Synthesis and characterization of novel quick-release propofol prodrug via lactonization

Article abstract of DOI:10.1016/j.bmcl.2013.01.034

The water-soluble derivatives of propofol have gained attention as a method to increase solubility of propofol. According to the principle of lactonization, the lead compound HX0969 was synthesized first and then the pharmacological features of HX0969 wer

Full text of DOI:10.1016/j.bmcl.2013.01.034

Bioorganic & Medicinal Chemistry Letters 23 (2013) 1813–1816
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis and characterization of novel quick-release propofol
prodrug via lactonization
⇑
Jun Yang, Wang Yin, Jin Liu, Yu Wang, Cheng Zhou, Yi Kang, Wen-Sheng Zhang
Laboratory of Anesthesia and Critical Care Medicine and Translational Neuroscience Center, West China Hospital, Sichuan University, Chengdu 610041, Sichuan, China
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 27 July 2012
Revised 19 December 2012
Accepted 11 January 2013
Available online 18 January 2013
The water-soluble derivatives of propofol have gained attention as a method to increase solubility of pro-
pofol. According to the principle of lactonization, the lead compound HX0969 was synthesized first and
then the pharmacological features of HX0969 were evaluated in a comparison with those of propofol in
the SD rats. Then, HX0969 disodium phosphate monoester (HX0969W) and glycine ester trifluoroacetic
acid salt (HX101230) were synthesized, and their pharmacological features were compared with those of
LusedraÒ, which has been recognized and marketed as a water-soluble prodrug of propofol since 2008.
The results showed that HX0969 could produce an anesthetic effect within a few seconds (3.6 3.0 s)
and its therapeutic index was 4.66 in the SD rat. The pharmacodynamic characteristics of HX0969W were
similar to those of the LusedraÒ. HX101230 could still produce an anesthetic effect within 60 s in the rats
though its therapeutic index was not so high (TI = 2.96). Therefore, our study has indicated that HX0969
is a potentially useful lead compound of propofol derivative. Its rapid anesthetic effect is probably asso-
ciated with lactonization.
Keywords:
Propofol prodrug
Lactonization
Synthesis
Quick-release
Ó 2013 Elsevier Ltd. All rights reserved.
Propofol is an intravenous anesthetic used frequently in clinical
practice. It has characteristics of rapid onset, rapid metabolization
and rapid elimination. The raw propofol is insoluble in water, so
the researchers have used some pharmaceutical methods to im-
prove its water solubility for intravenous administration.1 One fat
emulsion of propofol named DiprivanÒ has become a popular for-
mulation.2 Now, the researchers have been focused on the water-
soluble prodrugs of propofol because this fat emulsion has its dis-
advantages, for example, an injection pain and a bacterial infection
as well as some other adverse effects observed in the patient’s fat
metabolism.3 Some propofol solutions have successively been cre-
ated by the medicinal chemistry methods, and LusedraÒ is the
most outstanding propofol solution, which was firstly marketed
in the United States in 2008. Its active ingredient is fospropofol
disodium (GPI15715) (Fig. 1).4 GPI15715 can quickly release pro-
pofol, but its by-product, formaldehyde, has a safety risk for infu-
sion. In addition, the intermediate, hemiacetal of propofol, is
unstable and actually not available, so it cannot be modified conve-
niently as a lead compound.
the synthesis and the characterization of this kind of propofol pro-
drug, that is, HX0969. When HX0969 was synthesized, the stability
test of HX0969 in the rat plasma was conducted. The result showed
that HX0969 could release propofol quickly in the plasma (Fig. 2)
and no decomposition was observed in the water–ethanol and
emulsion.
In order to make sure whether lactonization or normal ester
hydrolysis is the mechanism for the propofol release from
HX0969, we measured butyrolactone, a theoretical breakdown
product of HX0969, when HX0969 was decomposed in the plasma.
However, butyrolactone was also rapidly metabolized to 4-
hydroxybutyraten (GHB) in the body, and the determination of
butyrolactone could be interfered by the matrix. If the dose of
butyrolactone was not great enough, butyrolactone would be diffi-
cult to be determined accurately in the plasma.6 So, the propofol
release mechanism was verified by the medicinal chemistry meth-
od in the following procedures: compounds 1 and 2 (analogues of
HX0969) were synthesized (Fig. 3), both of which were insoluble in
water. Compounds 1, 2 and HX0969 were prepared as an emulsion.
They were compared with DiprivanÒ, and their pharmacodynamic
characteristics were evaluated in the SD rats (Table 1).
To overcome the disadvantages of LusedraÒ, design of a prodrug
that can be decomposed quickly in vivo is an effective and efficient
strategy, that is, design of a novel quick-release propofol prodrug
via lactonization.5 So, the purpose of our study was to investigate
HX0969 could produce anesthesia in the SD rat very quickly (on-
set time = 3.6 3.0 s). Compound 1 injected in a dose of 40 mg/kg
through the tail vein could induce systemic convulsions rather than
anesthesia effects for 3.8 2.8 min in all the rats, which revealed its
severe toxicity. When the dose of compound 1 was decreased to
34 mg/kg, 3 of the 8 rats did not exhibit anesthesia effects or
⇑
Corresponding author. Tel.: +86 28 85164040; fax: +86 28 85164039.
E-mail addresses: wen_sheng.zhang@yahoo.com.cn, Zhang_ws@scu.edu.cn
(W.-S. Zhang).
0960-894X/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2013.01.034

