Method for Identifying Small Molecule Inhibitors of the Protein-protein Interaction Between HCN1 and TRIP8b

Ye Han; Kyle A. Lyman; Matt Clutter; Gary E. Schiltz; Quratul-Ain Ismail; Xiangying Cheng; Chi-Hao Luan; Dane M. Chetkovich

doi:10.3791/54540

JoVE Journal > Biochemistry

Bioquímica

Method for Identifying Small Molecule Inhibitors of the Protein-protein Interaction Between HCN1 and TRIP8b

Published: November 11, 2016

doi:

10.3791/54540

Ye Han*¹, Kyle A. Lyman*¹, Matt Clutter², Gary E. Schiltz³, Quratul-Ain Ismail¹, Xiangying Cheng¹, Chi-Hao Luan⁴, Dane M. Chetkovich^1,5

¹Davee Department of Neurology and Clinical Neurosciences,Feinberg School of Medicine, Northwestern University, ²Center for Molecular Innovation and Drug Discovery,Northwestern University, ³Department of Pharmacology,Feinberg School of Medicine, Northwestern University, ⁴High Throughput Analysis Laboratory, Department of Molecular Biosciences,Northwestern University, ⁵Department of Physiology,Feinberg School of Medicine, Northwestern University

Summary

The interaction between HCN channels and their auxiliary subunit has been identified as a therapeutic target in Major Depressive Disorder. Here, a fluorescence polarization-based method for identifying small molecule inhibitors of this protein-protein interaction, is presented.

Abstract

Hyperpolarization-activated cyclic nucleotide-gated (HCN) channels are expressed ubiquitously throughout the brain, where they function to regulate the excitability of neurons. The subcellular distribution of these channels in pyramidal neurons of hippocampal area CA1 is regulated by tetratricopeptide repeat-containing Rab8b interacting protein (TRIP8b), an auxiliary subunit. Genetic knockout of HCN pore forming subunits or TRIP8b, both lead to an increase in antidepressant-like behavior, suggesting that limiting the function of HCN channels may be useful as a treatment for Major Depressive Disorder (MDD). Despite significant therapeutic interest, HCN channels are also expressed in the heart, where they regulate rhythmicity. To circumvent off-target issues associated with blocking cardiac HCN channels, our lab has recently proposed targeting the protein-protein interaction between HCN and TRIP8b in order to specifically disrupt HCN channel function in the brain. TRIP8b binds to HCN pore forming subunits at two distinct interaction sites, although here the focus is on the interaction between the tetratricopeptide repeat (TPR) domains of TRIP8b and the C terminal tail of HCN1. In this protocol, an expanded description of a method for purifying TRIP8b and executing a high throughput screen to identify small molecule inhibitors of the interaction between HCN and TRIP8b, is described. The method for high throughput screening utilizes a Fluorescence Polarization (FP) -based assay to monitor the binding of a large TRIP8b fragment to a fluorophore-tagged eleven amino acid peptide corresponding to the HCN1 C terminal tail. This method allows 'hit' compounds to be identified based on the change in the polarization of emitted light. Validation assays are then performed to ensure that 'hit' compounds are not artifactual.

Introduction

Hyperpolarization-activated cyclic nucleotide-gated (HCN) channels are expressed in the heart and central nervous system where they play an important role in regulating membrane excitability¹. HCN channels have been implicated in the pathogenesis of Major Depressive Disorder (MDD)², which has led several groups to propose that limiting HCN channel function pharmacologically may be effective as a novel treatment for MDD³. However, directly targeting HCN channels is not viable because of their important role in the cardiac action potential⁴. Ivabradine, the only FDA approved HCN channel antagonist, is used for the treatment of heart failure to produce a bradycardic effect⁵. As such, there is a need for pharmacologic agents that limit HCN channel function exclusively in the central nervous system.

Tetratricopeptide repeat-containing Rab8b interacting protein (TRIP8b) is a brain specific auxiliary subunit of HCN channels that controls the surface expression and localization of HCN channels^6,7. Genetic knockout of TRIP8b causes a reduction of brain HCN channels⁷ without affecting HCN expression in the heart⁸. Interestingly, TRIP8b knockout mice spend less time immobile on the forced swim task and tail suspension task⁷, two commonly used screening tests for antidepressant efficacy^9-11. These results suggest that rather than directly targeting HCN channels with a small molecule antagonist of HCN channel function, disrupting the interaction between TRIP8b and HCN may be sufficient to produce antidepressant-like behavior.

TRIP8b binds to HCN at two distinct binding sites. The cyclic nucleotide binding domain (CNBD) of HCN interacts with a conserved domain of TRIP8b located N terminal to the TPR domains of TRIP8b^12,13. Although the residues of the CNBD that are involved in this interaction have been mapped¹⁴, the region of TRIP8b that is involved has not been narrowed down beyond an-80-amino acid fragment¹³. A second interaction occurs between the tetratricopeptide repeat (TPR) domains of TRIP8b and the C terminal tripeptide of HCN ('SNL' in HCN1, HCN2, and HCN4, but 'ANM' in HCN3)^3,12. The recently solved crystal structure¹⁵ of this C tail interaction revealed substantial structural similarity to the interaction between the peroxisomal import receptor, peroxin 5 (PEX5), and its interacting partners, containing type 1 peroxisomal targeting sequences (PTS1)¹⁶.

