br Statistics analysis br All data were presented as the
2.11. Statistics analysis
All data were presented as the mean standard error from at least three independent experiments. Statistical significance was tested by two-tailed Student's t-test or one-way ANOVA. Survival results were analyzed according to a Mantel-Cox curve with GraphPad Prism 7. The log-rank test was performed to determine statistically significant dif-ferences between diﬀerent groups. Statistical significance was set at *P < 0.05, and extreme significance was set at **P < 0.01 and
3. Results and discussion
3.1. Hierarchically modular assembly of TH-s-RSC
To proof our idea, we firstly prepared three functional modules. In Module A, the preset hooks of ADIBO-chol (Scheme S1 and S2) and ADIBO-DSPE (Scheme S3) were synthesized and characterized by 1H-Nuclear Magnetic Resonance (1H NMR) (Fig. S1) and Fourier Transform Infrared (FT-IR) spectroscopy. The Givinostat sensitive cationic lipid LHSSG2C14 was synthesized following our previous report . Then, the ADIBO-modified cationic liposome (AD-CL) with redox responsi-bility was prepared by thin-film dispersion method, which consisted of soy lecithin (S100), LHSSG2C14, ADIBO-Chol and ADIBO-DSPE in an optimal ratio of 15:15:9:8 (w:w:w:w). Afterwards, AD-CL was mixed with siHSP70 at the optimal N:P ratio (the molar ratio of nitrogen in AD-CL to phosphate of siHSP70) of 5:1 (mol:mol) via electrostatic in-teraction to form a condensed complex as Module A (Fig. S2). The particle size of Module A was 114 nm with a zeta potential of +15 mV, which was larger than AD-CL due to the encapsulation of siHSP70 confirmed by the agarose gel electrophoresis (Fig. 2A, Fig. S2). In ad-dition, Module B (N3-s-TRAIL, Scheme S4) was synthesized and char-acterized by matrix-assisted laser desorption/ionization time of flight mass spectrometry  (MALDI-TOF-MS, Fig. S3) and FT-IR spectro-meter (Fig. S4A), which indicated that each TRAIL molecule had con-jugated one or two N3 groups. And Module C (N3-HA, Scheme S5) was synthesized and characterized by 1H NMR (Fig. S5) and FT-IR spec-troscopy (Fig. S6).
To assemble the co-delivery vehicles, a hierarchically modular as-sembly strategy was developed based on the safe and facile copper-free click reaction between the ADIBO hooks on the surface of Module A and the azide groups of Module B and Module C. Firstly, Module A was incubated with Module B at an drug ratio of siHSP70:TRAIL at 7:1 (w:w). This optimized drug ratio showed the maximal synergistic eﬀect according to the MTT assay on the human lung adenocarcinoma epi-thelial cancer (A549) cells (Fig. S7). The resulting two-module assembly (T-s-RSC) had an average particle size of 123 nm and a zeta potential of +19 mV, which was larger and more positive than Module A due to the conjugation of positively-charged TRAIL (Fig. 2A, Fig. S8). Subse-quently, T-s-RSC whose unexploited ADIBO groups was approximately 98%, was further incubated with Module C at a molar ratio of N3 group: ADIBO group at 2:5 (mol:mol) (Fig. S9), thus obtaining the co-for-mulation (TH-s-RSC). After the decoration of HA, the particle size of TH-s-RSC was larger than T-s-RSC, and the zeta potential of TH-s-RSC was converted from positive to negative, which confirmed the suc-cessful assembly of HA (Fig. 2A). And the disappearance of the Journal of Controlled Release 304 (2019) 111–124
characteristic infrared absorption peak of azide group at 2111 cm−1 further indicated the occurrence of the copper-free click reaction among modules (Fig. S4). In addition, the transmission electron mi-croscope (TEM) examination also indicated the uniform spheroid structure of TH-s-RSC (Fig. 2B). It is noteworthy that no free siHSP70 was observed on the siRNA band during the process of hierarchically modular assembly (Fig. 2C and D), suggesting that this assembly ap-proach was favorable for fabricating the vectors of biopharmaceuticals.