Profiling nuclear cysteine ligandability and effects on nuclear localization using proximity labeling-coupled chemoproteomics

Summary

The nucleus controls cell growth and division through coordinated interactions between nuclear proteins and chromatin. Mutations that impair nuclear protein association with chromatin are implicated in numerous diseases. Covalent ligands are a promising strategy to pharmacologically target nuclear proteins such as transcription factors that lack ordered small-molecule binding pockets. To identify nuclear cysteines that are susceptible to covalent liganding, we couple proximity labeling (PL), using a histone H3.3-TurboID (His-TID) construct, with chemoproteomics. Using covalent scout fragments, KB02 and KB05, we identified ligandable cysteines on proteins involved in spindle assembly, DNA repair, and transcriptional regulation, such as Cys101 of histone acetyltransferase 1 (HAT1). Furthermore, we show that covalent fragments can affect the abundance, localization and chromatin association of nuclear proteins. Notably, the Parkinson disease protein 7 (PARK7) showed increased nuclear localization and chromatin association upon KB02 modification at Cys106. Together, this platform provides insights into targeting nuclear cysteines with covalent ligands.

Graphical Abstract

eTOC Blurb

Nuclear proteins constitute important targets for covalent therapeutics. Peng et al. couple proximity labeling and chemoproteomics to identify ligandable nuclear cysteines and chromatin-association changes driven by covalent ligands. Application of this technology reveals that covalent modification of Cys106 on PARK7 results in increased nuclear localization and chromatin association.

Introduction

The nucleus governs the regulation of cellular processes such as proliferation, differentiation, and apoptosis.¹ Precise coordination of the interactions between chromatin and transcription factors (TFs), chromatin remodelers, and other chromatin-associated proteins play a pivotal role in ensuring cell homeostasis.^2,3 Genetic mutations or external stimuli that disrupt these interactions lead to the onset of disease.⁴ Therefore, targeted therapies that modulate the activity and chromatin association of dysregulated nuclear proteins are warranted. Many chromatin-associated proteins, such as TFs, are recalcitrant to traditional non-covalent ligand discovery,⁵ primarily due to intrinsically disordered tertiary structures and a lack of small-molecule ligand-binding pockets.⁶ Covalent ligands can overcome the weak interactions present within shallow pockets by forming proximity-driven covalent linkages that stabilize protein-ligand complexes.^7,8 Recent efforts have generated covalent therapeutics targeting enzyme active sites, protein-protein interfaces, and allosteric pockets.^9-12 Additionally, covalent ligands with no direct functional effect can be modified to generate bifunctional molecules that act as proteolysis or deubiquitinase-targeting chimeras (PROTACs or DUBTACs).^13-15 Clinically approved covalent therapeutics predominantly target cysteine, but other residues, such as lysine and serine, have also been targeted.^16,17 Despite recent progress in covalent ligand development, few covalent ligand discovery programs have focused on nuclear proteins, and specifically, identifying functional ligands that alter the chromatin association of the protein targets.

Standard cysteine-targeted covalent-ligand discovery workflows utilize chemoproteomic platforms that simultaneously assess target engagement and selectivity within a complex proteome, such as the competitive isotopic tandem proteolysis-activity-based protein profiling (isoTOP-ABPP) workflow.^18-22 However, the discovery of covalent ligands for nuclear proteins using chemoproteomic methods is hampered by limited nuclear protein coverage in standard mass-spectrometry (MS) analyses of whole-cell lysates. This low coverage stems from the typically low abundance of many nuclear proteins, such as TFs, leading to the suppression of MS signals by highly abundant cytosolic proteins during MS analysis. Moreover, over 60% of nuclear proteins have multiple subcellular locations,²³ making it challenging to assess the effect of ligand treatment on the functionally relevant nuclear fraction. These challenges can be overcome by specifically enriching nuclear proteins during standard chemoproteomic workflows.

Typical nuclear proteome enrichment methods include cell fractionation by differential centrifugation, or partial membrane lysis using detergent.^24,25 However, these methods are time and labor-consuming, and lead to the loss of small soluble nuclear proteins.²⁶ Nuclear fractions can also be contaminated by other organelles like the endoplasmic reticulum (ER) due to attachment to the nuclear envelope. Additionally, the use of MS-incompatible detergents warrants extensive washing and buffer exchange steps to eliminate trace detergent. An alternative method is to fuse promiscuous cysteine-reactive probes to known nuclear-targeting sequences for direct covalent modification of nuclear proteins. For example, coupling the DNA-binding Hoechst moiety to a cysteine-reactive chloroacetyl group allowed enrichment of nuclear proteins, resulting in the identification of 58 nuclear proteins.²⁷ However, this strategy is limited by the low nuclear proteome coverage obtained, as well as the presence of the bulky nuclear-targeting group, which can bias the nuclear cysteine targets. Alternatively, proximity labeling (PL) using engineered biotin ligases enables unbiased enrichment of proteins within subcellular compartments.^28-30 The biotin ligases generate a reactive biotin intermediate, biotinoyl-5’-AMP, to biotinylate lysine residues within a 10 nm radius.³¹ The directed evolution of BioID, a mutant of E. coli biotin ligase,^28,32 to the more catalytically active TurboID, further enabled protein labeling with remarkable temporal and spatial resolution.²⁹ PL strategies have enabled proteomic profiling of diverse organelles such as mitochondria, as well as specific loci like the inner nuclear membrane and synaptic cleft.^33-35 Recently, we and others reported the combination of PL with cysteine-targeted chemoproteomics to assess the oxidation state of cysteines within the mitochondria.^36,37

Here, we report the generation of a stable histone H3.3-TurboID (His-TID) construct for enriching nuclear proteins. We demonstrate clear enrichment of nuclear proteins, including low-abundance TFs, in our PL-coupled MS analyses. For assessing cysteine ligandability, we performed competitive cysteine-labeling studies to rank cysteines by their sensitivity towards the previously reported scout fragments, KB02 and KB05,^20,38 revealing nuclear cysteines with high ligandability, including the transcription factor CEBPZ, the transcription activator MED23, and the nuclear structural protein NUMA1. Additionally, we demonstrate that Cys101 on the histone acetyl transferase 1 (HAT1) is selectively modified by the scout fragments.

Importantly, the standard competitive isoTOP-ABPP method assesses cysteine ligandability based on an occupancy-based readout that identifies high-stoichiometry ligand binding. However, high-stoichiometry binding events do not always translate to a functional outcome that modulates protein activity, localization, or stability. The function of many nuclear proteins relies on effective nuclear translocation and chromatin association. Therefore, we apply our His-TID PL method to evaluate the effects of covalent-ligand binding on protein nuclear localization and chromatin association. We demonstrate that the scout fragments can have multifaceted effects on the nuclear proteome, including modulation of protein abundance, nuclear localization and chromatin association. Notably, the Parkinson disease protein 7, PARK7, translocates to the nucleus and accumulates proximal to chromatin upon covalent liganding at Cys106. In summary, we report a platform that combines PL with chemoproteomics for assessing nuclear cysteine ligandability and the functional effects of cysteine liganding on nuclear localization and chromatin association.

Results

Generation and validation of His-TID expressing cell line

To generate a PL platform for investigating nuclear cysteine ligandability and changes in nuclear localization and chromatin association upon ligand binding, we fused TurboID to the C-terminus of nucleus-localized histone H3.3 to create a histone H3.3-TurboID (His-TID) construct. The histone core octamers (H2A, H2B, H3 and H4) assist in DNA packaging to form chromatin and contain highly post-translationally modified N-terminal tails that modulate chromatin structure and regulate gene expression.³⁹ Among the core histones, histone H3 is the most extensively modified, with over 20 characterized sites for epigenetic modifications.⁴⁰ Histone H3.3 is a variant that differs from canonical H3 by 3-4 amino acids. Unlike the canonical H3, which is expressed and incorporated into chromatin at S phase in a replication-dependent manner, H3.3 is expressed and incorporated into chromatin throughout the cell cycle.⁴¹ Additionally, incorporation of H3.3 into chromatin impairs higher-order chromatin folding, thereby generating an open chromatin structure that promotes gene activation.⁴² Given the characterized ~10 nm labeling radius of TID,³¹ our His-TID construct is expected to biotinylate nuclear proteins with a bias toward proteins that associate with chromatin at transcriptionally active loci (Figure 1A).

Figure 1. His-TID is expressed and active in the nucleus.

(A) Schematic illustration of the His-TID platform used to selectively biotinylate proteins localized within the cell nucleus.

(B) Workflow to generate polyclonal HeLa cells stably expressing His-TID (His-TID-P) via transfection and single-cell colony selection to obtain a monoclonal HeLa cell line with high levels of His-TID expression (His-TID-1).

(C) Western blot analysis of WT HeLa, His-TID-P, and His-TID-1 cell lysates (n = 1). Expression of His-TID construct was confirmed using anti-HA, and anti-GAPDH was used as a loading control.

(D) Immunofluorescence imaging of WT HeLa and His-TID-1 cells. Cell morphology and His-TID expression was monitored by bright field and anti-HA (green), respectively, and the nucleus was stained with DAPI (blue). Scale bars, 10 μm.

(E) Western blot analysis of whole cell, nucleus, chromatin crosslinking and enrichment samples using His-TID-1 cells (n = 2). Expression of His-TID construct was analyzed using anti-HA, and anti-Histone was used as a loading control.

(F) Western blot analysis of WT HeLa, His-TID-P, and His-TID-1 cells (n = 1). Cells were treated with DMSO or 500 μM biotin for 5 hours, and biotinylated proteins were visualized using strep-HRP. Anti-HA and anti-GAPDH antibodies were used to confirm the expression of His-TID and as a loading control, respectively. Biotinylation of the His-TID construct is denoted with asterisks and estimated molecular weight.

See also Figure S1 and S2.

To minimize variations in His-TID expression resulting from transient transfection, we generated a polyclonal HeLa cell line with stable His-TID expression (His-TID-P) using lentiviral transduction, which was then subjected to clonal selection to afford monoclonal populations with elevated His-TID expression (Figure 1B). Immunoblotting analysis of HA-tagged His-TID showed varied expression levels in polyclonal His-TID-P and different individual single-cell colonies (Figure S1). Single-cell colony 1 (His-TID-1) displayed the highest His-TID expression relative to His-TID-P and the other monoclonal populations (Figure 1C, Figure S1). Therefore, His-TID-1 was selected for all subsequent proteomic analyses.

