Profiling and Validation of Live-cell Protein Methylation with Engineered Enzymes and Methionine Analogues

Abstract

Protein methyltransferases (PMTs) regulate many aspects of normal and disease processes through substrate methylation, with S-adenosyl-L-methionine (SAM) as a cofactor. It has been challenging to reveal cellular protein lysine and arginine methylation because these modifications barely alter physical properties of target proteins and often are context-dependent, transient, and substoichiometric. To reveal bona fide methylation events associated with specific PMT activities in native contexts, we developed the live-cell Bioorthogonal Profiling of Protein Methylation (lcBPPM) technology, in which the substrates of specific PMTs are labeled by engineered PMTs inside living cells, with in situ synthesized SAM analogues as cofactors. The biorthogonality of this technology is achieved because these SAM analgoue cofactors can only be processed by the engineered PMTs —but not native PMTs — to modify the substrates with distinct chemical groups. Here, we describe the latest lcBPPM protocol and its application to reveal proteome-wide methylation and validate specific methylation events.

INTRODUCTION:

Protein methylation can modulate diverse biological processes, from gene transcription to DNA damage response (Berdasco and Esteller, 2019; Blanc and Richard, 2017; Luo, 2018; Scheer et al., 2019). Errors in this regulation have been implicated in many diseases, including cancer (Berdasco and Esteller, 2019; Blanc and Richard, 2017; Yang and Bedford, 2013). Protein lysine and arginine methylation occurs on histones and diverse non-histone substrates, and is catalyzed by protein methyltransferases (PMT), with S-adenosyl-L-methionine (SAM) as a cofactor (Luo, 2018). Elucidating the PMT-specific methylome, i.e. the collection of proteome-wide methylation events, is critical to understanding physiological and pathological roles of PMTs (Luo, 2012, 2018; Wang and Luo, 2013).

It is challenging to probe the methylome with conventional methods because of barely distinguishable electrostatic and steric properties of methylated versus unmethylated proteins (Luo, 2015a, b, 2018). Methylation does not affect the overall positive charge of lysine and arginine residues, and thus does not alter the electrostatic properties of the target proteins. Further the addition of a small methyl group does not significantly alter the size of target proteins. In addition, methylation often occurs adjacent to other complex posttranslational modifications, which complicates general recognition with antibodies (Arora et al., 2019; Beaver and Waters, 2016; James and Frye, 2016; Luo, 2015a, b, 2018; Shanle et al., 2017; Sudhamalla et al., 2017). Various approaches have been developed to identify substrates of PMTs, including proteomic perturbation with specific PMT inhibitors and substrate labeling of PMTs with protein microrrays (Luo, 2015a). These conventional approaches, though straightforward, require robust reagents for either enrichment or detection of methylated protein candidates. It has been challenging, however, to develop such enrichment or detection reagents of high quality (Luo, 2015a). While protein methylation can also be examined with biochemical assays with individually purified substrates(Luo, 2015a), this in vitro approach is limited by the lack of native context. Each of these methods thus has strengths and limitations but in general fall short of definitive information of PMT-substrate relation.

To definitively reveal the methylome of specific PMTs for functional annotation, our lab developed the Bioorthogonal Profiling of Protein Methylation (BPPM) technology (Fig. 1). Here, the substrates of specific PMTs are labeled by engineered PMTs using SAM analogue cofactors (Islam et al., 2012; Islam et al., 2011; Wang et al., 2011b). The key design of BPPM is to modify the cofactor-binding sites of the PMTs of interest by removing specific bulky amino acid residues, a bump-and-hole approach, and thus allow the engineered PMTs to accommodate bulky SAM analogue cofactors for substrate labeling (Islam et al., 2012; Islam et al., 2013; Islam et al., 2011; Wang et al., 2011b). Our previous BPPM protocols were developed to process SAM analogue cofactors, which cannot penetrate cell the membrane, to label PMT substrates with a moiety containing a terminal alkyne or azide group in a cell lysate (Blum et al., 2013a; Blum et al., 2013b). The alkyne or azide group can then be conjugated with other probes for in-gel fluorescence visualization and biotin-streptavidin enrichment. To overcome the membrane permeability issue of SAM analogues and, thus, be able to use BPPM in native cellular contexts, we subsequently developed live-cell BPPM (lcBPPM), which features an engineered methionine adenosyltransferase (MAT), which catalyzes the biosynthesis of SAM analogues from membrane-permeable methionine analogues, inside live cells (Wang et al., 2013). The resulting SAM analogues can then be utilized in-situ by engineered PMTs for lcBPPM (Fig. 2).

Figure 1.

Schematic description of the Bioorthogonal Profiling of Protein Methylation (BPPM) technology. SAM is used by native PMTs as a cofactor for substrate methylation, with S-adenosylhomocysteine as a byproduct. In BPPM, PMTs can be engineered to accommodate bulky SAM analogue cofactors, resulting in the modification of PMT substrates with distinct alkyl groups. Biorthogonality is achieved because these SAM analogue cofactors can only be processed by the engineered PMTs, for substrate labeling.

Figure 2.

Schematic description of live-cell Bioorthogonal Profiling of Protein Methylation (lcBPPM). In the context of BPPM, methionine analogues (in pink) can penetrate into the cells, likely by crossing the plasma membrane via specific amino acid transporters, and then be converted into SAM analogues by promiscuous methionine adenosyltransferase (MAT) variants. The resulting SAM analogues can then be recognized in situ by BPPM-feasible PMTs and used as cofactors to label PMT substrates (in blue). These labelled substrates can then be enriched and identified via mass spectroscopy or western Blot.

Through our efforts in the past decade, the BPPM protocols have been used to annotate PMT substrates with in vitro biochemical assays, cell lysates, and live cells (Ibáñez et al., 2010; Islam et al., 2011; Luo, 2012, 2015a, 2018; Wang et al., 2011a; Wang et al., 2013; Wang and Luo, 2013; Wang et al., 2011b). We have further modified the BPPM protocols for robustness, efficiency, and general applicability (Blum et al., 2013a; Blum et al., 2013b; Su et al., 2021). As a result, multiple versions of the BPPM protocols can be found in the literature, making it challenging for users to choose the most suitable protocol for their own application.

In the context of the merits of lcBPPM, we thus now describe a standard BPPM protocol with the latest optimized parameters as well as potential variations. Using two representative PMTs (GLP1 and PRMT1) as examples (Islam et al., 2013; Su et al., 2021; Wang et al., 2013; Zhang et al., 2015), we show how the lcBPPM protocol can be implemented to profile proteome-wide substrates and validate specific substrates of designated PMTs. We first describe the details for preparing and processing cells for lcBPPM (Basic Protocol 1, Fig. 3). We then describe our latest protocol for the alkyne-azide click reaction, which has been optimized for efficiency for labelling of lcBPPM substrates with biotin- or fluorophore-containing probes (Basic Protocol 2 and Alternate Protocol 1, Figure 3). Alternate Protocol 1 focuses on the fluorophore-tagged lcBPPM substrates for in-gel visualization, and is suitable to rapidly assess the labeling efficiency of a lcBPPM protocol with a small portion of the lcBPPM sample. It also allows users to optimize conditions before embarking on a large-scale experiment. In Basic Protocol 3, which has been developed for identification and validation of PMT substrates, the biotin-tagged lcBPPM substrates are enriched via streptavidin pull-down. The sample of Basic Protocol 3 can then be subject to two subsequent applications (Figure 3): LC-MS-MS, for uncovering methylome-wide PMT targets (Basic Protocol 4) and immunoblotting, for validating individual PMT targets (Basic Protocol 5). Basic Protocol 4 requires additional steps for sample processing for use in an MS instrument but is robust for revealing PMT targets in an unbiased manner. Basic Protocol 5, on the other hand, is straightforward for validating individual candidates, but is limited by its low throughput and the detection threshold of antibodies. Upon examining less-characterized PMTs, we strongly recommend including Alternate Protocol 1 for rough assessment of lcBPPM efficiency, followed by either Basic Protocol 4 for comprehensive target identification or Basic Protocol 5 for validation of individual PMT targets. For access to the precursor of the most commonly used SAM analogue in lcBPPM, we also include our latest protocol for gram-scale synthesis of Hey-Met (Support Protocol 1).

Figure 3.

General overview of the workflow and protocols described in this article.

STRATEGIC PLANNING

To reveal or validate the targets of specific PMTs in native contexts, three sets of critical reagents should be assessed in advance for lcBPPM feasibility : BPPM-feasible (engineered) PMTs paired with negative controls, a suitable MAT variant with expanded substrate specificity, and methionine analogues for the engineered PMTs and MAT variants. Among other variations in a lcBPPM protocol are the choice(s) of relevant cellular contexts, the methods to introduce engineered PMTs and MATs into relevant cellular contexts, tags on engineered PMTs, MATs, and candidate proteins to assess protein expression, and modules of characterizing BPPM-revealed target candidates. We discuss these below.

lcBPPM-feasible PMT variants and paired controls.

The vectors expressing lcBPPM-feasible PMT variants and their controls should be propagated prior to implementing this protocol. Here, we paired the vectors for the GLP1 Y1211A variant and the PRMT1 M48G variant with either a vector encoding a catalytically-dead version or with an empty vector as controls, respectively, to showcase the lcBPPM protocol (Islam et al., 2012; Su et al., 2021; Wang et al., 2013; Wang et al., 2011b; Zhang et al., 2015). Users can refer to our published BPPM strategies for details on how to engineer protein lysine methyltransferases (PKMTs: G9a and GLP1) containing a SET domain, and protein arginine methyltransferases (PRMTs: PRMT1 and PRMT3) embedded with the canonical Rossmann-fold-like topology of the class I methyltransferases, to process bulky SAM analogue cofactors such as Ab-SAM, EnYn-SAM, Pob-SAM, and Hey-SAM for substrate labeling (Guo et al., 2014; Islam et al., 2012; Islam et al., 2013; Luo, 2012; Wang and Luo, 2013; Wang et al., 2011b).(Guo et al., 2014; Islam et al., 2012; Islam et al., 2011; Qi et al., 2012; Wang et al., 2011b) (Chen et al., 2019; Guo et al., 2014; Islam et al., 2012; Islam et al., 2013; Islam et al., 2011; Su et al., 2021; Wang and Luo, 2013; Wang et al., 2011b; Zhang et al., 2015). As mentioned, here, we matched the Y1211A variant of GLP1 with the catalytically dead Y1142A variant as a control, and the M48G variant of PRMT1 with an empty vector, as a control. Among other choices of lcBPPM controls are the use of parental PMTs with no lcBPPM activity, lcBPPM-inert PMT homologs, doxycycline-inducible Tet-on/off expression of lcBPPM-feasible PMTs, and methylation-inert substrate variants. These lcBPPM sample-control pairs can also be coupled with well-established SILAC (stable isotope labeling by amino acids) and TMT (tandem mass tag) labeling protocols for mass spectrometry-based quantitative analysis.