Although both interaction sites are required for HCN channel function, the interaction between the TPR domains of TRIP8b and the C terminal tripeptide of HCN1 serves as the dominant binding site and regulates HCN surface expression. Therefore, this interaction was chosen as the targeting site in this study. For the remainder of the manuscript, when a reference is made to the interaction between TRIP8b and HCN, it is this interaction that is being referred to. This interaction is recapitulated by a highly soluble fragment of TRIP8b corresponding to its conserved C terminal containing the TPR domains required for binding the C terminal tail of HCN (residues 241-602 of the 1a-4 isoforms of TRIP8b)³.

In order to develop a high throughput screen to identify small molecules capable of disrupting this interaction, a fluorescence polarization (FP)-based assay was employed¹⁷. Fluorescence polarization is based on the excitation of a fluorophore-tagged ligand with polarized light, and measuring the degree of polarization of the emitted fluorescence¹⁸. In the presence of a binding partner, the rotational motion of the fluorescent ligand is constrained and polarized light is emitted¹⁹. In the absence of a binding partner, the rotational motion of the ligand leads to the emission of depolarized light.

In the enclosed protocol, a method for the purification of N terminal-tagged (6xHis) TRIP8b (241-602) using Nickel-Nitrilotriacetic acid (Ni-NTA) beads is presented. A similar protocol was employed to purify the Glutathione-S-Transferase (GST)-tagged C terminal 40 amino acids of HCN1 (HCN1_c40) used in step 7 of the protocol. For space considerations, a detailed description of that procedure was omitted.

In steps 2 through 7 of the protocol, a high throughput screening workflow is presented (see Figure 1). Protein-protein interactions are a notoriously difficult target for high throughput screening and readers are advised to seek out additional resources on the topic²⁰.

Steps 2 and 3 of the procedure characterize the in vitro affinity of the purified TRIP8b (241-602) construct for a fluorescein isothiocyanate (FITC)-tagged eleven amino acid peptide corresponding to the C terminal tail of HCN1 (HCN1_FITC). Based on the crystal structure of the TRIP8b-HCN complex¹⁵, this eleven amino acid segment is sufficient to produce binding with TRIP8b (241-602). In step 2, the K_d of the interaction is measured by titrating TRIP8b (241-602) into a fixed concentration of HCN1_FITC. In step 3, an unlabeled version of the HCN peptide used in step 2 is titrated into a fixed concentration of both TRIP8b (241-602) and HCN1_FITC to examine if the FITC tag interferes with binding. These experiments are essential to selecting the appropriate concentrations of TRIP8b (241-602) and HCN1_FITC used in the high throughput screen.

The premise of the high throughput screen is that a small molecule capable of disrupting the interaction between TRIP8b (241-602) and HCN1_FITC will produce a decrease in polarized light. In step 4, the Z factor of the assay is calculated²¹ for a given concentration of TRIP8b (241-602) and HCN1_FITC to ensure that the assay is appropriate for high throughput screening (step 5). Steps 6 and 7 are validation assays to confirm that the hits identified in the primary high throughput screen are acting by disrupting the interaction between TRIP8b (241-602) and HCN1_FITC rather than through a nonspecific mechanism. In step 6, carboxytetramethylrhodamine (TAMRA)-labeled HCN1 peptide (HCN1_TAMRA) is used in an otherwise identical fluorescence polarization assay to filter fluorescent compounds that compromise the FP assay using the FITC tag. Step 7 utilizes a larger HCN1 C terminal fragment (HCN1_C40) and employs a bead-based proximity assay, which is based on the 'tunneling' of a singlet oxygen from a donor bead to an acceptor bead brought close to one another by interacting proteins²².

