Bump-and-Hole Engineering Identifies Specific Substrates of Glycosyltransferases in Living Cells

Abstract

Studying posttranslational modifications classically relies on experimental strategies that oversimplify the complex biosynthetic machineries of living cells. Protein glycosylation contributes to essential biological processes, but correlating glycan structure, underlying protein, and disease-relevant biosynthetic regulation is currently elusive. Here, we engineer living cells to tag glycans with editable chemical functionalities while providing information on biosynthesis, physiological context, and glycan fine structure. We introduce a non-natural substrate biosynthetic pathway and use engineered glycosyltransferases to incorporate chemically tagged sugars into the cell surface glycome of the living cell. We apply the strategy to a particularly redundant yet disease-relevant human glycosyltransferase family, the polypeptide N-acetylgalactosaminyl transferases. This approach bestows a gain-of-chemical-functionality modification on cells, where the products of individual glycosyltransferases can be selectively characterized or manipulated to understand glycan contribution to major physiological processes.

Keywords: O-glycosylation; bioorthogonal; chemical proteomics; glycosyltransferase; isoenzyme; mucin.

Graphical abstract

Figure 1

GalNAc-T Bump-and-Hole Engineering (A) GalNAc-Ts initiate O-GalNAc glycosylation. Transfer of GalNAc to a Ser or Thr side chain is followed by downstream glycan elongation. (B) The principle of bump-and-hole engineering. Engineered double-mutant (BH) GalNAc-Ts are paired with UDP-GalNAc analogs 1–4 to chemically tag GalNAc-T substrates that can be monitored by click chemistry. (C) Overview of the steps taken in this study toward GalNAc-T bump-and-hole engineering in the living cell. PG, protecting group.

Figure 2

Bump-and-Hole Engineering Conserves Folding and Substrate Binding of GalNAc-T2 (A) Crystal structure of BH-T2 at 1.8 Å superposed with WT-T2 (PDB: 2FFU). Bound EA2 substrate peptide is cyan (sticks), Mn²⁺ is magenta (sphere), and UDP is gray (sticks). Ligands are taken from BH-T2. For superposition with WT-T2 ligands, see Figure S1A. (B) Superposition of the UDP-sugar binding site of BH-T2 and WT-T2. Electron density is rendered at 1 σ and carved at 1.6 Å. (C) Depiction of UDP-GalNAc analog 1 in a co-crystal structure with BH-T2 at 3.1 Å and UDP-GalNAc in a co-crystal structure with WT-T2 (PDB: 4D0T) (Lira-Navarrete et al., 2014), as well as WT and mutated gatekeeper residues. (D) Substrate specificities of BH-T1 and BH-T2 as determined in an in vitro glycosylation assay with detection by SAMDI-MS. For comparison with WT-GalNAc-T glycosylation, see Figure S1. Data are from one representative out of two independent experiments. See also Figure S1D and Table 1.

Figure 3

Engineered GalNAc-Ts Localize to the Golgi Compartment and Glycosylate Protein Substrates (A) Expression construct for full-length GalNAc-Ts under the control of a Dox-inducible promoter. Inverted tandem repeats (ITRs) are recognized by Sleeping Beauty transposase. WT-T2 and BH-T2 were expressed by stably transfected HepG2 cells in a Dox-inducible fashion. (B) Fluorescence microscopy of HepG2 cells stably transfected with T2 constructs, induced with 0.2 μg/mL Dox, and subsequently stained. Inset: magnification of a single cell. Scale bar, 10 μm. (C) In vitro glycosylation of proteins in a membrane fraction by full-length GalNAc-Ts using UDP-GalNAc analogs. Data are from one representative out of two independent experiments. Experiments were repeated with the membrane fraction of non-transfected cells and soluble, purified GalNAc-Ts as an enzyme source. DIC, differential interference contrast; rtTA, reverse tetracycline transcriptional activator. See also Figure S2.

Figure 4

Substrate Delivery to the Cytosol of Living Cells (A) Schematic of substrate delivery. Non-permissive steps are indicated by crossed arrows. The epimerase GALE interconverts UDP-GlcNAc and UDP-GalNAc. (B) HPAEC-PAD traces of extracts from HEK293T cells stably expressing WT-AGX1 or mut-AGX1 and fed with the indicated compounds. Dashed boxes indicate retention times of standards in separate reference runs. The product of potential epimerization of 1 by GALE, compound 8, is marked with an arrowhead. Data are of one experiment and were repeated for compound 5 in HEK293T cells transiently transfected with AGX1 constructs, as well as stably transfected K-562 cells. Insert: epimerization to 8 is suppressed in GALE-deficient K-562 cells expressing mut-AGX1 and fed with 5, but not cells carrying a control single guide RNA (sgRNA). A reference trace of compound 1 is shown. Data are of one representative out of two independent experiments. See also Figure S3.

Figure 5

Selective Bioorthogonal Labeling of the Living Cell Surface with Bump-and-Hole Engineered GalNAc-Ts (A) GalNAc-T and AGX1 co-expression construct and workflow of cell surface labeling. Red star depicts a fluorophore. (B) Labeling analysis of K-562 GALE-KO cells by flow cytometry of MB488-picolyl azide labeled and intracellular VSV-G-stained cells. Data are represented as individual values from three independent experiments, mean ± SEM of MB488 median fluorescence intensity of VSV-G-positive cells. Statistical analysis was performed by two-tailed ratio paired t test. (C) Labeling analysis by in-gel fluorescence of PNGase F-treated lysates from metabolically labeled K-562 cells. In-gel fluorescence and Coomassie staining are from one gel, and expression analyses are from one separate western blot. Data are representative of three independent experiments. (D) Schematic of glycoprotein enrichment and on-bead digest. The bifunctional molecule 10 bears an acid-labile diphenyldisiloxane moiety. (E) Exemplary MS data: mass spectrum (HCD) of a fully elaborated glycopeptide from SERPIN5A (site Thr39) and further examples from T2-specific sites from STC2 (Thr28) and APOE (Ser308). (F) Upper panel: previous data on ApoAI^220-230 glycosylation in GalNAc-T1 and T2 KO HepG2 cells (Schjoldager et al., 2015); lower panel: glycosylation sites of GalNAc-T1 and T2 uncovered by bump-and-hole engineering. A, formic acid; MFI, mean fluorescence intensity. See also Figures S4 and S5 and Data S1 and S2.