Determination of the Substrate Specificity of Protein Kinases with Peptide Micro- and Macroarrays

Abstract

Elucidation of the key determinants for the phosphorylation site specificities of protein kinases facilitates identification of their physiological substrates, and serves to better define their critical roles in the signaling networks that underlie a multitude of cellular activities. Albeit with some apparent limitations, such as the lack of contextual information for secondary substrate-binding sites, the synthetic peptide-based approach has been adopted widely for the kinase specificity profiling studies, especially when they are used in an array format, which permits the screening of large numbers of potential peptide substrates in parallel. In this chapter, we present detailed protocols for determining protein kinase substrate specificity using an approach that involves both peptide microarrays and macroarrays. In particular, SPOT synthesis on macroarrays can be used to follow up on in silico predictions of protein kinase substrate specificity with predictive algorithms.

1 Introduction

Protein phosphorylation catalyzed by protein kinases is the most common posttranslational mechanism to transduce extracellular signals intracellularly in eukaryotic organisms. In humans, about 2 % of the protein-coding genes specify at least 536 known protein kinases [1], which may phosphorylate as many as one million phosphosites in the phosphoproteome, with over 200,000 phosphosites already confirmed experimentally (www.phosphonet.ca). Despite this apparent promiscuity amongst protein kinases, they still demonstrate rather restricted physiological substrate specificity such that signaling networks typically operate with high fidelity and exquisite control. Such selective activation/deactivation of a subset of intracellular signaling pathways mediated by protein kinases relies heavily on the ability of these enzymes to differentiate their bona fide substrates in vivo. Indeed, dysregulation of protein phosphorylation events is implicated in over 400 types of human diseases, including cancer, diabetes, cardiovascular, neurological, and immunological disorders. Mutations in protein kinase genes may affect their enzymatic activity, substrate specificity, location, and stability. As the genomes of more individuals become sequenced for diagnostic purposes, prediction of changes in these parameters will first require better knowledge of the subtle differences between protein kinases in these properties.

Several molecular mechanisms contribute to protein kinase-substrate pairing specificity, such as the co-expression, co-localization, and co-activation of the kinase and its substrates, and protein-protein interactions mediated through direct docking sites or indirectly mediated via adaptor/scaffolding proteins. However, the selectivity of a particular protein kinase for its substrates is influenced to a high degree by molecular recognition of the amino acid sequences surrounding the phosphorylation sites [2–6]. Exploring the specificity of protein kinases using synthetic peptides has been a fruitful endeavor in cell signaling research for half a century [7–9]. Still, only a relatively small fraction of the human protein kinases have been systematically and comprehensively studied for their substrate selectivity, leaving lots of “orphans” with unknown connections to the rest of the cell signaling apparatus.

Simple alignment of the amino acid residues in phosphorylation sites of known substrate proteins and synthetic peptide substrates of many protein kinases provides insights into the amino acid preferences of individual protein kinases. This defines what is termed as a consensus substrate recognition sequence for a protein kinase. Favored amino acid residues in substrates show up at the highest frequencies at particular positions surrounding the phospho-acceptor amino acid.

The vast majority of the eukaryotic protein kinases, referred to as “typical,” feature a homologous catalytic domain with approximately 250 amino acid residues, including about 16 in highly conserved subdomains. These kinase subdomains are particularly critical for binding the ATP substrate and catalytic activity. From careful alignment of the consensus recognition sequences of protein kinases with their catalytic domain sequences, it has been feasible to identify the substrate determining residues of typical kinases. To predict the consensus recognition sequences for 492 human protein kinases domains, we have developed algorithms that have been trained with empirical data for over 14,000 known kinase-protein substrate pairs and 8000 kinase-peptide substrate pairs [10]. The results from these in silico analyses have been posted with open access on our PhosphoNET website (www.phosphonet.ca). This data is an excellent starting place for testing peptide substrates for target protein kinases of interest that have not been previously well studied.

In recent years, development of high-throughput techniques has made it feasible to screen for the specificity of a protein kinase of interest against a large set of peptides spotted on cellulose membrane or microarray surfaces [11–14]. The SPOT technique allows flexible and inexpensive synthesis of large number of peptides on cellulose membranes [15–20]. These peptides can be released and immobilized on microarray slides, or used directly for phosphotransferase assays as peptide macroarrays. Different strategies can be taken to select a library of peptides for kinase profiling [21–23]. A knowledge-based peptide library of confirmed physiological phosphorylation sites can be chosen in a microarray screening, while sequences modified from known protein substrates are also applicable. Quantitative analysis of the microarray image permits derivation of a consensus substrate sequence, which can later be used as a wild-type template of a substitutive library to be tested on the macroarray. In other cases, combinatorial libraries are employed for de novo screenings on the SPOT membrane [24].

