrug solutions to test whether the prodrugs were stable enough for the consequent research or they were unstable for the prodrugs to be easily decomposed. The prepared solution (fospropofol, 20 mg/mL; HX0969W, 60 mg/mL; HX0969-Gly-F3, 10 mg/mL; HX0969-Ala-HCl, 10 mg/mL) was collected and stored for 24 hours at room temperature. Consequently, the solution prestored for 24 hours and the solutions prepared at 0 hour were respectively mixed with methanol (1:99). After the mixed solution was transferred into the Eppendorf tubes and was vibrated, the supernatant of these mixtures was withdrawn and analyzed by HPLC.
TEST OF THE PRODRUGS BIOCONVERSION IN RAT PLASMA, RHESUS MONKEY PLASMA, AND RAT HEPATIC MICROSOMES
Hydrolysis in Rat Plasma and in Rhesus Monkey Plasma
The concentration of the released propofol and the conversion percentage (%, mol/mol) of these prodrugs were determined in vitro in rat and rhesus monkey plasma. After sevoflurane anesthesia, blood was collected from the femoral artery of rats and from the forelimb vein of monkeys. Plasma was then prepared by centrifugation at 4000g for 10 minutes at 4°C.
HX0969, HX0969-Ala-HCl, HX0969-Gly-F3, HX0969W, and fospropofol were respectively added to the preheated (37°C, 30 minutes) rat plasma. Subsequently, these drug-plasma mixtures (each drug in plasma, 5 μM/L) were placed in a water bath at 37°C and were shaken for 5 hours at 60 rpm.
After incubation, 100-μL samples were taken and mixed with 300 μL acetonitrile for deproteinizing the plasma at the following time points: 0, 1, 3, 5, 7, 10, 15, 20, 30, 45 minutes, 1, 2, 3, and 5 hours. After collection, the mixtures were vortexed and centrifuged at 12,000g for 10 minutes. The supernatant (15 μL) was gathered for analysis. Analysis of propofol in the plasma was performed by HPLC.
Hydrolysis in Rat Hepatic Microsomes
To test whether the novel prodrugs were metabolized by specific plasma esterases or they were hydrolyzed by nonspecific esterases in the other tissues, the following hydrolysis tests in rat hepatic microsomes were performed in vitro.
HX0969, HX0969-Ala-HCl, HX0969-Gly-F3, HX0969W, and fospropofol were incubated with rat liver microsomes in the presence of nicotinamide adenine dinucleotide phosphate (NADPH). Microsomal reaction mixtures, containing 1.0 mmol/L NADPH, 0.5 mg/mL (total protein) microsomes, were added to Tris-HCl buffer (pH = 7.4). Prodrug solutions were added to microsomes reaction mixture with a final concentration of 5 μM/L. The reaction solutions were kept at 37°C, and the samples were taken at 0, 1, 3, 5, 7, 10, 15, 20, 30, 45 minutes, 1, 2, 3, 5 and hours. The reaction of 200 μL mixture was terminated at the above time points by adding 200 μL cold methanol containing internal standard. After centrifugation, the supernatant was immediately collected and analyzed by HPLC.
The extent of the prodrug hydrolysis into propofol was expressed by Equation-Conversion percentage (%, mol/mol), which equates “
Equation (Uncited)Image Tools
” and Cpro here represented the concentrations of propofol in the plasma, which were assayed by HPLC. Appendix B describes HPLC analysis.
In Vivo Anesthetic Activities of Propofol Prodrugs at Equipotent Doses
Rats were fixed in a steel chamber with a tail exit hole. The drugs were injected within 15 seconds through a lateral tail vein catheter (24-gauge, Terumo, Japan), followed by a 1-mL normal saline flush. After the injection, rats were removed from the chamber and placed in supine position. Loss of righting reflexes (LORR), defined as the rat’s failure to right itself after drug administration, indicated the drug action. The 50% effective dose (ED50) for LORR after bolus administration was determined by the up-and-down method.18
To determine whether the approximate onset time of the novel prodrugs was consistent with the propofol release rate in the hydrolysis study, 5 randomized groups of rats, 8 rats in each group, were respectively given an IV dose of propofol and 4 prodrugs in a dose of 1.5-fold ED50 (propofol, 9 mg/kg; fospropofol, 60 mg/kg; HX0969W, 81 mg/kg; HX0969-Gly-F3, 24 mg/kg; HX0969-Ala-HCl, 29 mg/kg). We then started to measure the onset time to LORR and the duration of LORR (defined as the time from drug action to the time when the rat was spontaneously righting itself).