MAT variants with expanded substrate specificity.

Users should have access to the vector to express the human MAT I117A variant before implementing this protocol. We and other laboratories have engineered MATs of multiple species for broad substrate specificity (Huber et al., 2020; Huber et al., 2016; Singh et al., 2014; Wang et al., 2013; Wang et al., 2014). Given the special need of our BPPM technology for S-alkyl SAM analogues, we compared the catalytic efficiency of known MAT variants to process S-alkyl methionine analogues (Wang et al., 2014). The I117A mutant of human MAT2A was identified to be most suitable for lcBPPM because of its broad specificity to S-alkyl methionine substrates and high efficiency to process them in addition to methionine (Wang et al., 2013; Wang et al., 2014).

Choices of lcBPPM-feasible methionine analogues.

Hey-Met, the biosynthetic precursor of Hey-SAM, should be prepared in gram-scale for the current protocol. Because of general applicability of Hey-SAM as a product of the I117A MAT2A variant and the substrates of the GLP1 Y1211A variant and the PRMT1 M48G variant, we chose Hey-Met, to exemplify the lcBPPM protocol described here. When choosing S-alkyl methionine analogues for lcBPPM, users should determine whether they can be effectively processed by native or mutated MATs (human MAT2A I117A variant in this demonstration) into the corresponding SAM analogues (Wang et al., 2013; Wang et al., 2014). These S-alkyl SAM analogues can then be processed by lcBPPM-feasible PMT variants (e.g. GLP1 Y1211A variant and PRMT1 M48G variant, as used here) to label the same set of substrates of the native PMTs. In addition, the S-alkyl moieties should contain a chemical handle for direct detection, enrichment, or further modification. In the course of developing the BPPM technology, we mainly focused on chemical moieties containing a terminal-alkyne or azide group (e.g. Ab-SAM, EnYn-SAM, Pob-SAM and Hey-SAM), given their readiness for further modification via a well-established click reaction (Fig. 4) (Islam et al., 2012; Islam et al., 2013; Islam et al., 2011; Wang et al., 2011b). Here, Hey-SAM can be yielded by the I117A MAT2A variant, and can be recognized as a cofactor by both the GLP1 Y1211A variant and the PRMT1 M48G variant.

Figure 4.

Structures of BPPM-feasible SAM analogue cofactors. Ab-SAM, EnYn-SAM, Pob-SAM, and Hey-SAM have demonstrated their uses as BPPM cofactors (Guo et al., 2014; Islam et al., 2012; Islam et al., 2013; Luo, 2012; Wang and Luo, 2013; Wang et al., 2011b).

Other variations of lcBPPM.

Besides lcBPPM-feasible PMT variants, MAT variants, and methionine analogues, it is important to be aware that PMT-associated methylation can be highly context-specific, and lcBPPM should thus be performed with cell lines relevant to the biology of the examined PMTs. Here, we use HEK293T cells for demonstration.

Besides transient transfection of the lcBPPM-feasible PMT and MAT constructs, these enzyme variants can be introduced through stable viral transduction with or without knock-out/knock-down or CRISPR/Cas9 editing of the parental genes. For perturbation of native PMTs, such as constitutive knock-down/out or CRISPR/Cas9 editing, caution should be taken not to affect PMT-associated biology or transform the cells in an unexpected manner. It is also convenient to add tags to exogenous PMT and MAT variants and thus assess the efficiency of their expression or even determine their expression in a temporal manner (e.g. using a dTag or mAID conjugate coupled with their respective degrader to control the fate of exogenous PMTs) (Nabet et al., 2018; Yesbolatova et al., 2020).

Here, we include a protocol for labeling PMT substrates with a terminal-alkyne group as a chemical handle. This moiety can be coupled with various azide-containing probes, via the copper-catalyzed azide-alkyne cycloaddition (CuAAC) click reaction, for further characterization. In the example outlined in this article, we focused on the use of a TARMA-azide dye for target visualization and a biotin-azide probe for streptavidin enrichment. Among the cleavable biotin-azide probes available via Click Chemistry Tools (www.clickchemistrytools.com), we chose the one containing a chemically cleavable diazo linker for ready release of candidate proteins from streptavidin beads with sodium dithionite (Fig. 5). To examine compartment-specific protein candidates, such as those in the nucleus, the mitochondrion, or in chromatin, compartment-specific proteins could be first isolated after the cell lysis step.

Figure 5.

Structures of cleavable biotin-azide probes from Click Chemistry Tool. Diazo biotin-azide was used in our protocol to label terminal-alkyne-containing BPPM candidates.

BASIC PROTOCOL 1: Live-cell labeling of substrates of protein methyltransferases GLP1 and PRMT1 with lcBPPM-feasible enzymes and SAM analogue precursors.

GLP1 methylates its canonical substrate histone H3 as well as diverse biologically relevant non-histone substrates such as DNMT3A (Chang et al., 2011), p53 (Huang et al., 2010), C/EBP-β (Pless et al., 2008), Reptin (Lee et al., 2010), and MyoD (Ling et al., 2012) without consensus substrate-recognizing sequences. Similarly, PRMT1, the PRMT paralogue responsible for 80% of cellular arginine methylation in humans (Blanc and Richard, 2017), acts on the histone substrates H3R2 and H4R3, and a plethora of non-histone substrates at both canonical RGG/RXR and other diverse sequence motifs (Su et al., 2021). To effectively reveal GLP1- and PRMT1-dependent methylation events, we have developed the BPPM technology, for which PMTs such as GLP1 and PRMT1 were engineered to process SAM analogue factors for substrate labeling (Islam et al., 2012; Islam et al., 2011; Wang et al., 2011b). Here, we have chosen GLP1 and PRMT1 to illustrate the protocol for live-cell BPPM (lcBPPM). The user will express BPPM-feasible the GLP1 Y1211A or PRMT1 M48G variants in the presence of the MAT2A I117A variant in HEK293T cells. Cells will then be fed with the BPPM-feasible methionine analogue Hey-Met for in-situ production of the SAM analogue cofactor Hey-SAM for target labeling. This protocol describes the stepwise details for processing the cell lysate and then precipitating the proteome for further analysis.(Blum et al., 2013a; Blum et al., 2013b). A large-scale protocol is provided for rigorous target identification and validation, as well as a small-scale protocol (instructions shown in parenthesis in each step, see below) for rapid assessment of lcBPPM’s feasibility. A summary of the approach is shown in Fig. 6.

Figure 6.

General description of Basic Protocol 1 for live-cell Bioorthogonal Profiling of Protein Methylation, paired with either negative controls or SILAC. PMT substrates (in blue) are labeled with engineered PMTs and Hey-SAM inside living cells. The resulting proteins are processed as described in the protocol steps. Among the options for negative controls, are, in grey, (a) catalytically-dead PMTs and a no-enzyme control, and (b) methylation-inert substrate mutants. In addition, control cells can be cultured in an isotopic medium for SILAC labeling for MS-based quantitative comparison (ratio of proteins in blue versus those in grey, (c)).

Materials:

• Human Embryonic Kidney 293T (HEK293T) cells (ATCC, cat. no. CRL-11268)

• Dulbecco’s Modified Eagle Medium (DMEM), high glucose, HEPES (Thermo Fisher Scientific, cat. no. 12430104) supplemented with 10% (v/v) Dialyzed Fetal Bovine Serum (FBS) (Thermo Fisher Scientific, cat. no. 26400–044) (DMEM-FBS)

• DMEM, high glucose, no glutamine, no methionine, no cysteine (Drop-out DMEM) (Thermo Fisher Scientific, cat. no. 21013024)

• L-glutamine (Thermo Fisher Scientific, cat. no. 25030081)

• L-cysteine (Sigma-Aldrich, cat. no.168149)

• In-house synthesized (E)-hex-2-en-5-ynyl homocysteine (Hey methionine analogue, Hey-Met) (Islam et al., 2013; Wang et al., 2013; Wang et al., 2014) (Support Protocol 1)

• pCDNA3-FLAG-MAT2A I117A plasmid or pCDNA3–6×His-MAT2A I117A plasmid (Su et al., 2021; Wang et al., 2013)

• pCDNA3-FLAG-GLP1 plasmids (the BPPM-active Y1211A variant, and the catalytically dead Y1142A variant as control) (Islam et al., 2013) or pCDNA3-PRMT1 plasmids (the BPPM-active M48G variant, and an empty vector as a control) (Su et al., 2021; Zhang et al., 2015)

• Lipofectamine 3000 Transfection Reagent (Invitrogen, cat. no. L3000)

• Opti-MEM Reduced Serum Medium (Thermo Fisher Scientific, cat. no. 11058021)

• Phosphate-buffered saline (PBS) (Invitrogen, cat. no. 10010–031)

• Trypsin-EDTA (0.25%) (Thermo Fisher Scientific, cat. no. 25200056)

• Lysis buffer (See Reagents and Solutions)

• Pierce Detergent Compatible Bradford Assay Kit (Thermo Fisher Scientific, cat. no. 23246)

• Methanol (Sigma Aldrich, cat. no. 322415)

• Distilled deionized water (ddH₂O)

• 37 °C, 5% CO₂ Cell culture incubator

• 25-cm² (T25) cell culture flasks (Falcon, cat. no. 430639)

• 150-cm² (150 mm) cell culture dishes (Falcon, cat. no. 353025)

• 15 mL Falcon tubes

• 50 mL Falcon tubes

• 1.5 mL microcentrifuge tubes

• Vortex

• UV-visible spectrophotometer

• Misonix Ultrasonic Liquid Processor Sonicator (or an equivalent sonicator)

• Refrigerated centrifuge with 4 °C, 3,000 × g capacity

• Refrigerated high-speed centrifuge with 4 °C, 21,000 × g capacity

Protocol steps:

Co-transfection of HEK293T cells with the engineered MAT2A and PMTs (BPPM-active variants, with either empty vectors or catalytically dead variants as controls) for live-cell labeling of PMT substrates

• 1
  Seed 5.0 ×10⁶ (or 10⁵for small scale preparation, see note) HEK293T cells with 20 mL (5 mL) DMEM-FBS medium in a 150 mm (T25) flask in an incubator at 37 °C, 5% CO₂.