In this chapter, we describe the procedures of kinase substrate determination with peptide macro- and microarrays. While they have been demonstrated to very successfully identify an overall preference of amino acids at a given position, one limitation of this technique is the potential loss of contextual information unique to proteins. This is, however, overweighed by the advantages of their flexibility, easiness, convenience, and ever-decreasing costs. We recommend proceeding from an in silico analysis such as offered by PhosphoNET or a high-content peptide microarray to define the promising candidate peptide substrates. Systematic replacements of the amino acids in these peptides by SPOT synthesis on macroarrays can then be used to further improve specific kinase recognition. A specific substrate peptide identified using the above methods can serve as a powerful investigative tool to assay protein kinase activity in vitro with high sensitivity and selectivity, and provide insight into the regulation of a target protein kinase in cell signaling networks.

2 Materials

2.1 SPOT Synthesis and Preparation of Peptide Microarrays ( See Note 1 )

• 1. Solvents: N,N′-dimethylformamide (DMF), methanol (MeOH), ethanol (EtOH), N-methylpyrrolidone (NMP), diethylether (DEE), dichloromethane (methylene chloride, DCM) (see Note 2 ).

• 2. Membranes for the peptide synthesis are prepared from filter paper Whatman 540 or Whatman 50 (Whatman) [18, 25].

• 3. Diisopropylcarbodiimide (DIPC, DIC; Fluka), N-Methylimidazole (NMI; Sigma), and Fmoc-β-alanine, Fmoc-glycine (GLS Biochem) (see Note 3 ).

• 4. Amino functionalization solution: For 12.5 ml, 782 mg Fmoc-β-alanine or 750 mg Fmoc-glycine prepared in DMF. Add 468 μl DIC and 396 μl NMI.

• 5. Staining solution: 0.002 % (w/v) bromophenol blue (BPB; Sigma) in MeOH.

• 6. Pentafluorophenyl ester-preactivated amino acid derivatives with protecting groups according to the Fmoc strategy (OPfp esters; EMD Millipore, GLS Biochem, and Bachem) [26, 27] (see Note 4 ).

• 7. Piperidine solution: 20 % Piperidine (Sigma) in DMF.

• 8. Capping solution: 2 % (v/v) acetic anhydride (Sigma) and 2 % (v/v) ethyl-diisopropylamine (DIPEA, DIEA; Sigma) in DMF.

• 9. Deprotection solution A: Trifluoroacetic acid (TFA) containing 4 % (v/v) DCM, 3 % (v/v) triisopropylsilane or triisobutylsilane (TIPS or TIBS; Fluka), 2 % (v/v) distilled water, 1 % (w/v) thioanisole (methyl phenyl sulfide; Alfa Aesar) (Important! see Note 5 ).

• 10. Deprotection solution B: 65 % (v/v) TFA, 3 % (v/v) TIPS or TIBS, 2 % (v/v) distilled water, 1 % (w/v) thioanisole, 29 % (v/v) DCM (Important! see Note 5 ).

• 11. Ammonia gas (Air Liquide).

• 12. Erie Epoxysilane Coated Microarray Slides (Thermo Scientific).

• 13. Peptide microarray printing solution: Dimethyl sulfoxide (DMSO ), phosphate-buffered saline (PBS/0.9 % NaCl in 10 mM phosphate buffer, pH 7.4).

2.2 Kinase Substrate Specificity Screening on Peptide Microarrays

• 1. Microarray blocking buffer: 1 % bovine serum albumin (BSA ) in 100 mM HEPES, pH 7.5.

• 2. Protein kinase buffer: 5 mM MOPS pH 7.2, 2.5 mM β-glycerol-phosphate, 5 mM MgCl₂, 1 mM EGTA , 0.4 mM EDTA , 0.5 mM dithiothreitol (DTT ) (see Note 6 ).

• 3. Tris-buffered saline (TBS): 50 mM Tris–HCl, 150 mM NaCl, pH 7.5.

• 4. Tris-buffered saline with Tween (TBST): 0.05 % Tween-20 in TBS.

• 5. ATP stock solution: 10 mM ATP in distilled H₂O.

• 6. 0.5 % Sodium dodecyl sulfate (SDS) in distilled H₂O.

• 7. Pro-Q Diamond phosphoprotein/phosphopeptide microarray stain and destain solution (Life Technologies) (see Note 7 ).