Protocol

1. Purification of TRIP8b (241-602) Protein

Transform the plasmid containing TRIP8b (241-602) in the bacterial protein expression vector pGS21³ into competent E. coli for protein expression according to manufacturer's instructions. Plate 300 µl of the culture on Luria Broth (LB)-agar with 5 µg/ml of Chloramphenicol and Ampicillin. Incubate the plate at 37 °C for 16 h.
The next day, pick a single colony to inoculate 50 ml LB with 50 µg/ml of Chloramphenicol and Ampicillin. Incubate the culture at 37 °C for 16 hr (with shaking).
Add 50 ml culture to 1 L of LB with 50 µg/ml of Ampicillin. Incubate at 37 °C with shaking.
Once the culture has reached an OD₆₀₀ reading of 0.8-1.2, change the temperature of the incubator to 18 °C and add IPTG (Isopropyl-Beta-D-Thiogalactoside) to a final concentration of 1 mM. Allow protein expression to proceed for 16 hr.
Spin down the E. coli at 6,000 x g for 15 min at 4 °C to pellet the bacteria. After removing the tubes from the centrifuge, keep the bacteria on ice for the remaining of the procedure. Resuspend the bacteria in 36 ml of Buffer A with 0.25 mM of phenylmethanesulfonyl fluoride (PMSF).
Sonicate the resuspended bacteria on ice using a flat bolt for 5-10 min, alternating between 30 sec 'on' and 30 sec 'off' at high power.
Centrifuge at 12,000 x g for 15 min at 4 °C. Centrifuge for an additional 15 min if the supernatant is not clear yet.
Apply the supernatant to a column of 2 ml Ni-NTA beads. Incubate for 60 min at 4 °C with gentle rocking.
Allow the unbound material to flow through the column and wash with 500 ml of Buffer A.
Wash the column with 250 ml of Buffer B.
Wash the column with 125 ml of Buffer A supplemented with 5 mM imidazole.
Elute the protein from the column with 20 ml Elution Buffer.
Add the eluted protein to a dialysis cassette (Molecular weight cutoff – 10 kDa) using a syringe. Dialyze the protein for 60 min in 4 L of cold Phosphate Buffered Saline (PBS) at 4 °C.
Move the dialysis cassette into a new 4 L bucket of cold PBS. Dialyze protein for 16 hr at 4 ^oC.
The next morning, check the protein concentration using a Coomassie protein assay kit, following the manufacturer's instructions. Concentrate the protein using a protein concentrator, following the manufacturer's instructions, if a concentration below 40 μM is observed. Aliquot and freeze at -80 °C for storage.

2. Small Scale Fluorescence Polarization Assay to Characterize the Interaction of the Two Protein Fragments

Thaw 200 μl of 40 μM TRIP8b (241-602) and add it to a 1.5 ml tube. Add 100 μl of PBS to 11 additional 1.5 ml tubes.
Perform serial dilutions by transferring 100 μl of TRIP8b (241-602) from the original tube to the next tube in the series, pipetting up and down, and repeating the process. This will produce a 12-point series of 2-fold dilutions of TRIP8b (241-602) ranging from 0.01-40 μM.
Prepare a 650 μl master mix of 0.1 μM HCN1_FITC (6.5 μl of a 10 μM aliquot) and 2 mM Dithiothreitol (DTT) in PBS.
Note: The master mix is 2x.
Set up 12 new 1.5 ml tubes. Add 50 μl of the master mix containing HCN1_FITCto each tube, and then add 50 μl from each of the 12 serial dilutions.
Note: This will generate 12 tubes, each containing 0.05 μM HCN1_FITC and half of the concentration of TRIP8b (241-602) from the original serial dilution.
Add the dilution series to an assay plate, 30 μl per well, in triplicate. Use a low binding solid black 384-well plate.
Spin the plate for 2 min at 900 x g at RT.
Read the plate using a microplate reader, following the manufacturer's instructions. Measure fluorescence polarization using the following measurement parameters: 485 nm excitation, 530 nm emission, 505 nm dichroic mirror, 100 msec integration time.
Plot the polarization (mP) values versus the TRIP8b (241-602) concentration on a logarithmic scale. Fit the data with the Hill equation to determine the affinity (K_d) of TRIP8b (241-602) for the labeled HCN1 peptide.
Note: For preparing the serial dilutions, a multi-well plate can also be used instead of tubes.

3. Examine the Protein-protein Interaction Using a Positive Control

Add 200 μl of unlabeled HCN1 peptide (200 μM) to a 1.5 ml tube. Add 100 μl of PBS to 11 additional 1.5 ml tubes. Perform serial dilutions by transferring 100 μl from the tube containing 200 μl of peptide to the first tube of 100 μl PBS. Repeat the process to generate 12 tubes with 2 fold dilutions of unlabeled HCN1 peptide.
Prepare a 650 μl master mix of 0.2 μM HCN1_FITCand 4 μM TRIP8b(241-602).
Note: This is a 2x solution.
Add 50 μl of master mix to 12 new 1.5 ml tubes.
Add 50 μl of each serial dilution to a 1.5 ml tube.
Load the black 384-well microtiter plate in triplicate, adding 30 μl per well.
Centrifuge the plate for 2 min at 900 x g at RT.
Read the plate using a microplate reader capable of FP. Use the same settings as described in step 2.7.
Calculate the affinity of the unlabeled peptide for TRIP8b(241-602) using the Cheng-Prusoff equation: K_d¹ = IC₅₀ / (1 + [L]/K_d²), where K_d¹represents the affinity of the unlabeled peptide for TRIP8b(241-602), IC₅₀is determined in the preceding step and represents the ability of the unlabeled ligand to displace the labeled ligand, [L] is the concentration of the labeled ligand in step 3.2, and K_d²is the affinity of TRIPb(241-602) for the labeled ligand.

