UV light-induced spatial loss of sialic acid capping using a photoactivatable sialyltransferase inhibitor

Abstract

Sialic acids cap glycans displayed on mammalian glycoproteins and glycolipids and mediate many glycan-receptor interactions. Sialoglycans play a role in diseases such as cancer and infections where they facilitate immune evasion and metastasis or serve as cellular receptors for viruses, respectively. Strategies that specifically interfere with cellular sialoglycan biosynthesis, such as sialic acid mimetics that act as metabolic sialyltransferase inhibitors, enable research into the diverse biological functions of sialoglycans. Sialylation inhibitors are also emerging as potential therapeutics for cancer, infection, and other diseases. However, sialoglycans serve many important biological functions and systemic inhibition of sialoglycan biosynthesis can have adverse effects. To enable local and inducible inhibition of sialylation, we have synthesized and characterized a caged sialyltransferase inhibitor that can be selectively activated with UV-light. A photolabile protecting group was conjugated to a known sialyltransferase inhibitor (P-SiaFNEtoc). This yielded a photoactivatable inhibitor, UV-SiaFNEtoc, that remained inactive in human cell cultures and was readily activated through radiation with 365 nm UV light. Direct and short radiation of a human embryonic kidney (HEK293) cell monolayer was well-tolerated and resulted in photoactivation of the inhibitor and subsequent spatial restricted synthesis of asialoglycans. The developed photocaged sialic acid mimetic holds the potential to locally hinder the synthesis of sialoglycans through focused treatment with UV light and may be applied to bypass the adverse effects related to systemic loss of sialylation.

Fig. 1. Graphical depiction of the UV-activatable sialylation inhibitor. After passive diffusion over the cell membrane and de-esterification, UV-SiaFNEtoc (4) can be activated with 365 nm UV light which cleaves the photolabile protecting group (PPG). Released SiaFNEtoc is converted into the active nucleotide sugar CMP-SiaFNEtoc that blocks Golgi-resident sialyltransferases and thus sialic acid capping. Without irradiation, the PPG remains intact and 4 is not recognized as substrate the sialic acid pathway. Sia = sialic acid, Gal = galactose, GlcNAc = N-acetyl glucosamine.

Scheme 1. Synthesis of caged inhibitor 4. (i) Selectfluor, DMF, H₂O, 64%; (ii) TMSNEt2, DCM, 80% (based on recovery); (iii) 2-nitrobenzyl bromide, CsF, DMF, 75%.

Fig. 2. 365 nm UV-light activation of 4. (A) Representative picture shows test tubes containing 4 irradiated with no, 312 nm, 365 nm, or 470 nm light for 900 sec. Brown coloration was observed only for 365 nm treatment. (B) Cell surface sialylation of HEK293 cells cultured for 3 days with increasing concentrations of 5 or 365 nm (300 sec) treated 4. (C) Sialylation of HEK293 cells treated for 3 days with increasing concentrations of non-irradiated 4 or 100 μM 5. (D) 5 was irradiated for 0–600 sec with a 312 nm light source and increasing concentrations were added to HEK293 cells for 3 days. Cell surface sialylation was quantified by flow cytometry using biotin-Pan-Lectenz-streptavidin Alexa Fluor 647 complexes. Bar diagrams show mean percentages Pan-lectenz binding (sialylation) ± SD normalized to untreated control cells of 2–3 representative experiments.

Fig. 3. Activation of 4 in HEK293 cells. (A) 4 was irradiated with 365 nm light for 0–900 sec prior to addition of 100 μM to HEK293 cells followed by 3 days incubation and analysis of sialylation. 100 μM of 5 was added as positive control. Representative image shows coloration of 4 after 0–900 sec treatment with 365 nm light on a 96-well flat bottom plate. (B) 100 μM 4 was added to HEK293 cell cultures for 4 hours and after washing, the cells were directly irradiated for 0–30 sec with different intensities of 365 nm light (100% intensity corresponds to maximum intensity of the light source). After UV treatment, the cells were cultured for 3 days, and cell surface binding of biotin-Pan-Lectenz-streptavidin Alexa Fluor 647 complexes was measured by flow cytometry. Bar diagrams show mean percentages lectin binding (sialylation) ± SD normalized to untreated control cells of 2–3 representative experiments.

Fig. 4. UV-induced asialoglycan synthesis in HEK293 cells. (A) Schematic representation (left) and pictures of experimental set-up (middle, right). HEK293 cells were grown as monolayer on cover slips in 6-well plates and a 365 nm UV source was installed below the sample with a paper filter to allow local irradiation of a ca. 5 × 5 mm area. HEK293 cells were pulsed with 150 μm with 4 for 4 hours followed by extensive washing to remove compound that has not been taken up. The coverslip was locally irradiated with 365 nm light for 30 seconds at maximum intensity and the cells were cultured for 48 h, fixed, and stained with PNA lectin that recognizes asialoglycans. (B) Representative confocal microscopy images show the irradiated area of the HEK293 monolayer pulsed with 4 and PNA (green) and DAPI (magenta) staining is shown at 10× magnification (left panels). The white rectangle indicates the irradiation border and a 100x magnification of that area is shown (right image).