2.3 SPOT Synthesis of Peptide Macroarrays and Optimization of Kinase Activity of Selected Peptides

• 1. Preparation of amino-alkyl ether-modified membranes: 70 % perchloric acid (Alfa Aesar), epibromohydrine (Fluka), 1,3-diaminopropane (Alfa Aesar), 4,7,10-trioxa-1,13-tridecanediamine (Fluka), and sodium methylate (sodium methoxide; Fluka) (see Note 1 ).

• 2. Membrane washing solution: A mixture of 50 ml of MeOH and 1 ml of 70 % aqueous perchloric acid.

• 3. Membrane activation solution: A mixture of 5 ml epibromohydrine and 500 μl of 70 % aqueous perchloric acid in 1,4-dioxane.

• 4. Coupling reagents: DIC and N-hydroxybenzotriazole (HOBt; EMD Millipore). Coupling reagents are only necessary when no preactivated amino acid derivatives are used (see Note 4 ).

• 5. Non-preactivated amino acids with protection groups according to the Fmoc strategy [27, 28] (EMD Millipore and GL Biochem); preactivated amino acid derivatives with protection groups according to the Fmoc strategy, e.g., OPfp esters [26] (see Note 4 ).

• 6. Macroarray buffer I: 5 mM MOPS, 2.5 mM β-glycerol-phosphate, 1 mM EGTA , 0.4 mM EDTA , 0.25 mM DTT (see Note 8 ).

• 7. Macroarray buffer II: Macroarray buffer I with 100 mM NaCl, 0.2 mg/ml BSA .

• 8. Macroarray buffer III: Macroarray buffer I with 100 mM NaCl, 1 mg/ml BSA , 5 mM MgCl₂, 50 μM ATP.

• 9. Macroarray blocking buffer: 5 % sucrose, 4 % skim milk in TBST.

• 10. Detection antibodies: Generic phosphoserine, phosphothreonine, or phosphotyrosine antibodies.

• 11. Horseradish peroxidase (HRP)-conjugated secondary antibody recognizing IgGs from the host species that the detection antibody was from.

• 12. Staining solution I (8.3 ml for 10 ml staining mixture): 100 mg NaCl, 0.5 ml 1 M Tris–HCl, pH 7.4, and 7.8 ml distilled H₂O.

• 13. Staining solution II (1.7 ml for 10 ml staining mixture): 5 mg 4-chloro-1-naphthol in 1.7 ml methanol. This solution must be prepared fresh shortly before use.

• 14. Hydrogen peroxide (H₂O₂), 30 %.

3 Methods

3.1 SPOT Synthesis of Free Peptide Amides on Amino-Acid Ester-Modified Cellulose Membranes and Preparation of Peptide Microarrays

The synthesis of the peptides for the preparation of peptide microarrays and macroarrays is carried out by using the SPOT technique [29]. The SPOT technique was initially developed for synthesis of relatively large number of peptides in parallel on the membrane, and the resulting peptide macroarray can then be screened directly [16, 30]. More recently, the SPOT synthesis has been used for the synthesis of large number of free peptides in small amounts [31]. These free peptides are then dissolved and then spotted onto microarray substrates for the preparation of peptide microarrays.

A higher reactivity of the cellulose surface can be achieved by transforming the hydroxyl groups into amino groups, e.g., by esterification of with amino acids [30, 32]. Due to the relative weak ester bond between the cellulose and amino acids, this membrane type is particularly useful for the synthesis of free peptides. The synthesis of the peptides on the cellulose can be carried out by using preactivated amino acids or by performing an in situ activation of the amino acids during the synthesis [19]. The cleavage of the ester bond is possible in strong basic environment (e.g., ammonia gas, hydroxide solutions). The treatment with ammonia gas is preferable, since it results in the free, dry peptides absorbed on the cellulose fibers. The SPOT synthesis can be carried out either automatically or manually. We use the peptide/SPOT synthesizer MultiPep (Intavis Bioanalytical Instruments) for SPOT synthesis. We provide here a description of manual synthesis for a better understanding of the general procedures.

For the synthesis of free peptides, it is recommended to use large spots by applying a reagent volume of 1 μl onto the membrane. To release the peptides from the membrane, punch out the spot areas before or after the final cleavage and transfer the discs into small vials. The peptides can be eluted from the membrane with the microarray print buffer if ammonia gas is used for cleavage. The peptides should already be in solution if basic solutions are used for cleavage.

If not noted differently elsewhere, washing and treatment steps are performed on a rocking shaker and all manual washing steps should be carried out for at least 3 min each time.