  • The figures and parameters shown in italic in parenthesis are for small-scale preparation and in-gel visualization of lcBPPM candidates (Alternate Protocol 1 Step 2).
• 2
  Culture the cells in the DMEM-FBS medium to 40% confluency (3~5 days).

  • While a regular medium is sufficient to reveal target proteins with MS coupled with a TMT tagging protocol, a regular medium can be paired with a SILAC medium for quantitative evaluation of BPPM targets versus nonspecific controls (Basic Protocol 4).
• 3
  In a 1.5 mL microcentrifuge tube, mix 25 μg (5 μg) of the MAT2A I117A plasmid, 25 μg (5 μg) of the PMT plasmid, and 80 μg (16 μg) of the P3000 Lipofectamine reagent in 1.25 mL (250 μL) of Opti-MEM (Mixture A). In another 1.5 mL microcentrifuge tube, mix 1.25 mL (250 μL) of Opti-MEM with 80 μL (16 μL) of Lipofectamine 3000 reagent (Mixture B). Incubate the two mixtures separately at ambient temperature (22 °C) for 5 minutes (min). Then, combine mixture A and B, pipetting up and down gently, and incubate at ambient temperature (22 °C) for an additional 20 min.

  • In parallel, also transfect the cells with the catalytically dead or empty vectors, as control. The following steps apply to all transfections. As an example, here, we paired the pCDNA3-FLAG plasmid of MAT2A I117A with the BPPM-active Y1211A variant of GLP1 and the catalytically dead Y1142A variant as a control(Islam et al., 2012; Islam et al., 2013; Islam et al., 2011); the pCDNA3–6×His MAT2A I117A plasmid with the BPPM-active M48G variant of PRMT1 and an empty vector as a control (Su et al., 2021; Wang et al., 2013; Wang et al., 2011b; Zhang et al., 2015). To validate the sites of methylation, the paired plasmids containing tagged PMT substrates and their methylation-inert Lys/Arg mutants are also transfected at this step (Fig. 6). The sample-control BPPM samples should be paired and processed in parallel thereafter.
• 4Add the plasmid mixture to the 40%-confluent cells (Step 2) and allow the transfection to proceed for 24 hours.
• 5
  Carefully aspirate the transfection medium and replace with 20 mL (5 mL) fresh DMEM-FBS medium to recover for 24–48 hours in the incubator.

  • We have determined optimal recovery time for these particular plasmids (tGLP1 transfection, 24 hours; PRMT1 transfection, 48 hours). These conditions may vary for other plasmids.
• 6Aspirate off the recovery medium and carefully wash the cells with 10 mL (3 mL) of PBS. Incubate the cells in 30 mL (5 mL) of Drop-out DMEM freshly supplemented with 0.584 g/L L-glutamine, 0.0626 g/L L-cysteine, and 10% (v/v) dialyzed FBS, for 30 min.
• 7
  Aspirate off the old Drop-out DMEM and replace with fresh 20 mL (5 mL) of Drop-out DMEM supplemented with L-glutamine, L-cysteine, and dialyzed FBS, plus Hey-Met.

  • For this particular example, the final concentration of Hey-Met is 1.0 mM for GLP1 or 0.5 mM for PRMT1 (Su et al., 2021; Wang et al., 2013). The concentrations of Hey-Met were optimized with the in-gel fluorescent labeling protocol for desired signal-to-noise ratios (Alternate Protocol 1). We recommend examining 0.5‒2.0 mM Hey-Met for labeling efficiency.(Wang et al., 2014)
• 8
  Incubate the cells for 8 hours at 37 °C, 5% CO₂.

  • For this particular example, the incubation time was optimized with the in-gel fluorescent labeling protocol for desired signal-to-noise ratios (Alternate Protocol 1).
• 9Carefully aspirate the medium without detaching the cells and gently wash the attached cells with 20 mL (5 mL) of prechilled (4 °C) PBS.
• 10Aspirate the PBS, add 10 mL (3 mL) of pre-warmed (37 °C) trypsin, and incubate the trypsin-treated cells in a 37 °C, 5% CO₂ incubator for 5 min.
• 11Add 20 mL (6 mL) of DMEM to dilute the trypsin and pipette-mix the mixture. Collect the 30 mL (9 mL) suspension in a 50 mL (15 mL) Falcon tube and centrifuge at 2,000 × g for 5 min at ambient temperature (22 °C).
• 12
  Aspirate the supernatant and add 5 mL (1 mL) PBS to resuspend the cell pellet. Transfer the suspension into a 15 mL microcentrifuge tube (a 1.5 mL tube) and centrifuge at ambient temperature (22 °C) at 2,000 × g for 5 min. Carefully aspirate the PBS. The resultant cell pellet is ready for lysis.

  • The cell pellet can be stored at ‒ 80 °C before use.

Preparation of cell lysate

• 13Lyse the cell pellet with 5 mL (1 mL) of lysis buffer (see Reagents and Solutions) and incubate the mixture on ice for 20 min.
• 14
  Sonicate the sample in an ice bath with a single pulse at 65% amplitude for 20 min.

  • Immerse the sample in ice during sonication to prevent heat-induced protein degradation or denaturation. The lysate becomes translucent after the pellet is fully lysed.
• 15
  Centrifuge the sample at 4 °C at 15,000 × g for 20 min and collect the supernatant.

  • The sample can be split into several batches for large-scale preparation.
• 16Measure the protein concentration of the lysate using the Pierce Bradford Assay Kit per the manufacturer’s instructions.
• 17
  Reserve an aliquot of ~200 μg of protein as the input sample for western blot (Basic Protocol 5, Step 3) and proceed with the rest for protein precipitation.

  • We recommend processing the lysate with minimal delay until the completion of click labeling (Basic Protocol 2 and Alternate Protocol 1).

Precipitation of alkyne-labeled proteome

• 18Add 2–3 volumes of pre-chilled methanol (‒ 80 °C) (~15 mL for large-scale; ~3 mL for small-scale) to the sample.
• 19Keep the mixture at ‒ 80 °C overnight (12~16 hours).
• 20Centrifuge the sample at 3,200 × g at 4 °C for 20 min.
• 21Carefully remove the methanol supernatant and add 5 mL (1 mL) of pre-chilled (‒ 80 °C) methanol. Mix with gentle vortexing.
• 22Centrifuge the sample at 3,000 × g for 20 min at 4 °C.
• 23Repeat Steps 21–22.
• 24
  Carefully remove the methanol supernatant and air dry the pellet to near completion (~20 min). The resultant protein pellet is ready for click labeling (Basic Protocol 2 or Alternate Protocol 1)

  • It is important NOT to completely dry the protein pellet to avoid potential difficulty in dissolving the pellet in subsequent steps. Process the sample with minimal delay. Roughly 500 μg and 5~10 mg protein can be yielded from one T25 flask and one 150 mm flask of HEK293T cells, respectively.

BASIC PROTOCOL 2: Click labeling of lcBPPM cell lysates with a biotin-azide probe

To prepare the protein pellets yielded from Basic Protocol 1 for subsequent identification and validation of targets of GLP1 and PRMT1, the BPPM candidates labeled with a terminal-alkyne group are subject to copper-catalyzed azide-alkyne cycloaddition (CuAAC) —a click reaction— with a biotin probe (Fig. 7). Here, we describe newly optimized protocols to prepare the cocktails for the click reaction. This recipe was developed from our previous protocols (Blum et al., 2013a; Blum et al., 2013b), with some modifications. For instance, sodium ascorbate replaces TCEP, used in previous protocols, as a reducing agent (TCEP is associated with higher background labeling for an unknown reason). TBTA is replaced with BTTP, which shows better aqueous solubility and is also commercially available. While a higher concentration of SDS (e.g. 4% (w/v)) used in the previous protocols makes the protein pellet readily dissolve, a high amount of SDS suppresses the click reaction in our assay conditions. The optimized SDS concentration for the click reaction should not exceed 1% (w/v). Upon completion of this protocol, users will have biotin-labeled target candidates, processed with the large-scale lcBPPM sample, for proteome-wide target enrichment (Basic Protocol 3). While the steps in the protocol are presented for processing one lcBPPM sample, individual lcBPPM samples should be paired with their controls and processed in parallel here and thereafter.

Figure 7.

General description of Basic Protocol 2 and Alternate Protocol 1. BPPM-revealed protein candidates (in blue) are subject to the azide-alkyne click reaction with either a biotin probe (Basic Protocol 2) or a fluorescent dye (Alternate Protocol 1). The biotin-labeled substrates can then be subject to streptavidin beads for enrichment for target identification and validation, while the fluorescently-labeled substrate candidates can then be visualized via in-gel fluorescence for rapid assessment of feasibility of the lcBPPM protocol.