Upon confirming stable His-TID expression, we demonstrated nuclear localization and successful biotinylation activity in His-TID-1 cells. In confocal immunofluorescence microscopy images, HA-tagged His-TID colocalized with the DAPI nuclear stain. In contrast, WT HeLa cells showed no appreciable fluorescence signal, confirming the lack of background in the immunofluorescence imaging (Figure 1D). To further demonstrate nuclear localization, cell lysates were fractionated into nuclear and chromatin fractions for immunoblotting analysis (Figure 1E, Figure S1). His-TID was confirmed to localize in the nucleus, and furthermore to associate with chromatin, suggesting that TID fusion does not significantly impair H3.3 function. Despite the high expression levels of His-TID, there was no growth defect observed in His-TID-1 cells (Figure S2), suggesting that chromatin incorporation of His-TID did not compromise cellular viability. Lastly, we sought to confirm biotinylation by His-TID and identify optimal conditions for maximal biotinylation signal. His-TID-1, His-TID-P, and WT HeLa cells were treated with varying concentrations of biotin (0-1 mM) for 1 hour, or 500 μM biotin for different time periods (1-5 hour) (Figure S1). Optimal biotinylation conditions were determined to be 500 μM biotin treatment for 5 hours (Figure 1F). As expected, the higher levels of His-TID expression in the His-TID-1 cells relative to His-TID-P translated to a higher biotinylation signal in this monoclonal cell line.

Identification of His-TID enriched proteins by mass spectrometry

After confirming expression, nuclear localization, and successful protein biotinylation using His-TID, we sought to identify the biotinylated proteins in the His-TID-1 cell line. Briefly, His-TID-1 cells were treated with either 500 μM biotin ((+) biotin) or DMSO ((−) biotin) for 5 hours, and the resulting cell lysates were incubated with streptavidin resin to enrich biotinylated proteins. After dithiothreitol (DTT) reduction, iodoacetamide (IA) capping of cysteines, and subsequent on-bead trypsin digestion, the resulting tryptic peptides were analyzed by LC-MS/MS to generate a list of protein and peptide identifications (Figure 2A; Data S1). A total of 523 and 2855 proteins were identified in His-TID-1 (−) and (+) biotin samples, respectively (Figure 2B). The significantly higher number of identifications in the (+) biotin sample confirmed successful His-TID-mediated biotinylation in the presence of biotin. Protein identifications in the (−) biotin sample likely originate from endogenously biotinylated proteins, as well as abundant proteins that non-specifically adhere to the streptavidin beads. Additionally, low levels of TID-mediated biotinylation in the (−) biotin sample is expected due to trace amounts of endogenous biotin in cells. The identified proteins were further analyzed by mining the Uniprot database for protein subcellular locations^43,44 and the human TF catalog⁴⁵ for the presence of TFs. In the His-TID-1 (+) biotin sample, 2130 proteins were annotated as nuclear-localized, including 335 TFs. In contrast, the His-TID-1 (−) biotin sample afforded 374 nuclear proteins, with only 22 TFs (Figure 2C). Lastly, the subcellular distribution of spectral counts for (+) biotin sample was ~86% nuclear and ~10% cytosolic, compared to ~38% nuclear and ~55% cytosolic in the (−) biotin samples (Figure 2D).

Figure 2. MS analysis confirms that nuclear proteins and peptides are enriched by His-TID proximity labeling.

(A) Schematic illustration of the workflow for proximity labeling and MS analysis.

(B) Venn diagram comparing the total proteins identified using His-TID-1 cells treated with 500 μM biotin (orange) and DMSO (blue) for 5 hours. Protein counts were generated as a sum of all identified proteins from 2 technical replicates of 2 biological replicates (n = 4).

(C) Quantification of proteins and cysteine-containing peptides assigned to total proteins, nuclear proteins, and transcription factors identified in the MS analysis of His-TID-1 (−)/(+) biotin samples.

(D) Categorization of spectral counts assigned to nuclear, cytoplasmic, mitochondrial, and ER proteins from the MS analysis of His-TID-1 (−)/(+) biotin samples.

(E) Panther GO-Slim annotation set analysis of the TOP 10 overrepresented molecular functions for proteins identified in His-TID-1 (−)/(+) biotin samples. Nuclear molecular functions are highlighted in red.

See Data S1 for detailed information.

Since we were particularly interested in identifying potential reactive and ligandable cysteines on nuclear proteins, we focused on the nuclear cysteines identified in our proteomic data. Typical isoTOP-ABPP analyses use a cysteine-reactive iodoacetamide-alkyne (IA-alkyne) probe to selectively enrich cysteine-containing peptides from a whole-cell tryptic digest.^18,19 However, due to the limited subset of proteins biotinylated by His-TID, we expected to identify a substantial number of nuclear cysteine-containing peptides directly from MS analysis of the tryptic digest. In a prior study using a mitochondria-localized TID construct, a large number of cysteine-containing peptides were identified without further enrichment for cysteines.³⁶ We therefore parsed the total list of identified peptides from the His-TID-enriched tryptic digest for cysteine-containing peptides. In total, 3784 cysteine-containing peptides were identified in the (+) biotin sample, with 3310 located on nuclear proteins, including 442 from TFs. In contrast, the (−) biotin sample only contained 332 cysteine-containing peptides, with 252 and 11 originating from nuclear proteins and TFs, respectively (Figure 2C).

Furthermore, Gene ontology (GO) enrichment analysis of molecular functions using the PANTHER GO-Slim^46-49 annotation set demonstrated that proteins in the His-TID-1 (+) biotin sample were enriched for nuclear functions, such as basal transcription machinery binding, snoRNA binding, and helicase activity (Figure 2E, Data S1). None of these nuclear-associated molecular functions were significantly enriched in the His-TID-1 (−) biotin sample (Figure 2E, Data S1). Together, these data confirm the enrichment of nuclear proteins, including TFs, using the His-TID-1 cell line.

We compared the coverage that we obtained to that using a nuclear localization sequence (NLS)-appended TID construct (NLS-TID) reported previously.²⁹ Use of NLS-TID identified 1455 proteins, including 1117 nuclear proteins and 182 TFs. Our His-TID analysis afforded higher coverage of both nuclear proteins and TFs. Additionally, we compared our cysteine coverage on nuclear proteins and TFs to the original isoTOP-ABPP analysis of whole-cell lysates. In isoTOP-ABPP analyses of various cancer cell lines, only 604 nuclear cysteines and 38 TF cysteines were identified.¹⁹ This is significantly lower than the coverage obtained in our His-TID cell line analysis. Lastly, cysteine identifications in whole-cell lysate isoTOP-ABPP studies can be attributed to different subcellular pools of a given protein , whereas cysteines identified using His-TID enrichment are specifically from the nuclear-localized pool. Together, our initial proteomic analysis of His-TID-1 cells confirm that His-TID enrichment provides improved coverage of nuclear-localized cysteines.

Assessing nuclear cysteine ligandability using scout fragments

Upon confirming the successful identification of nuclear cysteines using the His-TID-1 cell line, we sought to evaluate nuclear cysteine susceptibility to covalent modification using two previously reported cysteine-reactive scout fragments, KB02 and KB05.²⁰ These scout fragments have been shown to capture a large fraction of ligandable cysteines within a cellular proteome.

We assessed cysteine liganding by the scout fragments in cell lysates from biotin-treated His-TID cells. Biotinylated cell lysates instead of live cells were treated with the scout fragments to avoid changes in nuclear localization and subsequent biotinylation levels induced by ligand binding. Cysteine liganding was indirectly monitored in a competitive manner, whereby cysteine modification by the scout fragment results in reduced labeling by a highly reactive cysteine-capping agent, in this case N-ethylmaleimide (NEM). Briefly, His-TID cells were treated with biotin, lysed, and the resulting cell lysates exposed to either DMSO, or 100 μM KB02 or KB05. The resulting free thiols in DMSO and fragment-labeled lysates were capped with light and heavy NEM, respectively. These lysates were then combined, incubated with streptavidin resin to enrich nuclear proteins, and subjected to trypsin digestion. As before, MS analysis of the trypsin digests afforded protein and peptide identifications that were filtered to generate a list of NEM-labeled cysteine-containing peptides. The intensities of the light and heavy-NEM labeled peptide elution peaks were quantified to provide light:heavy (L:H) ratios. Cysteine residues that were not targeted by KB02 or KB05 would display L:H ratios of ~1, signifying equal NEM labeling of that cysteine in the DMSO and fragment-treated samples. In contrast, a cysteine that is modified by the scout fragments would generate L:H ratios >1, indicating a reduction in NEM labeling in the fragment-treated samples. The higher the L:H ratio, the greater the stoichiometry of the scout-fragment modification (Figure 3A, 3B).

Figure 3. MS analysis of His-TID lysates treated with the scout fragments KB02 and KB05 identifies liganding sites on nuclear proteins.

(A) Schematic illustration of the workflow to profile nuclear cysteine ligandability.

(B) Chemical structures of the scout fragments, KB02 and KB05, used for ligandability screening.

(C) Heatmap comparing L:H ratios of nuclear cysteine residues identified in His-TID-1 samples treated with DMSO, KB02, or KB05. Any L:H ratios > 5 were capped at 5, and cysteines with unidentified ratios were colored in gray. Data are presented as the mean value of 2 technical replicates of 2 biological replicates (n = 4).

(D) Heatmap of the 10 most liganded nuclear cysteine residues by KB02, or KB05. A gradient color scale represents the corresponding colors for L:H ratios from 1 to 5. Any L:H ratios > 5 were capped at 5, and cysteines with unidentified ratios were colored in gray.

(E) Fraction of total identified proteins that were found in DrugBank.

(F) Fraction of identified liganded proteins that were found in DrugBank.

(G) Functional classification of identified liganded proteins.

(H) Structure of histone acetyltransferase 1 (HAT1) (PDB = 2P0W) with binding sites of acetyl CoA and histone H4 tail, and the KB02-liganded cysteine Cys101 and non-liganded Cys27.

(I) Extracted ion chromatograms of the HAT1 Cys101-containing peptide from DMSO, KB02, and KB05 labeled samples (light; red trace, heavy; blue trace), with L:H ratios indicated below.

(J) Intact protein LC-MS analysis of WT and C101S HAT1 (aa20-341) labeled with DMSO (red) or KB02 (blue).

See also Data S1 and Figure S3.