1814
J. Yang et al. / Bioorg. Med. Chem. Lett. 23 (2013) 1813–1816
Figure 1. Decomposition of GPI15715.
Figure 5. The structure of HX0969 and HX101230.
Table 2
Pharmacodynamic characteristics of HX0969W, HX101230 and GPI15715
Compound ED50
(mg/kg)
LD50 (mg/kg)
TI
OTa (min)
DTa (min)
Figure 2. The decomposition rate of HX0969 in the plasma.
GPI15715
HX0969W
34.78 4.14 163.11 19.71 4.69 2.17 0.32 31.04 7.90
44.50 5.30 288.47 34.30 6.48 2.11 0.23 27.49 5.73
HX101230 14.42 1.98
42.68 5.90
2.96 0.26 0.16 12.74 2.84
Data are presented as mean SD in all the eight measurements except TI.
a
OT and DT, respectively stand for the onset time and the duration time, that is,
the time from the injection of the drug to the disappearance of the righting reflex
and the time from the disappearance of the righting reflex to the recovery of the
righting reflex, the dose of the drug is the two-fold of ED50 (GPI15715: 70 mg/kg,
HX0969W: 89 mg/kg, HX101230: 29 mg/kg) injected via the tail vein.
Figure 3. Three propofol derivatives.
Table 1
Pharmacodynamic characteristics of HX0969 and DiprivanÒ
Compound ED50 (mg/kg) LD50 (mg/kg) TI
OTa (s)
4.67 3.6 3.0 24.36 4.87
4.12 0.00b
8.41 1.47
DTa (min)
HX0969
17.48 2.11
6.57 0.91
81.55 9.85
27.08 4.25
DiprivanÒ
Data are presented as mean SD in all the eight measurements except TI (Thera-
peutic index).
a
OT and DT, respectively stand for the onset time and the duration time, that is,
the time from the injection of the drug to the disappearance of the righting reflex
and the time from the disappearance of the righting reflex to the recovery of the
righting reflex. The dosage was the two-fold of ED50 for each of the two drugs when
OT and DT were determined, that is, HX0969: 35 mg/kg, and DiprivanÒ: 13 mg/kg,
administered via the tail vein.
b
The onset during the injection.
Figure 6. The onset time and the duration time in the rats when the two-fold of
ED50 of each drug was administered. No significant difference was found in the
onset time between HX0969 and DiprivanÒ (P = 0.568). No significant difference
was found in the onset time between HX0969W and GPI15715 (P = 0.568). No
significant difference was found in the duration time between HX0969W and
GPI15715 (P = 0.171). Both the onset time and the duration time for HX101230
were shorter than those for GPI15715 (P < 0.01 for the onset time, P < 0.001 for the
duration time). ⁄P < 0.01, #P < 0.001
seizures, but 5 of the 8 rats still exhibited systemic convulsions.
Compound 2 did not produce anesthesia effects when used in a dose
of 200 mg/kg. These findings indicated that with no structure of lact-
onization, compounds 1 and 2 would not be safe or effective in the
productionof anesthesia. So, the lactonizationwas probably a mech-
anism for the rapid release of propofol from HX0969 (Fig. 4).
Figure 4. Decomposition of HX0969.