Materials:

• Protein pellet (Basic Protocol 1, Step 24)

• BTTP (Click Chemistry Tools, cat. no. 1414)

• CuSO₄-5H₂O (Sigma-Aldrich, cat. no. 203165)

• Sodium ascorbate (Sigma-Aldrich, cat. no. A7631)

• Diazo biotin-azide probe (Click Chemistry Tools, cat. no. 1041)

• Dimethylsulfoxide (DMSO) (Sigma-Aldrich, cat. no. D8418)

• Distilled deionized water (ddH₂O)

• Click reaction buffer (see Reagents and Solutions)

• Click dilution buffer (see Reagents and Solutions)

• Methanol (Sigma Aldrich, cat. no. 322415)

• 15 mL Falcon tubes

• 1.5 mL Microcentrifuge tubes

• Misonix Ultrasonic Liquid Processor Sonicator (or an equivalent sonicator)

• Vortex

• Shaker

Protocol steps:

Large-scale biotinylation of terminal-alkyne-modified proteome by click reaction

1. Prepare the following aqueous stock solutions: 20 mM CuSO₄, 40 mM BTTP, and 200 mM sodium ascorbate.

   • The CuSO₄ and sodium ascorbate solutions should be freshly prepared. The 40 mM BTTP stock solution can be prepared beforehand and stored at ‒ 20 °C until use.

2. To prepare the click cocktail, mix 200 μL of 40 mM BTTP and 200 μL of 20 mM CuSO₄, and incubate the resulting blue mixture at ambient temperature (22 °C) for 1 hour. Into this mixture, add 50 μL of the 200 mM sodium ascorbate solution, to yield 450 μL of a colorless premixed click cocktail.

3. Resuspend the protein pellet (the large-scale product of Basic Protocol 1, Step 24) in 1 mL of the click reaction buffer (see Reagents and Solutions).

   • Modest sonication with Misonix Ultrasonic Liquid Processor Sonicator may facilitate this resuspension process. Sonicate at the 65% setting with a single pulse for 30 seconds to fully dissolve the sample. Otherwise, continue with 30-second pulses until the sample completely dissolves.

4. Into this suspension, add 2.45 mL of the click dilution buffer (see Reagents and Solutions), for a final volume of 3.45 mL.

5. Add 450 μL of the pre-mixed click cocktail (Step 2) to the mix.

6. Add 100 μL of 10 mM diazo biotin-azide DMSO stock solution to yield 4 mL of the click reaction mixture, with a final SDS concentration of 1% (w/v).

7. Incubate this mixture at ambient temperature (22 °C) in the dark, with shaking for 2 hours.

   • Incubation in the dark is required, as there are light-sensitive components in the click reaction mixture.
8. Repeat Basic Protocol 1 Step 18‒24 (Precipitation of alkyne-labeled proteome) to yield a biotinylated protein pellet ready for Basic Protocol 3.

   • It is important NOT to completely dry the protein pellet to avoid potential difficulty in dissolving the pellet in subsequent steps, and to process the sample with minimal delay.

ALTERNATE PROTOCOL 1: Click labeling of small-scale lcBPPM cell lysates with a TARMA-azide dye for In-gel fluorescence visualization

To rapidly assess lcBPPM feasibility for substrates of your protein of interest (e.g. GLP1 and PRMT1) with small portion of the sample, users can follow the protocol described here for rapid visualization using the protein sample yielded via Basic Protocol 1 (Fig. 7). Here, the small-scale protein pellet yielded from Basic Protocol 1 (lcBPPM candidates) labeled with a terminal-alkyne group is subject to a CuAAC click reaction with a fluorescent TAMRA dye. The TAMRA-labeled PMT substrate candidates are resolved by SDS-PAGE and visualized by in-gel fluorescence. One challenge of this protocol is that non-specific labeling of the TAMRA dye causes high background. Therefore, the BPPM samples should be paired with controls for side-by-side comparison. This approach is suitable for rapid evaluation of the efficiency or optimization of conditions of live-cell BPPM labeling. For robust target identification and validation, we recommend more reliable detection methods: LC-MS-MS (Basic Protocol 4) and immunoblot (Basic Protocol 5).

Materials:

• Protein pellet from Basic Protocol 1, Step 24

• BTTP (Click Chemistry Tools, cat. no. 1414)

• CuSO₄-5H₂O (Sigma-Aldrich, cat. no. 203165)

• Sodium ascorbate (Sigma-Aldrich, cat. no. A7631)

• TAMRA azide dye (Click Chemistry Tools, cat. no. AZ109)

• Dimethylsulfoxide (DMSO) (Sigma-Aldrich, cat. no. D8418)

• Distilled deionized water (ddH₂O)

• Click reaction buffer (see Reagents and Solutions)

• Click dilution buffer (see Reagents and Solutions)

• Methanol (Sigma Aldrich, cat. no. 322415)

• Gel loading buffer (see Reagents and Solutions)

• 20× NuPAGE MOPS SDS Running Buffer (Thermo Fisher Scientific, cat. no. NP0001)

• Gel fixing buffer (see Reagents and Solutions)

• Coomassie Brilliant Blue R-250 Staining Solution (Bio-Rad, cat. no. 1610436).

• 1.5 mL Microcentrifuge tubes

• Vortex

• Shaker

• 4–12% Criterion XT Bis-Tris Protein Gel (Bio-Rad, cat. no. 3450123)

• SDS-PAGE Electrophoresis

• Amersham Biosciences Typhoon 9400 gel scannerMisonix Ultrasonic Liquid Processor Sonicator (or an equivalent sonicator)

Protocol steps:

Small-scale fluorescent labeling of terminal-alkyne-modified proteome by click reaction

1. Prepare the following aqueous stock solutions: 20 mM CuSO₄, 40 mM BTTP, and 100 mM sodium ascorbate.

   • The CuSO₄ and sodium ascorbate solutions should be freshly prepared. The 40 mM BTTP stock solution can be prepared beforehand and stored at ‒ 20 °C until use.

2. To prepare the click cocktail, containing 2 mM BTTP, 1 mM CuSO₄, and 2.5 mM sodium ascorbate, mix 5 μL of 40 mM BTTP and 5 μL of 20 mM CuSO₄, and incubate the resulting blue mixture at ambient temperature (22 °C) for 1 hour. Into this mixture, add 2.5 μL of the 100 mM sodium ascorbate solution, to yield 12.5 μL of a colorless premixed click cocktail solution.

3. Resuspend the protein pellet (the small-scale product of Basic Protocol 1, Step 24) in 100 μL of the click reaction buffer (see Reagents and Solutions).

   • Modest sonication may facilitate this resuspension process. Sonicate at the level of 65% with a single pulse for 30 seconds. This may be sufficient to achieve full resuspension (no cloudiness). If not, continue with 30 second pulses and check the sample until full resuspension is achieved.

4. Take a 25 μL aliquot of this suspension (save the remaining 75 μL for future use), and add 60 μL of click dilution buffer.

5. Add the 12.5 μL click cocktail solution from Step 2 to the mix.

6. Add 2.5 μL of 10 mM TAMRA-azide DMSO stock solution, to yield 100 μL of the click reaction mixture with a final SDS concentration of 1% (w/v).

7. Incubate this mixture at ambient temperature (22 °C) in the dark, with shaking for 2 hours.

   • Incubation in the dark is required, as there are light-sensitive components in the click reaction mixture. We have determined the optimal reaction time on the basis of overall signal-to-noise ratios obtained following Alternate Protocol 1 for our proteins of interest. For abundant PMT targets, the click reaction time can be shortened.

8. Repeat Basic Protocol 1, Step 18‒24 (Precipitation of alkyne-labeled proteome) to yield the TAMRA-labeled protein pellet.

9. Dissolve the pellet in 200 μL of gel loading buffer.

10. Spin down the sample at 15,000 x g for 15 min at ambient temperature (22 °C).

11. Take a 40 μL aliquot for SDS-PAGE and heat-denature at 98 °C for 5 min.

12. Load the denatured aliquot onto a 4–12% Bis-Tris gel and run in 1X MOPS buffer at 160V for 80 min.

    • Protect the gel box from light.
13. Fix the gel in gel fixing buffer for at least 1 hour, protected from light, with gentle shaking.

    • The gel fixing step also removes unreacted TAMRA dye.

14. Rinse the gel in H₂O three times for rehydration.

15. Image the gel using an Amersham Biosciences Typhoon 9400 gel scanner using the channel for the TAMRA-labeled proteome (approximate excitation/emission maxima ~546/579).

16. After collecting the fluorescent image, stain the gel with Coomassie brilliant blue for protein loading control. Submerge the gel at ambient temperature (22 °C) for 20 min with gentle shaking.

17. Destain the gel in water.

18. Image the gel with a regular scanner to obtain total protein stained with Coomassie blue.

BASIC PROTOCOL 3: Enrichment of biotinylated BPPM proteome with streptavidin beads

To reveal proteome-wide targets or validate specific targets of, for instance, GLP1 and PRMT1 with lcBPPM, the protocol in this section describes how the biotin-labeled proteome prepared via Basic Protocol 2 is enriched by streptavidin beads (Fig. 8). As a key feature of this protocol, users take advantage of a chemically cleavable linker in the biotin-azide probe (Fig. 5, Basic Protocol 2) to selectively release the biotin-labeled, streptavidin-sepharose-trapped proteins with sodium dithionite under mild conditions. This step is expected to reduce the background signal caused by streptavidin-sepharose-associated nonspecific binding. The completion of this protocol results in a lyophilized protein sample, which can then be used in two ways: to reveal proteome-wide target candidates by LC-MS-MS (Basic Protocol 4) or to validate target candidates of interest with western blot (Basic Protocol 5).

Figure 8.

General description of Basic Protocols 3–5. The biotin-labeled proteome (Basic Protocol 2) can be enriched with streptavidin beads and released from the beads by treating them with sodium dithionite (Basic Protocol 3). The resulting samples can either be subject to LC-MS-MS to reveal proteome-wide substrates candidates (Basic Protocol 4) or be processed with western blot against specific target candidates (Basic Protocol 5).