Each individual fragment-treated MS analysis yielded >2000 cysteine-containing peptides, which were filtered for L:H ratios with low statistical deviation in at least two replicates. An additional filtering step was performed to obtain a final list of cysteine-containing peptides with L:H ratios of ~1 in the control sample comparing two DMSO-treated cell lysates (Data S1). After data processing, the KB02 treatment dataset contained 639 cysteines derived from nuclear proteins, including 61 from TFs. Of these identified cysteines, 31 cysteines (~5%) displayed >50% modification by KB02 at 100 μM. Similarly, for KB05 treatment, 672 nuclear cysteines were identified, including 32 cysteines from TFs, and 36 (~5%) were shown to be >50% liganded by KB05 at 100 μM (Figure 3C). Importantly, the subset of cysteines susceptible to the covalent fragments varied between KB02 and KB05, confirming that the structure of the fragment mediates target selectivity. The selectivity of cysteines targeted by KB02 and KB05 is further illustrated by focusing on the 10 most liganded nuclear cysteines by each fragment (Figure 3D). Only Cys656 on SURP and G-patch domain-containing protein 2 (SUGP2), and Cys373 on DNA-dependent protein kinase catalytic subunit (PRKDC) were equally sensitive to both KB02 and KB05, while the remaining cysteines displayed preference for one of the two fragments. Selective targets of KB02 include Cys282 on transcription factor CCAAT/enhancer-binding protein zeta (CEBPZ), which showed 80% labeling (L:H = 5.3). CEBPZ transcriptionally stimulates the expression of heat shock protein 70 (HSP70),⁵⁰ and aberrant methylation of the CEBPZ promotor leads to decreased CEBPZ expression in acute myeloid leukemia (AML).⁵¹ Similarly, Cys961 on nuclear mitotic apparatus 1 (NUMA1) also displayed 80% labeling by KB02 (L:H = 5.1). NUMA1 is known to stabilize the mRNA of ubiquitin-conjugating enzyme E2 C (UBE2C), which results in activation of the Wnt/β-catenin signaling pathway in triple-negative breast cancer (TNBC).⁵² Selective targets of KB05 include Cys40 on mediator of RNA polymerase II transcription subunit 23 (MED23), a downstream regulator of the Ras-MAPK pathway, which showed over 70% labeling (L:H = 3.3). Levels of MED23 are increased in patients with KRAS-mutant lung cancer, and MED23 depletion increases survival in a mouse xenograft model of lung cancer.⁵³ Focusing on the 10 most liganded cysteines by each fragment (Figure 3D), less than half were identified in a previous competitive isoTOP-ABPP analysis using scout fragment-treated whole-cell lysates, demonstrating the expanded profiling of nuclear targets using His-TID enrichment.²⁰

In total, our proteomic coverage in the KB02 and KB05 liganding studies included 704 cysteine residues from 398 nuclear proteins. Of these nuclear proteins, only 87 (~22%) were listed in the DrugBank database, which collates the targets of ~15,000 clinically approved and discovery-phase small-molecule and biological drugs (Figure 3E, Data S1).⁵⁴ The total number of ligandable nuclear proteins, defined as containing a cysteine with >50% occupancy by either KB02 or KB05 at 100uM, was 51 (~13% of the total quantified proteins). Of these 51 liganded proteins, only 8 (~16%) were found in the DrugBank database (Figure 3F). The 51 liganded proteins displayed diverse functional categorization, with high representation of transcriptional regulators (~31%) and RNA processing proteins (~20%) (Figure 3G). Together, these data demonstrate that profiling cysteine ligandability within the His-TID enriched proteome can provide access to protein classes that are currently underrepresented in the DrugBank.

We further characterized labeling of the histone acetyltransferase 1 (HAT1) by the scout fragments. HAT1 acetylates histone H4 tails to enhance TF binding and activate gene transcription.⁵⁵ Overexpression of HAT1 is characteristic of many cancer types and is associated with enhanced expression of oncogenes.^56,57 HAT1 contains 7 cysteine residues, two of which (Cys101 and Cys27) were identified in our scout-fragment analysis (Figure 3H). Cys101 displayed ~80% and ~40% liganding by KB02 and KB05, respectively (Figure 3I). In contrast, Cys27 was not covalently modified by either of the fragments. To evaluate fragment labeling of HAT1 in vitro, truncated WT and C101S HAT1 (amino acids 20-341) were recombinantly expressed and purified from E. coli for intact protein LC/MS analysis. Upon KB02 treatment, WT HAT1 showed an expected mass shift of 203 Da, while C101S HAT1 did not display a shift in the observed mass (Figure 3J). This in vitro characterization confirms covalent modification of HAT1 by KB02. Importantly, despite the presence of other cysteines on this protein, KB02 labeling is highly selective for Cys101 as confirmed by the lack of modification observed for the C101S mutant.

Cys101 on HAT1 is located distal to the acetyl-CoA binding site and the histone H4 tail interacting domain (Figure 3H). Therefore, it is unlikely that ligand binding to Cys101 would directly impact the catalytic activity of HAT1. Accordingly, an activity assay that assessed HAT1-catalyzed transfer of the acetyl group from acetyl-CoA to a histone-tail peptide showed that KB02 treatment did not affect HAT1 activity (Figure S3). Similarly, mutation of Cys101 also did not affect catalytic activity. Together, our data confirm the high selectivity of KB02 for HAT1 Cys101 over other cysteines on the protein, however, binding of KB02 does not affect HAT1 acetylation activity in vitro.

Identifying proteins that display changes in chromatin association upon covalent fragment treatment

As demonstrated by KB02 binding to HAT1 Cys101, a high-stoichiometry liganding event does not always impact protein function. For many nuclear proteins, nuclear localization and chromatin association are essential to cellular function, and covalent liganding of cysteine has been shown to modulate chromatin binding on several nuclear proteins. For example, methylene quinuclidinone (MQ) covalently modifies cysteines on the p53 tumor suppressor and increases mutant p53 association with chromatin.⁵⁸ In contrast, covalent targeting of a cysteine on c-MYC disrupts chromatin association of the MYC/MAX dimer.⁵⁹ Therefore, we aimed to determine if the scout fragments could either directly or indirectly affect nuclear localization and protein association with chromatin by measuring the degree of biotinylation by His-TID (Figure 4A).

Figure 4. Treatment of cells with scout fragments leads to changes in nuclear localization and chromatin association.

(A) Illustration of potential effects on chromatin association mediated by ligand binding.

(B) Schematic representation of the PL-coupled TMT workflow to profile changes in biotinylation upon in situ ligand treatment.

(C) Volcano plots of changes in protein biotinylation upon KB02 or KB05 versus DMSO treatment. Proteins with significant changes in biotinylation are highlighted (log₂Fold Change > 0.5 (red) or < −0.5 (blue), p < 0.05). Data were presented as the mean value of 3 technical replicates (n = 3).

(D) Western blot analysis of the whole cell and streptavidin-bound fractions of BBX, SQSTM1, and PIR using His-TID-1 cells treated with DMSO, KB02, or KB05 prior to biotinylation (n = 3). Anti-HA was used as a loading control.

(E) Western blot analysis of BBX, SQSTM1, and PIR using cytosolic, nuclear, and chromatin fractions of HeLa cells treated with DMSO, KB02, or KB05 (n = 2 for cytosolic and nuclear fractions, n = 3 for chromatin fractions). Anti-Histone was used as a loading control.

See also Data S1 and Figure S4.

To monitor changes in protein biotinylation upon scout fragment treatment, His-TID-1 cells were treated with DMSO, KB02 or KB05 prior to biotin addition. Western-blot analysis demonstrated consistent biotinylation levels across the three samples with some noticeable changes in the banding patterns (Figure S4). To identify and quantify biotinylated proteins, we coupled our His-TID enrichment with TMT labeling for quantitative MS analysis. Briefly, His-TID-1 cells were incubated with 30 μM of KB02, KB05, or DMSO for 1 hour prior to 5-hour biotinylation. After streptavidin enrichment and on-bead trypsin digestion, the tryptic peptides were subjected to TMTsixplex isobaric labeling and subsequent LC-MS/MS analysis (Figure 4B). MS analysis quantified biotinylation levels on 1657 nuclear proteins, including 296 TFs (Figure 4C, Data S1). In general, the vast majority of proteins showed no change in biotinylation upon fragment treatment. KB02 treatment resulted in 16 and 7 proteins displaying significant increases and decreases in biotinylation, respectively. Similarly, KB05 treatment significantly increased and decreased biotinylation in 13 and 8 proteins, respectively.

To further verify the MS data, we focused on three nuclear proteins, transcriptional coregulator Pirin (PIR), autophogy-regulating protein Sequestosome 1 (SQSTM1), and HMG box transcription factor BBX (BBX). KB02 increased PIR biotinylation, whereas KB05 increased and decreased biotinylation of SQSTM1 and BBX, respectively. To verify these changes, biotinylated proteins from DMSO, KB02 or KB05-treated cells were enriched on streptavidin beadsand subjected to immunoblotting analysis. Consistent with the MS data, immunoblotting analysis confirmed increased biotinylation of PIR upon KB02 treatment and increased and decreased biotinylation of SQSTM1 and BBX, respectively, upon KB05 treatment (Figure 4D, Figure S4). Concurrent immunoblotting analysis of the lysates prior to streptavidin enrichment indicated that total protein levels of PIR and SQSTM1 were not affected by KB02 or KB05 treatment. Therefore, the observed changes in biotinylation appear to be driven by increased nuclear localization or chromatin association of these proteins upon fragment treatment. In contrast, total protein levels of BBX displayed a significant decrease, indicating that the observed decrease in biotinylation is primarily driven by a decrease in protein abundance of BBX upon KB05 treatment.

The degree of biotinylation by His-TID provides an indirect measure of nuclear localization and chromatin association. Therefore, we sought to directly monitor protein levels using nuclear isolation and chromatin enrichment prior to immunoblotting (Figure 4E, Figure S4). Nuclear fractions were purified using a mild detergent NP-40 to lyse cell membranes while keeping the nuclear envelope intact,^25,60 (Figure S4). Chromatin enrichment was performed by formaldehyde crosslinking, cell nuclei extraction, lysis, and subsequent chromatin precipitation.⁶¹ The chromatin crosslinking distance of formaldehyde is only 0.2 nm, such that only proteins in close proximity to DNA will be crosslinked.⁶² Immunoblotting analysis of PIR indicated an increase in the nuclear fraction upon KB02 treatment. No change in the chromatin-associated fraction was observed, indicating that the increased biotinylation signal for PIR was driven by an increase in nuclear localization and not chromatin association. In contrast, immunoblotting studies of SQSTM1 displayed no change in the nuclear fraction but showed a marked increase in the chromatin fraction upon KB05 treatment. Therefore, the increased biotinylation of SQSTM1 was driven by an increase in chromatin association. Lastly, immunoblotting studies of BBX confirmed the overall decrease in cellular BBX levels upon KB05 treatment, since the BBX signal from the cytosolic, nuclear, and chromatin fractions all decreased upon KB05 treatment. Together, these data confirm that KB02 and KB05 can affect overall protein levels, nuclear localization, or chromatin association. However, since these proteins were not identified in our ligandability study, further investigation is required to determine if these observed changes are due to direct fragment modification of these proteins, or a downstream effect of ligand binding.

