Redirecting Vesicular Transport to Improve Nonviral Delivery of Molecular Cargo

Abstract

Cell engineering relies heavily on viral vectors for the delivery of molecular cargo into cells due to their superior efficiency compared to nonviral ones. However, viruses are immunogenic and expensive to manufacture, and have limited delivery capacity. Nonviral delivery approaches avoid these limitations but are currently inefficient for clinical applications. This work demonstrates that the efficiency of nonviral delivery of plasmid DNA, mRNA, Sleeping Beauty transposon, and ribonucleoprotein can be significantly enhanced through pretreatment of cells with the nondegradable sugars (NDS), such as sucrose, trehalose, and raffinose. The enhancement is mediated by the incorporation of the NDS into cell membranes, causing enlargement of lysosomes and formation of large (>500 nm) amphisome-like bodies (ALBs). The changes in subcellular structures redirect transport of cargo to ALBs rather than to lysosomes, reducing cargo degradation in cells. The data indicate that pretreatment of cells with NDS is a promising approach to improve nonviral cargo delivery in biomedical applications.

Keywords: electroporation; gene delivery; genome editing; lysosomal degradation; vesicle trafficking.

Figure 1.. pTREAT enhances pDNA electrotransfer.

(a) Schematic of the pTREAT strategy. (b) Fluorescence images showing Cy5-pDNA colocalized with lysosomes in HEK293 cells 1 h after electrotransfer. Lysosomes were labeled with LAMP1-RFP and Lysotracker green (LTG) markers; cell nuclei were stained with Hoechst 33342. Scale bar, 10 μm. (c) Effects of chloroquine (CQ) pretreatment (125 μM, 24 h) on electrotransfer of pDNA encoding EGFP into HEK293 cells. Cell viability, eTE, Expression Level, Apparent Expression Level, and Effective Expression Level were determined 24 h after electrotransfer and were normalized by data from matched control groups. (d) Pretreatment of HEK293 cells with sucrose for 24 h enhanced pDNA electrotransfer in a dose-dependent manner. (e) Pretreatment of HEK293 cells with sucrose (100 mM, 24 h) allowed reduction of the EGFP pDNA dose without compromising electrotransfer efficiency 24, 48, and 72 h post pulsing. (f) Pretreatment of HEK293 cells with sucrose allowed reduction of the electric field strength for improving cell viability without compromising electrotransfer efficiency 24 h post pulsing. Error bars, SEM; n = 4-6; * P < 0.05, Mann-Whitney U test.

Figure 2.. Sucrose pretreatment inhibits lysosome function and protects electrotransferred pDNA from degradation.

(a) Images showing pDNA distribution in untreated and sucrose-treated cells. pDNA labeled with MFP488 (green) was electrotransferred into HEK293 cells pretreated with sucrose (100 mM, 24 h) or untreated control (Ctrl). Images were acquired 10 min and 240 min after electrotransfer. The average number of green particles per cells was used as a measure of pDNA level. (b) pDNA level in untreated and sucrose-treated HEK293 cells (n > 90) at different time points post pulsing. (c) Images showing sucrose-induced enlargement of lysosomes. HEK293 cells expressing LAMP1-RFP were pretreated with sucrose; and Cy5-pDNA was electrotransferred into the cells, followed by staining with LTG. Images were acquired 60 min after electrotransfer. (d) Quantification of sucrose-induced lysosome enlargement (n > 50). Colocalization coefficient was quantified as the square of the Pearson correlation coefficient. (e) Quantification of colocalization coefficient of LAMP1 and LTG signals (n > 60). (f) Quantification of colocalization coefficient of LTG and pDNA signals (n > 70). (g) Images showing Magic Red-based cathepsin B substrate (MR-CtB) digestion. HEK293 cells pretreated with sucrose or untreated were loaded with MR-CtB (1 μM for 1 h). CtB activity was quantified as the total cellular fluorescence intensity in the red channel (n > 90). (h) Western blot of cathepsin B (CtB) and GAPDH in untreated (Ctrl) and sucrose-treated groups. Scale bars, 10 μm; error bars, SEM; * P < 0.05, Mann-Whitney U test; ns, not significant.

