Key words

1 Introduction

Spatial and temporal regulation of mitogen activated protein kinases (MAPKs) is of essential importance for transducing environmental and developmental signals (growth factors, stress) into adaptive and programmed responses (differentiation, inflammation, apoptosis) [1, 2]. Although MAPKs are ubiquitously expressed, their activation is finely tuned in a cell-type specific and temporal manner [3]. The regulators that achieve this fine-tuning include: (1) upstream kinases, (2) downstream phosphatases, and (3) scaffolding proteins [4–6]. Over the years, many groups using a variety of techniques have advanced the field’s understanding of how the activities of MAPKs are finely controlled by MAPK regulatory proteins. Clearly, understanding how these proteins interact at a molecular level is also essential [7–15]. X-ray crystallography has provided key insights into the regulation of MAPKs by their interacting proteins by mostly determining structures of MAPKs bound to different peptides derived from interacting proteins [7]; e.g., for the MAPK p38α a single structure of a full-length interacting protein (MK2—a substrate that binds with a K d of 20 nM) has been determined [16]. Over the last ~10 years, biomolecular NMR spectroscopy has complemented and, more critically, expanded these efforts [17–22]. By doing so, it has become apparent that regulatory proteins adopt distinct conformations when bound to MAPKs and have different degrees of flexibility upon complex formation, which contributes to the observed differences in their biological functions. Here we review the use of biomolecular NMR spectroscopy and other biophysical techniques to study complexes between the MAPK p38α and its regulatory proteins (Table 1).

Table 1 NMR and SAXS studies of p38α with its regulatory proteins
Full size table

2 Materials

Prepare all solutions using ultrapure Water (Milli-Q water purification system, Millipore). Chemicals should be at least ACS grade. Prepare and store reagents/solutions at temperatures and conditions recommended by the manufacturer. Chemicals specific for NMR labeling, e.g., [13C]d-glucose, [13C]d-d7-glucose (1,2,3,4,5,6,6-d7), 15NH4Cl, D2O are typically purchased from Cambridge Isotope Laboratories (CIL) or Isotech (Sigma). Standard buffers are used for purification; when necessary, buffers are autoclaved when prepared in order to prevent unwanted proteolytic degradation. All buffers are stored at 4 °C and are filtered (0.22 μm PES filter, Millipore) prior to use. If possible uniform buffers should be used throughout all experiments. Furthermore, it is important to reduce the salt concentration and select the right buffer for NMR experiments as the sensitivity of cryoprobe significantly decreases with increasing amounts of salt (thus it is also better to use HEPES buffer than phosphate buffer) [23]. Buffers for ITC experiments must be degassed. NMR data analysis can be performed by a variety of software, including Cara (http://cara.nmr.ch), NMRVIEW [24], Sparky [25], and CCPN [26, 27].

  1. 1.

    M9 medium: Weigh 5.8 g Na2HPO4 anhydrous, 3.0 g KH2PO4, and 0.5 g NaCl. Add water to a volume of 900 mL. Mix and adjust pH to 7.1–7.3 with HCl. Volume up to 1 L with water. Sterilize by filtration and store at 4 °C.

  2. 2.

    Vitamin mix: BME mix (Sigma-Aldrich). Use directly for H2O-based medium. Lyophilize and resuspend in D2O for D2O-based medium.

  3. 3.

    Solution Q: In a final volume of 1 L of water, add 8 mL 5 M HCl (see Note 1 ), 5 g FeCl2·4H2O, 184 mg CaCl2·2H2O, 64 mg H3BO3, 18 mg CoCl2·6H2O, 4 mg CuCl2·2H2O, 340 mg ZnCl2, and 40 mg Na2MoO4·2H2O. Mix thoroughly. Aliquot into 2 mL Eppendorf tubes, freeze and store at −20 °C.

  4. 4.

    Glucose: Use [12C]d-glucose for unlabeled protein expression; [13C]d-glucose for 13C labeled protein expression; [13C]d-d7-glucose (1,2,3,4,5,6,6-d7) for fully deuterated protein expression.

  5. 5.

    Ammonium chloride: Use NH4Cl for unlabeled protein expression and 15NH4Cl for NMR-active protein expression.

  6. 6.

    Preparation of medium: All plastic ware must be autoclaved before use; for large-scale expression we prefer Ultra-Yield 2 L expression flasks (six-times baffled for high aeration; Thomson Instrument Company); 250 mL of media for each Ultra-Yield 2 L expression flask [28].

  7. 7.