Covalent modification of PARK7 promotes nuclear translocation

The protein that displayed the highest magnitude change in biotinylation levels was PARK7, with ~2.8-fold increase upon KB02 treatment (Figure 4C). PARK7 is a multifunctional protein that primarily serves as a cellular antioxidant due to a highly redoxsensitive cysteine (Cys106), which scavenges reactive oxygen species in the cell.⁶³ Mutations that inactivate PARK7 are associated with autosomal recessive Parkinson’s disease.⁶⁴ PARK7 resides in the cytosol under basal conditions, but can translocate to both mitochondria and the nucleus upon oxidative stress.⁶⁵ Mitochondrial PARK7 provides protection from mitophagy and maintains mitochondrial function under oxidative stress,^66,67 while nuclear PARK7 serves as a transcriptional co-activator that protects against neuronal apoptosis.⁶⁸ To corroborate our MS data, we performed immunoblotting studies to monitor PARK7 levels in whole-cell, nuclear and chromatin-associated fractions upon KB02 treatment. KB02 treatment does not affect overall cellular levels of PARK7 but significantly elevates PARK7 in both nuclear and chromatin fractions (Figure 5A, Figure S5). To test the dynamics of PARK7 nuclear translocation upon KB02 treatment, HeLa cells were treated with KB02 for 1 hour in serum-free media prior to being replenished with fresh media. The cells were then harvested at different time points and analyzed by immunoblotting after nuclear and chromatin fractionation (Figure 5B, Figure S5). While the overall protein levels of PARK7 remained constant over the 10-hour time period (Figure S5), nuclear and chromatin-associated PARK7 levels displayed an observable increase within 3 hours and 20 mins after the 1-hour KB02 treatment, respectively. Therefore, KB02-induced nuclear translocation and chromatin association of PARK7 occurs rapidly after KB02 treatment and is sustained after removal of KB02.

Figure 5. Covalent modification of PARK7 promotes nuclear translocation and chromatin association.

(A) Western blot analysis of PARK7 using cytosolic, nuclear, and chromatin-associated fractions of HeLa cells treated with DMSO, KB02, or KB05 (n = 2 for cytosolic and nuclear fractions, n = 3 for chromatin fractions). Anti-Histone was used as a loading control.

(B) Western blot analysis of PARK7 using nuclear and chromatin-associated fractions of HeLa cells harvested at different time points upon DMSO or KB02 treatment (n = 3). Anti-Histone was used as a loading control.

(C) Structure of Parkinson disease protein 7 (PARK7) with Cys106 (red) and surrounding amino acids (blue). (PDB = 2OR3).

(D) Western blot analysis of WT and C106A PARK7 levels in whole cell and chromatin-associated fractions in HeLa cells treated with DMSO or KB02 (n = 3). PARK7 expression was analyzed using anti-FLAG, while anti-Histone and anti-GAPDH were used as a loading control for chromatin crosslinking and whole cell samples, respectively.

(E) Chemical structures of KB02-alkyne and the non-electrophilic analog of KB02.

(F) Rhodamine fluorescence gel analysis of WT and C106A PARK7 sequentially treated with KB02-alkyne and rhodamine azide (n = 2). Anti-FLAG western blot was provided as loading controls.

(G) Rhodamine fluorescence gel analysis of WT PARK7 treated with increasing amount of non-electrophilic analog KB02 before KB02-alkyne labeling (n = 2). Anti-FLAG was provided as loading controls.

See also Figure S5.

As mentioned previously, the changes in nuclear localization and chromatin association observed upon fragment treatment could result from direct liganding of cysteines on the target protein or a downstream consequence of a liganding event. PARK7 contains 3 cysteine residues, Cys46, Cys53, and the redox-active Cys106. Previous proteome-wide liganding studies identified Cys106 on PARK7 (Figure 5C) as highly susceptible to KB02 labeling.²⁰ To confirm that PARK7 Cys106 is necessary for KB02-induced nuclear translocation, HeLa cells expressing WT or C106A PARK7 were treated with DMSO or KB02. After chromatin crosslinking and enrichment, WT PARK7 displayed a significant increase in chromatin association after KB02 treatment, whilst C106A PARK7 displayed no change (Figure 5D, Figure S5). To demonstrate the direct labeling of Cys106 by KB02, HeLa cells expressing WT or C106A PARK7 were treated with KB02-alkyne³⁸ (Figure 5E), followed by immunoprecipitation (IP) to enrich for WT and C106A PARK7. The enriched proteins were then subjected to copper (I)-catalyzed azide-alkyne cycloaddition (CuAAC) with rhodamine azide, for in-gel fluorescence analysis. WT PARK7 exhibited an intense fluorescent band corresponding to KB02-alkyne modified PARK7, while C106A PARK7 displayed negligible fluorescence (Figure 5F, Figure S5). To evaluate if non-covalent interactions contribute to KB02 modification of PARK7, WT PARK7 was incubated with increasing concentrations of a non-electrophilic analog of KB02, 6-methoxy-1,2,3,4-tetrahydroquinoline (Figure 5E), for 1 hour, followed by 10-min labeling with KB02-alkyne. A concentration-dependent decrease in KB02-alkyne labeling was observed in the presence of the non-electrophilic analog (Figure 5G, Figure S5), confirming the contribution of non-covalent interactions to KB02 binding. Together, our data demonstrates increased nuclear translocation and chromatin association of PARK7 upon KB02 treatment. This KB02-induced translocation of PARK7 is dependent on Cys106, as the C106A mutant does not show nuclear translocation. Cys106 of PARK7 is directly modified by KB02, resulting in increased nuclear localization and chromatin association.

Discussion

Covalent ligands display distinct benefits over non-covalent ligands, including increased duration of action, and ability to target sites that are intractable to non-covalent binders.⁹ Recent advances in chemoproteomic platforms have identified ligandable sites containing nucleophilic amino acids such as cysteine, lysine, and tyrosine.^16,17 Nuclear proteins, such as transcription factors, chromatin-modifying enzymes, and DNA damage-repair machinery, are important therapeutic targets for covalent ligand discovery. However, investigations into the ligandability of nuclear proteins are impeded by the relatively low abundance and the diverse subcellular localizations of these proteins. To specifically interrogate the ligandability of nuclear cysteine residues, we coupled chemoproteomic profiling with PL. Specifically, we generated a His-TID construct that biotinylates proteins located proximal to chromatin. MS studies of the biotinylated proteome identified 2130 nuclear proteins with 3310 cysteine-containing peptides. Notably, we obtained substantial coverage of TFs that are underrepresented in whole-cell lysate analyses, with 442 cysteine residues from 335 TFs. We ranked ~650 nuclear cysteines by sensitivity to each of the two scout fragments, and identified liganding events on diverse nuclear proteins, including Cys282 on the transcription factor CEBPZ, Cys961 on the mitotic apparatus protein NUMA1, and Cys40 on the transcriptional co-activator MED23. We confirmed that Cys101 on the chromatin-modifying enzyme HAT1 is labeled by KB02 with high selectivity. Although covalent liganding did not appear to modulate the in vitro catalytic activity of HAT1, if a selective ligand for Cys101 on HAT1 can be developed, it can be further elaborated to generate bifunctional molecules for targeted protein degradation. The identification of non-functional liganding sites on nuclear proteins can therefore have therapeutic potential through development of bifunctional molecules.

Focusing on liganding events that modulate protein function, we utilized our His-TID construct to identify changes in nuclear localization and chromatin association upon exposure of cells to KB02 and KB05. By coupling our PL platform with TMT labeling, we mapped the biotinylation changes of 1657 nuclear proteins after ligand treatment. Further interrogation demonstrated that covalent liganding can have diverse effects on the nuclear proteome. Covalent liganding can affect total protein levels (e.g. BBX), modulate nuclear localization (e.g. PIR), and modify chromatin association (e.g. SQSTM1). BBX is a transcription factor that regulates cell progression from G1 to S phase,⁶⁹ PIR acts as a transcriptional co-activator of NF-kB signaling under oxidative stress,⁷⁰ and SQSTM1 is a cargo receptor that shuttles polyubiquitinated proteins from the nucleus for proteasomal degradation.⁷¹ Further experiments are required to determine if these changes are directly or indirectly mediated by covalent ligand binding. Nonetheless, the ability to modulate the nuclear levels of these proteins with small-molecule ligands provide a tool to regulate cellular function.

Additionally, we demonstrate that covalent modification of Cys106 on PARK7 by KB02 induced nuclear translocation and chromatin association. Previous studies demonstrated that PARK7 oxidation induces nuclear translocation. Although PARK7 lacks an NLS sequence, PARK7 oxidation induces phosphorylation by the p38 regulated/activated kinase (PRAK). PRAK contains an NLS, which can facilitate nuclear translocation of PARK7.⁷² Nuclear PARK7 , accumulates at double-strand DNA break (DSB) sites, and interacts with PARP1 to initiate DNA repair under stress conditions.⁷³ Nuclear PARK7 also sequesters the cell apoptosis-mediating protein Daxx in the nucleus to prevent apoptosis under oxidative stress.⁷⁴ Unlike the well characterized antioxidant functions of PARK7 that require the presence of reduced and unmodified Cys106, it remains unknown whether it’s the transcriptional and cytoprotective functions also require Cys106. Covalent liganding of Cys106 could enhance the nuclear pool of PARK7 to further enable these scaffolding and transcriptional functions. Though the scout fragments used in this study are not selective for PARK7, there has been progress in developing more selective PARK7 ligands⁷⁵ that can be used to modulate PARK7 nuclear translocation.

Together, the His-TID PL platform combined with chemoproteomics enables the study of nuclear cysteine ligandability and changes in nuclear localization and chromatin association induced by ligand binding. This platform is highly versatile and can enrich nuclear proteins for various applications, including investigating changes in cysteine oxidation state, protein nuclear localization and chromatin association under various cellular conditions such as oxidative stress.

Limitations of the study

The His-TID chemoproteomic platform enables profiling of nuclear cysteine ligandability and modulation of nuclear localization and chromatin association upon ligand binding. One limitation is the promiscuity of the TID-mediated biotinylation, which is not strictly confined to the nucleus due to diffusion of the reactive biotin intermediates, and biotinylation by newly translated His-TID. Inhibiting protein synthesis prior to biotin treatment decreases these background biotinylation events.³⁷ A second limitation is that our cysteine ligandability studies were performed in cell lysates and not in live cells due to the challenges in accounting for disparate levels of biotinylation between untreated and fragment-treated proteomes. Future efforts will implent live-cell fragment treatment and account for protein biotinylation changes in our proteomic analyses. Third, proteins that display changes in chromatin association might not result from direct ligand labeling, but instead a downstream consequence of liganding. Further evaluation, as we performed for PARK7, is needed to determine if the observed change is a direct consequence of liganding. Lastly, the coverage of nuclear proteins, including TFs, can be improved with the use of more advanced MS instrumentation, and the coverage of nuclear cysteines can be improved with dual-enrichment strategies that give broader coverage in PL-coupled chemoproteomics.³⁷ Our data show the promise of coupling His-TID biotinylation with chemoproteomics to interrogate the nuclear proteome.

Star Methods

Resource Availability

Lead contact

Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead contact, Eranthie Weerapana (eranthie@bc.edu).

Materials Availability

All unique/stable reagents generated in this study are available from the lead contact with a completed Materials Transfer Agreement.

Data and code availability

• All data reported in this paper will be shared by the lead contact upon request. The mass spectrometry proteomics data have been deposited to ProteomeXchange Consortium via the PRIDE database (http://www.proteomexchange.org) and are publicly available as of the date of publication. Accession numbers are listed in the key resources table.

• This paper does not report original code.

• Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon request.

