N1-Propargylguanosine Modified mRNA Cap Analogs: Synthesis, Reactivity, and Applications to the Study of Cap-Binding Proteins.

The mRNA 5′ cap consists of N7-methylguanosine bound by a 5′,5′-triphosphate bridge to the first nucleotide of the transcript. The cap interacts with various specific proteins and participates in all key mRNA-related processes, which may be of therapeutic relevance. There is a growing demand for new biophysical and biochemical methods to study cap–protein interactions and identify the factors which inhibit them. The development of such methods can be aided by the use of properly designed fluorescent molecular probes. Herein, we synthesized a new class of m7Gp3G cap derivatives modified with an alkyne handle at the N1-position of guanosine and, using alkyne-azide cycloaddition, we functionalized them with fluorescent tags to obtain potential probes. The cap derivatives and probes were evaluated in the context of two cap-binding proteins, eukaryotic translation initiation factor (eIF4E) and decapping scavenger (DcpS). Biochemical and biophysical studies revealed that N1-propargyl moiety did not significantly disturb cap–protein interaction. The fluorescent properties of the probes turned out to be in line with microscale thermophoresis (MST)-based binding assays.


Introduction
A cap structure is present on 5 end of all eukaryotic mRNAs. It plays significant roles in gene expression and mRNA metabolism. The cap consists of 7-methylguanosine bound by a 5 ,5 -triphosphate bridge to the first nucleoside of the mRNA transcript [1]. The positively charged nucleobase and the negatively charged phosphate bridge are essential for retaining cap functionality in its interactions with cap-recognizing proteins [2][3][4]. In a cell, the cap is targeted by multiple proteins involved in mRNA-related processes such as eukaryotic translation initiation factor 4E (eIF4E), which is involved in translation initiation, and decapping scavenger (DcpS) or protein complex Dcp1/Dcp2, which participate in mRNA degradation. Certain cap-binding proteins have been identified as potential therapeutic targets. Elevated levels of eIF4E may induce tumorigenesis [5] and have been detected in numerous cancer cells [6,7]. Thus, eIF4E is a potential antitumor therapeutic target [8][9][10]. Furthermore, DcpS decapping enzyme, was found to be a molecular target for spinal muscular atrophy (SMA) [11]. Chemically modified cap analogs including fluorescent probes are useful tools for the study of cap-binding proteins. To this end, nucleotides labeled with antibodies [12], fluorescent tags [13][14][15][16], EPR tags [17], and radioisotopes [18] have been developed.
Compounds 22 or 23 were coupled with N7-methylguanosine-5 -O-monophosphate-P-imidazolide (m 7 GMP-Im) (24) in DMF in the presence of ZnCl 2 to obtain cap analogs 4 and 5, respectively. The final products (1)(2)(3)(4)(5) were purified by ion exchange chromatography and the yields were 36-58%. The products were re-purified by semipreparative RP-HPLC, isolated as ammonium salts, and used for labeling in the CuAAC reaction and biophysical studies. The structures and purities of all new compounds were confirmed by HRMS, 1 H NMR, and 31 P NMR (Supplementary Information).