Results of prodrug hydrolysis in vitro are presented as mean ± SEM, and n = 3 denotes the times of sampling (Table 1). The ED50 of the 5 compounds is reported with 95% confidence interval (CI) and it is determined by the up-and-down method,18 95% CI = log−1(X50 ± 1.96*Sx50), where X50 is the log dose of ED50, Sx50 is the standard error, and the constant 1.96 reflects the 0.05 α level. Onset time and the duration time data from in vivo tests are presented as median values (interquartile range) and were compared by the Kruskal-Wallis test. All pairwise comparisons among groups were performed by 2-group 2-sided Mann-Whitney test, with the Bonferroni correction; P < 0.0083 was considered statistically significant (0.0083 = 0.05/6, and 6 were the times of pairwise comparisons among the 4 groups) (Table 2). Analysis was performed by Statistical Package for Social Sciences (SPSS™), Windows versions 16.0 (SPSS, Chicago, IL).
Solubility and Stability
Solubility was tested in normal saline (pH = 7.4) at room temperature (25°C ± 0.5°C). The water solubilities of fospropofol, HX0969W, HX0969-Gly-F3, and HX0969-Ala-HCl were 461.46 ± 26.40 mg/mL, 189.45 ± 5.02 mg/mL, 49.88 ± 0.58 mg/mL, and 245.99 ± 4.83 mg/mL, respectively (mean ± SEM). The retention times were as follows: fospropofol 3.4 minutes, HX0969W 3.3 minutes, HX0969-Gly-F3 20.8 minutes, and HX0969-Ala-HCl 15.1 minutes.
The results indicated that all prodrugs had strong enough stability for research after 24-hour placement. The concentration of the decomposed propofol in all prodrug solutions was lower than the value indicated in the lower limit of quantification (5 ng/mL).
BIOCONVERSIONS OF PRODRUGS IN RAT PLASMA, RAT HEPATIC MICROSOMES, AND RHESUS MONKEY PLASMA
Hydrolysis in Rat Plasma
During the 5-hour testing period, the maximal conversion percentage (%, mol/mol) of the 5 prodrugs converting to propofol in rat plasma was as follows: fospropofol, 62.75% ± 2.31%; HX0969, 93.85% ± 1.50%; HX0969W, 43.32% ± 0.92%; HX0969-Gly-F3, 88.79% ± 1.37%; and HX0969-Ala-HCl, 88.74% ± 1.47% (mean ± SEM), and the concentrations of propofol released from the 5 prodrugs reached a steady state at about 5, 5, 5 hours, 1 and 1 minute, respectively (Fig. 6).
Hydrolysis in Rhesus Monkey Plasma
The maximal conversion percentage (%, mol/mol) of the 5 prodrugs converting to propofol in rhesus monkey plasma was as follows: fospropofol, 66.87% ± 1.77%; HX0969, 94.57% ± 2.04%; HX0969W, 48.13% ± 2.80%; HX0969-Gly-F3, 91.81% ± 0.96%; and HX0969-Ala-HCl, 82.08% ± 1.88% (mean ± SEM). The hydrolysis rates of the 2 novel amino acid prodrugs converting to propofol were greater than the hydrolysis rates of the 2 former phosphate prodrugs (Fig. 7).