J. Yang et al. / Bioorg. Med. Chem. Lett. 23 (2013) 1813–1816
1815
Figure 7. Decomposition of HX0969W and HX101230 in vivo.
Scheme 1. Reaction conditions: (i) Et3N; (ii) NaBH4/I2; (iii) MsCl/pyridine; (iv) Gly-Boc/DCC; (v) H3PO4/Et3N; (vi) NaOH; (vii) CF3COOH; (viii) ClCH2I/NaH.
A phosphate-linked form of HX0969 synthesized in our study
could also release propofol quickly, which was its sodium salt, a
water-soluble derivative of propofol named HX0969W. Then, the
trifluoroacetic acid salt of glycine ester of HX0969 was synthesized,
which was named HX101230 (Fig. 5). Both the compounds were
soluble in water. HX0969W, HX101230 and GPI15715 were formu-
lated into aqueous solutions and were administered to the SD rats
via the tail vein for the investigation on their pharmacodynamic
characteristics (Table 2).
Figure 8. The slow-release derivatives of propofol.
As a result, HX0969W was similar to GPI15715 in some of the
pharmacodynamic characteristics, including the duration time
(DT) (time from disappearance of righting reflex to recovery of
righting reflex) (P = 0.171). The lead compound HX0969 could pro-
duce a very rapid onset of anesthesia but no significant difference
was found when compared with DiprivanÒ (P = 0.568). HX101230
HX0969 under the action of alkaline phosphatase or esterase
in vivo; then, HX0969 released propofol by lactonization (Fig. 7).
Propofol hemisuccinate was synthesized from propofol using
succinic anhydride.7 HX0969 was synthesized by the sodium boro-
hydride–iodine system, an excellent method for reduction of car-
boxylic acid.8 HX0969W was synthesized from sulfonyl, which
was a good leaving group. GPI15715 was synthesized from propofol
using chloroiodomethane and phosphoric acid7 (Scheme 1). Chloroi-
odomethane was synthesized according to the method previously
reported.9
produced
a significantly more rapid anesthesia effect than
GPI15715 (P < 0.01) or HX0969W (P < 0.01), and its anesthesia
duration time was significantly shorter than that by GPI15715
(P < 0.001) or by HX0969W (P < 0.01), but the therapeutic index
of HX101230 was not high (TI = 2.96). The simulation results
showed that HX0969 could be modified conveniently to obtained
some water-soluble derivatives of propofol (Fig. 6). These deriva-
tives could produce anesthesia quickly in vivo. We speculated that
HX0969W and HX101230 were decomposed first to generate
The water solubility of propofol prodrugs has received enough
attention but not enough attention has been paid on the release
rate of propofol. There have been some notable failures, such as