Binding Affinities to eIF4E
The prime role of the mRNA cap is to participate in initiation of translation through binding with eIF4E. Therefore, we investigated how the substitution at the N1 position influences the interaction between the cap analogs and eIF4E. To this end, we performed binding competition experiments with a Py-labeled m 7 GTP analog as previously described [16]. Thence, we determined IC50 values and calculated binding constants values (KD) for the cap analog-eIF4E complexes (Table 1, Figure 2). It was found that cap analogs 1-5 have affinities similar to that of the native cap structure (m 7 Gp3G) ( Figure S1, Table S1). Additionally, the differences identified among studied compounds derived from triphosphate chain modifications were in agreement with previously published results [30][31][32]. Reactivity of the cap analogs containing N1-propargylguanosine in CuAAC was evaluated. To this end, probes (6)(7)(8)(9)(10)(11)(12)(13)(14) were obtained by click chemistry in the reaction between commercially available fluorescent tags azide analogs (25)(26)(27)(28) or biotin azide (29) [14] and cap analogs bearing the propargyl moiety (1-5) (Figure 1). An aqueous solution of cap analog (1-5) was mixed with a tag (25)(26)(27)(28)(29) dissolved in dimethyl sulfoxide (DMSO) at room temperature (RT) followed by the addition of aqueous solutions of CuSO 4 and sodium ascorbate. The concentrations of the tags used for labeling ranged from 100-190 mM. HPLC analysis revealed that under these conditions, the conversion rates were in the range of 51-100%. The reactions were quenched by the addition of sodium ethylenediaminetetraacetate/sodium bicarbonate solution (Na 2 EDTA/NaHCO 3 ), and the probes were purified by HPLC and isolated as ammonium salts. Using this approach, nine new probes were synthesized ( Figure 1). To assess how substitution at the N1-position of guanosine influences mRNA cap recognition by specific proteins and enzymes, we performed a series of experiments with unlabeled cap analogs (1-5) and fluorescently labeled probes (6-14).

Binding Affinities to eIF4E
The prime role of the mRNA cap is to participate in initiation of translation through binding with eIF4E. Therefore, we investigated how the substitution at the N1 position influences the interaction between the cap analogs and eIF4E. To this end, we performed binding competition experiments with a Py-labeled m 7 GTP analog as previously described [16]. Thence, we determined IC 50 values and calculated binding constants values (K D ) for the cap analog-eIF4E complexes (Table 1, Figure 2). It was found that cap analogs 1-5 have affinities similar to that of the native cap structure (m 7 Gp 3 G) ( Figure S1, Table S1). Additionally, the differences identified among studied compounds derived from triphosphate chain modifications were in agreement with previously published results [30][31][32]. Table 1. IC 50 and K D values determined at 30 • C for the binding affinity for murine eIF4E. The IC 50 values were determined from the data shown in Figure 2. The K D values were calculated from the IC 50 values using a previously reported equation [33].  Figure 2. Dose-response curves for compounds 1-5 and m 7 Gp3G plotted from a competition binding assay for meIF4E [16]. eIF4E (75 nM) and a pyrene-labeled fluorescent probe (10 nM) were incubated at 30 °C with the indicated concentrations of the investigated compounds (1-5 or m 7 Gp3G as a reference). IC50 and KD values from triplicate experiments are shown in Table 1.

Compound
The propargyl moiety introduced to the N1 position of guanosine had minimal effect on the interaction with eIF4E (Table S1). For this reason, these cap analogs are very promising as tools to study eIF4E. Table 1. IC50 and KD values determined at 30 °C for the binding affinity for murine eIF4E. The IC50 values were determined from the data shown in Figure 2. The KD values were calculated from the IC50 values using a previously reported equation [33].

Enzymatic Degradation Studies
Susceptibility of the new cap analogs to hydrolysis by human decapping scavenger enzyme (hDcpS) was investigated to assess the influence of N1 substitution on cap analogs recognition by DcpS. Caps 1-5 or m 7 Gp3G (30 µM each) were incubated with hDcpS (30 nM) at 30 °C. Then, reaction samples were thermally deactivated and analyzed by RP-HPLC. Under these conditions, cap analog 1 was efficiently cleaved by DcpS but more slowly than m 7 Gp3G. All analogs modified within the phosphate bridge (2)(3)(4)(5) were resistant to hDcpS under the applied conditions ( Figure 3). The propargyl moiety at the N1 position of a second nucleobase slightly decreased the susceptibility of the new cap analogs to hDcpS cleavage but prevented neither the recognition nor the regioselectivity of the cleavage (Table S2). . Dose-response curves for compounds 1-5 and m 7 Gp 3 G plotted from a competition binding assay for meIF4E [16]. eIF4E (75 nM) and a pyrene-labeled fluorescent probe (10 nM) were incubated at 30 • C with the indicated concentrations of the investigated compounds (1-5 or m 7 Gp 3 G as a reference). IC 50 and K D values from triplicate experiments are shown in Table 1.
The propargyl moiety introduced to the N1 position of guanosine had minimal effect on the interaction with eIF4E (Table S1). For this reason, these cap analogs are very promising as tools to study eIF4E.