Materials:

• Protein pellet from Basic Protocol 2, Step 8

• 50 mL Falcon tubes

• 15 mL Falcon tubes

• Click reaction buffer (see Reagents and Solutions)

• BCA Protein Assay kit (Thermo Fisher Scientific, cat. no. 23227)

• Pull-down dilution buffer (see Reagents and Solutions)

• High Capacity Streptavidin Agarose Resin (Thermo Fisher Scientific, cat. no. 20359)

• Phosphate-buffered saline (PBS) (Invitrogen, cat. no. 10010–031)

• PBS supplemented with 0.2% (w/v) sodium dodecyl sulfate (SDS)

• 250 mM Ammonium bicarbonate buffer (ABC buffer)

• Distilled deionized water (ddH₂O)

• 1% (w/v) SDS ddH₂O solution

• 1% (w/v) SDS ddH₂O solution supplemented with 75 mM 2-mercaptoethanol

• Elution buffer (see Reagents and Solutions)

• Liquid N₂

• 2-mL Dolphin microcentrifuge tubes (Sigma-Aldrich, cat. no. Z717533)

• Amicon Ultra Centrifugal Filter (3KDa, NMWL) (Millipore Sigma, cat. no. UFC500396)

• Refrigerated centrifuge with 4 °C, 3,000 × g capacity

• Refrigerated high-speed centrifuge with 4 °C, 21,000 × g capacity

• Vortex

• End-over-end rotator

• UV-visible Spectrophotometer

• Lyophilizer

• Misonix Ultrasonic Liquid Processor Sonicator (or an equivalent sonicator)

Protocol steps:

Resuspension of the protein pellet

• 1
  Dissolve the protein pellet product from Basic Protocol 2, Step 8 in 1 mL of the click reaction buffer (see Reagents and Solutions).

  • Modest sonication with Misonix Ultrasonic Liquid Processor Sonicator may facilitate this resuspension process. Sonicate at the 65% setting with a single pulse for 30 seconds to fully dissolve the sample. Otherwise, continue with 30-second pulses until the sample completely dissolves.

1. Measure protein concentration with the BCA Protein Assay kit, following the manufacturer’s instructions.

2. Transfer 5 mg of the protein sample into a 15 mL Falcon tube and add 2 mL of the pull-down dilution buffer (see Reagents and Solutions).

   • 4% (w/v) SDS in the click reaction buffer is essential to dissolve the protein pellet. The pull-down dilution buffer is then added to reduce the SDS concentration to around 1.5 % (w/v) for better biotin-streptavidin binding.

Preparation of streptavidin beads

• 2
  Transfer 100 μL of the High Capacity Streptavidin Agarose beaded resin into a 15 mL Falcon tube.

  • The binding capacity of the streptavidin resin is around 10 mg/mL.
• 3Add 5 mL of PBS and equilibrate the mixture for 10 min with end-over-end rotation.
• 4Spin down the beads at ambient temperature (22 °C) at 4,000 × g for 2 min.
• 5Carefully remove the supernatant, add 5 mL PBS, and equilibrate the resulting mixture for 10 min.
• 6Repeat Steps 6–7. Spin down the beads at ambient temperature (22 °C) at 4,000 × g for 2 min.
• 7Carefully remove the supernatant, and then resuspend the beads in 250 μL of the pull-down dilution buffer (see Reagents and Solutions).

Sample enrichment by streptavidin beads

• 8Add the 5 mg of dissolved protein sample (Step 3) to the streptavidin bead suspension from Step 9.
• 9Incubate the mixture at ambient temperature (22 °C) for 1 hour with end-over-end rotation.
• 10Centrifuge the mixture at 4 °C at 4,000 × g for 2 min.
• 11Carefully remove the supernatant and add 10 mL of PBS containing 0.2% (w/v) SDS.
• 12Wash the beads by inverting the tube several times and centrifuge the mixture at 4 °C at 4,000 × g for 2 min.
• 13Repeat Step 13–14 twice more.
• 14Carefully remove the supernatant and add 10 mL of the ABC buffer, invert the tube several times, and centrifuge at 4 °C at 4,000 × g for 2 min.
• 15Repeat step 16 once more.
• 16Carefully remove the supernatant, resuspend the beads in 1 mL of ABC buffer, and then transfer the suspension into a 2 mL Dolphin microcentrifuge tube.
• 17Centrifuge at 4 °C at 4,000 × g for 2 min and carefully discard the supernatant.

Protein elution from streptavidin beads

• 18
  Mix 250 μL of elution buffer (see Reagents and Solutions) with the streptavidin beads from Step 19.

  • The elution buffer must be made fresh.
• 19Incubate the mixture at ambient temperature (22 °C) for 30 min.
• 20Centrifuge at 4 °C at 4,000 × g for 2 min.
• 21Collect and save the supernatant at 4 °C until use in step 25.
• 22Add another 250 μL of elution buffer to the streptavidin beads, incubate at ambient temperature (22 °C) for another 30 min, and centrifuge at 4 °C at 4,000 × g for 2 min.
• 23Collect the supernatant and combine the supernatants of Steps 23 and 24.

Concentration of eluted protein

• 24Wash an Amicon Ultra Centrifugal Filter (3KDa, NMWL) with 500 μL of 1% (w/v) SDS ddH₂O solution. Spin down at ambient temperature (22 °C) at 6,000 × g for 30 min.
• 25Wash the centrifugal filter with 500 μL of ddH₂O and spin down at ambient temperature (22 °C) at 6,000 × g for 30 min.
• 26Repeat Step 26 once more.
• 27Add the 500 μL of the combined protein elute (Step 25) to the centrifugal filter. Spin down at ambient temperature (22 °C) at 6000 x g for 30 min.
• 28Invert the filter of the centrifugal unit and spin down at ambient temperature (22 °C) at 1,000 × g for 3 min.
• 29Set the filter straight right-side-up and add 50 μL of 1% (w/v) SDS ddH₂O supplemented with 75 mM 2-mercaptoethanol. Pipette up and down to loosen any remnants that are stuck to the side of the filter.
• 30Transfer the solution into a new collection tube, carefully invert the filter in the collection tube, and centrifuge at ambient temperature (22 °C) at 1,000 × g for 3 min.
• 31Repeat Step 31–32 with the same collection tube used in Step 32.
• 32
  Combine the protein eluates from Steps 30, 32, and 33. Snap-freeze the eluted protein in liquid N₂ and lyophilize to dryness. The resulting protein powder is ready for further analysis with Basic Protocol 4 to reveal proteome-wide target identification or Basic Protocol 5 to validate specific targets.

  • This will give a final volume of ~125–150 μL, and the protein powder can be stored at ambient temperature (22 °C) for a month or at ‒20 °C for 6 months.

BASIC PROTOCOL 4: Proteome-wide identification of BPPM targets with mass spectrometry

To reveal proteome-wide targets of the PMTs of interest (e.g. GLP1 and PRMT1) with lcBPPM in an unbiased top-down manner, the streptavidin-sepharose-enriched samples yielded via Basic Protocol 3 are subject to mass spectrometry (MS) The protocol described in the section is provided in the context of advancements of MS instrumentation, resulting in higher accuracy and sensitivity, to reveal more target candidates, in particular for low-abundant proteins. We note that, even if the in-gel fluorescence approach shows only a few distinct labeling bands (Alternate Protocol 1), MS can still be sufficient to reveal hundreds of candidate proteins. LC-MS-MS analysis may be further combined with the SILAC and TMT-labeling method for quantitative comparison (Fig. 8). Given diverse instrument settings and preferred protocols of different MS facilities, here we only describe key steps to prepare samples for MS analysis. We advised discuss specific needs with staff at MS core facilities.

Materials:

• Lyophilized protein powder from Basic Protocol 3

• 4× Laemmli Sample Buffer (Bio-Rad, cat. no. 1610747)

• 4–12% Criterion XT Bis-Tris Protein Gel (Bio-Rad, cat. no. 3450123)

• 20× NuPAGE MOPS SDS Running Buffer (Thermo Fisher Scientific, cat. no. NP0001)

• Acetonitrile

• Dithiothreitol (DTT) (Thermo Fisher Scientific, cat. no. PR-V3151)

• Iodoacetamide (Thermo Fisher Scientific, cat. no. AC122270050)

• Distilled deionized water (ddH₂O)

• Trypsin Gold (Promega, cat. no. V5280)

• 0.1% Formic acid in 50% acetonitrile

• 1.5-mL Microcentrifuge tubes

• SDS-PAGE Electrophoresis

• Speedvac

• Liquid Chromatography with tandem mass spectrometry (LC-MS-MS)

Protocol steps:

1. Dissolve the lyophilized protein powder (Basic Protocol 3, Concentration of eluted proteins) in 25 μL of 4× Laemmli Sample Buffer. Dilute the mixture with 75 μL ddH₂O and heat the sample at 70 °C for 5 min (or at 100 °C for 2 min) for complete denaturation.

2. Centrifuge the sample at 15,000 × g for 15 min at ambient temperature (22 °C).

3. Load 30 μL onto a 4–12% Bis-Tris gel and run an SDS-PAGE with 1× MOPS buffer (dilute the 20× buffer with ddH₂O) at 160 V for 80 min.

4. Cut each lane of the whole gel into seven equal pieces and place them in a 1.5 mL microcentrifuge tube.

5. Dehydrate the gel slices with 500 μL of acetonitrile for 10 min at ambient temperature (22 °C). Remove the acetonitrile and allow the gel slices to air dry.

6. Treat the cut gel bands with enough volume of 25 mM DTT to cover the gel slices (~100–150 μL total). Incubate the sample in the dark for 20 min at ambient temperature (22 °C).

   • Solution should be added as the gel slices rehydrate to keep them submerged.
7. Remove the DTT and cover the gel slices with 55 mM iodoacetamide in ammonium bicarbonate. Incubate the sample in the dark for 20 min at ambient temperature (22 °C).

   • If a TMT protocol is intended for use with these samples for quantitative analysis, ammonium bicarbonate must be replaced with an amine-free buffer.
8. Remove the solution and cover the gel slices with a trypsin solution (1 μg/80 μL, ~100–150 μL) in 50 mM ammonium bicarbonate to digest the proteins in the gel. Incubate at 37°C for 16 hours.

