Targeted Protein O-GlcNAcylation Using Bifunctional Small Molecules

Protein O-linked β-N-acetylglucosamine modification (O-GlcNAcylation) plays a crucial role in regulating essential cellular processes. The disruption of the homeostasis of O-GlcNAcylation has been linked to various human diseases, including cancer, diabetes, and neurodegeneration. However, there are limited chemical tools for protein- and site-specific O-GlcNAc modification, rendering the precise study of the O-GlcNAcylation challenging. To address this, we have developed heterobifunctional small molecules, named O-GlcNAcylation TArgeting Chimeras (OGTACs), which enable protein-specific O-GlcNAcylation in living cells. OGTACs promote O-GlcNAcylation of proteins such as BRD4, CK2α, and EZH2 in cellulo by recruiting FKBP12F36V-fused O-GlcNAc transferase (OGT), with temporal, magnitude, and reversible control. Overall, the OGTACs represent a promising approach for inducing protein-specific O-GlcNAcylation, thus enabling functional dissection and offering new directions for O-GlcNAc-targeting therapeutic development.


Materials and Methods:
Cell Cultures, DNA Constructs, and Reagents HEK293T cells and HeLa cells (kind gift from Prof. Sze Lok Cheng, CUHK) were maintained and cultured at 37 °C with 5% CO2 in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10 % fetal bovine serum and 1% penicillin and streptomycin.FKBP12 F36V -OGT vector was a gift from Walker's group (Harvard Medical School). 1 HTN-BRD4 plasmids were constructed by subclone from pcDNA4-TO-HA-Brd4FL (addgene: #31351), then homologous recombination (Vazyme C115) into pHTN HaloTag® CMV-neo Vector (Promega G7711).HTN-CK2α plasmids were constructed by subclone from pDB1 (CK2αlpha) (addgene: #27083), then homologous recombination into pHTN HaloTag® CMV-neo Vector.HTN-EZH2 plasmids were constructed by subclone from 3XMyc-His6-EZH2 plasmid from Prof. YangChao Chen (CUHK), then homologous recombination into pHTN HaloTag® CMV-neo Vector.Transient transfection was conducted according to the manufacture's protocol using Lip2000 TR (AboRo, RL0401).Plasmids site-directed mutagenesis was conducted according to the manufacture's protocol using Mut Express II Fast Mutagenesis Kit V2 (Vazyme, C214).Pulse-chase assay 1 X 10 5 HEK293T cells were seeded in each well of 12 well plate.After cells attached, transfection reagents and plasmids were prepared and added in DMEM.After 24 h transfection, media were replaced by fresh warm complete media with DMSO or compounds.After certain time of OGTACs treatment, media in wells were replaced by fresh warm media with 5 µM rhodamine ligand.The cells were incubated in 37 °C for 15 mins.Then, media were aspirated and 100 µL IP lysis buffer (Thermo Scientific 87788) supplemented with 20 µM OGA inhibitor (Thiamet G, Bidepharm BD571819) and 100X protease inhibitor (MedChemExpress, HY-K0010) (referred as complete IP lysis buffer in the later paragraph) was added to each well.The lysate was incubated on ice for 20 min, spun down at 14,000 RPM at 4°C for 15 mins, and the supernatant were collected for bicinchoninic acid (BCA) assay and normalized to a final 2 mg/mL.The lysates were denatured by SDS-loading buffer (Biorad #1610747) and 10 µL of each sample was loaded in SDS-PAGE.After running, in gel fluorescence at rhodamine channel (545/575 nm) was analysed by Bio-Rad ChemiDoc MP Imaging System.After image, the gel was transferred to PVDF membrane to check equal loading using anti-GAPDH antibody.

CETSA
HEK293T cells (2.5 X 10 6 cells) were seeded in 10 cm culture dish overnight for attachment.On the next day, cells were transfected with HTN-BRD4:fOGT=1:0.05 plasmid for 24 hours until ~90% confluence.After treatment with 5 µM compounds or DMSO, cells were collected and washed with PBS.Subsequently, cells were resuspended in 500 µL PBS and equally divided into 50 µL aliquots.The tubes were subjected to heat challenge at 37-53°C for 5 mins, followed by cooling on ice.Cells were lysed by three repeated freeze-thaw cycles with liquid nitrogen and water bath.Lastly, cells were centrifuged at 14000 RPM for 15 mins and supernatants were collected for Western Blot analysis.