4. Evaluate Assay Performance (Calculate Z Factor)

Prepare a master mix by making a 12 ml solution of FP buffer with 2 μM TRIP8b (241-602), 50 nM HCN1_FITC, and 1 mM DTT in FP Buffer.
Note: This will require 60 μl of 40 μM TRIP8b (241-602), 60 μl of 10 μM HCN1_FITC, and 1,080 μl of FP Buffer.
Add 30 μl of the master mix to each well of a black 384-well microtiter plate.
Add 1 μl of 1 mM unlabeled HCN1 peptide to 192 wells to serve as a positive control.
Add 1 μl of FP buffer to 192 wells to serve as a negative control.
Centrifuge the plate for 2 min at 900 x g at RT.
Read the plate using a microplate reader.
Calculate the Z factor for the assay using the following formula: Z = 1 – 3*(σ_pos-σ_neg)/(μ_neg-μ_pos) where σ_posand σ_neg are the standard deviations of the positive and negative control wells, and μ_posand μ_negrepresent the average signals in the positive and negative control wells.

5. High Throughput Screen

Thaw aliquots of TRIP8b (241-602) and HCN1_FITC and keep them on ice.
Prepare 10 ml of TRIP8b (2 μM) and HCN1_FITC (50 nM) in FP buffer.
Dispense 25 μl of the mixture into each well of a low binding 384-well black microtiter plate using a pipette.
For library screening, add compounds into each well of columns 3 to 22 (40 µM final concentration for each) using an acoustic liquid handler.
Note: Due to the relatively low hit rate observed with this assay, two different compounds were pooled into a single well. The two compounds from each active well were subsequently tested individually to identify which conferred the activity.
Add 100 nl of Dimethyl sulfoxide (DMSO) into each well of columns 1 and 23 of each plate as negative controls.
Add unlabeled HCN1 peptide into each well of columns 2 and 24 of each plate as positive controls.
Incubate the plates at RT for 2 h. Read plates or incubate at 4 °C for 16 h before reading.
Measure fluorescence polarization on a plate reader. Use the same settings as described in step 2.7.
Calculate percent inhibition by normalizing the signals using the average positive and negative controls from each plate as 100%. Use the equation XN =100% * ((X_neg – X)/(X_neg – X_pos)) where X_N is the normalized percent inhibition corresponding to the signal X, X_pos is the average signal from the positive control wells, and X_neg is the signal from the negative control wells.

6. Confirmation of Hits Using TAMRA-tagged HCN Peptide

Validate hits with over 50% inhibition using TAMRA-labeled HCN1 peptide (HCN1_TAMRA).
1. Prepare a master mix of 12 ml of 2 μM TRIP8b (241-602) and 50 nM HCN1_TAMRA in FP buffer.
2. Dispense 25 μl of the master mix into each well of a black 384-well microtiter plate.
3. Transfer serial dilutions of each active compounds into the respective wells. Make two fold dilutions such that the concentration of each hit ranges from 0.2 µM to 200 µM.
4. For the negative control reactions, transfer 100 nl of DMSO into each well. For the positive control add 100 nl of serially diluted unlabeled HCN1 peptide with final concentration ranging from 200µM to 0.1µM.
5. Incubate the plate at RT for 2 hr. Read plates or incubate for 16 h at 4°C before reading.
6. Measure fluorescence polarization (mP) using the following measurement parameters: 535 nm Excitation; 580 nm emission; 535 nm dichroic mirror, 20 msec integration time.
7. Calculate IC₅₀ values by fitting the concentration dependent FP data of each compound using a four-parameter nonlinear regression model²³.

7. Dose-response Validation of Bead-based Proximity Assay

Thaw 1 aliquot of His-tagged TRIP8b (241-602) (His-TRIP8b) and GST-tagged HCN1 (GST- HCN1_C40) fusion proteins and keep them on ice.
Combine His-TRIP8b with GST-HCN1_C40 in assay buffer at concentrations of 400 nM His-TRIP8b and 40 nM GST HCN1_C40. Prepare 300 µl of the mixture for each compound to be tested.
Note: Compounds are run in triplicate at 11 different concentrations plus a DMSO control (no compound).
For control wells, prepare 30 µl of His-TRIP8b (200 nM) without GST-HCN1C40 and then separately prepare 30 µl of GST-HCN1C40 (20 nM) without His-TRIP8b in assay buffer.
For each compound to be tested, add 7 μl of the His-TRIP8b:GST-HCN1_C40 mixture (prepared in step 7.2 above) to 36 wells of a plate using a 16-channel pipette. For testing a single compound, arrange the 12 different concentrations in rows A-L and the three replicates in columns 1-3. Add 7 µl of the His-TRIP8b solution (prepared in 7.3) to columns 1-3 of row M and 7 µl of GST-HCN1_C40 solution (prepared in 7.3) to columns 1-3 of row N. Briefly centrifuge the plate to ensure the liquids are at the bottom of each well and to remove any bubbles.
Dispense the hit compounds to be tested into the plate such that the concentration of each hit ranges from 200 µM (in row A) down to 0.1 µM (in row K). Dispense a volume of DMSO in rows L-N that matches the amount of compound dispensed in rows A-K.
Shake the plate for 2 min and incubate for 2.5 h at RT.
Dilute the anti-GST Acceptor beads 1:50 with assay buffer.
Add 3.5 µl of the diluted Acceptor beads to each well of the plate with a 16-channel pipette and mix by gently pipetting to avoid creating bubbles.
Incubate the plate for 1 hr at RT in the dark.
Dilute the Nickel Chelate Donor beads 1:50 with assay buffer. Do not expose the mixture to light.
Add 3.5 µl of the diluted Donor beads to each well of the plate with a 16-channel pipette and mix by gently pipetting to avoid creating bubbles.
Incubate the plate for 1.5 hr at RT in the dark.
Read on a plate reader. Use a plate reader with the following measurement parameters: 40 msec excitation time, 100 msec emission time, 0.1 mm detection height.
Calculate IC₅₀ values by fitting data for each compound using a four-parameter nonlinear regression model²³.