   • Other proteases can be used instead of or together with trypsin for optimal digestion of target proteins of interest.

9. Extract the digested peptides with 150 μL of 0.1% formic acid in 50% acetonitrile at 37°C for 30 min.

10. Collect the first extract using a pipet and store at room temperature. Repeat Step 8 for a second extraction.

11. Collect the second extract and combine with the first one (Step 9). Concentrate the combined extracted peptides to ~5 μL using a Speedvac. Analyze the sample with a Thermo LTQ-Orbitrap mass spectrometer to reveal protein identities (Islam et al., 2013).

    • Various LC-MS-MS instruments can be chosen here for optimal detection of target proteins of interest.

BASIC PROTOCOL 5: Validation of individual BPPM targets with western blot

Besides the MS-based top-down approach to reveal candidate methylome-wide PMT targets, the streptavidin-sepharose-enriched samples yielded via Basic Protocol 3 can also be used to probe candidate proteins of specific interest (Fig. 8). The protocol described here is equivalent to an immunoblot approach to detect target proteins with antibodies. Here, we use H3 and DUSP4 as the examples of BPPM-revealed PRMT1 targets. The protocol can be generally adapted to examine BPPM-revealed PMT targets.

Materials:

• Lyophilized protein powder from Basic Protocol 3

• 4× Laemmli Sample Buffer (Bio-Rad, cat. no. 1610747)

• 4–12% Criterion XT Bis-Tris Protein Gel (Bio-Rad, cat. no. 3450123)

• 20× NuPAGE MOPS SDS Running Buffer (Thermo Fisher Scientific, cat. no. NP0001)

• 20× TBST (tris-buffered saline + tween 20) buffer (Thermo Fisher Scientific, cat. no. 03-500-537)

• Precision Plus Protein Kaleidescope Prestained Protein Standards (Bio-Rad, cat. no. 1610375)

• anti-DUSP4/MPK-2 (Santa Cruz, cat. no. sc-17821)

• anti-Histone H4 (Cell Signaling, cat. no. 13919)

• Goat anti-rabbit IgG H&L (HRP) (abcam, cat. no. ab6721)

• Goat anti-mouse IgG H&L (HRP) (abcam, cat. no. ab6789)

• Distilled deionized water (ddH₂O)

• Immobilon Forte Western HRP substrate (Millipore, cat. no. WBLUF0100)

• Trans-blot Turbo Midi 0.2 μm Nitrocellulose membrane (Bio-Rad, cat. no. 1704159)

• Blotting-Grade Blocker (Bio-Rad, cat. no. 1706404)

• CL-XPosure Film (Thermo Fisher Scientific, cat. no. 34090)

• SDS-PAGE Electrophoresis reagents

• Trans-Blot Turbo Transfer System (Bio-Rad)

• Film developer

Protocol steps:

1. Dissolve the lyophilized protein powder (Basic Protocol 3) in 25 μL of 4× Laemmli Sample Buffer. Dilute the mixture with 75 μL ddH₂O and heat the sample at 70 °C for 5 min (or 100 °C for 2 min) for complete denaturation.

2. Centrifuge the sample at 15,000 × g for 15 min at ambient temperature (22 °C).

3. Load 30 μL of each sample onto a 4–12% Bis-Tris gel and run SDS-PAGE with 1× MOPS buffer (dilute the 20× buffer with ddH₂O) at 160 V for 80 min.

   • 2~5 % of the total input (the saved aliquot of Basic Protocol 1 Step 17) and the sample of negative controls (empty vectors or catalytically dead PMT variants in Basic Protocol 1 Step 3, or methylation-incompetent substrate variants with targeted Lys/Arg mutated) can be loaded side-by-side for quantification of the BPPM labeling efficiency and signal-to-noise ratios. Additional gels can be run in parallel to probe known substrates such as H3 as positive controls of the BPPM method or multiple targets of interest.

4. Transfer the SDS-PAGE-resolved proteins to a Trans-blot Turbo Midi 0.2 μm Nitrocellulose membrane with the Trans-Blot Turbo Transfer System.

5. Incubate the nitrocellulose membrane in blocking buffer (2.5% Blotting-Grade Blocker in 1× TBST) for 30 min on a shaker at ambient temperature (22 °C).

6. Cut the membrane in half, between the 25kDa and the 37 kDa markers, to preserve the expected H4 band (~10 kDa) and the expected DUSP4 band (~42 kDa).

7. Blot the upper membrane with 5 mL of anti-DUSP4 (1:1000 in blocking buffer), and the lower membrane with 5 mL anti-H4 (1:1000 in blocking buffer), overnight at 4 °C with shaking.

8. Wash the blotted membranes with ~5 mL 1× TBST for 10 min three times at ambient temperature (22 °C).

9. Incubate the upper membrane with 5 mL goat anti-mouse (1:5000 in blocking buffer) and the lower membrane with 5 mL goat anti-rabbit (1:5000 in blocking buffer). Shake the membranes for 1 hour at ambient temperature (22 °C).

   • An HRP-conjugated secondary antibody was used in our experiments. Other antibody conjugates can be considered for optimal detection of target proteins.

10. Wash the resulting membranes with ~5 mL of 1× TBST for 10 min three times.

11. Cover the membranes with a minimal amount of HRP substrate, expose the treated membranes to a film in a dark room for ~30 s, and develop the film to visualize the blotted target protein.

    • The exposure time shown here has been optimized for these individual targets. For other proteins, this will need to be empirically determined.

SUPPORT PROTOCOL 1: Gram-scale synthesis of Hey-Met

Hey-SAM is used as the SAM analogue cofactor in the lcBPPM protocol described in this article to label substates of GLP1 and PRMT1. Because Hey-SAM is not membrane-permeable, we have circumvented this membrane penetration issue by pairing it with an engineered MAT2A for in-situ production of Hey-SAM with Hey-Met, the membrane-permeable biosynthetic substrate precursor of Hey-SAM, for live-cell labeling. Hey-Met is likely internalized by cells via amino acid transporters of broad specificity. In a methionine depleted medium, Hey-Met is readily converted into Hey-SAM by the MAT2A variant. This protocol outlines the key steps to synthesize gram-scale Hey-Met from commercially available starting materials with high yield (Fig. 9). While this protocol is straightforward for a synthetic chemist, we encourage those inexperienced with chemical synthesis to go through the details with experienced chemists before its implementation.

Figure 9.

Synthesis of Hey-Met.

Materials:

• (E)-pent-2-en-4-yn-1-ol (ASW MedChem, cat. no. DH-33159)

• S-Benzyl-L-homocysteine (BOC Sciences, cat. no. 7689-60-3)

• p-Toluenesulfonyl chloride (Alfa Aesar, cat. no. A14547–30)

• Sodium metal (Sigma-Aldrich, cat no. 262714–5g)

• Liquid ammonia

• Ammonia cylinder

• Anhydrous ethanol (Decon Laboratories Inc, cat no. DSP-MD.43)

• Potassium hydroxide (Sigma-Aldrich, cat no. 8143531000)

• Distilled deionized water (ddH₂O)

• 5% Ammonia hydroxide (Fisher Scientific, cat no.628-16) in ddH₂O

• Diethyl ether (Sigma-Aldrich, cat no, 309966-1l)

• 0.1 N hydrochloric acid (Sigma-Aldrich, cat no. 2104-50ml)

• 1.0 N hydrochloric acid (Sigma-Aldrich, cat no. H9892-100ml)

• NaCl (Sigma-Aldrich, cat no. S9888-1kg)

• Anhydrous MgSO₄ (Sigma-Aldrich, cat no. 208094-500g)

• 0.2% Ninhydrin in 16:3:1 (v/v/v) ethanol/acetic acid/S-collidine

• Acetonitrile (Fisher Scientific, cat no. A998-4)

• Trifluoroacetic acid (Sigma-Aldrich, cat no. 8082600101)

• 4:1:1 (v/v/v) n-Butanol/acetic acid/ddH₂O

• Dry-ice ethanol bath

• Argon

• Silica gel thin-layer chromatography (TLC)

• Dowex 50 (H⁺) ion-exchange column

• Rotary evaporator

• Reversed-phase High performance liquid chromatography (HPLC) equipped with a Waters^™ XBridge Prep C18 Column 5.0 μm OBD, 19 × 150 mm)

• Lyophillizer

Protocol steps:

Synthesis of (E)-pent-2-en-4-ynyl tosylate (Wang et al., 2014)

• 1Into 10 mL of diethyl ether, add 165 mg (2.0 mmol) of (E)-pent-2-en-4-yn-1-ol and 480 mg (2.4 mmol) of p-toluenesulfonyl chloride at −5 °C with an ice-salt bath.
• 2Add 560 mg (10 mmol) of potassium hydroxide and stir this mixture with a magnetic stirrer bar at −5 °C with an ice-salt bath for 1 hour.
• 3Remove the reaction mixture from the ice-salt bath and stir at ambient temperature (22 °C) overnight.
• 4Quench the reaction with 20 mL of ice-cold ddH₂O water, allow the organic phase to separate from the aqueous phase, and collect the upper organic phase with an Erlenmeyer flask.
• 5Extract the bottom aqueous phase with 10 mL of diethyl ether twice and combine all organic phase extractions.
• 6Wash the combined organic extraction stepwise with 15 mL of ddH₂O, 1 N HCl, and brine, respectively.
• 7Add around 5 g of anhydrous MgSO₄ until the organic extraction becomes clear, and then dry the solution for 3 hours.
• 8
  Remove volatiles in the organic extraction with a rotary evaporator under vaccum and then weight the oil residue of (E)-pent-2-en-4-ynyl tosylate.

  • The desired product (E)-pent-2-en-4-ynyl tosylate is obtained as a light brown solid, with 85% yield, and is directly used as a starting material in Step 17. If required, its purity is confirmed by ¹H-NMR.

Synthesis of Hey-Met (Wang et al., 2014).