Hydrolysis in Rat Hepatic Microsomes
The maximal conversion percentage (%, mol/mol) of the 5 prodrugs converting to propofol in rat hepatic microsomes was as follows: fospropofol, 77.96% ± 0.84%; HX0969, 94.53% ± 1.17%; HX0969W, 71.95% ± 0.61%; HX0969-Gly-F3, 96.34% ± 0.94%; and HX0969-Ala-HCl, 93.74% ± 1.51%. The concentrations of propofol released from the 5 prodrugs reached a steady state at about 7, 1, 15, 1, and 1 minutes, respectively (Fig. 8).
Compared with hydrolyses performed in plasma, this experiment showed that all prodrugs had a greater release rate (within 15 minutes) and a larger conversion extent (maximal conversion percentage >70%) to propofol in rat hepatic microsomes (Table 1), which indicated that these prodrugs could be hydrolyzed in the presence of the plasma enzyme system and hepatic enzymes.
Anesthetic Activities of Prodrugs In Vivo
ED50 with 95% CI of the 5 compounds was as follows: propofol, 5.99, 5.36–6.71 mg/kg; fospropofol, 40.36, 33.04–45.99 mg/kg; HX0969W, 53.89, 50.86–57.08 mg/kg; HX0969-Ala-HCl, 19.2, 18.27–20.18 mg/kg; and HX0969-Gly-F3, 15.75, 14.26–17.39 mg/kg. The onset time to LORR and the time duration of the 5 drugs at their equipotent doses (1.5-fold ED50) were measured (Table 2). Compared with the previous prodrugs of fospropofol and HX0969W, the 2 improved compounds (HX0969-Gly-F3 and HX0969-Ala-HCl) had a much shorter onset time to action (Fig. 9) and a much shorter duration. Meanwhile, the 2 novel amino acid prodrugs were used in smaller doses when compared with fospropofol and HX0969W (Table 2).
Compared with the merits and drawbacks of propofol and phosphate prodrugs reported in the pertinent literature, the application of amino acid that was found to be valuable for increasing water solubility of parent drugs19–22 was considered in our present prodrugs design. We synthesized 2 prodrugs based on the introduction of alanine and glycine and hypothesized these improved prodrugs could retain the main advantages of propofol (rapid onset) and fospropofol (water solubility). In this study, a series of experiments was performed to investigate the prodrug solubility and stability as well as the rate for releasing propofol in different physiological media (rat plasma, rhesus monkey plasma, and rat hepatic microsomes). Some other in vivo experiments were performed in this study to determine whether their anesthetic activities corresponded to the results obtained by the previous in vitro experiments.
The water solubility of HX0969-Gly-F3 (49.88 ± 0.58 mg/mL) seemed lower than the other prodrugs, but this solubility was sufficient for the research and pharmaceutical demands. HX0969-Ala-HCl had a good solubility in normal saline (245.99 ± 4.83 mg/mL). All prodrugs that we investigated were so stable that the decomposed propofol seemed to be negligible (lower than detection limits).
The hydrolysis in the rat plasma confirmed that 2 novel amino acid prodrugs (HX0969-Gly-F3 and HX0969-Ala-HCl) released propofol more rapidly than the 2 phosphate ester prodrugs (fospropofol and HX0969W). The maximal conversion percentage (%, mol/mol) at the first time point indicated that the 2 amino acid prodrugs could release a large amount of propofol in a very short period of time. However, compared with the test results in rat plasma, the consequent test results in monkey plasma showed that all the prodrugs exhibited a relatively unsatisfactory release rate and amount of propofol. Such different enzymatic activities in the plasma might have resulted from interspecies differences. Various esterases including carboxylesterase (CES), paraoxonase, and butyrylcholinesterase are found in blood. The existence of the esterases and their expression levels are different among species. For instance, paraoxonase and butyrylcholinesterase exist in almost all the species. However, CES is only abundant in rabbits, mice, and rats. In addition, the plasma of monkeys, dogs, and humans does not contain CES.23 All differences in plasma esterases and their expression levels may result in different hydrolase activities. Considering the characteristics of the human plasma,24 the monkey plasma was considered to be a suitable alternative in our present study. Although the 4 prodrugs could not release propofol as rapidly as what was found in the test of rat plasma, the results of this test indicated that the 2 former phosphate prodrugs were inferior to the amino acid prodrugs in the release rate and amount of propofol.