Representative Results

To avoid copyright issues with our previous publication, a TAMRA-tagged probe HCN1_TAMRA was used to generate Figures 2 and 3. Note that this substitution did not make an appreciable difference in the results, and the protocols are identical to those outlined above with HCN1_FITC. To assess the interaction with HCN1_TAMRA, TRIP8b (241-602) was titrated into a fixed concentration of HCN1_TAMRAusing the protocol outlined in step 2 (Figure 2). Next, the experiment outlined in step 3 was performed and unlabeled HCN1 peptide was titrated into a fixed concentration of TRIP8b (241-602) and HCN1_TAMRA (Figure 3).

To verify that the assay was suitable for high throughput screening, its performance was next examined by the protocol in step 4; and a Z factor of 0.89 was obtained, indicating that the assay was ready for high throughput screening (Figure 4). Then, a 20,000 compound small molecule library was screened by the procedure outlined in step 5. All compounds that have percentage inhibition above 50% were then tested in the second FP assay with HCN1_TAMRA (step 6). Finally, the confirmed hits were tested in the bead-based proximity assay (step 7, Figure 5). One hit compound, NUCC-5953, was identified as a result of these experiments.

Figure 1: HTS Workflow. Schematic describing the workflow for high throughput screening. Diagram proceeds from the top to bottom, with each triangle representing a step in the protocol. Please click here to view a larger version of this figure.

Figure 2: K_d of Protein-protein Interaction. (A) Schematic showing the experiment described in step 2. As TRIP8b (241-602) is titrated into a fixed concentration of HCN1_TAMRA, the TPR domains of TRIP8b (in gray) bind to the 11 amino acid peptide (in black, with terminal 'SNL' highlighted). (B) As the concentration of TRIP8b (241-602) increases, more HCN1_TAMRAmolecules become bound and polarization increases (K_D= 0.320.01μM). Error bars denote standard deviation. Please click here to view a larger version of this figure.

Figure 3: IC₅₀ of Positive Control. (A) Schematic demonstrating experimental paradigm for step 3. As unlabeled HCN1 peptide is titrated into a fixed concentration of TRIP8b (241-602) and HCN1_TAMRA, the labeled peptide is displaced. (B) Note that the signal decreases as the concentration of unlabeled HCN1 increases, indicating displacement of HCN1_TAMRA(IC₅₀= 1.070.08μM) . Error bars denote standard deviation. Please click here to view a larger version of this figure.

Figure 4: Representative Results from HTS. (A) Results from the high throughput screen performed using the protocol outlined in step 5. Each point on the graph represents a single compound. The X and Y coordinates are determined by the percent inhibition in each run. (B) Results from the screen plotted with positive and negative controls (see legend). Each compound's average percent inhibition (across two runs of the assay) is plotted on the Y axis. (C) Results from confirmatory FP experiments using a TAMRA-labeled HCN1 peptide. Compounds showing greater than 50% inhibition in step 5 were then used in step 6. As in (A), the X and Y coordinates are determined by the percent inhibition in two runs of the assay. Reproduced from Han et al.³. Please click here to view a larger version of this figure.

Figure 5: Bead-based Proximity Assay. (A) Schematic showing the bead-based proximity assay. TRIP8b (241-602) contains an N terminal hexahistidine tag that binds Nickel Chelate donor beads (circle with stripes). HCN1_C40contains an N terminal GST tag that binds acceptor beads. When brought into proximity by the interaction of the TPR domains of TRIP8b and the C terminal tripeptide of HCN1, excitation of the donor bead by 680 nm wavelength light produces singlet oxygen. This singlet oxygen transfers energy to acceptor beads within a defined radius and leads to the emission of light. (B) Titration of increasing concentrations of the hit compound (NUCC-5953) into a fixed concentration of TRIP8b (241-602) and HCN1_C40. Reproduced from Han et al.³. Error bars denote standard deviation. Please click here to view a larger version of this figure.

Discussion

Because of its potential as a therapeutic target in MDD²⁴, there has been considerable interest in pharmacological approaches that antagonize HCN channel function in the central nervous system⁴. However, these efforts have been stalled by the important role of HCN channels in cardiac pacemaking and the risk of arrhythmia²⁵. We reasoned that disrupting the interaction between HCN and its brain specific auxiliary subunit, TRIP8b⁸, might be sufficient to produce antidepressant-like effects without affecting cardiac HCN channels³. This hypothesis was bolstered by the observation that mice lacking TRIP8b exhibit antidepressant-like behavior⁷. Targeting this interaction could become a new paradigm for the treatment of MDD.