• 9Place 255 mg (1.0 mmol) of S-benzyl-L-homocysteine into a round-bottom flask connected to an ammonia cylinder.
• 10Place the flask into a dry-ice ethanol bath.
• 11Add 20 mL of liquid ammonia to yield a colorless solution.
• 12
  Divide a total of 50 mg (2.2 mmol) of sodium metal into five batches and add them batch by batch (one batch every two minutes).

  • After adding the sodium metal, the colorless solution will gradually turn dark blue, and then becomes colorless.
• 13Remove the dry-ice ethanol bath to evaporate the ammonia.
• 14Apply a flow of argon, and then a vacuum to remove residual ammonia to yield the desired intermediate as a white powder.
• 15Dissolve the white powder in 10 mL of anhydrous ethanol and cool the mixture to 0 °C.
• 16Add 1.05 mmol of (E)-pent-2-en-4-ynyl tosylate (the product of Step 8) dissolved in 3 mL of anhydrous ethanol.
• 17Stir the reaction at 0 °C for 1 hour, then at ambient temperature (22 °C) overnight.
• 18Remove the volatile ethanol with a rotary evaporator under reduced pressure and then dissolve the rest mixture in 5 mL ddH₂O.
• 19Load this mixture onto a Dowex 50 (H⁺) ion-exchange column. Wash the column with ddH₂O until the pH value of the washout reaches 7.0.
• 20Elute the Dowex column (diameter × height: 2 × 10 cm²) with two column volumes of 5% ammonia hydroxide.
• 21
  Monitor the eluted products by silica gel TLC developed by 4:1:1 v/v/v n-BuOH/AcOH/H₂O.

  • The desired product is detected by treating the TLC plate with 0.2% ninhydrin in 16:3:1 v/v/v 96% EtOH/AcOH/s-collidine followed by heating.
• 22Combine the eluent containing the desired product and concentrate with a rotary evaporator to yield a white solid as a crude product.
• 23Fully dissolve the crude product with 1 mL of 0.1 N HCl, followed by further purification with reversed-phase HPLC, using a 10% ‒ 60% gradient of acetonitrile in 0.1% trifluoroacetic acid as the elute.
• 24
  Combine the desired fractions and lyophilize to yield the desired product Hey-Met as a white powder (yield of 58%).

  • Purity was confirmed by ¹H-NMR (500MHz, D2O): 2.13–2.19(m, 1H), 2.22–2.28(m, 1H), 2.52(t, 1H, J = 2.6Hz), 2.66(t, 2H, J = 7.5Hz), 3.00–3.02(m, 2H), 3.23(dd, 2H, J = 7.2, 0.7Hz), 4.14(t, 1H, J = 6.3Hz), 5.64–5.70(m, 1H), 5.75–5.81(m, 1H).

REAGENTS AND SOLUTIONS:

Unless otherwise indicated, use deionized distilled water (ddH₂O) in all recipes and protocol steps.

Click dilution buffer

• 50 mM Triethanolamine (1 M stock, pH 7.4)

• 150 mM NaCl (2M stock)

  • Store ingredients above at ambient temperature up to 6 months

• cOmplete^™, EDTA-free Protease Inhibitor Cocktail (Roche, cat. no. 4693132001) (added immediately prior to its use)

Click reaction buffer

• 50 mM Triethanolamine (pH = 7.4) (1 M stock)

• 150 mM NaCl (2M stock)

• 4% (w/v) sodium dodecyl sulfate (SDS)

  • The solution containing the three ingredients above can be stored at ambient temperature (22 °C) for up to 6 months.

• cOmplete^™, EDTA-free Protease Inhibitor Cocktail (added immediately prior to use)

Elution buffer

• 250 mM ammonium bicarbonate

• 1% (w/v) sodium dodecyl sulfate (SDS)

  • Store above ingredients at ambient temperature for up to 6 months

• 200 mM sodium dithionite (Sigma-Aldrich, cat. no. 1.06507) (add immediately prior to its use)

Gel loading buffer

• 40 mM Tris HCl (pH = 6.8) (1 M stock)

• 10 mM Ethylenediaminetetraacetic acid (ETDA)

• 10% (v/v) Glycerol

• 10% (v/v) β-Mercaptoethanol (β-ME)

• 2% (w/v) SDS

  • No dye should be in this buffer, to avoid interference with in-gel fluorescent signals.

Gel fixing buffer

• ddH₂O (50%)

• Methanol (40%)

• Acetic Acid (10%)

Lysis buffer

• 50 mM HEPES (pH = 8.0) (1 M stock)

• 0.005% (v/v) Tween-20

  • Store above ingredients at 4 °C up to 6 months

• 1 mM TCEP (Sigma-Aldrich, cat. no. C4706) (add immediately prior to its use)

• cOmplete^™, EDTA-free Protease Inhibitor Cocktail (added immediately prior to its use)

Pull-down dilution buffer

• 50 mM Triethanolamine (from 1 M stock, pH 7.4)

• 150 mM NaCl (from 2M stock)

• 1% (v/v) Brij O10 (Sigma-Aldrich, cat. no. P6136)

  • Store above ingredients at ambient temperature for up to 6 months

• cOmplete^™, EDTA-free Protease Inhibitor Cocktail (added immediately prior to its use)

COMMENTARY

BACKGROUND INFORMATION:

With S-adenosyl-L-methionine (SAM) as the methyl donor cofactor, protein lysine methyltransferases (PKMTs) and protein arginine methyltransferases (PRMTs) catalyze protein lysine and arginine methylation, respectively (Berdasco and Esteller, 2019; Blanc and Richard, 2017; Scheer et al., 2019). The human genome encodes more than 60 putative PKMTs and 9 PRMTs (Luo, 2018; Yang and Bedford, 2013). PKMTs and PRMTs can act on histones and on non-histone protein substrates (Luo, 2018; Yang and Bedford, 2013). Dynamic methylation states can influence diverse physiological and pathogenic functions. (Blanc and Richard, 2017; Chang et al., 2011; Huang et al., 2010; Lee et al., 2010; Ling et al., 2012; Luo, 2018; Pless et al., 2008; Su et al., 2021). Dissecting the complex methylome network in a native context is thus essential to annotate the roles of PMTs in normal and disease settings.

Protein lysine or arginine methylation does not significantly alter the electrostatic state and size of target proteins (Luo, 2018). This little difference caused by lysine and arginine methylation presents a great technical challenge to illuminate cellular methylomes (Luo, 2015a, 2018). Because many lysine and arginine methylation events are dynamic, substoichiometric, and of low abundance, an enrichment step is essential to capture methyllysine- and methylarginine-containing peptides for top-down analysis of the cellular methylome (Luo, 2015a, 2018). The most common method to enrich proteome-wide methylation events is to rely on anti-methyllysine or anti-methylarginine antibodies (Luo, 2015a, 2018). However, it is challenging to develop high-quality, sequence-independent pan-antibodies against diverse methyllysine/methylarginine-containing peptides.

To unambiguously dissect complex PMT-methylome networks, we formulated the live-cell Bioorthogonal Profiling of Protein Methylation (lcBPPM) technology (Luo, 2018; Wang and Luo, 2013). In a stepwise manner, we first engineered multiple PKMTs and PRMTs, such as GLP1 and PRMT1, to process bulky SAM analogues such as Ab-SAM, EnYn-SAM, Hey-SAM, and Pob-SAM as cofactors for substrate labeling (Guo et al., 2014; Islam et al., 2012; Islam et al., 2013; Luo, 2012; Wang and Luo, 2013; Wang et al., 2011b). In comparison with the native SAM cofactor, these BPPM-feasible SAM analogue cofactors contain the characteristic allylic sulfonium for double activation (Islam et al., 2012; Islam et al., 2013; Wang et al., 2013). Interestingly, while the allylic moiety is essential for engineered PMTs to process SAM analogues, the cofactor activities of these SAM analogues for engineered PMTs are comparable to that of SAM (Bothwell and Luo, 2014; Islam et al., 2013; Wang et al., 2011b). This observation argues that the target labeling is defined by multiple factors such as substrate deprotonation, which are determined by PMTs rather than the structures of cofactors. While the double-activated SAM analogue cofactors are not as stable as SAM (Bothwell et al., 2012; Bothwell and Luo, 2014), their half-life time is sufficient for enzymatic labeling of PMT substrates in vitro and inside living cells (Bothwell and Luo, 2014; Islam et al., 2013; Wang et al., 2011b).

To overcome the poor membrane permeability of SAM analogues, we then engineered methionine adenosyltransferase (MAT) for promiscuous biosynthesis of bulky SAM analogues from membrane permeable methionine analogues (Wang et al., 2013; Wang et al., 2014). The resulting SAM analogues can then be utilized by engineered PMTs as cofactor surrogates to label substrates with distinct chemical moieties inside live cells. The merit of this approach lies in its ability to tag a terminal-alkyne(azide)-containing moiety to PMT substrates for enrichment and characterization via a “click” reaction, unambiguously identifying distinctly-modified targets for individual (engineered) PMTs. The robustness of our live cell BPPM (lcBPPM) technology allows massively parallel methylome profiling of 80% PMT subfamilies (unpublished data). The lcBPPM technology could be coupled with SILAC- or TMT-based MS for quantitative target identification, or immunoblot-based detection for individual target validation (Fig. 8), as exemplified with GLP1 and PRMT1.

It is worth noting that while PMTs were engineered to maximally maintain the catalytic efficiency and native modes to engage substrates (Islam et al., 2013), we envision the BPPM technology, in the context of conventional methods, as a gap-bridging approach to reveal bona fide PMT substrates in a more effective and definitive manner. Although related approaches have been developed for kinases and acetyltransferases, our ability to generate SAM analogs in live cells offers an advantage over the in vitro approaches so far applied in those cases.

CRITICAL PARAMETERS:

Among critical parameters for lcBPPM are the dosing amount and time window of BPPM-feasible methionine analogues, the quality of the reagents for a click reaction, and the various buffer conditions to process protein samples.