Enzymatic Degradation Studies
Susceptibility of the new cap analogs to hydrolysis by human decapping scavenger enzyme (hDcpS) was investigated to assess the influence of N1 substitution on cap analogs recognition by DcpS. Caps 1-5 or m 7 Gp 3 G (30 µM each) were incubated with hDcpS (30 nM) at 30 • C. Then, reaction samples were thermally deactivated and analyzed by RP-HPLC. Under these conditions, cap analog 1 was efficiently cleaved by DcpS but more slowly than m 7 Gp 3 G. All analogs modified within the phosphate bridge (2)(3)(4)(5) were resistant to hDcpS under the applied conditions ( Figure 3). The propargyl moiety at the N1 position of a second nucleobase slightly decreased the susceptibility of the new cap analogs to hDcpS cleavage but prevented neither the recognition nor the regioselectivity of the cleavage (Table S2). Fluorescently labeled cap analogs that serve as DcpS substrates can be considered as activity probes. We performed an enzymatic cleavage test with 30 nM hDcpS and 30 µM probe to evaluate whether fluorescently labeled derivatives of cap analog 1 (probes 6-9) can be used to monitor enzymatic activity via changes in fluorescence emission. For probes 6-8, the fluorescence intensity at the emission maximum did not significantly change following complete cleavage by DcpS (Figures 4  and S2). For the Py-labeled probe 9, the fluorescence intensity at the emission maximum increased by 50% after complete DcpS cleavage (Figures 4c and S3). Nevertheless, certain previously developed probes provided higher sensitivity than the one formulated here [15,16]. Modest changes in fluorescence emission upon the enzymatic cleavage of compounds 6-9 were independently confirmed with the nonspecific pyrophosphatase PDE-I ( Figure S4). The lack of fluorescence changes upon enzymatic cleavage suggests that the fluorescent labels attached to the N1 position of guanosine are relatively insensitive to alterations in the local environment. This characteristic is desirable in fluorescence polarization (FP) and microscale thermophoresis (MST) binding assays. Therefore, in the next step, we tested whether compounds 6-14 could serve as binding probes for either eIF4E or DcpS.

Evaluation of N1 Fluorescently Labeled Cap Analogs as Probes for Microscale Thermophoresis
First, we performed MST direct binding experiments with fluorescein-labeled probes 6, 11, or 12 and murine eIF4E (meIF4E). Fluorescently labeled cap analogs that serve as DcpS substrates can be considered as activity probes. We performed an enzymatic cleavage test with 30 nM hDcpS and 30 µM probe to evaluate whether fluorescently labeled derivatives of cap analog 1 (probes 6-9) can be used to monitor enzymatic activity via changes in fluorescence emission. For probes 6-8, the fluorescence intensity at the emission maximum did not significantly change following complete cleavage by DcpS ( Figure 4 and Figure S2). For the Py-labeled probe 9, the fluorescence intensity at the emission maximum increased by 50% after complete DcpS cleavage ( Figure 4c and Figure S3). Nevertheless, certain previously developed probes provided higher sensitivity than the one formulated here [15,16]. Modest changes in fluorescence emission upon the enzymatic cleavage of compounds 6-9 were independently confirmed with the nonspecific pyrophosphatase PDE-I ( Figure S4). Fluorescently labeled cap analogs that serve as DcpS substrates can be considered as activity probes. We performed an enzymatic cleavage test with 30 nM hDcpS and 30 µM probe to evaluate whether fluorescently labeled derivatives of cap analog 1 (probes 6-9) can be used to monitor enzymatic activity via changes in fluorescence emission. For probes 6-8, the fluorescence intensity at the emission maximum did not significantly change following complete cleavage by DcpS (Figures 4  and S2). For the Py-labeled probe 9, the fluorescence intensity at the emission maximum increased by 50% after complete DcpS cleavage (Figures 4c and S3). Nevertheless, certain previously developed probes provided higher sensitivity than the one formulated here [15,16]. Modest changes in fluorescence emission upon the enzymatic cleavage of compounds 6-9 were independently confirmed with the nonspecific pyrophosphatase PDE-I ( Figure S4). The lack of fluorescence changes upon enzymatic cleavage suggests that the fluorescent labels attached to the N1 position of guanosine are relatively insensitive to alterations in the local environment. This characteristic is desirable in fluorescence polarization (FP) and microscale thermophoresis (MST) binding assays. Therefore, in the next step, we tested whether compounds 6-14 could serve as binding probes for either eIF4E or DcpS.