A detailed protocol for identifying small molecule inhibitors of the interaction between the TPR domains of TRIP8b and the C terminal tripeptide of HCN1 is presented above. To facilitate its general application, we here describe our reasoning that led to the development of this assay. Although TRIP8b binds to HCN subunits at two locations, we focused on the interaction between the TPR domains of TRIP8b and the C terminal tail of HCN. The structural elucidation of this interaction by x-ray crystallography revealed a deep pocket formed by the TPR domains of TRIP8b around the C terminal of HCN¹⁵. The second TRIP8b-HCN interaction occurs between an 80 amino acid stretch of TRIP8b (located N terminal to the TPR domains) and the cyclic nucleotide binding domain of HCN. This interaction occurs across many different amino acids of each protein and constitutes a diffuse surface with fewer well defined intermolecular interactions²⁶. Based on these structural observations, we reasoned that the interaction between the TPR domains of TRIP8b and the C terminal of HCN is more susceptible to disruption by a small molecule inhibitor.

Initially, a larger fragment of TRIP8b was used in the assay based on the assumption that a longer construct may have allosteric regulatory sites and increase the chance for success. Full length TRIP8b was initially used, but this approach was limited by protein aggregation, degradation, and sensitivity to freeze-thaw cycles. Subsequently, two intermediately sized TRIP8b constructs, a longer one (residues 219-602) and a shorter one (residues 259-602), were tested before an intermediate construct (residues 241-602) was selected. Although each of the four truncations described above contain the TPR domains relevant for binding to the C terminal of HCN, only the intermediate length clone (241-602) was sufficiently stable for use in screening assays. In particular, we were able to obtain large quantities of the protein after purification by Ni²⁺ affinity chromatography without additional separation steps.

After choosing a suitable fragment of TRIP8b and HCN, we next determined the affinity of the labeled HCN1 peptide for TRIP8b(241-602). As a general rule, an accurate measurement can only be made if the concentration of the ligand is substantially below the K_D of the binding partner²⁸. In our case, we used 50 nM of labeled HCN1 peptide for the experiment in step 2, and obtained a K_D of 0.320.01μM. For more experimental design considerations regarding protein-protein interactions, readers are referred to one of several excellent reviews on the topic^27-30.

Once we determined the K_D of the labeled HCN1 peptide for TRIP8b(241-602), we then determined if the label itself interferes with binding. In step 3, we obtained a IC₅₀ value of 1.070.08 μM by titrating an unlabeled HCN1 peptide into a fixed concentration of TRIP8b(241-602) to displace the labeled peptide. Combined with the K_D of the labeled peptide for TRIP8b(241-602), and applying the Cheng-Prusoff equation, we then estimated the affinity of the unlabeled peptide for TRIP8b(241-602) as K_D= 0.93μM. This is in close agreement with the affinity of TRIP8b(241-602) for the labeled peptide, and suggests that labeling the peptide did not substantially affect its affinity to TRIP8b. An important feature of the fluorescence polarization-based screen is its excellent signal to noise ratio as indicated by a Z factor of 0.89. With other parameters the same, the magnitude of the change in FP signal is dictated by the size of the binding partner (TRIP8b (241-602)) and the size of the fluorophore-labeled ligand. A change in polarization (from 44 to 224 mP) observed with the 11 amino acid peptide ligand between its free and bound states with the ~42 kDa TRIP8b (241-602) protein is sufficient to reproduce the targeted interaction and provides a sufficient dynamic range for library screening.

One of the challenges of any high throughput screening assay is distinguishing true 'hit' compounds from artifactual changes in signal intensity. In fluorescence polarization-based screens, it is common for many compounds to either interact directly with the fluorophore or fluoresce on their own. To circumvent these issues, a two-stage screening procedure using two distinct fluorophores is described above. This procedure reduces the likelihood that fluorescent compounds and compounds interacting directly with the fluorophore will advance through the screening process. In addition to binding the fluorophore, some compounds act as 'aggregators' in vitro and lead to nonspecific changes in the fluorescence polarization signal. These compounds are thought to form detergent-sensitive micellar structures and inhibit protein-protein interactions^31,32. To mitigate these effects, it is important that the fluorescence polarization buffer used in the high throughput screening protocol outlined above includes a detergent (Triton), although additional validation with other detergents should be considered.

It should be noted that there are several important limitations of the screening procedure described above. Although TRIP8b binds to HCN in two locations, the screen described above examines only one interaction site and will not identify small molecules targeting the CNBD interaction. Similarly, compounds that allosterically modulate TRIP8b binding to HCN by interacting with TRIP8b in the N terminal region will not be identified as hits. Both of these considerations are the result of using a smaller TRIP8b fragment (see above), which ensures reproducibility in the screening assays outlined in the procedure. In order to avoid these limitations, future efforts may be directed at using a cell based screen incorporating full length HCN and TRIP8b constructs^33,34. This sort of screen may rely on high throughput electrophysiological methods in order to identify compounds capable of disrupting the interaction between HCN and TRIP8b, and limiting the expression of HCN channels at the cell surface. Of note, approaches such as these would need to include counter-screens to ensure that small molecules were not directly limiting HCN channel function as antagonists, as this could lead to off target effects in vivo.