Key Resources Table

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Antibodies
HA-Tag (C29F4) Rabbit mAb (1:1000)	Cell Signaling Technology	Cat# 3724; RRID:AB_1549585
HA-Tag (6E2) Mouse mAb (1:1000)	Cell Signaling Technology	Cat# 2367; RRID:AB_10691311
GAPDH (14C10) Rabbit mAb (1:1000)	Cell Signaling Technology	Cat# 2118; RRID:AB_561053
Streptavidin-HRP (1:2000)	Cell Signaling Technology	Cat# 3999; RRID:AB_10830897
Anti-Rabbit IgG, HRP-linked Antibody (1:2000)	Cell Signaling Technology	Cat# 7074; RRID:AB_2099233
Anti-Mouse IgG (H+L), F(ab')2 Fragment (Alexa Fluor® 488 Conjugate) (1:500)	Cell Signaling Technology	Cat# 4408; RRID:AB_10694704
ProLong® Gold Antifade Reagent with DAPI	Cell Signaling Technology	Cat# 8961
Histone H3 (D1H2) XP® Rabbit mAb (1:2000)	Cell Signaling Technology	Cat# 4499; RRID:AB_10544537
Calreticulin (D3E6) XP® Rabbit mAb (1:1000)	Cell Signaling Technology	Cat# 12238; RRID:AB_2688013
ATPIF1 (D6P1Q) XP® Rabbit mAb (1:1000)	Cell Signaling Technology	Cat# 13268; RRID:AB_2798167
BBX Antibody (1:1000)	Novus Biologicals	Cat# NBP2-38633
Dj 1 Antibody (1:1000)	Cell Signaling Technology	Cat# 2134; RRID:AB_2160260
SQSTM1/p62 Antibody (1:1000)	Cell Signaling Technology	Cat# 5114; RRID:AB_10624872
Pirin (1E8) Rat mAb (1:1000)	Cell Signaling Technology	Cat# 9777; RRID:AB_2268318
Anti-Rat IgG, HRP-linked Antibody (1:2000)	Cell Signaling Technology	Cat# 7077; RRID:AB_10694715
Anti-DYKDDDK Tag (FLAG) Rabbit mAb (1:1000)	Cell Signaling Technology	Cat# 14793; RRID:AB_2572291
Chemicals, peptides, and recombinant proteins
DMEM High Glucose Medium	Cytiva	Cat# SH30243.01
RPMI 1640 Medium, No Phenol Red	Gibco	Cat# 11835030
Dulbecco's Phosphate-Buffered Salt Solution 1X (DPBS)	Corning	Cat# 21031CM
Fetal Bovine Serum (FBS)	R&D Systems	Cat# S11150
Antibiotic-Antimycotic Solution	Gibco	Cat# 15240062
HyClone™ Trypsin Protease	Cytiva	Cat# SH30042.01
Phusion™ High-Fidelity DNA Polymerase	New England Biolabs	Cat# M0530
NotI-HF®	New England Biolabs	Cat# R3189
EcoRI-HF®	New England Biolabs	Cat# R3101
T4 DNA Ligase	New England Biolabs	Cat# M0202
Gateway™ LR Clonase™ II Enzyme Mix	Invitrogen	Cat# 11791020
Gibson Assembly Cloning Kit	New England Biolabs	Cat# E5510S
QuikChange Lightning Site-Directed Mutagenesis Kit	Agilent Technologies	Cat# 210518
Q5® Site-Directed Mutagenesis Kit	New England Biolabs	Cat# E0554S
Puromycin Dihydrochloride	Santa Cruz Biotechnology	Cat# sc-108071
Lipofectamine 2000 Reagent	Invitrogen	Cat# 11668027
Thiazolyl Blue Tetrazolium Bromide	Thermo Scientific Chemicals	Cat# 158990010
Polybrene	Millipore Sigma	Cat# TR1003G
DMSO	Thermo Scientific	Cat# 610971000
D-Biotin	Chem-Impex International	Cat# 000335G
Thermo Scientific™ Pierce™ 16% Formaldehyde (w/v), Methanol-free	Thermo Scientific	Cat# 28906
Normal Goat Serum	Cell Signaling Technology	Cat# 5425
Albumin from Bovine Serum	Millipore Sigma	Cat# 9048468
Triton™ X-100	Sigma-Aldrich	Cat# 93443
Tween™ 20	Sigma-Aldrich	Cat# P1379
Trichloroacetic Acid (TCA)	Sigma-Aldrich	Cat# T9159
Pierce™ Streptavidin Agarose	Thermo Scientific	Cat# 20353
SDS	Bio-Rad	Cat# 1610302
Urea	Thermo Scientific	Cat# 424585000
UltraPure™ Dithiothreitol (DTT)	Invitrogen	Cat# 15508013
Iodoacetamide (IA)	Thermo Scientific	Cat# 122270050
Sequencing Grade Modified Trypsin	Promega	Cat# V5111
Formic Acid, 98%	Millipore Sigma	Cat# FX04405
2-Chloro-1-(6-methoxy-1,2,3,4-tetrahydroquinolin-1-yl)ethan-1-one (KB02)	Sigma-Aldrich	Cat# 912131
N-(4-Bromophenyl)-N-phenylacrylamide (KB05)	Sigma-Aldrich	Cat# 911798
Tris	Bio-Rad	Cat# 1610719
Ethylenediaminetetraacetic Acid (EDTA)	Thermo Scientific	Cat# A1071336
Tris(2-carboxyethyl)phosphine Hydrochloride (TCEP)	Sigma-Aldrich	Cat# 4706
N-Ethylmaleimide (NEM)	Sigma-Aldrich	Cat# 04259
N-Ethylmaleimide (Ethyl-D5, 98%)	Cambridge Isotope Laboratories	Cat# DLM-6711-PK
Triethylammonium Bicarbonate Buffer (TEAB)	Sigma-Aldrich	Cat# T7408
EPPS	Thermo Scientific	Cat# J61476AK
TMTsixplex™ Isobaric Label Reagent Set	Thermo Scientific	Cat# 90061
Pierce™ Hydroxylamine, 50%	Thermo Scientific	Cat# 90115
Sep-Pak C18 Classic Cartridge	Waters	Cat# WAT051910
Acetonitrile, Optima™ LC/MS Grade	Fisher Scientific	Cat# A955-4
Water, Optima™ LC/MS Grade	Fisher Scientific	Cat# W64
IPTG, Dioxane-free	Thermo Scientific	Cat# R0392
Imidazole	Sigma-Aldrich	Cat# I2399
Glycerol	Sigma-Aldrich	Cat# G5516
2-Mercaptoethanol	Fisher Scientific	Cat# O3446I-100
Phenylmethylsulfonyl Fluoride (PMSF)	Thermo Scientific	Cat# 36978
HisPur™ Ni-NTA Superflow Agarose	Thermo Scientific	Cat# 25214
Zeba™ Spin Desalting Columns, 7K MWCO	Thermo Scientific	Cat# 89890
Amicon™ Ultra-4 Centrifugal Filter Units	Millipore Sigma	Cat# UFC801024
Halt™ Protease Inhibitor Cocktail (100X)	Thermo Scientific	Cat# 87786
Sodium Butyrate	Thermo Scientific	Cat# A1107906
NP-40 Surfact-Amps™ Detergent Solution	Thermo Scientific	Cat# 85124
Formaldehyde, 37% (w/w) in water	Thermo Scientific	Cat# 119690010
Glycine	Thermo Scientific	Cat# A138160C
RNase A	Thermo Scientific	Cat# EN0531
KB02Yne	Sigma-Aldrich	Cat# 925225
6-Methoxy-1,2,3,4-Tetrahydroquinoline	Thermo Scientific Chemicals	Cat# H63002MD
Pierce™ IP Lysis Buffer	Thermo Scientific	Cat# 87787
Pierce™ Universal Nuclease for Cell Lysis	Thermo Scientific	Cat# 88700
Pierce™ Anti-DYKDDDDK Magnetic Agarose	Thermo Scientific	Cat# A36797
Pierce™ 3x DYKDDDDK Peptide	Thermo Scientific	Cat# A36805
TAMRA Azide, 5-Isomer	Lumiprobe	Cat# A7130
Tris[(1-benzyl-1H-1,2,3-triazol-4-yl)methyl]amine (TBTA)	Sigma-Aldrich	Cat# 678937
Coomassie Brilliant Blue G-250	TCI America	Cat# B31935G
DMEM High Glucose Medium	Cytiva	Cat# SH30243.01
Dulbecco's Phosphate-Buffered Salt Solution 1X (DPBS)	Corning	Cat# 21031CM
Fetal Bovine Serum (FBS)	R&D Systems	Cat# S11150
Recombinant Human WT HAT1 (20-341)	This Study	N/A
Recombinant Human C101S HAT1 (20-341)	This Study	N/A
Recombinant Human WT PARK7	This Study	N/A
Recombinant Human C106A PARK7	This Study	N/A
Critical commercial assays
DC Protein Assay	Bio-Rad	Cat# 500-0116
Histone Acetyltransferase Activity Assay Kit (Colorimetric)	Abcam	Cat# ab65352
Deposited data
Proteomics data, deposited to PRIDE	This Study	PXD043732
Experimental models: Cell lines
HEK293T	ATCC	Cat# CRL-3216; RRID:CVCL_0045
HeLa	ATCC	Cat# CCL-2; RRID:CVCL_0030
His-TID HeLa	This Study	N/A
His-TID-1 HeLa	This Study	N/A
Oligonucleotides
Oligonucleotides used for PCR and cloning	Azenta	See Table S1
Recombinant DNA
pDONR221	Invitrogen	Cat# 12536017
V5-TurboID-NES_pCDNA3	Branon, et al. 201829	RRID:Addgene_107169
pLenti CMV Puro Dest (w118-1)	Campeau, et al. 200976	RRID:Addgene_17452
psPAX2	Didier Trono	RRID:Addgene_12260
pCMV-VSV-G	Stewart, et al. 200377	RRID:Addgene_8454
H3.3B (H3F3B) (NM_005324) Human Tagged ORF Clone	Origene Technologies	Cat# RC202257
pDONR221 TurboID-HA	Kisty, et al. 202336	N/A
pLenti kozak-TurboID-HA	Kisty, et al. 202336	N/A
pLenti kozak-Histone-TurboID-HA	This Study	N/A
Human HAT1 ORF Clone	GenScript	Cat# OHu16522
pET-28a(+)	Novagen	Cat# 69864-3
pET-28a(+)-HA-HAT1-AA20-341	This Study	N/A
pET-28a(+)-HA-C101S HAT1-AA20-341	This Study	N/A
pcDNA3.1 (+) Mammalian Expression Vector	Invitrogen	Cat# V79020
Human PARK7 ORF Clone	Genscript	Cat# OHu26877
pcDNA3.1+_FLAG_PARK7	This Study	N/A
pcDNA3.1+_FLAG_C106A PARK7	This Study	N/A
pDONR221	Invitrogen	Cat# 12536017
Software and algorithms
Proteome Discoverer	Thermo Fisher	RRID:SCR_014477
Thermo Xcalibur	Thermo Fisher	RRID:SCR_014593
Prism 9	GraphPad	RRID:SCR_002798
PANTHER GO Slim	Mi, et al. 201947	RRID:SCR_004869
MagTran	Amgen	N/A

Experimental Model and Study Participant Details

Cell lines

HEK 293T (CRL-3216, Homo sapiens, female, embryonic kidney) and HeLa cells (CCL-2, Homo sapiens, female, cervix) were obtained from ATCC. Cells were grown at 37°C under 5% CO₂ in DMEM/High glucose with L-glutamine, sodium pyruvate (Cytiva) supplemented with 10% fetal bovine serum (R&D Systems), 100 units/mL of penicillin, 100 μg/mL of streptomycin, and 0.25 μg/mL of amphotericin B (Gibco). Cell lines were not authenticated prior to use since they were obtained from ATCC.