3.1.1 Preparation of Esterified Membranes for Free Peptide Synthesis

This procedure is always carried out manually as follows:

• 1. For the amino functionalization, cut a piece of filter paper to the desired size. For the modification of a sheet with a size of 10 cm × 15 cm (that fits on a tray of the synthesizer and can accommodate 9 × 12 = 118 spots), prepare the 12.5 ml of amino functionalization solution. For larger membranes, use more reagent solutions accordingly.

• 2. Transfer the mixture into a chemically resistant box with lid (stainless steel or polypropylene) and place the filter paper into the liquid. Avoid air bubbles trapped underneath the paper. The membrane should be completely soaked in the solution.

• 3. Let the membrane react with the reaction mixture in the closed box overnight (see Note 9 ).

• 4. After the treatment, wash the membrane at least three times with DMF. If necessary the membrane can be stored at −20 °C for a few weeks until needed (see Note 10 ).

• 5. For the Fmoc deprotection, treat the membrane twice with piperidine solution for at least 5 min each time.

• 6. Wash the membrane at least four times with DMF, followed by washing at least twice with MeOH or EtOH.

• 7. To perform an optional staining [32] (see Note 11 ): Treat the membrane with staining solution for at least 2 min until the filter paper shows a homogeneous blue color. If the staining is insufficient, repeat with fresh staining solution.

• 8. After staining, wash the membrane at least twice with MeOH or EtOH, until the liquid remains colorless.

• 9. After thorough air-drying, the membrane is ready for the first coupling (see Note 12 ).

3.1.2 Preparation of Coupling Solutions

Generally, two different methods can be used for the preparation of coupling solutions.

The first method is using preactivated Fmoc- protected amino acid derivatives (e.g., pentafluorophenyl esters) [16]. The advantage of this method is the use of a single type of reagent, which simplifies the preparation of the solutions and reduces the chances of human errors. Unfortunately, activated esters are commercially available for standard amino acids only. Preactivated derivatives of other amino acids have to be synthesized in advance [26]. This second approach is in situ activation of protected Fmoc-amino acid derivatives. Since for the synthesis of free peptides for the microarray production only standard amino acids are used, we only describe the first method in this chapter.

Most solutions of preactivated amino acids are prepared by dissolving the corresponding amino acid derivatives in NMP at a concentration of 0.3 M. Due to poor solubility, the serine derivative must be dissolved in amine-free DMF. Except for the arginine derivative, all solutions can be used for at least 1 week if the stock solutions are stored at or below −20 °C. Replace amino acid solutions of the previous day with fresh amounts from the stock solutions each day (see Note 4 ). Due to the instability of the dissolved preactivated arginine derivative, it must be prepared fresh every day.

3.1.3 SPOT Synthesis on the Cellulose Membrane

This protocol describes the SPOT synthesis on large spots. Usually, a pipetting robot is used to deliver the coupling solutions; but for a small number of large spots the solution can be pipetted manually.

To locate the large spots after synthesis, mark the center of each spot at the start or during the synthesis with a pencil. For peptide macroarrays, the edges of the entire area should be marked in the same way. Since the application of 1 μl of coupling solution results in spots with a diameter of about 7 mm, the distance between the centers of two spots should be at least 8 mm.

• 1. For amino acid coupling, the synthesis runs from the C- to the N-terminus. Deliver the desired volumes of activated amino acid solutions to the corresponding positions on the membrane. For the first coupling cycle, use 1 μl amino acid solution (large spots for microarray production) or about 0.1 μl (small spots for macroarrays). For all other steps, use 20 % more of amino acid solution in order to cover the entire spot area.

• 2. After the delivery cycle allow the reagents to react for at least 20 min.

• 3. A repeat of the spotting is recommended in order to achieve a higher coupling yield.

• 4. Capping: This step reduces the number of side products by acetylation of unreacted amino groups. At the first coupling cycle place the membrane face down in a box filled with an appropriate amount of capping solution. Do not leave air bubbles under the membrane. Do not shake!

• 5. Renew the capping solution after about 5 min and let it react for another 20 min.

• 6. For all subsequent cycles treat the membrane with capping solution twice for 5 min.

• 7. To remove the Fmoc- protecting group, wash the membrane with DMF four times.

• 8. Treat with piperidine solution twice for 5 min each time.

• 9. Wash at least four times with DMF.

• 10. Wash at least twice with MeOH or EtOH. Do not perform this step for the final Fmoc removal.

• 11. Staining (optional) [33] (see Note 11 ): Stain the membrane with staining solution in a box while shaking and leave it in until the spots are stained sufficiently. If necessary, renew the solution (see Note 13 ).

• 12. For destaining, wash the membrane at least twice with MeOH or EtOH until the wash solution remains colorless.