Although HCN1, HCN2, and HCN4 all have a conserved 'SNL' C terminal peptide, the residues N terminal to this tripeptide vary substantially and likely affect the TRIP8b binding affinity. This raises the possibility that a small molecule hit obtained by screen or designed based on other hit compounds may provide HCN isoform specificity in disrupting the TRIP8b-HCN interaction. In the original screen, NUCC-5953 was identified as the first small molecule capable of disrupting the interaction between TRIP8b and HCN1³. Future work with this assay may identify additional small molecule inhibitors with desirable chemical properties for drug development.

Divulgaciones

The authors have nothing to disclose.

Acknowledgements

This work was supported by National Institutes of Health Grant R21MH104471 and R01MH106511 (D.M.C.), Brain Research Foundation SG 2012-01 (D.M.C.), Northwestern University Clinical and Translational Sciences Institute 8UL1TR000150 (Y.H.), Chicago Biomedical Consortium HTS-004 (Y.H. and D.M.C.), and National Institutes of Health Grant 2T32MH067564 (K.L.). A part of this work was performed by the Northwestern University Medicinal and Synthetic Chemistry Core (ChemCore) at the Center for Molecular Innovation and Drug Discovery (CMIDD), which is funded by the Chicago Biomedical Consortium with support from the Searle Funds at the Chicago Community Trust and Cancer Center Support Grant P30 CA060553 from the National Cancer Institute awarded to the Robert H. Lurie Comprehensive Cancer Center. The high throughput screen work was performed in the High Throughput Analysis Laboratory which is also a core facility of the Robert H. Lurie Comprehensive Cancer Center.

Materials

Ni-NTA agarose	Qiagen	30210
Dialysis cassette	ThermoFisher	66456
Isopropyl b-D-1-thiogalactopyranoside	Sigma-Aldrich	I5502-1G
384 Well Black plate	Corning	3820
Proxiplate	Perkin-Elmer	6008289
Anti-GST Acceptor beads	Perkin-Elmer	6760603C
NiChelate	Perkin-Elmer	AS101D
pGS21-a	Genscript	SD0121
PMSF	Sigma-Aldrich	10837091001
Coomassie Kit	ThermoFisher	23200
Protein concentrator	ThermoFisher	88527
Perkin Elmer Enspire Multimode Plate reader	Perkin-Elmer	#2300-001M
BL21 (DE3) Competent Cells	Agilent	200131