Figure 3.. Sucrose treatment alters autophagic and endocytic pathways.

(a) Super-resolution confocal images of LC3 and electrotransferred pDNA in HEK293 cells. TMR-pDNA was electrotransferred into cells expressing LC3-GFP 24 h after sucrose treatment (NDS). Ctrl: untreated control. Images were acquired 30 min after electrotransfer. White arrow, LC3 puncta; yellow arrow, pDNA in large LC3+ vesicles. (b) Percent of HEK293 cells with large LC3+ vesicles per image, and percent of vesicles containing electrotransferred pDNA. Error bars, SEM; n = 5-8; * P < 0.05, Mann-Whitey U test. Images of endosomal markers: (c) Rab5-GFP, (d) Rab6-GFP, (e) Rab7-GFP, and (f) Rab11-GFP in untreated and sucrose-treated HEK293 cells. (g) Super-resolution confocal images showing colocalization of LC3-GFP with Rab5-RFP, Rab6-RFP, Rab7-RFP and LAMP1-RFP in HEK293 cells, respectively. Cells were pretreated with sucrose (100 mM, 24 h). (h) Images showing colocalization of LC3-GFP and different markers in HEK293 cells. The cells were pretreated with sucrose followed by incubation with TMR-dextran (250 μg/mL, 1 h), lysotracker red (LTR) (50 nM, 5 min), or MR-CtB (1 μM, 1 h) before acquiring images. Scale bars, 10 μm.

Figure 4.. NDS-induced changes in HEK293 cells are reversible.

(a) Images showing effects of different sugar treatments on formation of large LC3+ vesicles. LC3-GFP-expressing cells were treated with the indicated sugars at 100 mM for 24 h, then were stained with LTR (50 nM, 5 min). Cell nuclei were stained with Hoechst 33342. (B-E) HEK293 cells were treated with sucrose at 100 mM for 24 h then washed and incubated with invertase (0.5 mg/mL) or sucrose (100 mM) in DMEM or with DMEM alone for 8 h. The control cells (Ctrl) were unexposed to sucrose. (b) Images showing effects of invertase on the large LC3+ vesicles. (c) Images showing effects of invertase on lysosomes in cells expressing LAMP1-RFP. After sucrose treatment, cells were incubated for 8 h in DMEM with or without invertase, and then stained with LTG (50 nM, 5 min) before image acquisition. Normalized lysosome size and colocalization coefficient of LAMP1 and LTG signals were determined by image analysis (n > 30). (d) Images showing effects of invertase treatment on cathepsin B (CtB) activity in HEK293 cells. After sucrose treatment, cells were incubated for 8 h in DMEM with or without invertase, and then incubated with MR-CtB (1 μM, 1 h). CtB activity was quantified as the total red fluorescence intensity relative to the DMEM alone group. (e) Effects of invertase treatment on electrotransfer efficiency. After sucrose treatment, cells were incubated for 8 h in DMEM with or without invertase, then pDNA for EGFP was electrotransferred into cells. eTE, Expression Level, and Apparent expression level were measured 24 h after electrotransfer. Error bars, SEM; n = 4–7; * P < 0.05, Mann-Whitney U test. Scale bars, 10 μm.

Figure 5.. ALB formation involves non-canonical autophagy and vesicle fusion.

