Nanomolar Protein Thermal Profiling with Modified Cyanine Dyes

Protein properties and interactions have been widely investigated by using external labels. However, the micromolar sensitivity of the current dyes limits their applicability due to the high material consumption and assay cost. In response to this challenge, we synthesized a series of cyanine5 (Cy5) dye-based quencher molecules to develop an external dye technique to probe proteins at the nanomolar protein level in a high-throughput one-step assay format. Several families of Cy5 dye-based quenchers with ring and/or side-chain modifications were designed and synthesized by introducing organic small molecules or peptides. Our results showed that steric hindrance and electrostatic interactions are more important than hydrophobicity in the interaction between the luminescent negatively charged europium-chelate-labeled peptide (Eu-probe) and the quencher molecules. The presence of substituents on the quencher indolenine rings reduces their quenching property, whereas the increased positive charge on the indolenine side chain improved the interaction between the quenchers and the luminescent compound. The designed quencher structures entirely altered the dynamics of the Eu-probe (protein-probe) for studying protein stability and interactions, as we were able to reduce the quencher concentration 100-fold. Moreover, the new quencher molecules allowed us to conduct the experiments using neutral buffer conditions, known as the peptide-probe assay. These improvements enabled us to apply the method in a one-step format for nanomolar protein–ligand interaction and protein profiling studies instead of the previously developed two-step protocol. These improvements provide a faster and simpler method with lower material consumption.


Materials and instrumentation
The  dissolved in 100 μL of water and mixed with the probe peptide (0.5 mg) in 100 μL of the labeling buffer.Reaction solution was incubated at room temperature (RT) for 18 h.Eu-probe purification was carried out using reversed-phase adsorption chromatography, Dionex ultimate 3000 LC system from Thermo Fischer Scientific, Dionex, and Ascentis RP-amide C18 column from Sigma-Aldrich, Supelco Analytical under the following conditions: eluent system containing 50 mM TEAA pH 7.0:ACN 100%, linear gradient (1 ml/min from 10:90 to 50:50 in 17 min).After purification, Eu-probe concentration was determined based on the Eu(III)ion concentration by comparing observed TRL-signal to a commercial Eu(III)-standard (DELFIA), assuming that the probe peptide and Eu(III)-chelate are in 1:1 ratio.

Quenchers Preparation
9][10] Quencher 2. First step: 2,3,3-Trimethylindolenine (160 µL, 1 mmol) and iodobutane (136 µL mL, 1.2 mmol) were added to a microwave reaction vial equipped with a micro magnetic stir bar.The microwave vial then was sealed with a camp and heated in microwave oven at 155 °C for 30 min.After cooling, the resulting residue was washed with ether/aceton, however, unlike all the procedures we could not isolate the desired product as a solid but an oily sticky material, which was used as a starting material in the second step of the reaction.
Second step: a mixture of the crude product of the first step (68.6, 0.2 mmol), bis-iminium salt (34.3 mg, 0.13 mmol), NaOAc (28.9 mg, 0.4 mmol) and acetic anhydride (1.7 mL) was placed in sealed microwave vessel with stirring bar.Sealed vessel was heated in in microwave oven at 150 °C for 20 min.Reaction mixture was diluted with diethyl ether (10-20 mL) and filtered in vacuo.Solid was washed twice with diethyl ether (5 mL).A clean filter flask was attached to the funnel and the resulting solid was dissolved with dichloromethane (10-15 mL) leaving unreacted sodium acetate crystals on the filter funnel.The filtrate was transferred to a clean round bottom flask and dichloromethane was removed with a rotary evaporator.The blue/green crude was formed after solvent removal.The crude product was purified by reverse-phase HPLC.MS calculated for C 33 H 43 N 2 + 467.34 found 467.32.
Quencher 3. First step: 2,3,3-Trimethylindolenine (32 µL, 0.2 mmol) and benzyl bromide (47.5 µL mL, 0.4 mmol) were added to a microwave reaction vial equipped with a micro magnetic stir bar.The microwave vial then was sealed with a camp and heated in microwave oven at 130 °C for 30 min.After cooling, the resulting residue was washed with ether/aceton, however, unlike all the procedures we could not isolate the desired product as a solid but an oily sticky material, which was used as a starting material in the second step of the reaction.
Second step: a mixture of the crude product of the first step (65.8, 0.2 mmol), bis-iminium salt (25.9 mg, 0.1 mmol), NaOAc (22 mg, 0.3 mmol) and acetic anhydride (1.3 mL) was placed in sealed microwave vessel with stirring bar.Sealed vessel was heated in in microwave oven at 150 °C for 20 min.Reaction mixture was diluted with diethyl ether (10-20 mL) and filtered in vacuo.Solid was washed twice with diethyl ether (5 mL).A clean filter flask was attached to the funnel and the resulting solid was dissolved with dichloromethane (10-15 mL) leaving unreacted sodium acetate crystals on the filter funnel.The filtrate was transferred to a clean round bottom flask and dichloromethane was removed with a rotary evaporator.The blue crude was formed after solvent removal.The crude product was purified by reverse-phase HPLC.

