A vital sugar code for ricin toxicity

Abstract

Ricin is one of the most feared bioweapons in the world due to its extreme toxicity and easy access. Since no antidote exists, it is of paramount importance to identify the pathways underlying ricin toxicity. Here, we demonstrate that the Golgi GDP-fucose transporter Slc35c1 and fucosyltransferase Fut9 are key regulators of ricin toxicity. Genetic and pharmacological inhibition of fucosylation renders diverse cell types resistant to ricin via deregulated intracellular trafficking. Importantly, cells from a patient with SLC35C1 deficiency are also resistant to ricin. Mechanistically, we confirm that reduced fucosylation leads to increased sialylation of Lewis X structures and thus masking of ricin-binding sites. Inactivation of the sialyltransferase responsible for modifications of Lewis X (St3Gal4) increases the sensitivity of cells to ricin, whereas its overexpression renders cells more resistant to the toxin. Thus, we have provided unprecedented insights into an evolutionary conserved modular sugar code that can be manipulated to control ricin toxicity.

Figure 1

Loss of Slc35c1 and Fut9 protects cells from ricin toxicity. (A) Randomly mutagenized single-cell mESC clones were exposed to ricin (2 ng/ml) for 10 days and ratios of GFP+/mCherry+ cells were measured. Isolated, ricin-resistant, mutant clones were then analyzed via inverse PCR and their integration sites were determined. All clones were found to harbor the gene trap in sense orientation at the indicated intronic sites (green arrows) of either Fut9 (asterisks) or Slc35c1 (black triangles). (B) Survival of mESCs harboring a gene trap in either Fut9 or Slc35c1 in sense (KO) or antisense (WT) orientation in the presence of the indicated concentrations of ricin. Alamar Blue cell viability assay was used to determine cell survival. Data are representative of three independent experiments. (C) Independent Fut9 and Slc35c1 mutant (KO) and reverted WT mESC sister clones were grown in the presence or absence of ricin (8 ng/ml). Representative images are shown. Scale bar, 100 μm. (D) Mixed populations of unlabeled Slc35c1 WT and mutant (KO) mESCs were exposed to different concentrations of ricin for 3 days. The amount of fucose (detected by AAL) and Lewis X (SSEA-1, CD15) expressing cells was monitored by immunofluorescence microscopy (upper panels) and flow cytometry (lower panels). Scale bar, 50 μm.

Figure 2

Slc35c1 mutant MEFs and intestinal organoids show increased resistance to ricin. (A) Slc35c1 WT and KO MEFs were cultured in the presence or absence of ricin. Cell survival was determined after 3 days by Alamar Blue. Data are shown as mean ± SD (n = 3) and are representative of three independent experiments. ^*P < 0.05, ^**P < 0.01, ^***P < 0.001 (Student's t-test). (B) Representative images of Slc35c1^+/+ and Slc35c1^−/− MEFs in the presence or absence of ricin (10 ng/ml) for 2 days. Scale bar, 50 μm. (C) WT and Slc35c1 mutant MEFs were stained for the presence of fucose-containing glycans using AAL and microscopically analyzed. Scale bar, 25 μm. (D) Murine intestinal organoids isolated from control Slc35c1^+/+ and mutant Slc35c1^−/− mice were treated with ricin (8 ng/ml) for 5 days and the numbers of surviving, intact organoids after ricin exposure were assessed. The percentage of surviving organoids compared to untreated gut organoids was determined. Data are shown as mean ± SD (n = 3). Representative data from five different experiments are shown. (E) Slc35c1^+/+ and Slc35c1^−/− intestinal organoids were treated with different doses of ricin for 5 days and stained for markers (Epcam to detect epithelial cells; AAL to detect fucosylated glycans; and UEA to specifically detect α1,3-fucosylated glycan structures) to assess organoid integrity. DAPI was used as a nuclear counterstain. Representative images are shown. Scale bar, 50 μm.

Figure 3

Loss or inhibition of fucosylation confers ricin resistance. (A) Control (Slc35c1^control) as well as mutant (Slc35c1^mut) HDFs were treated with ricin (4 ng/ml) and ℒ-fucose (10 mM) for 2 days. Their morphology and structural integrity were monitored by light microscopy. Scale bar, 50 μm. (B) Cell viability of Slc35c1^mut and Slc35c1^control HDFs, supplemented with ℒ-fucose (10 mM) for 24 h and treated with different doses of ricin for 48 h, was determined using Alamar Blue. Data are shown as mean ± SD of triplicate cultures. (C) Toxicity of ricin (after 48-h treatment) in Slc35c1^+/+ and Slc35^−/− MEFs cultured in the presence or absence of fucose (10 mM), was determined by Alamar Blue assay. Data are shown as mean ± SD of triplicate cultures. (D) Human HL-60 cells were treated with the indicated concentrations of the fucosylation inhibitor 2F-peracetyl fucose (FI-1) for 3 days, stained for SSEA-1 (CD15) and analyzed via flow cytometry. Normal SSEA-1 expression in HL-60 cells (control) as well as isotype-matched control cells (unstained) are shown. (E) HL-60 cells were pretreated with different concentrations of FI-1 and then exposed to different doses of ricin. Survival rates were assessed by Alamar Blue. Representative data of three independent experiments are shown. (F) The number of intact, surviving intestinal organoids was determined and the overall survival with and without the fucosylation inhibitor 2-deoxy-2-fluorofucose (FI-2, 250 μM) was assessed. Data are shown as mean ± SD of triplicate cultures and are representative for three different experiments showing similar results. (G) WT mouse intestinal organoids were pretreated with FI-2 and exposed to ricin (8 ng/ml) for 5 days. Representative images of organoids are shown. Scale bar, 50 μm.