(a) Effects of autophagy activator and inhibitor on electrotransfer. pDNA for EGFP was electrotransferred into HEK293 cells pretreated with either an autophagy activator, rapamycin (100 nM, 24 h), or inhibitor, 3-MA (5 mM, 24 h). eTE was measured 24 h after electrotransfer. (b) Images showing effects of rapamycin on ALB formation in HEK293 cells. The cells were treated with rapamycin (100 nM, 24 h) or trehalose (100 mM, 24 h), then washed and stained with LTR (50 nM, 5 min). Cell nuclei were stained using Hoechst 33342. (c) Effects of mTOR activator MHY1485 (10 μM, 24 h) pretreatment on electrotransfer of pDNA into HEK293 cells. (d) Images showing effects of MHY1485 on LC3+ and Rab7+ vesicle formation. HEK293 cells expressing LC3-GFP or Rab7-RFP were pretreated with MHY1485 (10 μM, 24 h), then washed and stained with LTR or LTG (50 nM, 5 min) before image acquisition. (e) Effects of siRNA mediated knockdown of autophagy markers on electrotransfer. HEK293 cells were treated with siRNAs targeting Atg5, Atg9, or LC3 for 48 h before pDNA for EGFP was electrotransferred into the cells. eTE and Expression Level were measured 24 h after electrotransfer and normalized using data from cells treated with scramble siRNA (Ctrl). Western blot was used to confirm gene knockdown. (f) Images showing dynamics of LC3 incorporation into ALB membranes after electrotransfer. TMR-pDNA (10 μg/mL) was electrotransferred into HEK293 cells expressing LC3-GFP; and the cells were continuously imaged under a microscope. The bottom panel shows the change in fluorescence intensity with time within the yellow box around an ALB. (g) Images showing the dynamics of LC3 incorporation into the membrane of an ALB containing dextran. HEK293 cells expressing LC3-GFP were incubated with TMR-dextran (250 μg/mL, 1 h). The bottom panel shows the change in fluorescence intensity with time within the yellow box around the ALB. Indicated time points are relative to the time when the first image was taken (0 min). (h) Effects of siRNA-mediated knockdown of Rab7 and Vps39 on electrotransfer. HEK293 cells were treated with siRNA targeting Rab7 or Vps39, then with sucrose (100 mM, 24 h). Thereafter, pDNA for EGFP was electrotransferred into the cells. eTE was measured 24 h after electrotransfer. Western blot analysis was used to determine Rab7 level; Vps39 level was measured with qPCR. (i) Effects of siRNA mediated knockdown of Rab7 and Vps39 on sucrose-induced ALB formation. HEK293 cells expressing LC3-GFP were treated with siRNAs targeting Rab7 or Vps39, then incubated with sucrose (100 mM, 24 h) before image acquisition. Error bars, SEM; n = 4-10; * P < 0.05, Mann-Whitey U test; ns, not significant.

Figure 6.. Application of pTREAT to gene delivery and genome editing.

(a) Effects of sucrose treatment on electrotransfer of mRNA. mRNA for EGFP (10 μg/mL) was electrotransferred into HCT116 cells pretreated with sucrose (100 mM, 24 h) or untreated (Ctrl). After 24 h, electrotransfer efficiency was determined by flow cytometry. Right panel, histogram showing the EGFP fluorescence intensity in control and sucrose treated groups. Histograms are representative of at least three independent experiments. (b) Effects of sucrose treatment on delivery of Sleeping Beauty (SB) transposon system for EGFP expression. The system was electrotransferred into HEK293 cells pretreated with sucrose (100 mM, 24 h) or untreated. eTE was determined 3 and 7 days after electrotransfer. (c) Effects of sucrose treatment on plasmid-based genome editing. An sgRNA targeting TRAC and a pDNA for expressing Cas9 were electrotransferred into HEK293 cells pretreated with sucrose (100 mM, 24 h) or untreated. 3 days after electrotransfer, the editing efficiency was determined by DNA sequencing. (d) Effects of sucrose treatment on RNP-based genome editing. The RNP for editing TRAC was electrotransferred into primary human T cells pretreated with sucrose or untreated. 6 days after electrotransfer, the editing efficiency was determined by DNA sequencing. (e) Results from ICE analysis confirming the outcome of TRAC editing in primary human T cells. Error bars, SEM; n = 4-10; * P < 0.05, Mann-Whitey U test; ns, not significant.