To explore the possible metabolite approach of these 2 novel amino acid prodrugs, hydrolysis tests in rat hepatic microsomes were performed in vitro. We expected that the results could lead to a clearer answer concerning the hydrolysis mechanism of these compounds. The results showed that all prodrugs in the present experiment could be hydrolyzed and propofol could then be released in both plasma and hepatic microsomes. The hydrolysis rates of the 5 prodrugs were greater in the presence of hepatic enzymes, which indicated that these prodrugs could be hydrolyzed by nonspecific esterases. The relevant enzymes might exist not only in plasma but also in the liver. In summary, the current in vitro tests might not completely simulate the real environment in animal bodies; therefore, further studies should investigate the pharmacokinetics of these prodrugs.
In a subsequent study, we investigated the general anesthetic activities of the prodrugs in rats.
The results of the in vivo tests at the equipotent dose demonstrated that the IV injection of HX0969-Ala-HCl or HX0969-Gly-F3 had a more rapid onset (within 1 minute) than the 2 former water-soluble prodrugs (fospropofol and HX0969W). Compared with the 2 former phosphate prodrugs (fospropofol and HX0969W), these 2 improved prodrugs have a shorter duration of action. In addition, when compared with the 2 phosphate ester prodrugs, the 2 improved amino acid prodrugs could be used in a smaller dose to achieve the sedative effect rapidly. These findings indicated that the 2 novel amino acid prodrugs had rapid rate and high efficiency in releasing the parent drug propofol.
The results of hydrolysis tests in rat plasma indicated that the improved amino acid prodrugs had a higher amount of hydrolysis and a greater release rate than their lead compound (HX0969). That was not consistent with what we had hypothesized (the phosphate ester prodrug HX0969W, the lead compound (HX0969), was released first, and propofol was released by means of cyclization). One possibility was that the chemical bonds of these improved prodrugs broke in another way, in which the prodrugs could release the parent drug directly. Introduction of the amino acid group accelerated the rate of release. However, our present study was lacking in evidence to verify this hypothesis because of the absence of the related mass spectrum study. This remains an open question for further study.
In addition, 1 product of the lead compound HX0969 should be butyrolactone in theory, which will be converted to 4-hydroxybutyric acid quickly in vivo. As an endogenous substance in mammals, hydroxybutyric acid exhibits low toxicity and can be metabolized into the Krebs cycle.25,26 The previous study showed that IV administration of 4-hydroxybutyric acid sodium (200 mg/kg) might induce sedation in rats.27 One previous study, which focused on the relationship between sleep duration and the concentration of 4-hydroxybutyric acid or gamma-hydroxybutyric acid (GHB) in rat body fluid, found that a dose of 400 to 800 mg/kg 4-hydroxybutyric acid was administered in that experiment.28 However, a previous study indicated that the release of 4-hydroxybutyric acid was approximately only 24 mg/kg when administered with the 2-fold of ED50 for HX0969w in rats,14 which was much lower than the dose of 4-hydroxybutyric acid for a sleep-inducing effect. Thus, in the present study, we speculated that the amount of GHB produced from these prodrugs might be negligible to affect the results of our present tests in rats.
In a word, the improved design of propofol prodrugs in this study is valuable. The introduction of the amino acid group provides the prodrugs favorable properties, for example, good water solubility, a more satisfactory rate of parent drug release, and higher anesthetic efficacy. Theoretically, this novel design can avoid production of formaldehyde and phosphate ester. The adverse effects induced by accumulation of phosphate and formaldehyde, especially pruritus and paresthesia, cannot be estimated in the current experiments. To perform a relevant study in future is necessary.