Evaluation of N1 Fluorescently Labeled Cap Analogs as Probes for Microscale Thermophoresis
First, we performed MST direct binding experiments with fluorescein-labeled probes 6, 11, or 12 and murine eIF4E (meIF4E). The lack of fluorescence changes upon enzymatic cleavage suggests that the fluorescent labels attached to the N1 position of guanosine are relatively insensitive to alterations in the local environment. This characteristic is desirable in fluorescence polarization (FP) and microscale thermophoresis (MST) binding assays. Therefore, in the next step, we tested whether compounds 6-14 could serve as binding probes for either eIF4E or DcpS.

Evaluation of N1 Fluorescently Labeled Cap Analogs as Probes for Microscale Thermophoresis
First, we performed MST direct binding experiments with fluorescein-labeled probes 6, 11, or 12 and murine eIF4E (meIF4E).
To this end, fluorescent probes (25 nM) were mixed with increasing concentrations of meIF4E and microscale thermophoresis was measured for each sample ( Figure S5). Baseline corrected normalized fluorescences (∆F n ) determined from the MST traces were plotted as a function of protein concentration. K D values could then be calculated from the resulting binding curves ( Figure 5). Probes 6 and 12 showed similar affinities for meIF4E, which were higher than the affinity of probe 11 (Table 2). These findings align with those determined for the parent compounds 1, 2, and 3 (Table 1). Additionally, this suggests that labeling at the N1 position of guanosine does not interfere with already known effects of phosphate bridge modification.  Table 2.
To this end, fluorescent probes (25 nM) were mixed with increasing concentrations of meIF4E and microscale thermophoresis was measured for each sample ( Figure S5). Baseline corrected normalized fluorescences (ΔFn) determined from the MST traces were plotted as a function of protein concentration. KD values could then be calculated from the resulting binding curves ( Figure 5). Probes 6 and 12 showed similar affinities for meIF4E, which were higher than the affinity of probe 11 (Table  2). These findings align with those determined for the parent compounds 1, 2, and 3 (Table 1). Additionally, this suggests that labeling at the N1 position of guanosine does not interfere with already known effects of phosphate bridge modification.
With these promising results in mind, we decided to use the same approach for studying DcpS enzyme. For this purpose, cleavage-resistant probes 11 and 12 were used. Both probes were highaffinity ligands for DcpS. Notably, probe 12 had 2.5-fold greater affinity for DcpS than probe 11 (KD = 14 and 38 nM, respectively). This is in agreement with previously reported data on the influence of triphosphate chain modifications on cap-DcpS interaction [30][31][32].