Immunoprecipitation
For each IP reaction, 2.2 X 10 6 HEK293T cells were seed in 10 cm dish.After cells attached, transfection reagents and plasmids were prepared and added in DMEM.
After 24 h transfection, media were replaced by fresh warm complete media with DMSO or probes.For washout study, after 4 h treatment, media with DMSO or OGTAC-4 (500 nM) were replaced by fresh media with DMSO or AP1867 (50 µM) for further incubation.After certain time of OGTACs treatment, cells were collected by cold PBS and lysed by 200 µL complete IP lysis buffer.The lysate was incubated on ice for 20 min, spun down at 14,000 RPM at 4°C for 15 mins, and the supernatant were collected for bicinchoninic acid (BCA) assay and normalized to a final 2 mg/mL.For input, 20 µL of diluted sample was reacted with equal volume of 10 µM TMR ligand in IP lysis buffer and rotate gently at RT for 15 mins.The remaining samples were subjected to protein A/G magnetic beads (MedChemExpress, HY-K0202), which prebinding with protein target protein antibody.For samples for LC-MS/MS or HTN-BRD4 mutant constructs, Halo-Trap Magnetic Agarose (Proteintech, otma) was used to incubate with protein lysates.After gentle rotation at 4 °C for at least 16 h, the beads were washed with IP lysis buffer and boiled in 40 µL 2X SDS-loading buffer (Biorad #1610747) for 5 mins to elute proteins from beads.Eluted samples were directly subjected into WB analysis.To get stronger signal of RL2, we used anti-mouse-HRP (Cell signaling, #7076) as secondary antibody for RL2 and anti-rabbit-DyLight 488 (Invitrogen, #35552) or IRDye® 680RD Goat anti-Rabbit (LI-COR, 926-68071) as secondary antibody for total target proteins.To clearly illustrate the overlay between the signal of RL2 and the total target proteins, we displayed the signal of RL2 and the corresponding total proteins in two colours on the same blot using Bio-Rad Image Lab software 6.1.0.Specifically, for Figure 2, Figure 4C&D, Figure S1, Figure S3, Figure S10, Figure S13B, Figure S17B, IRDye® 680RD Goat anti-Rabbit was used for target proteins and displayed in red, while the corresponding RL2 signal was displayed in green; for other remaining blots, anti-rabbit-DyLight 488 was used for target proteins and displayed in green, while the corresponding RL2 signal was displayed in red.

Chemoenzymatic labelling and mass shift assay
For each chemoenzymatic labelling reaction, 2 X 10 5 HEK293T cells were seed to each well of 6 well plate.After cells attached, transfection reagents and plasmids were prepared and added in DMEM.After 24 h transfection, media were replaced by fresh warm complete media with DMSO or probes.

Mass Spectrometry
For MS sample preparation, cell pellets were treated and collected same as the method in immunoprecipitation.HaloTrap Magnetic Agarose (proteintech, otma) was separated as 25 µL aliquot for each reaction.The beads were washed three times with 500 µL IP lysis buffer, followed by complete removal of supernatant.Then, cell lysates were added to each tubes and incubated at 4 °C for 4 h with gentle rotation.After incubation, beads were intensively washed with IP lysis buffer and eluted in 2 X SDSloading buffer heated at 95 °C for 5 mins.The samples were submitted to SDS-PAGE and stained with Coomassie brilliant blue solution, and the bands around 250 kDa were cut out, washed three times by ddH2O, destained in 100 µL destaining buffer for 20 mins at 25 °C and this process was repeated once.The gel was then washed by 100% ACN for 15 mins and dried.Then tris(2-carboxyethyl) phosphine (TCEP) was added to reduce protein for 30 mins at 25 °C, followed by addition of iodoacetamide (IAA) solution for 30 mins reaction at 25 °C in dark.The supernatant was discarded and wash by 100% ACN, freeze dried.The sample was then digested by Trypsin solution for 20 h at 37 °C.After centrifugation, supernatant was collected and digested peptides were extracted by the following steps.Extraction solution was added and incubated for 20 mins at 25 °C; this step was repeated once and 100% ACN was added to extracted again.All extracted fractions were combined and freeze dried and desalt by C18 columns.The elutes from columns were freeze dried and injected to MS.

Chemical synthesis
NMR spectra were acquired on Bruker 400 & 500 NMR spectrometer, running at 400 MHz for 1 H and Bruker 500 NMR spectrometer at 126 MHz for 13 C respectively. 1 H NMR spectra were recorded at 400 MHz in CDCl3, using residual CHCl3 as the internal standard. 13C NMR spectra were recorded at 126 MHz in CDCl3 using residual CHCl3 as the internal standard.Reactions were monitored by thin layer chromatography and the products were purified using preparative thin layer flash chromatography (ALUGRAM Xtra, 818333).Mass spectrometry was performed on Agilent LC-MS/MS system consisted of two Agilent 1290 series pumps and auto-sampler, coupled with 6430 triple quadrupole mass spectrometer equipped with and ESI source (Agilent Technologies, Inc., Santa Clara, CA, USA).Unless otherwise noted, analytical grade solvents and commercially available reagents were used without further purification.Unless otherwise noted, chemical starting materials are purchased from Bide pharm without further purification.

Figure S12 .Figure S13 .CFigure S14 .
Figure S12.Evaluation of O-GlcNAc inducing effects at wider concentration range of OGTAC-4 on HTN-BRD4 by IP-WB method.The quantification was conducted by immunoblot signal of RL2/HTN-BRD4 as the mean ± s.e.m. of n = 2 biologically independent experiments.For 3 nM of OGTAC-4, the experiment was conducted once.

Figure S16 .Figure S17 .
Figure S16.Immunoblotting analysis of the O-GlcNAcylation inducing effect of truncated fOGT (tfOGT) on global celluar proteins.The quantification was conducted by immunoblot signal of pan-RL2/GAPDH as the mean ± s.e.m. of n = 2 biologically independent experiments.TR, transfection reagent.

Figure S18 .
Figure S18.Immunoblotting analysis of global O-GlcNAcylation level during reversibility study.The quantification was calculated by immunoblot signal of total RL2 relative to GAPDH as the mean ± s.e.m. of n = 3 biologically independent experiments.Statistical significance was calculated by multiple unpaired t-tests.ns, not significant.

Figure S19 .Figure S20 .
Figure S19.Proposed rationale for the activity difference between OGTACs and why stable ternary complex does not mean efficient induction of O-GlcNAcylation.