Referencias

Biel, M., Wahl-Schott, C., Michalakis, S., Zong, X. Hyperpolarization-activated cation channels: from genes to function. Physiol Rev. 89 (3), 847-885 (2009).
Shah, M. M. HCN1 channels: a new therapeutic target for depressive disorders?. Sci Signal. 5 (244), pe44 (2012).
Han, Y., et al. Identification of Small-Molecule Inhibitors of Hyperpolarization-Activated Cyclic Nucleotide-Gated Channels. J Biomol Screen. 20 (9), 1124-1131 (2015).
Postea, O., Biel, M. Exploring HCN channels as novel drug targets. Nat Rev Drug Discov. 10 (12), (2011).
Stieber, J., Wieland, K., Stöckl, G., Ludwig, A., Hofmann, F. Bradycardic and proarrhythmic properties of sinus node inhibitors. Mol Pharmacol. 69 (4), 1328-1337 (2006).
Santoro, B., Wainger, B. J., Siegelbaum, S. A. Regulation of HCN channel surface expression by a novel C-terminal protein-protein interaction. J Neurosci. 24 (47), 10750-10762 (2004).
Lewis, A. S., et al. Deletion of the hyperpolarization-activated cyclic nucleotide-gated channel auxiliary subunit TRIP8b impairs hippocampal Ih localization and function and promotes antidepressant behavior in mice. J Neurosci. 31 (20), 7424-7440 (2011).
Heuermann, R. J., et al. Reduction of thalamic and cortical Ih by deletion of TRIP8b produces a mouse model of human absence epilepsy. Neurobiol Dis. 85, 81-92 (2016).
Steru, L., Chermat, R., Thierry, B., Simon, P. The tail suspension test: A new method for screening antidepressants in mice. Psychopharmacology. 85 (3), 367-370 (1985).
Petit-Demouliere, B., Chenu, F., Bourin, M. Forced swimming test in mice: a review of antidepressant activity. Psychopharmacology. 177 (3), 245-255 (2004).
Cryan, J. F., Mombereau, C., Vassout, A. The tail suspension test as a model for assessing antidepressant activity: Review of pharmacological and genetic studies in mice. Neurosci Biobehav Rev. 29 (4-5), 571-625 (2005).
Han, Y., et al. Trafficking and gating of hyperpolarization-activated cyclic nucleotide-gated channels are regulated by interaction with tetratricopeptide repeat-containing Rab8b-interacting protein (TRIP8b) and cyclic AMP at distinct sites. J Biol Chem. 286 (23), 20823-20834 (2011).
Hu, L., Santoro, B., Saponaro, A., Liu, H., Moroni, A., Siegelbaum, S. Binding of the auxiliary subunit TRIP8b to HCN channels shifts the mode of action of cAMP. J Gen Physiol. 142 (6), 599-612 (2013).
Saponaro, A., et al. Structural basis for the mutual antagonism of cAMP and TRIP8b in regulating HCN channel function. Proc Natl Acad Sci U S A. 111 (40), 14577-14582 (2014).
Bankston, J. R., Camp, S. S., DiMaio, F., Lewis, A. S., Chetkovich, D. M., Zagotta, W. N. Structure and stoichiometry of an accessory subunit TRIP8b interaction with hyperpolarization-activated cyclic nucleotide-gated channels. Proc Natl Acad Sci U S A. 109 (20), 7899-7904 (2012).
Gatto, G. J., Geisbrecht, B. V., Gould, S. J., Berg, J. M. Peroxisomal targeting signal-1 recognition by the TPR domains of human PEX5. Nat Struct Biol. 7 (12), 1091-1095 (2000).
Owicki, J. C. Fluorescence polarization and anisotropy in high throughput screening: perspectives and primer. J Biomol Screen. 5 (5), 297-306 (2000).
Smith, D. S., Eremin, S. A. Fluorescence polarization immunoassays and related methods for simple, high-throughput screening of small molecules. Anal Bioanal Chem. 391 (5), 1499-1507 (2008).
Roehrl, M. H. A., Wang, J. Y., Wagner, G. A general framework for development and data analysis of competitive high-throughput screens for small-molecule inhibitors of protein-protein interactions by fluorescence polarization. Bioquímica. 43 (51), 16056-16066 (2004).
Arkin, M. R., Wells, J. A. Small-molecule inhibitors of protein-protein interactions: progressing towards the dream. Nat Rev Drug Discov. 3 (4), 301-317 (2004).
Zhang, J., Chung, T., Oldenburg, K. A Simple Statistical Parameter for Use in Evaluation and Validation of High Throughput Screening Assays. J Biomol Screen. 4 (2), 67-73 (1999).
Eglen, R. M., et al. The use of AlphaScreen technology in HTS: current status. Curr Cheml Genomics. 1 (1), 2-10 (2008).
Lehninger, A. L., Nelson, D. L., Cox, M. M. . Lehninger Principles of Biochemistry. , (2000).
Kim, C. S., Chang, P. Y., Johnston, D. Enhancement of dorsal hippocampal activity by knockdown of HCN1 channels leads to anxiolytic- and antidepressant-like behaviors. Neuron. 75 (3), 503-516 (2012).
Wahl-Schott, C., Biel, M. HCN channels: structure, cellular regulation and physiological function. Cell Mol Life Sci. 66 (3), 470-494 (2009).
DeBerg, H. A., Bankston, J. R., Rosenbaum, J. C., Brzovic, P. S., Zagotta, W. N., Stoll, S. Structural Mechanism for the Regulation of HCN Ion Channels by the Accessory Protein TRIP8b. Structure. 23 (4), 734-744 (2015).
Weiss, J. N. The Hill equation revisited: uses and misuses. FASEB J. 11 (11), 835-841 (1997).
Prinz, H. Hill coefficients, dose-response curves and allosteric mechanisms. J Chem Biol. 3 (1), 37-44 (2009).
Pollard, T. D. A Guide to Simple and Informative Binding Assays. Mol Biol Cell. 21 (23), 4061-4067 (2010).
Rossi, A. M., Taylor, C. W. Analysis of protein-ligand interactions by fluorescence polarization. Nat Protoc. 6 (3), (2011).
Ryan, A. J., Gray, N. M., Lowe, P. N., Chung, C. -. W. Effect of Detergent on “Promiscuous” Inhibitors. J Med Chem. 46 (16), 3448-3451 (2003).
McGovern, S. L., Caselli, E., Grigorieff, N., Shoichet, B. K. A common mechanism underlying promiscuous inhibitors from virtual and high-throughput screening. J Med Chem. 45 (8), 1712-1722 (2002).
Hertzberg, R. P., Pope, A. J. High-throughput screening: new technology for the 21st century. Curr Opin Chem Biol. 4 (4), 445-451 (2000).
Sundberg, S. A. High-throughput and ultra-high-throughput screening: solution-and cell-based approaches. Curr Opin Biotechnol. 11 (1), (2000).

Play Video

PDF

DOI

DOWNLOAD MATERIALS LIST

Citar este artículo

Han, Y., Lyman, K. A., Clutter, M., Schiltz, G. E., Ismail, Q., Cheng, X., Luan, C., Chetkovich, D. M. Method for Identifying Small Molecule Inhibitors of the Protein-protein Interaction Between HCN1 and TRIP8b. J. Vis. Exp. (117), e54540, doi:10.3791/54540 (2016).