APPENDIX A: SYNTHESIS OF COMPOUNDS
All materials and reagents were obtained from commercial suppliers and were used without further purification. HX0969-Gly-F3 and HX0969W were prepared, as described previously, as a white powder and a white crystal, respectively. Fospropofol was prepared, as described previously, as a white crystal.
General Synthetic Procedures
All the reactions described were monitored by thin-layer chromatography (TLC) using aluminum sheets precoated with MERCK Silica gel 60 F254 (Darmstadt, Germany). Samples were visualized by UV light. 1H,13C NMR-spectra were recorded on a Bruker Avance II spectrometer (Bruker, Rheinstetter, Germany) operating at 300 or 400 MHz at 25°C. Tetramethylsilane (TMS) for CDCl3 samples was used as an internal reference, and no internal reference was used for D2O samples. The splitting pattern abbreviations are as follows: s = singlet; d = doublet; t = triplet; q = quadruple; quintex = quin; sex = sextex; hep = heptet; m = multiplet. Mass was determined using an API 3000 LC-Ms/Ms (Applied Biosystems/MDS SCIEX, Foster City, CA, USA).
HX0969 and HX0969W
2,6-Diisopropylphenyl 4-Hydroxybutanoate (HX0969)
Sodium borohydride (0.84 g, 22 mmol) was suspended in anhydrous tetrahydrofuran (15 mL). The reaction flask was cooled with an ice bath. A solution of propofol hemisuccinate (6 g, 21.5 mmol) in anhydrous tetrahydrofuran (30 mL) was slowly added dropwise to the reaction solution for 30 minutes. Then, the mixture was stirred at 0°C for 1 hour. A solution of iodine (2.76 g, 10.8 mmol) in anhydrous tetrahydrofuran (20 mL) was slowly added dropwise to the reaction solution for 40 minutes. Then, the mixture was stirred at 0°C for 30 minutes. The reaction solution was evaporated to dryness. Then, ethyl acetate (50 mL) was added and filtered. The filtrate was washed with a saturated solution of sodium bicarbonate (40 mL), washed with water (2 × 40 mL), dried over MgSO4, and evaporated to dryness to produce a colorless oil. Purification was performed on a silica gel chromatography column (20% ethyl acetate in cyclohexane) to yield HX0969 as a colorless oil (4.4 g, 16.7 mmol, 77.7%).
1H NMR (CDCl3, 600 MHz) δ: 1.195 (12H, d, J = 7.2 Hz), 1.620 (1H, s), 2.053 (4H, quin, J = 6.6 Hz), 2.774 (2H, t, J = 7.2 Hz), 2.899 (2H, quin, J = 6.6 Hz), 3.780 to 3.800 (2H, t, J = 6 Hz), 7.150 to 7.259 (3H, m). 13C NMR (CDCl3, 400 MHz) δ: 22.75, 23.70, 27.62, 30.77, 61.96, 123.93, 126.53, 140.28, 145.50, 172.57. ESI-Ms: (M+1). HRMS: for C16H24O3·H+, calcd 265.1804, found 265.1808.
HX0969 (8.9 g, 33.7 mmol) was dissolved in pyridine (16 mL).The reaction flask was cooled with an ice bath. Methane sulfonyl chloride (4.8 g, 41.9 mmol) was added to the reaction solution and stirred for 30 minutes. The reaction temperature was kept <10°C. No HX0969 was detected by TLC. The reaction solution was poured into water (90 mL), and the pH was adjusted to 1 with concentrated hydrochloric acid. The mixture was extracted with ethyl acetate (60 mL). The organic layer was washed with water (2 × 50 mL), dried over Na2SO4, and filtered. The filtrate was evaporated to dryness and recrystallized with cyclohexane to produce 2,6-diisopropylphenyl 4-(methylsulfonyloxy)butanoate as a white solid (6.34 g, 18.5 mmol, 55.2%, mp: 50°C–51°C). 1H NMR (CDCl3, 300 MHz) δ: 1.206 (12H, d, J = 6.9 Hz), 2.185 to 2.273 (2H, m), 2.797 to 2.927 (4H, m), 3.034 (3H, s), 4.376 (2H, t, J = 6 Hz), 7.152 to 7.259 (3H, m). 13C NMR (CDCl3, 300 MHz) δ: 22.78, 23.50, 24.43, 27.51, 29.61, 37.23, 68.56, 123.89, 126.57, 140.10, 145.26, 171.08. HRMS: for C17H26O5S·Na+, calcd 365.1399, found 365.1391.
