Lipidation of polyethylenimine-based polyplex increases serum stability of bioengineered RNAi agents and offers more consistent tumoral gene knockdown in vivo

Abstract

Recently we have established a novel approach to produce bioengineered noncoding RNA agents (BERAs) in living cells that carry target RNAi molecules (e.g., siRNA and miRNA) and thus act as "prodrugs". Using GFP-siRNA-loaded BERA (BERA/GFP-siRNA) as a model molecule, this study was to define the in vitro and in vivo knockdown efficiency of BERAs delivered by liposome-polyethylenimine nanocomplex (lipopolyplex or LPP). Compared to in vivo-jetPEI® (IVJ-PEI) and polyplex formulations, LPP offered greater protection of BERA/GFP-siRNA against degradation by serum RNases. Particle sizes and zeta potentials of LPP nanocomplex remained stable over 28 days when stored at 4 °C. Furthermore, comparable levels of BERA/GFP-siRNA were delivered by LPP and IVJ-PEI to luciferase/GFP-expressing human SK-Hep1-Luc-GFP or A549-Luc-GFP cells, which were selectively processed into target GFP-siRNA and subsequently knocked down GFP mRNA and protein levels. In addition, LPP-carried BERA/GFP-siRNA was successfully delivered into xenograft tumors and offered more consistent knockdown of tumoral GFP mRNA level in an orthotopic hepatocellular carcinoma (HCC) SK-Hep1-Luc-GFP xenograft mouse model, while IVJ-PEI formulation showed larger variation. These findings demonstrated that lipidation of polyplexes improved serum stability of biologic RNAi molecules, which was efficiently delivered to orthotopic HCC tissues to knock down target gene expression.

Keywords: Bioengineering; Cancer; Lipopolyplex; Mouse model; Orthotopic HCC; RNA delivery; siRNA.

Fig. 1.

Schematic illustration of the preparation of BERA/GFP-siRNA-loaded LPP. Cationic liposome was prepared with DOTAP and cholesterol (Step 1), and polyplex was made separately by mixing BREA/GFP-siRNA and branched polyethylenimine (bPEI10k) solutions (Step 2). BERA/GFP-siRNA-loaded LPP nanocomplex was produced after the incubation of polyplex with liposomes, followed by post-insertion of DSPE-PEG₂₀₀₀ (Step 3).

Fig. 2.

Comparison of particle size (nm) and surface Zeta potential (mV) values between polyplex (A; complex between bPEI10k and BERA/GFP-siRNA) and LPP (B; lipid coated polyplex) formulations at various N/P ratios. Values are mean ± SD (N = 3 per group).

Fig. 3.

Metabolic and storage stability of biologic siRNA-loaded LPP nanocomplex. (A) Transmittance variance of BERA/GFP-siRNA-loaded LPP and IVJ-PEI in 50% FBS did not change significantly over 24 h. (B) Urea-PAGE analyses of isolated RNAs demonstrated that LPP provided better protection of BERA/GFP-siRNA molecules against degradation by serum RNases, as compared to polyplex only or IVJ-PEI formulations. Naked RNAs in the absence and presence of FBS were used as controls. 0–24 h and 0–10 min indicates the incubation times. (C) Particle sizes and polydispersity index (PDI) as well as (D) zeta potentials of BERA/GFP-siRNA-loaded LPP were unchanged over 28 days of storage at 4 °C. Values are mean ± SD (N = 3 per group).

Fig. 4.

Delivery of BERA/GFP-siRNA and its knockdown efficiency in human carcinoma cells in vitro. (A) GFP fluorescence imaging demonstrated the knockdown of GFP by BERA/GFP-siRNA in SK-Hep1-Luc-GFP cells, delivered by Lipofectamine 3000 (LF3000), in vivo-jet PEI (IVJ) or LPP nanocomplex. This was further confirmed by measuring GFP fluorescence intensity (B) and mRNA levels (C) in SK-Hep1-Luc-GFP cells, attributable to the successful release of GFP-siRNA (D) from precursor GFP-siRNA levels (E). Likewise, delivery of BERA/GFP-siRNA and subsequent knockdown of GFP were shown in A549-Luc-GFP cells (F, G, H and I). Cells were treated with 5 nM BERA/GFP-siRNA for 72 h. GFP fluorescence intensity was measured at Ex/Em = 488/509 nm and RNA levels were determined by qPCR with gene specific primers. Values are mean ± SD of triplicate treatments (N = 3 per group). *P < 0.05, ^**P < 0.01, and ^***P < 0.001.

Fig. 5.

Delivery and effectiveness of BERA/GFP-siRNA in orthotopic HCC xenograft mouse models in vivo. (A) GFP-siRNA was successfully delivered to orthotopic HCC tumor tissues, leading to significantly lower levels of GFP mRNA. (B and C) Target siRNA molecules were also distributed to mouse lung and liver tissues. Orthotopic HCC xenograft mouse models were established after grafting SK-Hep1-Luc-GFP cells into the left liver lobe. Mice were treated i.v. with 1 mg/kg of BERA/GFP-siRNA-loaded LPP every other day for 4 times. IVJ-PEI formulated BERA/GFP-siRNA was used for comparison. GFP-siRNA and mRNA levels were determined by qPCR assays. Values are mean ± SD (N = 3 per group). *P < 0.05, ^**P < 0.01, and ^***P < 0.001.