Discussion
In the present study, we aimed to develop new mRNA cap analogs amenable to click chemistry and convertible into potential binding or activity probes. We synthesized m 7 Gp3G dinucleotides bearing a "clickable" propargyl group handle at the N1 position of guanosine. Certain compounds were further modified within the triphosphate chain to modulate their susceptibility to enzymatic degradation. Utility of the probes in the investigation of cap-binding proteins and cap-specific enzymes was evaluated with eIF4E and DcpS, respectively. The N1-propargyl group did not significantly decrease the binding affinity for eIF4E significantly. Moreover, HPLC-monitored enzymatic degradation experiments showed that the N1 modification only slightly impaired recognition by DcpS. The N1-propargyl compounds were then converted into fluorescent probes  Table 2. With these promising results in mind, we decided to use the same approach for studying DcpS enzyme. For this purpose, cleavage-resistant probes 11 and 12 were used. Both probes were high-affinity ligands for DcpS. Notably, probe 12 had 2.5-fold greater affinity for DcpS than probe 11 (K D = 14 and 38 nM, respectively). This is in agreement with previously reported data on the influence of triphosphate chain modifications on cap-DcpS interaction [30][31][32].

Discussion
In the present study, we aimed to develop new mRNA cap analogs amenable to click chemistry and convertible into potential binding or activity probes. We synthesized m 7 Gp 3 G dinucleotides bearing a "clickable" propargyl group handle at the N1 position of guanosine. Certain compounds were further modified within the triphosphate chain to modulate their susceptibility to enzymatic degradation. Utility of the probes in the investigation of cap-binding proteins and cap-specific enzymes was evaluated with eIF4E and DcpS, respectively. The N1-propargyl group did not significantly decrease the binding affinity for eIF4E significantly. Moreover, HPLC-monitored enzymatic degradation experiments showed that the N1 modification only slightly impaired recognition by DcpS. The N1-propargyl compounds were then converted into fluorescent probes with CuAAC. Characterization of the probes by emission spectroscopy revealed that the labels attached to the N1 position are comparatively insensitive to fluorescence intensity changes in response to protein binding or cleavage. Therefore, the labels are useful for experiments monitored by fluorescence intensity changes and for biophysical binding assays based on FP measurements or MST. To demonstrate this, we performed MST binding experiments using both eIF4E and DcpS in which we plotted reproducible binding curves and determined K D . For most compounds, the K D were in the low nanomolar range (Table 2). To the best of our knowledge, the present study is the first example of the application of MST in the analysis of ligand binding to DcpS enzyme. Although compounds 1-5 do not possess extraordinary inhibitory properties in vitro experiments, the probe 12 tightly bound DcpS (K D = 14 nM) and is, therefore, the preferred candidate for the development of an MST-based competitive binding assay to be used in the discovery and further in vitro evaluation of DcpS inhibitors. Studies on the specificity of N1-propargyl containing analogs for other cap binding proteins and decapping enzymes or possibility to incorporate them into RNA are underway.

Starting Materials and Chemical Reagents
Starting materials were purchased from Sigma-Aldrich Corp. (St. Louis, MO, USA) unless specified otherwise. Tetrabutylammonium iodide was purchased from Fluka Honeywell (Mexico City, Mexico) and imidazole was purchased from EMD Merck Millipore (Billerica, MA, USA). For syntheses under anhydrous conditions, solvents were dried over 4 Å molecular sieves for 24 h. Dichlorophosphorylphosphorimidoyl trichloride was prepared as described previously [34] and used in liquid form for subsequent reactions as problems with its crystallization occurred.