Method Details

Cloning of pLenti Histone-TurboID-HA

Initial parent vector pDNOR221 MCS-TurboID-HA was generated through PCR amplification of V5-TurboID-NES_pCDNA3 (Addgene 107169) with attb1/attb2 sites incorporated for Gateway cloning.³⁶ Full-length Histone H3.3B (NM_005324) was amplified from H3.3B human tagged ORF clone (Origene) with kozak sequence incorporated at the 5’-end. Primers for H3.3B PCR amplification were listed in Table S1. The H3.3B amplicon and pDNOR221 MCS-TurboID-HA were digested by NotI and EcoRI (New England Biolabs (NEB)) and then ligated using T4 DNA ligase (NEB) to construct pDONR Histone-TurboID-HA, which was further cloned into pLenti CMV Puro DEST (Addgene) by Gateway LR Clonase (Invitrogen). Plasmid constructs were transformed into TOP10 competent bacterial cells by chemical transformation.

Generation of HeLa Cell Line Stably Expressing Histone-TurboID-HA

HEK 293T cells under passage 15, grown in DMEM with 10% heat-inactivated fetal bovine serum (HIS), were used to generate lentiviruses. At 70-80% confluency, HEK 293T cells were transfected with 2 μg vsv-g, 2 μg psPAX2, and 2 μg pLenti Histone-TurboID-HA using 24 μL Lipofectamine 2000 (Invitrogen). Cell media was replaced with 5 mL DMEM containing 10% HIS 24 hours after transfection. Cell media was filtered through a 0.45 μm filter and combined with 5 mL DMEM + 10% HIS and 10 μL of 10 mg/mL polybrene 48 hours after transfection. Filtered media containing lentiviruses was added to HeLa cells at 20% confluency and was replaced with 10 mL complete DMEM media with 1 μg/mL puromycin 24 hours after transduction. Transduced HeLa cells were grown under puromycin selection for 3 days to allow for selection for cells that stably express Histone-TurboID (His-TID).

Single-Cell Colony Selection

Polyclonal HeLa cells expressing His-TID (His-TID-P) were resuspended and diluted to 5 cells/mL. Cell density was measured using a table-top cell counter. Cell suspension was added to 96-well cell culture plates with 100 μL per well. Cells in wells containing single cell colonies were allowed to reach 100% confluency before being further propagated and subjected to Western blot analysis.

Biotin Labeling of Cells Stably Expressing His-TID

HeLa cells stably expressing His-TID were grown to full confluency in 10 cm cell dishes and treated with either increasing concentrations (0, 10, 100, 500, 1000 μM) of biotin in DMSO for 1 hour or 500 μM biotin with increasing incubation times (0, 1, 2, 3, 4, 5 hours) at 37°C. Biotin treatment was quenched by 3 ice-cold DPBS (Corning) washes.

Lysate Preparation

Cells were harvested by scraping and washed with ice-cold DPBS. Cell pellets were resuspended in DPBS and lysed by ultrasonic tip sonication (Cole Parmer) at the amplitude of 75% for 30 pulses. Cell lysates were subjected to centrifugation at 14,000 rpm for 10 min at 4°C. DC protein assay (Bio-Rad) was performed to determine protein concentrations.

Immunoblotting

Protein samples were separated by 12.5% SDS-PAGE gels at 160 volts for 70 min and transferred to nitrocellulose membranes at 75 volts for 100 min. Membranes were washed with Tris-buffered saline and 0.1% Tween 20 (Sigma-Aldrich) (TBST) for 3 times. Membranes were blocked with 5% (w/v) non-fat dry milk in TBST at room temperature for 1 hour, washed with TBST 3 times, and incubated with respective primary antibodies overnight at 4°C. After 3 washes with TBST, the membranes were incubated with HRP-conjugated secondary antibodies for 1 hour at room temperature. For biotin blots, membranes were blocked with 5% BSA (Millipore Sigma) in TBST overnight at 4°C, and incubated with streptavidin-HRP (Cell Signaling Technology) in 5% BSA/TBST for 1 hour at room temperature. Membranes were developed with HRP chemiluminescent substrate (Thermo Scientific) and imaged using a ChemiDoc MP imaging system (Bio-Rad).

Immunofluorescence

Cells were grown in 35 mm dishes with glass bottom (MatTek) until 80% confluency. After 3 washes with DPBS, cells were fixed by incubating with 4% formaldehyde (Thermo Scientific) for 15 min at room temperature. The fixed cells were rinsed 3 times with DPBS for 5 min and blocked with 5% normal goat serum (Cell Signaling Technology) in DPBS with 0.3% Triton X-100 (Sigma-Aldrich) for 1 hour at room temperature. After blocking, cells were incubated with HA-tag mouse mAb (Cell Signaling Technology) diluted in DPBS with 0.1% BSA and 0.3% Triton X-100 overnight at 4°C. After 3 rinses with DPBS, cells were incubated with Alexa Fluor488-conjugated anti-mouse secondary antibody (Cell Signaling Technology) for 2 hours at room temperature in the dark. After final DPBS washes, cells were applied with Prolong Gold antifade reagent with DAPI (Cell Signaling Technology) and cured overnight at room temperature. Samples were covered with cover slips, sealed for long term storage at 4°C, and protected from light before imaging with a Leica TCS SP5 scanning confocal microscope.

Cell Proliferation Assay

HeLa or His-TID-1 cells were seeded in 96-well plates at the density of 1x10⁴ cells/mL and MTT assay was conducted every day for 5 consecutive days. 110 μL clear RPMI with 0.5 mg/mL MTT was added to each well and cells were incubated at 37°C for 4 hours. 100 μL of 0.1 g/mL SDS/H₂O with 0.01 M HCl was added to each well, pipetted to mix well, and incubated with cells overnight at room temperature. Absorbance was taken at the wavelength of 570 and 650 nm. There were 3 biological replicates per cell line per day, and the absorbance values were normalized to Day 0 to generate proliferation fold change.

Trypsin Digest of Proximity Labeling Samples

2 mg of protein lysate from His-TID-1 cells treated with DMSO or 500 μM biotin for 5 hours was precipitated by trichloroacetic acid (TCA) (Sigma-Aldrich) and stored at −80°C for 1 hour or overnight. Proteins were pelleted at 14,000 rpm for 10 min and resuspended in 500 μL ice-cold acetone before centrifugation at 5,000 rpm for 10 min at 4°C. Pellets were resuspended in 1 mL 1.2% SDS (Bio-Rad) in DPBS by sonication and then combined with 5 mL DPBS containing 100 μL of washed streptavidin agarose resin (Thermo Scientific) for rotation at 4°C overnight. Solutions were rotated at room temperature for 2-3 hours to solubilize SDS and the resin was pelleted by centrifugation at 1,400 g for 3 min. The pelleted resin was incubated with 0.2% SDS/DPBS for 10 min at room temperature followed by 3 washes with DPBS and 3 washes with water (5 mL each). After the final wash, the resin was resuspended in 500 μL 6 M urea (Thermo-Scientific) with 25 μL 200 mM dithiothreitol (DTT, Invitrogen) and heated at 65°C for 20 min with resuspension every 10 min. 25 μL of 400 mM iodoacetamide (IA, Thermo Scientific) was added to the mixture and the reaction was incubated at 37°C for 30 min with rotation. The sample was then combined with 950 μL DPBS and spun down to remove the supernatant. The resin was further resuspended in 200 μL 2 M urea with 2 μL 100 mM CaCl₂ and 2 μg trypsin (Promega), and proteins were digested overnight at 37°C with rotation. The supernatant was collected, and the resin was washed 3 times with 50 μL DPBS to generate a combined final volume of 350 μL, with 17.5 μL formic acid (Millipore Sigma) added to quench the trypsin digestion.

Liquid Chromatography-Tandem Mass Spectrometry (LC-MS/MS) for Trypsin Digest Samples

LC-MS/MS was performed on an Orbitrap Exploris 240 mass spectrometer with Xcalibur version 4.4 (Thermo Fisher Scientific) coupled to a Dionex UltiMate 3000 RSLCnano system. MS samples were desalted using Sep-Pak C18 cartridges (Waters) and dried on SpeedVac. Desalted peptides were resuspended in 100% water buffer A (100% H₂O and 0.1% formic acid) and loaded onto an Acclaim PepMap 100 C18 loading column (Thermo Scientific) for further elution onto an Acclaim PepMap RSLC column (Thermo Scientific), where the peptides were separated with a 2-hour gradient ranging of 5 to 25% buffer B (20% H₂O, 80% acetonitrile, and 0.1% formic acid) in buffer A at a flow rate of 0.3 μL/min. The spray voltage was set to 2.1 kV. One full MS1 scan (120,000 resolution, 350-1,800 m/z, RF lens 65%, automatic gain control (AGC) target 300%, automatic maximum injection time, profile mode) was obtained every 2 s with dynamic exclusion (repeat count 2, duration 10 s), isotope exclusion (assigned), and apex detection (30% desired apex window) enabled. A variable number of MS2 scans (15,000 resolution, AGC 75%, maximum injection time 100 ms, centroid mode) were obtained between each MS1 scan based on the highest precursor masses, filtered for monoisotopic peak determination, theoretical precursor isotopic envelope fit, intensity (5E4), and charge state (2-6). MS2 analysis included the isolation of precursor ions (isolation window 2 m/z) followed by higher-energy collision dissociation (HCD) (collision energy 30%).

LC-MS/MS Proteomics Analysis for Trypsin Digest Samples

LC-MS/MS data were analyzed on Thermo Proteome Discoverer version 2.4 and searched using SequestHT and Percolator⁷⁸ algorithms against Homo sapiens proteome database from Uniprot Protein Knowledgebase (UniProtKB)⁷⁹. The protease for data analysis was specified as trypsin with a maximum of 2 missed cleavages. Peptide precursor tolerance was set to 10 ppm with a fragment mass tolerance of 0.02 ppm. Cysteine alkylation (+57.021) was set as a static modification, methionine oxidation (+15.995), acetylation (+42.011), and methionine loss (+131.040) of the protein N-terminus were set as dynamic modifications. The false discovery rate (FDR) was set to 1% for peptide identification. Characterization of nuclear localization and transcription factor (TF) were determined by Uniprot database,⁷⁹ and the human transcription factor catalog,⁴⁵ respectively.

