Cholesterol Conjugates of Small Interfering RNA: Linkers and Patterns of Modification

Abstract

Cholesterol siRNA conjugates attract attention because they allow the delivery of siRNA into cells without the use of transfection agents. In this study, we compared the efficacy and duration of silencing induced by cholesterol conjugates of selectively and totally modified siRNAs and their heteroduplexes of the same sequence and explored the impact of linker length between the 3' end of the sense strand of siRNA and cholesterol on the silencing activity of "light" and "heavy" modified siRNAs. All 3'-cholesterol conjugates were equally active under transfection, but the conjugate with a C3 linker was less active than those with longer linkers (C8 and C15) in a carrier-free mode. At the same time, they were significantly inferior in activity to the 5'-cholesterol conjugate. Shortening the sense strand carrying cholesterol by two nucleotides from the 3'-end did not have a significant effect on the activity of the conjugate. Replacing the antisense strand or both strands with fully modified ones had a significant effect on silencing as well as improving the duration in transfection-mediated and carrier-free modes. A significant 78% suppression of MDR1 gene expression in KB-8-5 xenograft tumors developed in mice promises an advantage from the use of fully modified siRNA cholesterol conjugates in combination chemotherapy.

Keywords: MDR1 gene; chemical modifications; cholesterol conjugate; duration of silencing; nuclease resistance; siRNA.

Figure 1

Schematic structure of the conjugates.

Figure 2

Silencing of MDR1-GFP gene expression in KB-8-5-MDR1-GFP cells by siRNA conjugates. Flow cytometry data obtained at 72 h following 100 nM conjugate transfection by Lipofectamine 2000 (A) or 5 μM conjugate delivery in a carrier-free mode (B). C—control, untreated cells. Gray bars—21/21 selectively modified siRNA, blue bars—19/21 selectively modified siRNA, green bars—19/21 siRNA with fully modified antisense strand, red bars—21/21 siRNA with fully modified antisense strand. Mean values (±SD) and statistical significance of differences from control (*—p < 0.05, **—p < 0.01), calculated using Student’s t-test from the results of three independent experiments, are shown in the figure.

Figure 3

Silencing activity of anti-MDR1 siRNAs with different modification patterns and their 5′-cholesterol conjugates. (A,B) The kinetics of the silencing of MDR1-GFP gene expression in KB-8-5-MDR1-GFP cells by siRNAs and their conjugates. Flow cytometry data obtained at 3–21 days following 100 nM conjugate transfection by Lipofectamine 2000 (A) or 5 μM conjugate delivery in a carrier-free mode (B). Mean values (±SD) and statistical significance of differences from control (*—p < 0.05), from siMDR1 (#—p < 0.05), and from Ch(6)-siMDR1 (%%—p < 0.01), calculated using Student’s t-test from the results of three independent experiments, are shown in the figure. (C) Silencing of MDR1 mRNA expression by 7.5 μg/g of cholesterol-modified siMDR1 in KB-8-5 xenograft tumor in SCID mice 4 days after IV injection (n = 3–5). Statistical significance of differences in qRT-PCR data were obtained with the Mann–Whitney U test.