• 13. Air-dry the membrane before continuing with the next coupling cycle (see Note 12 ).

• 14. Building up the peptide chain: Except for the last coupling cycle, repeat steps 1–12. For the last coupling cycle, carry out only steps 1 and 7!

• 15. Upon the removal of the last Fmoc-protecting groups, wash the membrane at least four times with DMF followed by washes with DCM three times.

• 16. Air-dry the membrane.

• 17. Final side-chain deprotection: Treat the membrane with 20 ml of deprotection solution A. The membrane must always be submerged in the deprotection solutions. Keep the box tightly closed. Do not shake!

• 18. After 30 min, pour out the solution very carefully (see Note 14 ). If the membrane is already soft, remove the residual solution completely with a pipette.

• 19. Treat the membrane with at least 20 ml of deprotection solution B for 3 h in the closed box without shaking.

• 20. Remove the solution carefully with a pipette.

• 21. Wash the membrane first with DCM at least five times (see Note 14 ) and DMF twice followed by MeOH twice.

• 22. Dry the membrane in a fume hood.

3.1.4 Cleavage of the Peptides from Membrane as Free Peptide Amides

The method described here involves the exposure of the entire dry membrane or the punched-out spots to ammonia gas that breaks the ester bond between the peptides and cellulose and forms a C-terminal amide (see Note 15 ).

• 1. Place the dry membrane or the punched-out spots in a glass desiccator (see Note 16 ).

• 2. Set the desiccator under vacuum.

• 3. Fill the desiccator with ammonia gas.

• 4. To replace most of the air by ammonia repeat steps 2 and 3 at least twice.

• 5. Let the reaction proceed overnight.

• 6. Open the desiccator under a fume hood (Attention! ammonia gas is highly corrosive and irritant!). Let the gas dissipate for at least 30 min.

• 7. Transfer the discs into wells of microtiter plates (MTPs) or vials followed by addition of adequate amount of DMSO to dissolve the adsorbed free peptides (see Note 17 ). The resulting peptide stock solutions can be kept at −20 °C for long-term storage.

The amounts of peptides on amino acid ester-modified membranes for preparing peptide microarrays should range from 340 to 400 nmol/cm². If a common hole puncher is used, the peptide amount recovered typically ranges from 100 to 120 nmol. The purity of the peptides may vary significantly, ranging from 40 to 91 % [34, 35], depending on their sequences [36] (see Fig. 1).

Fig. 1

Example of two HPLC images representing different kinase substrate peptides: The top image shows an optimum purity, while the bottom one represents a more typical example of purity

3.1.5 Preparation/Printing of Peptide Microarrays

To prepare peptides for microarray printing, the peptide stock solutions are diluted with three times PBS and 15 μl of each is plated into 384-well microtiter plates according to the final layout on the arrays. The epoxysilane-coated microarray slides from Thermo-Fisher are employed as the substrates for spotting without pretreatments. Printing is carried out by a microarrayer equipped with quill pins for peptide delivery. The printed arrays are left on the arrayer overnight under the relative humidity of 50 %. Orientation markers and appropriate control peptides are integrated for quality assessment and array orientation. Printed peptide microarrays are stored in vacuum at −80 °C until use.

3.2 Kinase Substrate Specificity Screening on Peptide Microarrays

In the following protocol, we have generalized the conditions to make it suitable for most purified protein kinases that have detectable activity in vitro. The concentration of the kinase, reaction buffer, and the incubation time for the kinase assay on a peptide microarray slide can vary from kinase to kinase and may depend on the specific activity of the kinase. We recommend a negative control experiment performed in parallel without the kinase or with a kinase-dead mutant. A phosphoprotein/phosphopeptide sensor dye, Pro-Q Diamond stain can be used for probing with high sensitivity and low false-positive rate [37]. The consensus substrate sequence identified from microarray data analysis will be used for further optimization or individual verification.

• 1. Block the slide in a microarray reaction tray with the microarray blocking buffer for 1 h at room temperature with agitation (see Note 18 ).

• 2. Wash the slide with TBST three times, 2 min each time.

• 3. Take out the slide from TBST using forceps, and allow excess liquid to slip off. Clip on to the slide a microarray incubation chamber.

• 4. Make up a proper volume of kinase reaction mix by diluting purified active protein kinase in protein kinase buffer, and supplement with 100 μM ATP. Load the solution into the incubation chamber.

• 5. Incubate the reactions for 2 h at 30 °C in a shaking incubator. Keep the microarray in a humidity chamber if necessary (see Note 19 ).