Data analysis
In all assays, the signal/background ratio (S/B) was calculated as µmax/µmin, where µmax is the signal of thermally denatured protein and µmin is the signal of native protein.The data were analyzed using Origin 2016 software (Origin Lab, Northampton, MA) with the standard sigmoidal fitting function: For two-phase curves, T m values were determined separately for each phase.Significantly lower signal was observed with Q7 than Q1.This can be explained by the influence of either steric interferences of the larger side chain groups or high hydrophobicity of Q2-Q3.Beckford et al. 11 suggested that steric hindrance of side chain groups is a determining factor in binding between cyanine dyes and human serum albumin (HSA).In the case of Q7, the electrostatic interaction between positively charged side chain amine and the negatively charged peptide of the Eu-probe compensated for the side chain steric hindrance.In contrast, the presence of substituents in the indolenine rings of Q4-Q led to low quenching ability, which may be due to steric hindrance of substituents and/or electrostatic repulsion between negatively charged sulfate groups and the peptide of the Eu-probe.Obviously, the negative charge of Q6 also diminishes Euprobe/quencher interaction.Black column represents Eu-probe TRL-signal without a quencher.

Figure S5.
Comparison of S/B ratios in Protein-Probe antibody denaturation assays using the Eu-probe (1 nM) and quenchers Q1-Q10 (1 µM).Q7 exhibited the highest S/B ratio.Q8 showed a lower S/B ratio than Q9, indicating that the presence of a hydroxyl group on the indolenine side chain (Q8) was less effective.The hydroxyl group does not carry a positive charge at the assay pH of 4, unlike the amino group, leading to a weaker interaction with the Eu-probe.Furthermore, eliminating the amino group but retaining the C6 alkyl chain structure (Q10) of Q7 caused a significant decline in the S/B ratio compared to Q7.The S/B ratio was calculated by dividing the mean TRL-signal measured with the denatured sample at 85 °C by that of the intact sample in the antibody assay.                          in the one-step Peptide-Probe assays using the Eu-probe (6 nM) and Q14 (15 nM) in HEPES buffer (10 mM, pH 7.4, 30 mM NaCl, 0.01% Triton X-100).At high protein concentration, the profiles are clearly one-phasic while at lower concentrations the profiles are two-phasic indicating fine-structural melting profiling.It is also obvious from the measured data that more fine-structural information is obtained at reduced protein concentration.Within the aims of the current study, we did not further investigate this.However, this may originate from dimeric structure of the protein as the dimer dissociation constant is at the low nanomolar range indicating that the proteins were as dimers at all measured concentrations. 20Moreover, p53 Y220C had a lower T m value and higher T m (phase 2 vs. phase 1) compared to the other p53 proteins, which may reflect the effect of this specific mutation on protein stability and structural properties. 21The biphasic data are interesting as all current methodologies measure thermal profiles at elevated concentrations above 1000 nM concentration -apparently missing some fine-structural information leading to concentration-driven T m value data interpretation.

Figure S1 .
Figure S1.(A) Two-step Protein-Probe protocol: 1 st cycle: Intact protein (8 µL) is heated at desired temperature, Eu-probe and quencher 1 (65 µl, pH 4) are added and TRL-signal is measured.2 nd cycle: addition of fresh intact protein (8 µL), Euprobe and quencher (65 µl, pH 4) to new wells and heated followed by TRL measurement.Subsequent cycles: The same cycling continues until the protein is fully denatured.(B) One-step Peptide-Probe protocol: Intact protein (5 µL) and the Euprobe and Q14 (15 µl, pH 7.5) are heated at desired temperature and measured for TRL-signal.The same wells are further heated at higher temperature and remeasured until protein is fully denatured.