    NMR tubes: Different NMR tubes are available for measurements which differ both in diameter (1, 1.7, 3, and 5 mm [outdated: 8 and 10 mm]; 5 mm is the size most commonly used) and shape (distinct shapes can partially compensate the reduced sensitivity of high salt samples in cryo-probes). Shigemi has developed a set of tubes that use solvent-susceptibility matched glass and thus allows for lower sample volumes (~40 % reduction ), with similar tubes are also available from New Era Scientific.

3 Methods

3.1 Production of p38α and Regulatory Protein Tyrosine Phosphatases for NMR Studies

p38α and its regulatory proteins are large proteins (≥35 kDa) for biomolecular NMR spectroscopy [29, 30]. In order to analyze proteins using NMR spectroscopy, they must be uniformly labeled with NMR-active nuclei. NMR-active nuclei have a nuclear spin, preferably with a spin number of ½ (1H, 13C and 15N nuclei have spin numbers of ½). Labeling is achieved by providing labeled precursors (e.g., [13C]-d-glucose, 15NH4Cl) during protein expression in a variety of expression systems, including Escherichia coli (E. coli) [28], yeast [31], insect cell [32], and human expression systems [33].

However, in order to overcome the broad line-widths that are characteristic of proteins ≥35 kDa, MAPKs must be expressed in D2O-based medium and TROSY [34] versions of all 2D and 3D experiments must be recorded using a high-field NMR spectrometer (800–1000 MHz 1H Larmor frequency) [35]. Deuteration reduces the dipole–dipole 1H–1H relaxation, as deuterium has a ~7-fold lower gyromagnetic ratio than protons [36]. However, as most NMR experiments rely on the excitation and detection of 1H (largest gyromagnetic ratio; 99.9 % natural abundance), deuteration limits the available experimental approaches for such proteins. Most 1HN will rapidly back-exchange during protein purification in H2O-based buffers. However, H/D exchange can be slow in β-sheets and in hydrophobic pockets where access of water or strong hydrogen bonds can significantly limit the exchange rates. If possible, protein refolding either by chemicals or pressure in H2O-based buffers can be helpful; however, refolding of larger signaling enzymes, such as p38α or tyrosine phosphatases, is not readily achievable. For many years expression of deuterated proteins was limited to the E. coli expression system, but recently it has become economically feasible to express 2H-labeled proteins in insect cells [37]. To achieve the necessary high expression yields, E. coli cells must be adapted for growth in D2O-based medium.

3.1.1 Fresh Transformation and Colony Expression Screening

  1. 1.

    Transform the expression plasmid into the appropriate E. coli expression cell line (see Notes 2 and 3 ).

  2. 2.

    Screen multiple (three to five) colonies for robust expression using a small scale growth and induction (i.e., β-d-1-thiogalactopyranoside [IPTG]) in Luria broth (LB) medium (3–5 mL). Allow expression to proceed for 2 h at 37 °C and evaluate the expression levels of the different colonies using sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and Coomassie staining.

3.1.2 D2O Adaption (see Table 2 and Fig. 1)

Table 2 D2O adaption
Full size table
Fig. 1
figure 1

Schematic of the D2O adaption protocol. See text for details

Full size image
  1. 1.

    20 μL of uninduced LB growth is transferred into 3 mL of freshly prepared 0 % D2O medium and grown for ~12–15 h (37 °C; 250 rpm shaking).

  2. 2.

    50 μL of the 0 % D2O culture is added to 3 mL of 30 % D2O and grown for ~12–15 h (37 °C; 250 rpm shaking).

  3. 3.

    50 μL of the 30 % D2O culture is added to 3 mL of 50 % D2O and grown for ~12–15 h (37 °C; 250 rpm shaking).

  4. 4.

    100 μL of the 50 % D2O culture is added to 3 mL of 70 % D2O and grown for ~12–15 h (37 °C; 250 rpm shaking).

  5. 5.

    500–2000 μL (depending on protein) of the 70 % D2O culture is added to 100 mL of 100 % D2O medium in a 500 mL Ultra-Yield expression flask (see Notes 4 and 5 ) and grown for ~12–15 h (37 °C; 250 rpm shaking).