• 6. Carefully remove the incubation chamber. Wash the slide with 0.5 % SDS twice, 5 min each time.

• 7. Wash the slide with TBST seven times, 5 min each time to remove SDS.

• 8. Block the slide in the microarray blocking buffer again for 1 h at room temperature.

• 9. Stain the slide with Pro-Q Diamond stain for 1 h at room temperature in the dark.

• 10. Destain with the Pro-Q Diamond destain solution three times, 15 min each time.

• 11. Rinse the slide in H₂O once. Wash in H₂O for 15 min.

• 12. Dry the slide using N₂ flow (see Note 20 ).

• 13. Scan in a microarray scanner at a wavelength of 532 nm or 543 nm (see Note 21 ). Examples of obtained images are presented in Fig. 2.

  Fig. 2

  

  Examples of protein kinase substrate peptide microarray images: Each kinase substrate peptide was printed in triplicate and phosphorylated by preparations of purified active protein kinases on epoxysilane-coated microarray slides. The phosphorylation of the peptides was visualized with Pro-Q Diamond stain. In these images, the first grid of peptide spots on peptide microarrays phosphorylated in vitro by the protein kinases FGR, p38-gamma MAP kinase, TAOK2, and Pim1 are shown for comparison. Orientation markers are printed in each of the four corners

• 14. Use microarray image analysis software to archive quantitative data and generate a consensus peptide sequence.

3.3 SPOT Synthesis of Peptide Macroarrays for Substrate Sequence Optimization

Peptide macroarrays are a very useful tool for the optimization of lead sequences identified from peptide microarrays or phage libraries [38]. The use of microarrays may be considered if the amount of the kinase is limited. Several strategies are described for the optimization of peptides on peptide macroarrays, e.g., substitution analysis (systematic substitution of all residues of the sequence by other amino acids), truncation analysis (systematic reduction of the length of a peptide), and loop scan (systematic variation/insertion of cyclization of a peptide) [20]. Signals can be detected by radioactivity [39, 40], fluorescence [40], chemiluminescence, or colorimetric staining. In this chapter, we only describe the colorimetric detection due to the simplicity in setup and the lower background compared to fluorescence-based detection as result of autofluorescence of peptides.

3.3.1 Preparation of Amino-Alkyl Ether-Linked Membranes for Probing Directly on the Membrane as Peptide Macroarrays

In contrast to those modified by amino acids linked via an ester bond, for the SPOT synthesis of peptide macroarrays the use of amino-alkyl ether-linked membranes [19, 20] is recommended, since they keep the peptides stably attached to the cellulose. This stability also allows the storage at −20 °C for several months. For that reason, it makes sense to prepare a large membrane (e.g., 19 cm × 29 cm) and cut it into several smaller pieces if needed.

• 1. Wash the membrane with about 50 ml of membrane washing solution for several minutes. Then air-dry the membrane.

• 2. Activate the membrane by treatment with 50 ml of membrane activation solution in a closed box.

• 3. After 3 h, wash the membrane once with EtOH or MeOH for about 15 min.

• 4. Leave the activated membrane above in about 60 ml of a 50–75 % solution of 1,3-diaminopropane (CAPE) or 4,7,10-trioxa-1,13-tridecanediamine (TOTD, trioxa, DIPEG) in DMF overnight. The concentration of the amines affects the density of the amino groups on the membrane surface.

• 5. On the next day, wash the membrane three times with DMF, twice with water, and then three times with MeOH [41, 42].

• 6. After treatment with a methanolic suspension of 5 M sodium methylate, wash the membrane with MeOH three times and water five times followed by MeOH three times.

• 7. Staining (optional) [33]: Wash the membrane at least three times with MeOH or EtOH for at least 30 s each.

• 8. Treat the membrane with staining solution for at least 2 min until the filter paper shows a homogeneous blue color (see Note 22 ). If staining is insufficient, renew the staining solution.

• 9. After staining, wash the membrane at least twice with MeOH or EtOH, until the wash solution remains colorless.

• 10. Dry the membrane with the airstream of a fume hood (see Note 12 ). The membrane is ready for the first coupling or storage.

• 11. For the direct use, cut the required size of the membrane and store the remaining sheet. The amino functionality of amino-alkyl ether-linked membranes is typically in the range of 350–700 nmol/cm².

3.3.2 Preparation of Coupling Solutions

As mentioned in Subheading 3.1.2, two different methods are commonly used for the preparation of coupling solutions. The first method uses Fmoc-protected amino acid pentafluorophenyl esters, which has been described in detail elsewhere [15]. For nonstandard amino acids, e.g., phosphoamino acid derivatives, the preactivated derivatives need to be synthesized. Alternatively, an in situ activation of protected Fmoc-amino acid derivative can be used. We elaborate further on the latter method here, as the inclusion of phosphopeptides as positive controls is usually necessary in kinase activity assays.