Chromatography and Optical Density Measurements
The nucleotides (1-5, 17, 22, and 23) were purified by ion-exchange chromatography on a DEAE Sephadex A-25 (HCO 3 − form) column. The column was loaded with reaction mixture and washed with water until the eluate no longer precipitated in the presence of AgNO 3 solution to elute solvents and salts that do not bind to the column. Nucleotides were eluted with triethylammonium bicarbonate (TEAB) gradients in deionized water: 0-0.7 M for nucleoside monophosphates, 0-1.0 M for nucleoside diphosphates, and 0-1.2 M for nucleoside triphosphates. The collected fractions were analyzed spectrophotometrically at 260 nm and fractions containing the desired nucleotides were analyzed by RP-HPLC and pooled. Yields were calculated on the basis of milliunit optical density measurements (mODU, absorbance of solution multiplied by its volume in ml) of the isolated products and their corresponding starting materials (nucleotides or nucleotide P-imidazolide derivatives). Optical measurements were performed in 0.1 M phosphate buffer pH 7.0. Afterwards, evaporation under reduced pressure with repeated additions of 96% and then 99.8% ethanol was carried out to decompose TEAB and remove residual water. The nucleotides were isolated as triethylammonium (TEAH + ) salts. The final products were re-purified on semipreparative RP-HPLC. The products, after repeated freeze-drying, were isolated as ammonium salts.
UV-detection was performed at 254 nm for nucleotides or at the absorption maximum wavelengths of specific dyes.
The HPLC methods used in the present study were as follows: •

NMR and MS Analyses
The structure and purity of probes were confirmed by high-resolution mass spectrometry with negative or positive electrospray ionization (HRMS (−) ESI or HRMS (+) ESI). Nucleotide derivative structures were confirmed by high-resolution mass spectrometry with negative or positive electrospray ionization (HRMS (−) ESI or HRMS (+) ESI) and 1 H NMR, 13 13 C NMR, and 31 P NMR chemical shifts were reported in ppm and referenced to respective internal standards: Sodium 3-(trimethylsilyl)-2,2 ,3,3 tetradeuteropropionate (TSP) and 20% phosphorus acid in D 2 O. Signals in 1 H NMR dinucleotide spectra were assigned on the basis of their 2D NMR spectra (gDQCOSY, gHSQCAD).

Determination of the eIF4E-Cap Complex Dissociation Constants K D
A previously described pyrene fluorescence intensity binding method was used to determine the dissociation constants of the nonfluorescent ligands binding to the eIF4E protein in a competitive binding experiment [16]. The experiments were performed in 96-well, black, non-binding assay plates with point fluorescence measurements (λ exc = 345 nm; λ em = 378 nm). Each well contained a buffer (50 mM Hepes/KOH pH = 7.2 containing 100 mM KCl and 0.5 mM EDTA), 10 nM of pyrene-labeled 7-methylguanosine pentaphosphate probe, ligand 1-5 or m 7 Gp 3 G (half-log dilutions of C lig from 100 µM to 0.003 µM), and 75 nM of eIF4E protein. The reagents and added protein were pre-incubated for 15 min at 30 • C and stirred at 300 rpm. Measurements were performed in three different temperatures: At 20, 30, and 37 • C. A previously derived Equation (1) [35] was used to calculate the inhibition (dissociation) constant K I of a competitive ligand according to the dependence of the recorded fluorescence intensity on the inhibitor concentration.
where IC 50 is the inhibitor concentration required to replace 50% of the fluorescent probe from the protein binding site, [P t ] is the total protein concentration, [PL] 50 is the protein-probe complex concentration at 50% inhibition, [L] 50 is the free probe concentration at 50% inhibition, and [P] 0 is the free protein concentration without inhibition.
In order to determine IC 50 value, we used Origin ® 2017 (OriginLab Corp., Northampton, MA, USA) software to fit the curve described by the derived Equation (2) to the measured fluorescence intensity values for various inhibitor concentrations.
where C lig. is the concentration of non-fluorescent ligand. The plots were created in GraphPad Prism v. 7.00 (GraphPad Software, La Jolla, CA, USA).

