Oligonucleotide Phosphorothioates Enter Cells by Thiol-Mediated Uptake

Abstract

Oligonucleotide phosphorothioates (OPS) are DNA or RNA mimics where one phosphate oxygen is replaced by a sulfur atom. They have been shown to enter mammalian cells much more efficiently than non-modified DNA. Thus, solving one of the key challenges with oligonucleotide technology, OPS became very useful in practice, with several FDA-approved drugs on the market or in late clinical trials. However, the mechanism accounting for this facile cellular uptake is unknown. Here, we show that OPS enter cells by thiol-mediated uptake. The transient adaptive network produced by dynamic covalent pseudo-disulfide exchange is characterized in action. Inhibitors with nanomolar efficiency are provided, together with activators that reduce endosomal capture for efficient delivery of OPS into the cytosol, the site of action.

Keywords: cellular uptake; dynamic covalent chemistry; oligonucleotides; phosphorothioates.

Figure 1

CLSM images of HeLa Kyoto cells after 2 h incubation with DNA 2 (A) and OPS 1 without (B) and with (C) preincubation with 3 (10 μM); red, 1, 2 (Cy5); blue, Hoechst 33 342; scale bar: 50 μm. R=nucleobases, sequence: AGGTCCCCATACACCGAC.

Figure 2

Inhibitors of the uptake of OPS 1. A) Dose response curves for 4 (brown) and 10 (orange). I _T: Average fluorescence intensities per cell ± SEM, normalized against that without the addition of an inhibitor (I _T(0)=1), with fit to Hill equation. B) Comparison of MICs against OPS 1 and ETP 6 a (upward and rightward arrows: MIC > c _MAX; downward arrow: MIC<c _MIN), with trend line to guide the eye.

Figure 3

A) A tentative dynamic covalent exchange mechanism for the thiol‐mediated uptake of OPS. B) Normalized HPLC traces for the exchange of 13 (A=adenosyl) with 14 a (top to bottom: 14 a, 1:1, 10:1, 100:1 13/14 a and 100:1 after addition of 100 equiv. of TCEP), and mass spectrum of 15 b. Conditions: 100 μM 14 a, PBS, pH 7.4; *=reduced 14 a.

Figure 4

CLSM images of HeLa Kyoto cells after 2 h incubation at 37 °C with A) non‐modified OPS 1 and B) activated OPS I (500 nM 1, 500 μM 3, PBS, pH 7.4, 30 min, 25 °C, followed by centrifugal filtration), scale bar=50 μm. C) HCHT data showing normalized I _T, i.e., fluorescence intensity per cell I of the whole cell (T) for OPS 1 activated with 3 (500 μM, with purification after activation), 5 (500 μM, with purification), 7 (25 μM), 8 (20 μM), 10 (500 μM), 11 (50 μM) and 12 (50 μM), divided by fluorescence intensity per cell I ₀ of non‐activated OPS 1. Data are average values from > two sets of experiments ± SEM and analyzed by one‐way ANOVA compared to that without activation (* P<0.033; ** P<0.0021). D) I _T for OPS 1 (500 nM) activated with 8 (c varied) with and without purification after activation. E) Normalized fluorescence intensities in whole cells (I _T) vs. those in punctate emission (I _M) of OPS I activated with 3 (dark blue) relative to nonactivated OPS 1 (light blue). F) I _T−I _M for OPS 1 activated with 3, 5, 7, 8, and 12. G) Normalized HPLCs of 13 with 0.1, 0.2, 0.5, 1.0, 4.0 and 10 equiv. 12 (bottom to top). H) Same for 5 a with 0.0, 0.5, 1.0, 10 and 100 equiv. 17.