• 1. Dissolve Fmoc-amino acids in 0.9 M HOBt solution (in NMP) to a concentration of 0.45 M. Except for the arginine derivatives, these solutions can be stored at −20 °C for at least a week. Use fresh aliquots of the prepared amino acid/HOBt solutions on a daily basis (see Note 4 ).

• 2. To these solutions add a fresh prepared mixture of 20 % DIC in NMP at a ratio of 3:1 (e.g., for final volume of 100 μl of in situ-activated amino acid solution mix 75 μl of HOBt/amino-acid solution with 25 μl of 20 % DIC/NMP) (see Note 23 ). This solution is ready to use.

3.3.3 SPOT Synthesis of the Peptide Macroarray

The SPOT synthesis of a peptide macroarray follows a similar protocol as described in Subheading 3.1.3. Due to the much smaller volume delivered to each spot (0.1 μl instead of 1.0 μl) and the large number of peptides, manual synthesis of peptide macroarrays is not recommended. At minimum, delivery of the amino acid solutions should be carried out by a robotic system. Since the coupling protocol may vary between the different spotting devices, no general protocol is described here but the instructions are available in Subheading 3.3.2 above. Ether-modified membranes are usually a bit sturdier than esterified membranes, which makes them much easier to handle during the final TFA treatment (Subheading 3.1.3, step 17). After this step, the membrane is ready for use or can be stored at or below −20 °C.

3.3.4 Probing of the Peptide Macroarray

If the synthesized macroarray is to be used immediately after the TFA treatment, skip the drying step. Otherwise, soak the dried membrane in MeOH or EtOH for 10 min prior to probing.

• 1. Block the membrane in macroarray buffer II at room temperature overnight.

• 2. Incubate the membrane in macroarray buffer III for 1 h at 30 °C in a shaking incubator.

• 3. Dilute the purified active protein kinase in the kinase assay buffer supplemented with 100 μM ATP. The final concentration of the kinase varies according to its specific activity.

• 4. Incubate the membrane in the solution for 0.5–2 h at 30 °C in a shaking incubator.

• 5. Wash the membrane in TBST three times, 10 min each time at room temperature.

• 6. Block unspecific binding sites with the macroarray blocking buffer for 2 h at room temperature.

• 7. Wash the membrane once in the wash buffer for 5 min.

• 8. Make up selected generic phospho-Ser/Thr/Tyr antibody with macroarray blocking buffer following the supplier’s recommendation.

• 9. Incubate the membrane in the antibody solution overnight at 4 °C with agitation.

• 10. Wash the membrane in wash buffer three times, 10 min each time.

• 11. Incubate the membrane with the HRP-conjugated secondary antibody in macroarray blocking buffer for 2 h at room temperature.

• 12. Wash the membrane in wash buffer three times, 10 min each time.

• 13. Wash the membrane in TBS three times, 5 min each time.

• 14. Mix staining solution I with staining solution II. Immediately before use, add 5 μl of 30 % H₂O₂.

• 15. Treat the membrane with the above staining mixture until the signals are fully developed.

• 16. Stop the reaction by washing thoroughly with water. Scan the wet membrane using a flatbed scanner and save the resulting image (see Note 24 ). An example of macroarray image generated is shown in Fig. 3.

  Fig. 3

  

  Example of a substitutional analysis on a protein kinase peptide substrate macroarray: From a wild-type substrate peptide sequence of human JAK kinase (first column on the macroarray), amino acids at each position were substituted with other amino acids systematically. Active JAK kinase was used to phosphorylate the macroarray membrane. The phosphorylation signal of the each peptide spot was detected with the 4G10 monoclonal phosphotyrosine-specific antibody as described in Subheading 3.3

4 Notes

• 1. Solvents and reagents used here are toxic, corrosive, irritant, and/or flammable and can be hazardous. Therefore, precautions while handling them should be taken.

• 2. The quality of solvents for washing steps should be at least of ACS grade. Solvents for dissolving reagents for the synthesis must be amine and water free. Due to the risk of decomposition under the influence of light, organic solvents, with the exception of MeOH and EtOH, should be stored in the dark. The water used should always be distilled/deionized.

• 3. Due to linear structure of the molecule and the resulting flexibility, β-alanine is commonly used for amine functionalization of the filter paper by esterification. More recently, due to the higher loading on the membrane and lower stability during cleavage, glycine is used to functionalize membranes for free peptide synthesis [32]. While other amino acids may be used, a higher risk of losing functionality is expected as a result of the lower chemical stability of the ester bond between the amino acid and cellulose. The amino functionalization solution is prepared prior to use.