Susceptibility to DcpS Hydrolysis
Enzymatic reactions with human DcpS were carried out in 50 mM Tris/HCl, 200 mM KCl, 0.5 mM EDTA, pH = 7.6 buffer at 30 • C for 30 nM hDcpS and 30 µM of a cap analog studied. As a control, m 7 Gp 3 G was used. Cap analogs concentration was determined spectrophotometrically (absorption coefficient used in calculations is equal to 22,600 mL/mmol/cm in 0.1 M phosphate buffer pH = 7.0). For 1 and m 7 Gp 3 G, 3 independent experiments were performed. Aliquots were terminated by heat inactivation for 3 min at 95 • C. Samples were analyzed with RP-HPLC using method E.

hDcpS and SVPDE Hydrolysis Monitoring
Human DcpS was expressed and purified as described previously [16]. Enzyme was stored at −80 • C at 10 µM concentration (monomer). Enzyme hDcpS was preincubated at 30 • C for 15 min before usage. Enzymatic reactions with hDcpS enzyme were conducted using 30 nM hDcpS (dimer) and 100 nM compound in 50 mM Tris/HCl, 200 mM KCl, 0.5 mM EDTA, pH = 7.6 buffer at 30 • C. Reaction progress was monitored by recording emission spectra, upon excitation at the wavelength characteristic for a fluorophore studied, using Cary Eclipse (Agilent Technologies, Santa Clara, CA, USA) equipped with a xenon lamp in a quartz cuvette (10 mm × 4 mm).
PDE-I from Crotalus adamanteus (EC 3.1.4.1) venom was purchased as a lyophilized solid from Sigma-Aldrich Corp. (St. Louis, MO, USA). The PDE-I was dissolved in a storage buffer (110 mM Tris/HCl pH 8.9 buffer containing 110 mM NaCl, 15 mM MgCl 2 , and 50% glycerol) to prepare a 1 mg/mL solution and then stored at −20 • C. Before the assay, the enzyme was diluted to 100 µg/mL with 50 mM Tris/HCl, pH = 8.0 buffer. Enzymatic reactions with PDE-I were conducted for 100 ng/mL and 100 nM compound in 50 mM Tris/HCl, 5 mM MgCl 2 , pH = 8.0 buffer at 30 • C. Reaction progress was monitored by recording the emission spectra at the excitation wavelength of a fluorophore in a Cary Eclipse (Agilent Technologies, Santa Clara, CA, USA) equipped with xenon lamp using a quartz cuvette 10 × 4 mm. Probes 11 and 12 were dissolved in MST buffer (50 mM HEPES/KOH, 100 mM KCl, 0.5 mM EDTA, pH = 7.2, 0.2% Tween 20) to obtain a 50 nM stock solution. In study of a probe 11, human DcpS (hDcpS) was dissolved in MST buffer in a 16 point 1:1 dilution series in a 2.0 µM to 0.061 nM concentration range. Equal volumes (10 µL) of hDcpS and the probe were mixed together to obtain the following assay concentrations: 25 nM probe and 1.0 µM to 0.030 nM hDcpS. In study of a probe 12, hDcpS was dissolved in MST buffer in a 16 point 2:1 dilution series in a 700 nM to 1.60 nM concentration range. Equal volumes (10 µL) of hDcpS and the probe were mixed together to obtain the following assay concentrations: 25 nM probe and 350 nM to 0.80 nM hDcpS.
For both assays, samples were loaded into standard Monolith™ NT.115 Series Capillares (NanoTemper Technologies, Cambridge, MA, USA). MST measurements were performed on Monolith NT.115 instrument (NanoTemper Technologies, Cambridge, MA, USA) at 25 • C. Measurements parameters were set as follows: 40% LED Blue power, Medium MST power. Data were obtained for three independently pipetted measurements and analyzed in PALMIST v. 1.4.4 [36] using a 1:1 binding model to calculate K D , using the signal from an MST-on time of 20 s (for eIF4E assay: Cold region start, −3 s; hot region start, 0.5 s; for DcpS assay: Cold region start, −3 s; hot region start 19 s), confidence intervals were calculated with error surface projection (ESP) method. Plots presented were generated using GUSSI v. 1.4.2 [37].