Sodium 4-(2,6-Diisopropylphenoxy)-4-Oxobutyl Phosphate (HX0969W)
Triethylamine (40 mL, 288 mmol) and 85% phosphate (28 g, 243 mmol) were added to N,N-dimethylformamide (DMF) (110 mL). 2,6-Diisopropylphenyl 4-(methylsulfonyloxy)butanoate (10 g, 29.2 mmol) was added to the reaction solution and stirred at 80°C for 2 hours. The reaction solution was poured into water (1000 mL). The pH of the solution was adjusted to 1 with concentrated hydrochloric acid. The mixture was extracted with ethyl acetate (500 mL).The organic layer was evaporated to dryness, and the pH of the residue was adjusted to 9 with 30% sodium hydroxide solution. Then, acetone (150 mL) was added to precipitate a white solid. The mixture was filtered, and the filter cake was dissolved in water (30 mL). The solution was then decolorized with 0.3% activated carbon at 50°C for 2 minutes and filtered under standard atmospheric pressure. The filtrate was poured into acetone (500 mL) to precipitate a white crystal. This was kept in a refrigerator at 4°C overnight, filtered, and dried in vacuum at 50°C for 2 hours to yield HX0969W as a white crystal (8 g, 20.6 mmol, 70.5 %, mp >200°C). 1H NMR (CDCl3, 400 MHz) δ: 1.070 (12H, d, J = 6.8 Hz), 1.955 to 2.011 (2H, m), 2.777 to 2.867 (4H, m), 3.761 (2H, q, J = 6.4 Hz), 7.198 to 7.255 (3H, m).13C NMR (CDCl3, 400 MHz) δ: 22.48, 25.91, 27.11, 30.48, 63.17, 124.40, 127.39, 140.85, 144.66, 176.09. HRMS: for C16H23Na2O6P·H+, calcd 389.1106, found 389.1100.
HX0969 (15 g, 57 mmol) and Boc-L-gly (10 g, 57 mmol) were added to dichloromethane (50 mL). A catalytic amount of 4-dimethylaminopyridine (DMAP) was added to the reaction solution. A solution of N,N’-dicyclohexylcarbodiimide (DCC) (11.8 g, 57 mmol) in dichloromethane (30 mL) was slowly added dropwise to the reaction solution with stirring. No HX0969 was detected by TLC. The mixture was filtered, and the filtrate was evaporated to dryness to obtain yellow oil. Purification was performed on a silica gel chromatography column (20% ethyl acetate in cyclohexane) to yield HX0969-Gly-Boc as a colorless oil (18.5 g, 44 mmol, 77.2%). 1H NMR (CDCl3, 400 MHz) δ: 1.195 (12H, d, J = 6.8 Hz), 1.462 (9H, s), 2.147 (2H, quin, J = 6.8 Hz), 2.732 (2H, t, J = 7.2 Hz), 2.868 (2H, quin, J = 6.8 Hz), 3.949 (2H, d, J = 5.2 Hz), 4.292 (2H, t, J = 6.4 Hz), 5.033 (1H, s), 7.149 to 7.264 (3H, m).13C NMR (CDCl3, 400 MHz) δ: 24.05, 26.91, 27.59, 28.30, 30.45, 42.41, 64.14, 80.07, 123.95, 126.58, 140.22, 145.43, 155.72, 170.36, 171.35. HRMS: for C23H35NO6·H+, calcd 422.2543, found 422.2535.