• 4. Reagents must be protected from moisture. To avoid condensation from air humidity, reagent containers should be kept unopened for approximately 30 min after storage in fridge or freezer before use.

• 5. Do not pour the mixture into the TFA! Otherwise, it could heat up to a dangerous level! Mix the additives first and then pour the TFA into that mixture.

• 6. Protein kinase buffer is usually made as 5× or 10× stock solution without DTT and stored at −20 °C. DTT is dissolved in DMSO to a stock concentration of 1 M and frozen in small aliquots. Before use, dilute the 5× or 10× protein kinase buffer with dH₂O and add DTT to a final concentration of 0.5 mM.

• 7. Pro-Q Diamond destaining solution can also be made with 50 mM sodium acetate in 20 % acetonitrile (v/v), pH 4.

• 8. Macroarray buffer I is usually made as 5× or 10× stock solution without DTT and stored at −20 °C. DTT should be freshly added before use.

• 9. For more consistent physicochemical behavior, it is recommended after the functionalization overnight to treat the membrane with an appropriate amount of capping solution for at least 20 min.

• 10. Loss of functionality is possible after long-term storage. For long-term storage, −80 °C is recommended. Wash the membrane at least three times with MeOH (or EtOH) and let it air-dry prior to storage. Any traces of amines or traces of water on the membrane could lead to complete loss of activity within a relative short period of time! Place the membrane in a sealed plastic bag. To proceed with the synthesis after storage, treat the membrane once with DMF for about 20 min.

• 11. The staining has no detrimental effect on peptide synthesis process itself. Free amino groups of coupled amino acids are stained blue. Thus, it is an indicator for the completion of the previous coupling step. Since unreacted amino groups are acetylated during the capping step, the spots remain colorless.

• 12. For quick drying, the membrane can be further washed twice with DEE.

• 13. The staining of the spots must not be too strong, since a high amount of absorbed BPB could lead to an incorporation of some dye into the peptide. In that case, the removal of the dye is difficult and may affect detection after incubation. Due to varying acidity of the coupled amino acids and the built-up peptide chain, differences in the intensity of the staining of the spots are normal: for example, spots where aspartic acid or cysteine as last coupled amino acids may show no or very little staining, whereas alanine and lysine show usually a deep blue color.

• 14. During TFA treatment, the membrane may become very soft. Do not pour the liquid directly onto the membranes as it may destroy it. Shake very gently and do not try to lift the membrane out until, after several washing steps, it becomes harder and less likely to be torn.

• 15. If it is necessary to yield a free carboxyl group at the C-terminus, do not treat the membrane with ammonia gas. Instead, punch out the spots, transfer them into vials, and treat them with an aqueous strong basic solution such as 50 % ammonium hydroxide, 1 % triethylamine, or 1 M sodium hydroxide solutions [18, 39]. This solution needs to be exchanged for microarray printing buffer before microarray production.

• 16. Since many desiccators made from plastics are not inert to ammonia gas, it is strongly recommended to use glass desiccators only.

• 17. To reduce possible contamination with side products that are in higher concentration at the rim of the spot, the diameter of the punched-out membrane discs should be smaller than the spot diameter.

• 18. All incubations and washes are performed in microarray reaction trays on a horizontal shaker or tilt rocking platform at low speed.

• 19. A humidity chamber keeps the slide from drying up during the 2-hour incubation, especially when the reaction volume is small. We suggest the speed of the shaking incubator be set below 100 rpm. High shaking speed could cause increase of background.

• 20. If using N₂ flow, dry the slide quickly to avoid water markers in the background when scanning. An alternative method is to dry the slide by spinning in a swing-bucket centrifuge.

• 21. Pro-Q Diamond stain has excitation maxima of 555 nm and emission maxima of 580 nm [37]. Please refer to the user manual of Pro-Q Diamond phosphoprotein/phosphopeptide microarray stain from Molecular Probes for detailed information.

• 22. In contrast to esterified membranes, the background of TODT and CAPE membranes from the dye staining may disappear slowly over several coupling cycles. Nevertheless, it should not affect the quality of the synthesis.

• 23. To avoid clogging of the needle by urea during automatic spotting, shake the in situ-activated amino acid solutions for about 30 min; then centrifuge the mixtures and transfer the supernatants into vials of the synthesis rack.

• 24. Scan the membrane when it is wet. The staining could fade during drying and change color, making the spots undistinguishable.