Figure 4

Fucosylation controls ricin toxicity via intracellular trafficking. (A) A modified version of the ricin toxin that harbors one specific sulfation (RS1) and three mannosylation (RS2) sites in the catalytically active RTA subunit allows for quantification of the amount of intracellular toxin trafficking to the Golgi and the ER, using radioactively-labeled isotopes. (B, C) Slc35c1 WT and KO mESCs were incubated with engineered RTA (see scheme in A) to quantify ricin trafficking. The total amount of ricin (total RTA), as well as the amount of the sulfated, or sulfated and mannosylated form, was determined by western blot or autoradiography, respectively (C), and quantified (B). Data are shown as mean ± SD (n = 3). Each data set is representative of two independent experiments. ^*P < 0.05, ^**P < 0.01, ^***P < 0.001 (Student's t-test). (D) The rate of protein synthesis was determined after 3 h of ricin exposure in Slc35c1 control (WT) and Slc35c1 mutant (KO) sister mESCs, using radioactively-labeled isotopes. Data are shown as mean ± SD of triplicate cultures. (E) Addition of galactose (200 μM) or Lewis X (200 μM), but not fucose (200 μM) increased viability of ricin-treated cells. Experiments were repeated three times; representative data are shown. NS, not significant; RTA, ricin toxin A subunit; RTB, ricin toxin B subunit.

Figure 5

A vital sugar code for ricin toxicity. (A) Proposed sugar code for ricin toxicity. α1,3-fucose residues of Lewis X structures (SSEA-1⁽⁺⁾) impair α2,3-sialylation of terminal galactoses (i.e., sialyl Lewis X formation, SSEA-1⁽⁻⁾), leading to enhanced exposure of terminal galactoses and ricin binding. Absence of fucosylation allows more efficient sialylation of terminal galactoses and thus is assumed to inhibit ricin binding. GlcNAc (N-acetylglucosamine), Gal (Galactose), NeuNAc (N-Acetylneuraminic acid, one type of sialic acid). (B) Slc35c1 mutant and WT MEFs were stained for α2,3-sialic acid using MALII and counterstained with DAPI to image nuclei. Scale bar, 50 μm. (C) HL-60 cells were treated with the fucosylation inhibitor 2F-peracetyl fucose (FI-1, 100 μM) or the sialylation inhibitor 3F_ax-peracetyl Neu5Ac (SI, 250 μM) for 3 days. The amount of the fucose containing, un-sialylated epitope SSEA-1 (CD15) was determined via flow cytometry. SSEA-1 expression of vehicle-treated cells (control) as well as an isotype-matched control (isotype control) is shown. (D) HL-60 cells were pretreated with inhibitors of fucosylation (FI-1, 100 μM) or sialylation (SI, 250 μM) and exposed to different amounts of ricin thereafter. The survival of the cells was determined using Alamar Blue. Data are representative of three independent experiments. (E, F) Slc35c1 wild type (WT) and mutant (KO) mESCs (E) and MEFs (F) were treated with SI (250 μM) and their sensitivity to ricin was assessed using Alamar Blue. Data in D-F are shown as mean ± SD of triplicate cultures. Experiments were repeated three times with similar results. ^*P < 0.05, ^**P < 0.01, ^***P < 0.001; NS, not significant (Student's t-test).

Figure 6

St3Gal4 determines susceptibility or resistance to ricin. (A) Mixed populations of cells that harbor a reversible gene trap in St3Gal3, St3Gal4 or St3Gal6, as well as parental WT control cells, were subjected to ricin treatment for 3 days. The ratio of mutant (sense, GFP) to WT (antisense, mCherry/Cre) cells was assessed via flow cytometry. Data are shown as mean ± SD of triplicate cultures. (B) Human near haploid KBM7 cells were used to generate ST3GAL4 KO clones using CRISPR/Cas9. Genomic PCR and sequencing of mutagenized clones, as well as control cells, showed appropriately targeted loci (8-bp or 11-bp deletions). Exons are indicated as black boxes. The guide RNAs were designed to delete sequences in exon 6 (arrow). Two different mutant clones were generated (ST3GAL4 KO-1, KO-2). (C) ST3GAL4 mutant and control human KBM7 cells were stained for sialyl Lewis X (CD15s) and analyzed via flow cytometry. (D) ST3GAL4 KO-1 and KO-2 KBM7 cells, as well as control cells, were treated with different amounts of ricin and their viabilities were determined. Data are representative of three independent experiments. (E, F) mESCs were infected with a doxycycline-inducible expression construct coding for ST3GAL4 together with mCherry, or an empty control vector coding for mCherry only. (E) Mixed populations of infected and uninfected, as well as control cells, were treated with doxycycline and exposed to ricin (4 ng/ml) for 9 days. The percentage of mCherry+ cells was monitored over time by flow cytometry. Data are shown as mean ± SD of triplicate cultures. (F) An ST3GAL4-overexpressing mESC clone as well as an empty vector control were treated with various concentrations of ricin and their viabilities were determined using Alamar Blue staining. Data are shown as mean ± SD of triplicate cultures.