2,6-Diisopropylphenyl 4-(2-Aminoacetoxy) Butanoate Trifluoroacetic Acid Salt (HX0969-Gly-F3)
HX0969-Gly-Boc (13.2 g, 31.3 mmol) was dissolved in dichloromethane (20 mL). Trifluoroacetic acid (20 mL) was added to the solution, which was stirred at room temperature for 2 hours. The reaction solution was evaporated to dryness to produce yellow oil. Purification was performed on a silica gel chromatography column (30% ethyl acetate in cyclohexane) to yield HX0969-Gly-F3 as a white solid (5.6 g, 12.9 mmol, 41.2%, mp: 110°C–112°C). 1H NMR (CDCl3, 300 MHz) δ: 1.187 (12H, d, J = 6.8 Hz), 2.109 (2H, quin, J = 6.8 Hz), 2.698 (2H, t, J = 7.4 Hz), 2.862 (2H, quin, J = 6.8 Hz), 3.848 (2H, s), 4.295 (2H, t, J = 6.2 Hz), 7.139 to 7.259 (3H, m), 8.509 (3H, s). 13C NMR (CDCl3, 300 MHz) δ: 23.02, 23.71, 27.54, 30.16, 40.20, 65.35, 116.46 (q, J = 1161 Hz), 123.90, 126.57, 140.20, 145.40, 162.31 (q, J = 140 Hz), 167.63, 171.40. HRMS: for C18H27NO4·H+ (not contain CF3COOH), calcd 322.2018, found 322.2012.
2,6-Diisopropylphenyl 4-(2-(Tert-Butoxycarbonyl amino)Propanoyloxy)Butanoate (HX0969-Ala-Boc)
HX0969 (5 g, 18.9 mmol) and Boc-L-Ala (3.58 g, 18.9 mmol) were added to dichloromethane (120 mL). A catalytic amount of DMAP was added to the reaction solution. A solution of DCC (3.9 g, 18.9 mmol) in dichloromethane (30 mL) was slowly added dropwise to the reaction solution with stirring. No HX0969 was detected by TLC. The mixture was filtered, and the filtrate was evaporated to dryness to obtain yellow oil. Purification was performed on a silica gel chromatography column (20% ethyl acetate in cyclohexane) to yield HX0969-Ala-Boc as a colorless oil (6.02 g, 13.8 mmol, 73.0%).
1H NMR (CDCl3, 300 MHz) δ: 1.181 (12H, d, J = 6.9 Hz), 1.378 to 1.439 (12H, m), 2.130 (2H, quin, J = 6.6 Hz), 2.722 (2H, t, J = 7.8 Hz), 2.861 (2H, quin, J = 6.9 Hz), 4.260 (2H, t, J = 6.3 Hz), 7.123 to 7.260 (3H, m). 13C NMR (CDCl3, 300 MHz) δ: 18.47, 22.85, 24.46, 27.45, 28.20, 30.21, 49.13, 63.92, 79.71, 123.80, 125.68, 140.08, 145.310, 155.02, 171.21, 173.23. HRMS: for C24H37NO6·Na+, calcd 458.2511, found 458.2507.
2,6-Diisopropylphenyl 4-(2-Aminopropanoyloxy)Butanoate Hydrochloride (HX0969-Ala-HCl)
HX0969-Ala-Boc (3 g, 6.9 mmol) was dissolved in dichloromethane (30 mL). Hydrogen chloride gas was led into the solution, which was stirred at room temperature for 2 hours. The reaction solution was evaporated to dryness to produce yellow oil. Purification was performed on a silica gel chromatography column (30% ethyl acetate in cyclohexane) to yield HX0969-Ala-HCl as a white solid (1.1 g, 3 mmol, 43.5%, mp: 79°C–80°C). 1H NMR (CDCl3, 400 MHz) δ: 1.185 (12H, d, J = 6.9 Hz), 1.772 (3H, d, J = 6.3 Hz), 2.145 (2H, t, J = 6.3 Hz), 2.728 (2H, t, J = 7.2 Hz), 2.861 (2H, quin, J = 6.8 Hz), 4328 (3H, s), 7.132 to 7.260 (3H, m), 8.786 (3H, s). 13C NMR (CDCl3, 400 MHz) δ: 16.25, 22.96, 23.77, 23.97, 27.69, 30.44, 49.50, 65.49, 124.04, 126.70, 140.30, 145.51, 170.11, 171.42. HRMS: for C19H30ClNO4·H+, calcd 336.2186, found 336.2183.