Statistical Analysis of Overrepresented Gene Ontology (GO) Molecular Functions Using PANTHER

PANTHER overrepresented GO^46-49 tests were performed on His-TID (+)/(−) biotin datasets in search of molecular functions from PANTHER GO-Slim annotation set. Statistical difference of overrepresentation for each molecular function was determined by Fisher’s exact test with a Bonferroni correction.

Competitive Chemoproteomic Analysis of Scout Fragments

2 mg cell lysates from His-TID-1 cells treated with 500 μM biotin for 5 hours were labeled with 100 μM functionalized fragments (2-Chloro-1-(6-methoxy-1,2,3,4-tetrahydroquinolin-1-yl)ethan-1-one (KB02), N-(4-bromophenyl)-N-phenylacrylamide (KB05), Sigma-Aldrich),²⁰ or DMSO at room temperature for 2 hours. Proteins were precipitated with 10% TCA at −80°C for 1 hour or overnight and pelleted by centrifuging at 14,000 rpm for 10 min. Pellets were resuspended in 500 μL ice-cold acetone and centrifuged at 5,000 rpm for 10 min. 80 μL denaturing alkylation buffer (DAB) (6 M urea, 200 mM Tris-HCl (Bio-Rad), 10 mM EDTA (Thermo Scientific), 0.5% SDS, pH 8.5) was added to resolubilize protein pellets and 4 μL of 50 mM tris(2-carboxyethyl)phosphine hydrochloride (TCEP, Sigma-Aldrich) was added to reduce oxidized cysteines at 37°C for 5 min. The reactions were diluted with 120 μL DAB and 4 μL of 500 mM N-ethyl maleimide (light-NEM, Sigma Aldrich) or D5-NEM (heavy-NEM, Cambridge Isotope Laboratories) was added to DMSO-treated or fragment-labeled samples, respectively. Alkylation reactions were allowed for 2 hours at 37°C in the dark with shaking, and diluted 3-fold with 500 μL H₂O, then 5-fold with 2.5 mL ice-cold acetone to precipitate proteins at −20°C for 2 hours or overnight. Proteins were pelleted by centrifuging at 4,500 g for 30 min, and the pellets were resuspended in 500 μL ice-cold acetone followed by another spin at 4,500 g for 10 min. 0.5 mL of 1.2% SDS/DPBS was used to resolubilize proteins and DMSO-treated (light NEM-alkylated) samples were combined with fragment (KB02 or KB05)-labeled (heavy NEM-alkylated) samples. Combined protein samples were added to 5 mL DPBS with 100 μL washed streptavidin agarose resin and incubated at 4°C overnight with rotating. Samples were rotated at room temperature for 2-3 hours to resolubilize SDS and the supernatant was discarded after centrifugation at 1,400 g for 3 min. Resin was resuspended in 5 mL 0.2% SDS/DPBS, rotated for 10 min at room temperature, and washed 3 times with DPBS and 3 times with H₂O. Resin was resuspended in 200 μL 2 M urea in DPBS with 1 mM CaCl₂ and 2 μg trypsin to digest at 37°C overnight with rotating. The supernatant after digestion was combined with 3x DPBS washes of the resin and 17.5 μL formic acid were added to quench trypsin digestion. The samples were run on Orbitrap Exploris 240 mass spectrometer with Xcalibur version 4.4 and analyzed using Thermo Proteome Discoverer V2.4 as described previously for trypsin digested samples with minor adjustments. Cysteine alkylation by either light NEM (+125.048) or heavy NEM (+130.079) were included as dynamic modifications for two independent searches, and cysteine alkylation by IA was excluded from static modifications. Characterized nuclear localization, TF, and DrugBank annotation was determined by Uniprot database,⁷⁹ the human transcription factor catalog,⁴⁵ and the DrugBank database,⁸⁰ respectively. L:H ratios of 0.01 and 100 were considered only when presented in all 4 replicates per condition. Average L:H values were calculated for each peptide, and peptides with relative standard deviations > 50% were removed from further analysis. The MS data were collected as 2 technical replicates each from 2 independent biological replicates.

Cloning of pET28a_HAT1 (aa 20-341)

cDNA encoding amino acids 20-341 of Homo sapiens HAT1 (NM_003642.3) was amplified from HAT1 cDNA ORF clone (GenScript) and cloned into pET28a vector using Gibson Assembly Cloning Kit (New England Biolabs). Site-directed mutagenesis was performed to obtain C101S HAT1 mutant using QuikChange Lightning Site-Directed Mutagenesis Kit (Agilent). Primers for pET28a_HAT1 (aa20-341) cloning and mutagenesis were listed in Table S1. The plasmids were transformed into BL21 (DE3) competent cells by chemical transformation.

Expression and Purification of human HAT1 (aa 20-341)

HAT1 expression and purification was adapted from Wu, et al.⁸¹ and Delaney, et al.⁸² Protein expression was induced by 1 mM IPTG (Thermo Scientific) in LB media when cells were grown to OD 0.6 at 37°C and shaken at 25°C for 20 hours. Cells were harvested and suspended in lysis buffer (50 mM Na-phosphate, 250 mM NaCl, 5 mM imidazole, 5% glycerol, 2 mM β-mercaptoethanol and 1 mM PMSF, pH 7.4) and lysed using ultrasonic tip sonication at 30% amplitude for 120 pulses. The lysate was centrifuged at 14,000 rpm for 10 min and the supernatant was loaded onto Ni-NTA superflow agarose (Thermo Scientific). The column was washed 5 times with 2 mL column washing buffer (20 mM Tris-HCl, 250 mM NaCl, 5% glycerol, 30 mM imidazole, and 1 mM PMSF, pH 8.0). Proteins were eluted with 5x 1 mL elution buffer (20mM Tris-HCl, 250 mM NaCl, 5% glycerol, 500 mM imidazole, and 1 mM PMSF, pH 8.0). Eluents containing HAT1 were combined, buffer exchanged into storage buffer (25 mM HEPES, 70 mM NaCl, 0.75 mM NaH₂PO₄, 5% glycerol, pH 7.4) using Zeba spin desalting columns (Thermo Scientific), and concentrated with Amicon ultra centrifugal filters (Millipore Sigma). The protein was then aliquoted, flash frozen by liquid nitrogen, and stored at −80°C for long term storage.

HAT1 Labeling and LC/MS Analysis

HAT1 in storage buffer was diluted to 0.3 mg/mL and reduced with 1 mM TCEP at room temperature for 1 hour. WT and C101S HAT1 solutions were treated with either DMSO or 2 mM KB02 for 1 hour at room temperature and buffer exchanged into DPBS to remove excess KB02. Protein concentrations were determined by DC protein assay and 400 ng of protein was used for whole protein mass spectrometry (ESI-LC/MS). ESI-LC/MS was performed on a 1260 Agilent Infinity Series HPLC coupled with a 6230 Agilent TOF mass spectrometer and the data was analyzed using MagTran. Anticipated masses for analyzed proteins were calculated with the loss of initiator methionine.

HAT1 Activity Assay

HAT1 activity assay was conducted using histone acetyltransferase activity assay kit (Abcam). DMSO or KB02 labeled WT or C101S HAT1 was buffer exchanged into water and diluted to 0.2 mg/mL. 40 μL of this solution were added to 96-well plate (water as a negative control) and mixed with 68 μL assay mix containing 50 μL 2x HAT1 buffer, 5 μL HAT substrate I, 5 μL HAT substrate II, and 8 μL NADH generating enzyme. The reaction was incubated at 37°C for 4 hours. The absorbance at 440 nm was measured every 5 min using a Synergy Neo2 hybrid multimode reader (Agilent). The assay was conducted twice with 3 replicates each time and two-tailed, paired t-test was used to calculate p-values between 2 conditions.

In Situ Fragment Labeling

WT HeLa cells or HeLa cells stably expressing His-TID from single cell colony 1 (His-TID-1) were treated with 30 μM of functionalized fragments KB02, KB05, or DMSO for 1 hour in serum free DMEM at 37°C. Cell media was then aspirated and replenished with complete DMEM. For streptavidin pulldown, His-TID-1 cells were further treated with 500 μM biotin for 5 hours at 37°C. For nuclei isolation and chromatin pulldown, WT HeLa cells were further grown for 1, 3, 5, 7, and 10 hours before harvested.

Streptavidin Pulldown Using Lysates from In Situ Labeling

2 mg cell lysates from His-TID-1 cells labeled with DMSO, KB02, or KB05 in situ were precipitated with 10% (w/v) TCA at −80°C for 1 hour or overnight. Samples were pelleted at 14,000 rpm for 10 min and then resuspended in 500 μL ice-cold acetone for centrifugation at 5,000 rpm for 10 min at 4°C. The pellet was resolubilized in 1 mL 1.2% (w/v) SDS/DPBS and incubated with 5mL DPBS containing 100 μL washed streptavidin agarose resin at 4°C overnight. Samples were rotated at room temperature for 2-3 hours to resolubilize SDS and then centrifuged at 1,400 g for 3 min to pellet down streptavidin resin. Pelleted resin was resuspended in 5 mL 0.2% (w/v) SDS/DPBS and rotated at room temperature for 10 min before washed with DPBS and H2O for three times each. Resin was resuspended in 500 μL 6 M urea/DPBS and 25 μL 200mM DTT was added to reduce oxidized cysteines at 65°C for 20 min with resuspension every 10 min. 25 μL 400mM IA was added to alkylate newly-reduced cysteines at 37°C for 30 min with rotating. 950 μL DPBS was added to each tube and the resin was spun down to discard the supernatant. For immunoblotting, resin was boiled in 100 μL SDS-PAGE buffer at 95°C for 10 min and the supernatant was collected. For TMT labeling, the resin was further washed 3 times with DPBS to remove residual urea. 200 μL of 100 mM triethylammonium bicarbonate (TEAB, Sigma-Aldrich) in DPBS containing 1 mM CaCl₂ and 2 μg trypsin were added to each tube and the samples were digested at 37°C with rotating overnight. Upon retrieving the supernatant, the resin was washed with 3x 50 μL of DPBS and the combined supernatant generated a final sample volume of 350 μL. 17.5 μL formic acid (5% final concentration) were added and samples were dried on SpeedVac.