Figure S2 .
Figure S2.HPLC trace (monitored at 254 nm) of Q2.Gradient from 40% to 75% MeCN in 16 min.The retention time of the desired product was 13.3 min.

Figure S3 .
Figure S3.HPLC trace (monitored at 254 nm) of Q3.Gradient from 40% to 75% MeCN in 16 min.The retention time of the desired product was 12.2 min.

Figure S4 .
Figure S4.Comparison of quenching efficiency of Q1-Q7 with the Eu-probe: quenchers (1 µM) were mixed with the Euprobe (1 nM) in citrate-phosphate buffer, pH 4, and their quenching ability was determined by measuring their TRL-signal.Significantly lower signal was observed with Q7 than Q1.This can be explained by the influence of either steric interferences of the larger side chain groups or high hydrophobicity of Q2-Q3.Beckford et al.11 suggested that steric hindrance of side chain groups is a determining factor in binding between cyanine dyes and human serum albumin (HSA).In the case of Q7, the electrostatic interaction between positively charged side chain amine and the negatively charged peptide of the Eu-probe compensated for the side chain steric hindrance.In contrast, the presence of substituents in the indolenine rings of Q4-Q led to low quenching ability, which may be due to steric hindrance of substituents and/or electrostatic repulsion between negatively charged sulfate groups and the peptide of the Eu-probe.Obviously, the negative charge of Q6 also diminishes Euprobe/quencher interaction.Black column represents Eu-probe TRL-signal without a quencher.

Figure S6 .
Figure S6.HPLC trace (monitored at 254 nm) of Q8.Gradient from 30% to 80% MeCN in 20 min.The retention time of the desired product was 14.2 min.

Figure S7 .
Figure S7.HPLC trace (monitored at 254 nm) of Q9.Gradient from 30% to 80% MeCN in 20 min.The retention time of the desired product was 12.6 min.

Figure S8 .
Figure S8.HPLC trace (monitored at 254 nm) of Q10.Gradient from 30% to 80% MeCN in 22 min.The retention time of the desired product was 21.3 min

Figure S10 .
Figure S10.HPLC trace (monitored at 254 nm) of Q12.Gradient from 0% to 80% MeCN in 20 min.The retention time of the desired product was 13 min.

Figure S11 .
Figure S11.HPLC trace (monitored at 254 nm) of Q 13.Gradient from 0% to 80% MeCN in 20 min.The retention time of the desired product was 12.6 min.

Figure S12 .
Figure S12.HPLC trace (monitored at 254 nm) of Q 14. Gradient from 0% to 80% MeCN in 20 min.The retention time of the desired product was 12 min.

Figure S13 .
Figure S13.Comparison of S/B ratios in Protein-Probe antibody denaturation assays using the Eu-probe (1 nM) and quenchers Q1, Q7, Q11-Q14 (1 µM).The S/B ratio was calculated by dividing the mean TRL-signal measured with the denatured sample at 85 °C by that of the intact sample in the antibody assay.

Figure S16 .
Figure S16.S/B ratios of Protein-Probe antibody denaturation assays using the Eu-probe (1 nM) and Q13 (1 μM) across various buffer conditions (phosphate-citrate pH 4-7; HEPES 10 mM pH 7.5; Tris 10 mM pH 7.5), all supplemented with 0.01% Triton X-100.The S/B ratio was calculated by dividing the mean TRL-signal from the denatured sample at 85 °C by that of the native antibody sample kept at room temperature.

Figure S18 .
Figure S18.S/B ratios of Protein-Probe antibody denaturation assays using the Eu-probe and Q14 at different concentrations (1.56-200 nM) in various buffer conditions (phosphate-citrate pH 4, 5, 7; HEPES 10 mM pH 7.5; Tris 10 mM pH 7.5; H 2 O) all supplemented with 0.01% Triton X-100.The S/B ratio was calculated by dividing the mean TRL-signal from the denatured sample at 85 °C by that of the native antibody sample kept at room temperature.

Table S1 .
Melting temperature (T m ) values of bovine carbonic anhydrase (BCA II) using different methods and conditions.

Table S2 .
Comparison of Thermal Profile Data for p53 wt , p53 R273C , and p53 Y220C Using the One-Step Peptide-Probe and DSF Methods.The T m values calculated with the DSF method were lower than those obtained with the one-step Peptide-Probe method.b Two-phasic curves were obtained with the one-step Peptide-Probe method. a