GPI15715 was prepared, as described previously,29 as a white crystal. 1H NMR (D2O, 300 MHz) δ: 1.142 (12H, d, J = 6.9 Hz), 3.387 (2H, hep, J = 6.8 Hz), 5.202 (2H, d, J = 7.5 Hz), 7.123 to 7.211 (3H, m). 13C NMR (D2O, 300 MHz) δ: 23.32, 26.27, 93.38, 124.33, 125.95, 143.09, 150.19. (M+1). HRMS: for C13H19Na2O5P·H+, calcd 333.0844, found 333.0852.
APPENDIX B: HIGH-PERFORMANCE LIQUID CHROMATOGRAPHY ANALYSIS
All prodrugs were detected by high-performance liquid chromatography (Agilent 1100 series, Agilent Technologies, Santa Clara, CA, USA) with Ultraviolet (UV) detector (Agilent 1100 series G13158 DAD). The elution was performed by using mobile phase consisting of methanol-water-tetrabutylammonium hydroxide (63:37:0.5 for HX0969-Ala-HCl and HX0969-Gly-F3; 70:30:0.5 for fospropofol and HX0969W), with triethylamine (0.1%, v/v), and phosphoric acid was used for pH adjustment. The flow rate was 1.0 mL/min. The UV absorbance detector was set at 260 nm. The retention time of HX0969-Ala-HCl and HX0969-Gly-F3 was 20.8 and 15.1 minutes, respectively.
Analysis of propofol was performed by high-performance liquid chromatography (Waters 2695 separation system, Waters, Milford, MA, USA) with a fluorescence detector (Waters 2475, Waters). The mobile phase consisted of acetonitrile–water (60:40) with a flow rate of 1.2 mL/min. The excitation and emission wavelengths were 276 and 310 nm, respectively. The internal standard was thymol (400 ng/mL). The retention time of propofol and internal standard was 7.0 and 3.7 minutes, respectively.
Supernatant (10 μL) was directly injected into a Zorbax Eclipse XDB-C18 column (4.6 × 150 mm, 5-μm particle size, Agilent), with the guard column packed with C18 (Phenomenex, Torrance, CA, USA). The temperature of column and autosampler were set at 25°C and 4°C, respectively. The volume of injection was 10 μL.
Name: Bing-Chen Lang, MS.
Contribution: This author helped design and conduct the study, analyze the data, and write the manuscript.
Attestation: Bing-Chen Lang has seen the original study data, reviewed the analysis of the data, approved the final manuscript, and is the author responsible for archiving the study files.
Name: Jun Yang, PhD.
Contribution: This author helped design and conduct the study, analyze the data, and write the manuscript.
Attestation: Jun Yang has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.
Name: Yu Wang, MS.
Contribution: This author helped design and conduct the study and analyze the data.
Attestation: Yu Wang has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.
Name: Yun Luo, MS.
Contribution: This author helped conduct the study.
Attestation: Yun Luo has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.
Name: Yi Kang, BS.
Contribution: This author helped conduct the study.
Attestation: Yi Kang has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.
Name: Jin Liu, MD.
Contribution: This author helped design the study and write the manuscript.
Attestation: Jin Liu has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.
Name: Wen-Sheng Zhang, MD.
Contribution: This author helped design the study and write the manuscript.
Attestation: Wen-Sheng Zhang has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.
This manuscript was handled by: Marcel E. Durieux, MD, PhD.
The authors would like to thank Ling-Hui Yang, PhD, (Laboratory of Anesthesia and Critical Care Medicine, Translational Neuroscience Center, West China Hospital, Sichuan University, Chengdu, P.R. China) for the assistance in reviewing the manuscript.