TMT Labeling MS Preparation and Analysis

70 μL 200 mM EPPS buffer, pH 8.5 (Thermo Scientific) and 30 μL acetonitrile was added to each tube for peptide resuspension. Digested peptides from in situ labeling-streptavidin pulldown were labeled using TMTsixplex (Thermo Scientific) by adding each 100 μL resuspended samples to 5 μL 20 μg/μL individual isobaric TMT tags. The mixture was incubated at room temperature for 1 hour in the dark, and then 5 μL 5% hydroxylamine (Sigma Aldrich) was added to the solution and incubated for 15 min in the dark to quench the reaction. 5 μL formic acid was added to each condition and all conditions were combined. The combined sample was dried on SpeedVac overnight, and the dried peptides were resuspended in 500 μL of high-pH buffer A containing 95% water, 5% acetonitrile, and 10 mM ammonium bicarbonate.⁸³ The resuspended sample was loaded onto a manual injection loop connected to an Agilent 1100 Series HPLC. The injected peptides were further separated on a 25-cm Agilent Extend C18 column using a 60-min gradient from 20 to 35% high-pH buffer B containing 10% water, 90% acetonitrile, and 10 mM ammonium bicarbonate.⁸³ Fractions were collected into a 96-deep-well plate using a Gilson FC203B fraction collector, and samples in every sixth well were combined resulting in six pooled fractions to be dried on SpeedVac. The samples were run on Orbitrap Exploris 240 mass spectrometer with Xcalibur version 4.4 and analyzed using Thermo Proteome Discoverer V2.4 as described previously with minor adjustments. TMT 6plex (+229.163) modification on lysine residues and protein N terminus was set as static modifications in addition to cysteine alkylation by IA (+57.021). Protein ratios for each channel versus the control channel were calculated from the grouped protein abundances and median-normalized. Only nuclear proteins that were presented in all 6 channels and 3 replicates were considered for further analysis. Average ratios were calculated and proteins with relative standard deviations > 50% were removed from the dataset. Two-tailed, paired t-test was used to calculate p-values. The MS data was collected as 3 technical replicates from 1 biological replicate.

Nuclear Isolation of In Situ Labeled Cells

Preparations for nuclear enriched samples were adapted from Wang, et al.⁶⁰ Cell pellets of WT HeLa cells labeled with DMSO or KB02 in situ were resuspended in 10x volume of nuclear isolation buffer (NIB) containing 15 mM Tris-HCl pH 7.5, 60 mM KCl, 15 mM NaCl, 5 mM MgCl₂, 1 mM CaCl₂, 250 mM sucrose, 1X HALT protease inhibitor cocktail (Thermo Scientific), 10 mM sodium butyrate (Thermo Scientific), and 0.3% NP-40 (Thermo Scientific) and incubated on ice for 20 min with inversion every 10 min. Lysed cells were centrifuged at 250 g for 5 min at 4°C, and the pelleted nuclei were washed twice with 500 μL NIB without NP-40.

Chromatin Enrichment of In Situ Labeled Cells

Preparations for chromatin crosslinking and enrichment were adapted from Kustatscher, et al.⁶¹ After in situ labeling, WT HeLa cells were washed with DPBS twice, and 12 mL of 1% (w/v) warm formaldehyde (Thermo Scientific) in DPBS were added per 150 mm tissue culture plate. The plates were incubated at 37°C for 10 min with rocking every 5 min. Glycine was added to a final concentration of 0.25 M to halt the cross-linking reaction and the plates were incubated at room temperature for 5 min with rocking every 2.5 min. Plates were rinsed with DPBS twice before harvesting the cells. Harvested cells were resuspended in 1 mL ice-cold cell lysis buffer (25 mM Tris-HCl, pH7.4, 0.1% (v/v) Triton X-100, 85 mM KCl, 1X Halt protease inhibitor cocktail) and centrifuged at 2,300 g for 10 min at 4°C. The pelleted crude cell nuclei were resuspended in 500 μL cell lysis buffer with 200 μg/mL RNase A (Thermo Scientific) and incubated at 37°C for 15 min with rotating before centrifugation at 2,300 g for 10 min at 4°C to pellet cell nuclei. The cell nuclei pellet was resuspended in 500 μL SDS buffer (50 mM Tris-HCl, pH 7.4, 10 mM EDTA, 4% (w/v) SDS, 1X Halt protease inhibitor) and incubated at room temperature for 10 min. 1.5 mL urea buffer (10 mM Tris-HCl, pH 7.4, 1 mM EDTA, 8 M urea) were added and mixed thoroughly by inverting the tube several times. The sample was centrifuged at 16,100 g for 30 min at room temperature and the pellet was washed one more time with 0.5 mL SDS buffer and 1.5 mL urea buffer. The pellet was resuspended in 2 mL SDS buffer and centrifuged at 16,100 g for 30 min at room temperature to remove urea. The pellet was resuspended in 200 μL storage buffer (10 mM Tris-HCl, pH 7.4, 1 mM EDTA, 25 mM NaCl, 10% glycerol, 1X Halt protease inhibitor cocktail) and sonicated at 75% amplitude for 30 pulses on ice. The sample was then centrifuged at 16,100 g for 30 min at 4°C and the supernatant was transferred to determine the protein concentration. The protein solutions were normalized to 2 mg/mL and mixed with 2X SDS-PAGE buffer to heat at 98°C for 30 min to reverse the formation of cross-links for further immunoblotting analysis.

Cloning of WT and C106A PARK7

Full-length human WT PARK7 (NM_007262.5) sequence was amplified from Homo sapiens PARK7 cDNA ORF clone (Genscript) using Phusion high fidelity DNA polymerase and cloned into pcDNA3.1 (+) (Addgene) via restriction enzyme digestion and T4 DNA ligation. Site-directed mutagenesis was performed to obtain pcDNA-C106A PARK7 using Q5 site-directed mutagenesis kit (New England Biolabs). Primers for pcDNA3.1_PARK7 cloning and mutagenesis were listed in Table S1.

Transient Transfection for Expressing WT and C106A PARK7

HeLa cells were seeded at 35% confluence in complete DMEM the day before transfection. 125 μL serum free DMEM was added to two tubes with 10 μL Lipofectamine 2000 or 10 μg pcDNA in each tube. Two tubes were combined and incubated at 37°C for 15 min. The mixture was then added dropwise to HeLa cells at 70% confluence. Fresh complete DMEM was replenished 20 hours post-transfection.

In Situ Labeling of Transfected Cells and Immunoprecipitation

48 hours after transfection, WT HeLa cells and HeLa cells transiently expressing WT or C106A PARK7 were treated with 30 μM KB02yne (Millipore Sigma) for 1 hour in serum free DMEM. Cell media was replaced with complete DMEM, and cells were further grown for 2 hours. Cell media was then aspirated and cells were washed with DPBS for 3 times. 1 mL Pierce IP lysis buffer (Thermo Scientific) was added dropwise to each 10-cm plates with 0.5 μL nuclease (Thermo Scientific) and incubated at 4°C for 5 min. Cell lysate was collected and centrifuged at 14,000 rpm for 20 min at 4°C. 50 μL anti-DYKDDDDK magnetic agarose (Thermo Scientific) was washed with 500 μL IP lysis buffer for 3 times and then incubated with cell lysate supernatant at room temperature for 1 hour with rotating. The magnetic agarose was collected using a magnetic separation rack and washed twice with 500 μL DPBS and H₂O, respectively. The agarose was resuspended in 100 μL 1.5 mg/mL Pierce 3X DYKDDDDK peptide (Thermo Scientific) in PBS and rotated at room temperature for 15 min to elute bound proteins. The eluent was further subjected to click chemistry and rhodamine fluorescence gel analysis.

Non-Covalent Binding Competition Assay

An increasing amount of non-covalent KB02 (0, 150, 300, 600 μM) was incubated with eluent from immunoprecipitation using HeLa cells overexpressing WT PARK7 for 1 hour before 30 μM covalent KB02-alkyne labeling for 10 min. Excessive amount of non-covalent KB02 and KB02-alkyne was removed using P-6 gel columns, and the eluted proteins were further subjected to click chemistry and rhodamine fluorescence gel analysis.

Click Chemistry and Rhodamine Fluorescence Gel

50 μL of the eluent from immunoprecipitation was used for rhodamine azide click chemistry reaction by mixing with 1 μL 5 mM 5-TAMRA azide (Lumiprobe), 1 μL 50 mM TCEP, 3 μL 1.7 mM tris[(1-benzyl-1H-1,2,3-triazol-4-yl)methyl]amine (TBTA, Millipore Sigma), and 1 μL 50 mM CuSO₄. The reaction was incubated at room temperature for 1 hour with vortex every 15 min and quenched by mixing with equal amount of 2X SDS-PAGE buffer. Samples were run on a 12.5% SDS-PAGE gel at 90 volts for 120 min and visualized using a ChemiDoc MP imaging system. The gel was Coomassie-stained and imaged after overnight destain.

Quantification and Statistical Analysis

All MS data were obtained from triplicates or quadruplicates. Quantifications of L:H ratios of PL-competitive chemoproteomic studies and PL-TMT labeling datasets were expressed as the average value of the replicates with relative standard deviation < 50%. Two -tailed, paired t-tests were used to calculate p-values of sample versus negative control datasets in PL-TMT labeling. For cell proliferation assay, data was collected from 3 biological replicates and presented as mean (SD). P-values for cell proliferation assay were calculated using a two-tailed, unpaired t-test. For HAT1 activity assay, data was collected as 3 biological replicates each from 2 individual experiments and presented as mean (SD). P-values for HAT1 activity assay were calculated using a two-tailed, paired t-test. Significance is indicated as follows: *p <0.05, **p <0.01, ***p <0.001, ns for not significant. Details of each individual statistical analyses are provided within each figure legend or Data S1.

Supplementary Material

1

Data S1. Excel file containing lists of all proteins and peptides identified in each mass spectrometry analysis, and Gene Ontology analyses. Related to Figure 2-4.

Highlights.

• Proximity labeling-coupled chemoproteomics assesses nuclear protein ligandability

• Changes in nuclear localization reveals functional effects of cysteine liganding

• PARK7 displays enhanced nuclear localization upon liganding of Cys106

Significance.

Nuclear proteins are pivotal to cell homeostasis through coordinated interactions with chromatin. Perturbations to these interactions contribute to the pathogenesis of various diseases. Small-molecule ligands that modulate the chromatin association of nuclear proteins have therapeutic utility. Specifically, covalent ligands targeting nucleophilic amino acids, such as cysteine, can target nuclear proteins lacking well-structured ligand-binding sites. The low abundance of many nuclear proteins, including transcription factors, hinder chemoproteomic studies. Here, we couple a His-TID-based proximity labeling (PL) approach with chemoproteomics to identify nuclear cysteines that are susceptible to the cysteine-reactive scout fragments KB02 or KB05. The identified ligandable nuclear proteins possess diverse functional categorization including transcription regulation, RNA processing, and DNA repair, which are underrepresented in the DrugBank. We demonstrate selective liganding of Cys101 of the chromatin-modifying enzyme, HAT1, by the covalent fragments. The His-TID-based platform can also assess functional changes mediated by covalent ligands, including protein abundance, nuclear localization and chromatin association. Importantly, PARK7 was shown to translocate to the nucleus and accumulate proximal to chromatin upon liganding at Cys106. Together, our studies combine PL with chemoproteomics to assess nuclear cysteine ligandability and investigate changes in nuclear localization and chromatin association mediated by cysteine liganding.

Inclusion and Diversity

We support inclusive, diverse, and equitable conduct of research.