In the examples discussed in this protocol, transfection efficiency and thus expression of the Y1211A variant of GLP1 (Islam et al., 2013), the M48G variant of PRMT1 (Wang et al., 2011b), and the I117A variant of MAT2A (Wang et al., 2013) were optimized for these plasmids and HEK293T cells, and such optimization is critical for successful lcBPPM. These conditions are subject to change if different constructs and cell lines are used. In regard to the intracellular enzymatic activity of the I117A variant of MAT2A, we have examined this construct to process effectively Hey-Met into Hey-SAM inside live HEK293T cells with a MS-based assay (Wang et al., 2013). In addition, the lcBPPM feasibility of the I117A variant of MAT2A and the M48G variant of PRMT1 has been proved with other cell lines. In contrast, the enzymatic activity of Y1211A variant of GLP1 for BPPM was examined only in vitro and with HEK293T cells. To assure the efficiency of lcBPPM, we strongly recommended conducting rigorous biochemical and cell-lysate-based in vitro BPPM assays with engineered PMTs, well-characterized substrates, and SAM analogue cofactors before implementing the lcBPPM protocol. For the PMTs without well-annotated substrates, BPPM-feasible variants can be developed on the basis of closely related PMT homologues and then validated with Alternate Protocol 1 in a small scale (Islam et al., 2013; Su et al., 2021; Wang et al., 2013). Although it is ideal for lcBPPM to be conducted at a level of activity of the engineered PMTs comparable to that of the native PMT, we found that it is more beneficial to maximize the intracellular activity of a BPPM-feasible PMT variant through overexpression of the most active mutant, for maximal target labeling. The labelled candidates can be revealed by MS with high confidence, ranked quantitatively according to their SILAC or TMT ratios, and prioritized for further validation.

To optimize the dose and time window of Hey-Met for cell treatment, we relied on small-scale, in-gel fluorescence visualization readouts (Alternate Protocol 1) by varying concentrations (0.2‒2.0 mM) and incubation time (4‒24 hours) to identify a plateau of high signal-to-noise ratios. In most of so-far optimized lcBPPM protocols, cells are treated with 0.2‒1.0 mM for 4‒12 hours in a methionine-deficient medium. The optimal detection of a specific PMT substrate is a function of both the labeling efficiency of the associated lcBPPM apparatus and the rates of production and degradation of the target protein, and thus can only be determined case by case.

Although the conditions for the CuAAC click reaction have been well documented in the literature, a rigorous practice of Basic Protocol 2 is essential for the success of lcBPPM. To set up a click reaction, pre-incubation of CuSO₄ with the BTTP ligand is necessary. In addition, CuSO₄ and sodium ascorbate need to be freshly prepared; old solutions significantly impact the efficiency of the CuAAC click reaction under our conditions. In our prior protocol (Blum et al., 2013a; Blum et al., 2013b), the copper ligand TBTA was used in the place of BTTP. Great caution should be exercised if replacing TBTA or BTPP with other copper ligands, which require extensive testing. Here, we used sodium ascorbate rather than TCEP as a reducing agent of our click reaction, to better suppress cystine-involved background labeling. In addition, we noticed that salt concentrations and buffer choices affect the efficiency of the click reaction to various degrees. We recommend the conditions documented in the current protocol unless validated otherwise.

Another critical parameter of lcBPPM is the concentration of SDS in various buffers. In general, a high concentration of SDS (e.g. 4%) is used to dissolve protein pellet. However, 4% SDS in a buffer impairs biotin-streptavidin interaction (Basic Protocol 3). In addition, we found that the presence of more than 2% SDS greatly impairs the efficiency of the click reaction. Therefore, we dilute 4% SDS to 1% for click reactions (Basic Protocol 2) and to 1.5% for streptavidin enrichment (Basic Protocol 3).

Among other important practices for successful lcBPPM are the complete washing of protein pellet with cold methanol before a click reaction to remove unconsumed Hey-SAM, and after a click reaction to remove unreacted azide probes; the methanol in the protein pellet should not be fully removed because of the difficulty in solubilizing the otherwise fully-dried protein pellet. Occasionally, modest sonication is needed to facilitate dissolving protein pellet.

For potential troubleshooting, we recommend sparing portion of the samples after Basic Protocol 1 as the loading control of the proteome (Basic Protocol 1 Step 17) and after Basic Protocol 2 for the quality control of the click reaction (Basic Protocol 2 Step 8). See next section.

TROUBLESHOOTING:

See Table 1 for common problems with the protocols, their causes, and potential solutions

Table 1.

Troubleshooting

Problem	Possible Cause	Solution
Low signal	Poor plasmid expression	Check plasmid quality and/or increase amount
	Poor labeling	Increase cofactor concentration/incubation time
	Low target protein concentration in lysate	Increase starting cell number and/or cell lysate for click reaction
		Enrich cellular compartment(s) associated with target proteins, e.g. nuclear fraction, before cell lysis to enrich nuclear proteins
High background	Non-specific labeling	Reduce cofactor concentration and/or incubation time

UNDERSTANDING RESULTS:

lcBPPM candidate proteins can be revealed through Alternate Protocol 1, Basic Protocol 4, and Basic Protocol 5. For the readout of Alternate Protocol 1, one should focus on distinct or stronger fluorescent bands that appear in a lcBPPM sample relative to its negative control (right lanes versus left lanes of each panel in Fig. 10). For the readout of Basic Protocol 4, MS-identified PMT substrates are expected to be enriched in a lcBPPM sample versus its negative control(s). Such MS readouts can be based on increased spectral counts of candidate peptides or higher isotopic ratios of sample versus control peptides when a SILAC or TMT method is implemented. Because of identification of several hundred candidate proteins and relatively high variation of their MS intensities, at least two replicates of MS samples should be obtained for comparison; at least three replicates of SILAC or TMT multiplex sample-control pairs are needed for rigorous statistical analysis. In the current protocol, trypsin was used to digest samples for MS analysis by cleaving the amide bonds after Lys/Arg sites (Basic Protocol 4 Step 8). In our experience, monomethylation of Lys and Arg has a minimal effect on trypsin to recognize these sites. In contrast, the cleavage efficiency of trypsin at Lys and Arg sites and the labeling by TMT at Lys sites are inhibited by higher degrees of methylation. Therefore, higher degrees of methylation rather than monomethylation are expected to lead to unique longer peptides with methylation sites buried in middle. For the readout of Basic Protocol 5, one should focus on positive or stronger WB bands in the lcBPPM sample in comparison with the corresponding positions in its negative controls without engineered PRMT1 (left two lanes versus right two lanes in Fig. 11a) or with a methylation-dead DUSP4 variant (the right lane versus the left lane in Fig. 11b). The labeling efficiency of lcBPPM can also be estimated from the intensity ratios of candidate proteins in lcBPPM samples (Basic Protocol 5) versus their total inputs (Basic Protocol 1 Step 17).

Figure 10.

Sample data for Alternate Protocol 1. The in-gel fluorescence signals of Hey-SAM-labeled PMT substrates were obtained with BPPM-feasible PMT variants: Y1211A GLP1(left), Y1154A G9a (middle), and M48G PRMT1 (right). The BPPM samples with engineered PMTs exhibit a stronger signal, reflective of more labelled proteins, (the right columns in each gel) than the control samples (the left columns in each gel). Y1211A GLP1 data is adapted from Islam et al., 2013. Y1154A G9a is adapted with permission from Wang et al., 2013 (Copyright 2013, American Chemical Society). M48G PRMT1 data is adapted from Su et al., 2021.

Figure 11.

Representative data of Basic Protocol 5. Histone H4 and DUSP4 (endogenous and HA-tagged) were labeled by BPPM-feasible M48G variant of PRMT1 (both V1 and V2 isoforms) and Hey-SAM in NB4 cells following the lcBPPM protocol. The resulting proteins were further modified by the diazo biotin-azide probe (Basic Protocol 2), enriched with streptavidin beads (Basic Protocol 3), and blotted with the respective antibodies (Basic Protocol 5). The lcBPPM procedure allow the target proteins of interest (native histone H4 and DUSP4) to be pulled down by streptavidin beads. In contrast, decreased readout in the western blot signals are expected for (a) negative controls without engineered PRMT1 and (b) with methylation-dead DUSP4 mutants. Figure is adapted from (Su et al., 2021).

TIME CONSIDERATIONS:

Roughly, a week is needed to prepare lcBPPM samples and conduct the follow-up analysis: 3~4 days for Basic Protocol 1, 17 hours for Basic Protocol 2, 18 hours for Basic Protocol 4 or 5, and 20 hours for Alternate Protocol 1.

It takes 3~4 days for live-cell labeling of PMT substrates (Basic Protocol 1). The duration of Basic Protocol 1 is mainly determined by the time it takes cells to recover after transfection of plasmids and to label targets after adding the methionine analogue (12~24 hours for each process). While harvested cells can be stored at ‒80 °C for a few days, cell lysates should be prepared in 2 hours and be immediately processed for protein precipitation (Basic Protocol 1 Step 24). For the methanol-precipitated protein pellet (Basic Protocol 1 Step 18‒24), overnight precipitation should be used for the highest protein recovery. The subsequent washing steps take ~1.5 hours.

It takes 17 hours to complete the click protocol (Basic Protocol 2): 1.5 hours to prepare the click reagents, 2 hours for the click reaction, and 13.5 hours to work up the click reaction. The resulting protein pellet can be stored at ‒20 °C for a week and at ‒80 °C for three months. With the samples obtained from Basic Protocol 2, enrichment of biotinylated lcBPPM proteome with streptavidin beads can be completed in 7 hours (Basic Protocol 3). The pulldown sample can be stored at ‒80 °C for three months. With the lcBPPM pellet obtained via Basic Protocol 3, sample preparation for LC-MS-MS analysis (Basic Protocol 4) or immunoblot of candidate proteins (Basic Protocol 5) can be completed in 18 hours. In-gel fluorescence visualization of the lcBPPM proteome can be completed in 20 hours (Alternate Protocol 1).