NMR-Based Fragment Screening in a Minimum Sample but Maximum Automation Mode

We conduct nuclear magnetic resonance studies. This allows us to determine the structures of proteins, RNA, and DNA in solution. These biomolecules are the constituents, are the components of our cells, and our protocols that we present here is concerned with the screening of small molecules and their binding to these targets.

From these molecules, our medicinal chemistry and structural biology can derive new drugs that combat diseases. Main advantage of our technique is that we have full control over the quality of the molecular targets, like the proteins, and we have full control over the quality of the small molecule fragments and drugs, and we can also screen proteins for binding of a wide range of affinities, which is an advantage compared to other methodologies. To prepare screen samples for NMR measurement, use a sample preparation robot to distribute 768 compounds into eight 96-well plates to obtain 64 mixtures containing 12 fragments to a final concentration of 4.2 millimolar per mixture.

Transfer the biomolecular target of interest, diluted in an appropriate screening buffer, into the appropriate number of barcoded three-millimeter NMR high-throughput sample changer tubes, and use the robot to transfer 10 microliters of each ligand mixture into each tube of target biomolecule. Then, have the robot mix the solutions thoroughly. For NMR acquisition, load the sample into the spectrometer.

Open the spectrometer software, then, select the parameter set and pulse sequences for ligand-based experiments. For all the listed experiments, select excitation sculpting as the water suppression. For fluorine-19 screening, select both 1D and T2 experiments.

Select the spectral width to 220 parts per million, the excitation frequency to 140 parts per million, and the analysis time between one to five hours. For T2, the CPMG time should alternate between 5 and 100 milliseconds. Record the T1r and T2 experiments.

Record the saturation transfer difference as a pseudo 2D. To process the two single 1D spectra, use the AU program ProcStd function, with or without the relax option. The WaterLOGSY is a single 1D that should be phased with a negative for the solvent signal.

To analyze the hydrogen screening data, follow the instructions for the fragment-based screening tool to store the biomolecular magnetic resonance NMR data from screening campaigns, such that each screening mixture has its directory in which a subdirectory contains the different experiments measured on the sample. Store the reference spectra, having all the data saved from the samples without the biomolecular target, but with the mixtures and the single compound in different directories. Create a direct path to the directory containing the acquired data, and select the NMR directory in which all the mixtures should have a distinct directory.

Make sure the CSV, fragment screen, XML document, and BAC files are also copied into the NMR directory of the data. To use the fragment-based screening tool for a screened sample, drag the fragment-based screening project symbol into the middle of the analysis window. The symbol should appear if the previously saved datasets were copied into it.

The fragment-based screening options window should automatically open. Select a CSV cocktail file containing the names of the mixes, the names of each fragment, and the division of each fragment into the mixes. Define a reference ligand spectra folder with all the measured spectra of the single fragments, and define a reference blank experiment folder, which usually contains the datasets of the mixes without the investigated target.

To define the investigated spectra and spectra display colors, open the Spectra Types tab and set the spec type according to the process data. In the Display Layout tab, define the spectra that will be compared according to the spec types. Then, click OK to start the project.

While the data is being processed, a separate window with the table summarizing all the cocktail mixes and the ligands of each mix will open. Double-click on a cell to open the respective datasets. Before assigning binders, ensure that the reference peaks match each other and have the same chemical shift.

If differences are observed, open the Process tab, and use the serial processing option to correct them. For the first analysis on the fluorine-19 mixtures, open the Analyze tab and select the integrate function. Make sure that a clear integration region for the corresponding fluorine-19 signal is defined for each fragment in the mixture.

Save export integration regions to export the integration file for future use, and save any used integration files in the appropriate installation directory. For fluorine-19 data, open a dataset with or without the investigated target, and under the Analyze tab, click Integrate and Read Import Integration Regions to load the corresponding integration file into the current spectrum. Then, click Save and return to locate a list of the integrated regions in the Integrals tab, and copy this information into a spreadsheet for downstream analysis.

The molecular structure or constitution of the ligands in the fragment library is analyzed using molecular confidence software as demonstrated, and the results are presented as a graphical output. An orange color indicates that the fragment exhibits an inconsistency in structure or concentration. Green-colored wells indicate that the fragment is consistent.

In this representative analysis, 103 fragments containing one or several fluorine groups from the in-house library were divided into five mixes of 20 to 21 fragments per mixture. Fluorine-19 transverse relaxation experiments were measured for each mixture that applied CPMG pulse trains. Thresholds are useful for defining binder, weak binder, or non-binder in the mixture, as observed in these typical thiamine protein tyrosine kinase A and RNA spectra.

In this representative analysis, hit one showed a T2 decrease of about 50%and a chemical shift greater than or equal to six hertz. The WaterLOGSY did not exhibit a significant change in signal to be counted as positive. However, as two out of three experiments were positive, this fragment was counted as a hit.

For hit two, the T2 showed a decrease of approximately 80%in signal intensity. A clear signal change was also observed for the WaterLOGSY. The chemical shift was not enough in this experiment, but as the two previous criteria were positive, it was still counted as a hit.

The final goal of such methods is to develop new drugs or high-affinity inhibitors of, for example, enzymes, and in this, medicinal chemistry takes over and synthesize these new molecules, but structure biology, the techniques you apply here can guide medicinal chemistry and make that approach way faster.