Polymeric micelles loading with ursolic acid enhancing anti-tumor effect on hepatocellular carcinoma

Ursolic acid (UA) is widely found in many dietary plants, which has been proved to be effective in cancer therapy. But unfortunately its hydrophobic property limits its clinical application. Polymer micelles (PMs) are constructed from amphiphilic block copolymers that tend to self-assemble and form the unique core-shell structure consisting of a hydrophilic corona outside and a hydrophobic inner core. PMs could entrap the hydrophobic substance into its hydrophobic inner core for solubilizing these poorly water-soluble drugs and it is widely applied as a novel nano-sized drug delivery system. This study aimed to develop the drug delivery system of UA-loaded polymer micelles (UA-PMs) to overcome the disadvantages of UA in clinical application thus enhancing antitumor effect on hepatocellular carcinoma. UA-PMs was prepared and characterized for the physicochemical properties. It was investigated the cell-growth inhibition effect of UA-PMs against the human hepatocellular carcinoma cell line HepG2 and human normal liver cell line L-02. UA-PMs was evaluated about the in vivo toxicity and the antitumor activity. We took a diblock copolymer of methoxy poly (ethylene glycol)-poly(L-lactic acid) (mPEG-PLA) as carrier material to prepare UA-PMs by the thin-film dispersion method. MTT assay and wound-healing assay were investigated to assess the inhibition effect of UA-PMs against HepG2 cells on cell-growth and cell-migration. Further, we chose KM mice for the acute toxicity experiment and assessed the antitumor effect of UA-PMs on the H22 tumor xenograft. UA-PMs could markedly inhibit the proliferation and migration of HepG2 cells. In vivo study showed that UA-PMs could significantly inhibit the growth of H22 xenograft and prolong the survival time of tumor-bearing mice. It demonstrated that UA-PMs possess great potential in liver cancer therapy and may enlarge the application of UA in clinical therapy.


Acute toxicity assessment of UA-PMs in mice 35
The acute toxicity of UA-PMs in vivo was evaluated on KM mice, and the animals were 36 randomly divided into eleven groups (n=10). The negative group received only the vehicle 37 (sterile normal saline) by the intraperitoneal injection. The other nine groups were treated 38 intraperitoneally with blank PMs (500 mg/kg) and UA, UA-PMs at various doses (125, 250, 39 500, 750 mg/kg) at one time with 5-FU as the positive control. All groups were administered 40 in a volume of 0.5 mL. In the following week, mice were keenly observed for the adverse 41 effects. The weight of the mice was also measured regularly as an indicator of acute toxicity.

42
Blood samples were collected from the orbital venous plexus of mice on day -1 (before 43 treatment), and on day 3, day 7 post-administration for liver and kidney function detection of

59
The physical stability study indicates that UA-PMs are stable with an average drug-leakage 60 rate of 2.5% during the storage within 30 days. However, the average drug-leakage rate is 61 increased to 18% at the day 60 of storage and obvious precipitates were formed at the end of 62 90-day storage with an average drug-leakage rate of 48%. Meanwhile, we also found in the 63 acceleration test, when UA-PMs were stored at the condition of 25°C ± 2°C/60% RH ± 5% 64 RH, the average drug-leakage rate is 1.5% after 5-day storage, 10% after 15-day storage.

65
After a 30-day storage, the solution became un-clear and the obvious precipitates were formed.

66
Therefore, UA-PMs should be prepared with freezing dry technology for long-term storage.

67
For media stability test, there is no bursting release and the accumulative percentage of 68 leakage from UA-PMs at the different time points of 0.5, 1, 2, 6, 12, 14 and 24 h are 0.00 ± 69 5 0.03 %, 0.00 ± 0.59 %, 19.15 ± 1.21 %, 30.06 ± 1.29 %, 41.21 ± 3.32 %, 46.13 ± 3.31 % and 70 53.22 ± 2.15 %, respectively. This result is consistent with that from the in vitro release study, 71 in which UA-PMs was not mixed with culture media containing 10% fetal bovine serum. That 72 means UA-PMs shows the same drug-release profile with or without serum presence, which 73 indicated that UA-PMs could keep stable in cell culture media, the blood-mimicking  heart-toxicity to the tested mice.

Histopathological analysis 111
The histopathological study on heart, liver, spleen, lung and kidney was conducted to observe 112 whether there are pathological changes for estimating the potential toxicity of UA-PMs. As 113 described in Figure S2, significant muscular rupture was found in the heart section of mice

120
For both lung and kidney, the histopathological result was consistent with the detection 121 about viscera index, which indicated that UA-PMs showed no toxic effects in these two 122 8 organs. Comparing the results about spleen toxicity to mice after treatment with UA 500 123 mg/kg, UA 750 mg/kg and UA-PMs 750 mg/kg, we found the result from viscera index was 124 not consistent with that from histopathological analysis, in which the spleen index was 125 obvious increased (Supplement table S1) but the histopathological section was normal 126 (Supplement Figure S2). This phenomenon may because the spleen index is not sure to be 127 relative with spleen injury but it has relationship with the immune function as mentioned in 128 discussion.

129
From these results, we can see that UA-PMs at a dose of 125, 250 and 500 mg/kg 130 showed no toxic effect on normal mice and these doses may be the better choices in our 131 further experiments, but it still needs to be confirmed via the following study about antitumor 132 activities.