1816
J. Yang et al. / Bioorg. Med. Chem. Lett. 23 (2013) 1813–1816
propofol disodium phosphate esters and propofol amino acid ester
derivatives10,11 (Fig. 8). These prodrugs were unable to meet clini-
cal requirements due to their slow propofol release rate. If the pro-
pofol release was not rapid enough, the anesthetic dosage had to
be increased, which inevitably led to an accumulation of the anes-
thetic. Thus, the main advantages of propofol, including the rapid
recovery, would be lost. As a drug development project, we con-
sider it a failure.
of the rat and the concentration of 4-hydroxybutyraten (GHB) in
the rat body fluid when 400–800 mg/kg of 4-hydroxybutyric acid
was administered.17 The release of 4-hydroxybutyric acid was only
24 mg/kg when administered with the two-fold of ED50 for
HX0969w in the rat, which was much lower than the dose of 4-
hydroxybutyric acid for the sleep-inducing effect on the rat re-
ported in the literature; therefore, we have believed that the by-
product of HX0969W used for anesthesia has no toxicity or a lower
toxicity.
The main reason for the slow release of propofol from the pro-
drugs obtained by the modification of the phenolic hydroxyl group
is the slow hydrolysis rate of the pro-moiety due to the steric hin-
drance with two ortho-isopropyl groups. When dynamic force is
strong enough to promote decomposition of the prodrug, the pro-
pofol release rate would become rapid. For example, when
GPI15715 was decomposed to a hemiacetal, the unstability of the
intermediate would become a strong dynamic force to induce the
rapid propofol release. HX0969 can produce anesthesia in the SD
rat by a rapid release of propofol. Since the primary hydroxyl group
of HX0969 is not severely hindered, a variety of second pro-moie-
ties such as phosphate and amino acid ester can be attached to in-
crease the water-solubility of HX0969. Such pro-moieties can be
smoothly hydrolyzed by enzymes in vivo to generate HX0969,
which is quickly converted to propofol probably via lactonization.
The breakdown product of HX0969 is butyrolactone in theory,
which will be converted to 4-hydroxybutyric acid within a short
period of time in vivo, as an endogenous substance in mammals,
4-hydroxybutyric acid has a low toxicity and can be metabolized
into the Krebs cycle.12,13 There has been no report about the acute
toxicity of butyrolactone injected through the rat tail vein. How-
ever, there has been a report about the effect of butyrolactone on
the local cerebral glucose utilization in the rat after its tail vein
injection, but no significant changes were found in the rat behavior
in the 75 or 150 mg/kg dose group.14 Additionally, there has been a
report about the effect of butyrolactone on the activity of the rat
after the administration of butyrolactone 100 or 200 mg/kg, but
no toxicity was reported.15 The release of butyrolactone was only
20 mg/kg when administered with the two-fold of ED50 for
HX0969w in the rat, which was much lower than the dose of buty-
rolactone reported in the literature; therefore, we speculated that
butyrolactone released from HX0969w might not cause any signif-
icant toxicity. Butyrolactone will be transformed to 4-hydroxy-bu-
tyric acid in a short time in vivo, the latter is a clinically used drug.
A report showed that an intravenous administration of 4-hydroxy-
butyric acid sodium 200 mg/kg induced sleep in the rat.16 Another
report was focused on the relationship between the sleep duration
HX0969 was found to be sensitive to the plasma but no quick
decomposition was detected in the emulsion or the aqueous etha-
nol solutions without enzymes (Supplementary data). But the ex-
act kinds of enzymes that decompose HX0969 or HX0969W
remain to be established.
Acknowledgment
This study was supported by a Grant from the 973 program (No.
2005CB522601), Beijing, China.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmcl.2013.01.
034.
References and notes
1. Baker, M. T.; Naguib, M. Anesthesiology 2005, 103, 860.
2. Sebel, P. S.; Lowdon, J. D. Anesthesiology 1989, 71, 260.
3. Picard, P.; Tramèr, M. R. Anesth. Analg. 2000, 90, 963.
4. Schywalsky, M.; Ihmsen, H.; Tzabazis, A.; Fechner, J.; Burak, E.; Vornov, J.;
Schwilden, H. Eur. J. Anaesthesiol. 2003, 20, 182.
5. Gomes, Paula; Vale, Nuno; Moreira, Rui Molecules 2007, 12, 2484.
6. Palmer, R. B. Toxicol. Rev. 2004, 23, 21.
7. Hwndler Sheldon, S.; Sanchez Robert, A.; Zielinski, Jan. U.S. Patent 6,254,853,
2001.
8. Bhaskar Kanth, J. V.; Mariappan, P. J. Org. Chem. 1991, 56, 5964.
9. Altabev, N.; Snith, R. D.; Suratwala, S. I. Chem. Ind. 1973, 331.
10. Banaszczyk, M. G.; Carlo, A. T.; Millan, V.; Lindsey, A.; Moss, R.; Carlo, D. J.;
Hendler, S. S. Anesth. Analg. 2002, 95, 1285.
11. Gallop, M. A.; Xu, F.; Cundy, K. C.; Sasikumar, V.; Woiwode, T. W. U.S. Patent
20,050,004,381, 2005.
12. Waszkielewicz, A.; Bojarski, J. Pol. J. Pharmacol. 2004, 56, 43.
13. Doherty, J. D.; Roth, R. H. J. Neurochem. 1978, 30, 1305.
14. Wolfson, L. I.; Sakurada, O.; Sokoloff, L. J. Neurochem. 1977, 29, 777.
15. Hampel, H.; Hapke, H. J. Arch. Int. Pharmacodyn. Ther. 1968, 171, 306.
16. Depoortere, H.; Rousseau, A.; Jalfre, M. Rev. Electroencephalogr. Neurophysiol.
Clin. 1977, 7, 153 [Article in French].
17. Roiko, S. A.; Felmlee, M. A.; Morris, M. E. Drug Metab. Dispos. 2012, 40, 212.