3.1.3 D2O Large Scale Expression

25 mL of the 100 % D2O culture is added to 225 mL of 100 % D2O medium in a 2 L Ultra-Yield flask and the cells grown to OD600 of ~0.6–0.8 (37 °C; 250 rpm shaking with the appropriate antibiotics). Cells are induced using the appropriate induction agent (i.e., IPTG) and then expressed at appropriate temperature (e.g., 18 °C; 18–20 h, 250 rpm shaking for p38α [see Note 6 ]). Large scale expression of [2H,15N]-p38α is done in M9 medium supplemented with 15NH4Cl (1 g per 1 L medium) in 99.8 % D2O, and [2H,13C,15N]-p38α is expressed in M9 medium supplemented with 15NH4Cl (1 g per 1 L medium) and [13C]d-d7-glucose (4 g per 1 L medium) in 99.8 % D2O.

3.1.4 D2O Recycling

D2O can be readily recycled and reused with a slightly (~10–15 %) lower degree of deuteration, which is done in many laboratories.

  1. 1.

    Following the pelleting of the cell mass after the large-scale D2O expression, the supernatant is collected and stored (RT or −20 °C; see Note 7 ).

  2. 2.

    Used D2O is distilled in a standard glass apparatus, with a controlled heating mantle and a 4 L distillation flask with two inlets (see Note 8 ). Aluminum foil is used to lower the boiling temperature. Regular boiling stones are used to ensure to ensure uniform boiling.

  3. 3.

    During distillation, the distillation flask will accumulate debris. Used D2O is added to the flask until 20–40 L has been distilled. At this point the flask is either cleaned or replaced.

  4. 4.

    Distilled D2O is perfectly clear; however, it has nasty ammonia-like smell. To make the distilled D2O suitable for culturing bacteria (E. coli will not grow on the freshly distilled, but nasty smelling D2O), the distilled D2O must be treated with ~20 g of acid-washed activated charcoal (Sigma) per 1 L of distilled D2O under vigorous stirring for 12–24 h. The activated carbon is removed by filtering through a Buchner filter.

  5. 5.

    If necessary, the recycled D2O can be filtered through a 2 μm bottle-top filter to remove any remaining debris/charcoal. At this point, the D2O is completely clear and does not have any detectable odor. The recycled D2O can then be used identically to fresh D2O. Mass spectrometry will confirm a lower rate of deuteration (~85 %) (see Note 9 ).

3.2 Interaction with Kinase Interaction Motif Protein Tyrosine Phosphatases (KIM-PTPs) and MAPK Protein Tyrosine Phosphatases

Rapid assessment of a protein–protein interaction can be achieved using size exclusion chromatography (SEC) by simply comparing the retention volume (or time) of the free and complex samples. SEC media varies widely in quality and size, leading to different numbers of theoretical plates and thus resolution. For protein studies, Superdex or Sephacryl (GE Healthcare) are the most commonly used media due to their high inertness to nonspecific interactions and their excellent separation capabilities. An earlier eluting peak indicates a higher molecular weight complex and can be readily compared with the SEC of the individual components (proteins). The best readout is SDS-PAGE, where both proteins should be present throughout the peak with similar intensities. In order to measure a full thermodynamic profile of a protein–protein interaction, isothermal titration calorimetry is the best method. Its drawback is the need for rather high amounts of protein. Malvern (VP-ITC and ITC200) and TA Instruments (nano-ITC) are the main manufacturers of these instruments.

  1. 1.

    All ITC experiments should be performed at the same temperature.

  2. 2.

    All proteins and/or peptides must be in the identical buffer (see Note 10 ).

  3. 3.

    All proteins should be purified using SEC (e.g., Superdex 75 26/60; GE Healthcare) immediately prior to the experiment.

  4. 4.

    To carry out the ITC experiment, titrant (10 μL per injection) is injected into the sample cell over a period of 20 s with a 250 s interval between titrations to allow for complete equilibration and baseline recovery. About 28 injections are delivered during each experiment (example given for a VP-ITC system), and the solution in the sample cell is stirred at 307 rpm to ensure rapid mixing.

  5. 5.

    Data analysis, Origin: A nonlinear least-squares algorithm and the titrant and sample cell concentrations are used to fit the integrated heat per injection to an equilibrium binding equation, providing values of the stoichiometry (n), change in enthalpy (ΔH) and the binding constant (K). Subsequently, ΔG and ΔS are readily calculated. Data are most commonly analyzed with a one-site binding model assuming a binding stoichiometry of 1:1. Typically, Origin 7.0 software with a specific ITC module is used for data evaluation.

  6. 6.

    Data analysis, NITPIC [38] and SEDPHAT [39]: NITPIC is a stand-alone program that is used to integrate ITC data from a variety of instruments and prepare the data for analysis using SEDPHAT, a program designed to analyze multiple datasets from different biophysical methods, including ITC.

3.3 NMR Spectroscopy

NMR spectroscopy is one of the best techniques to measure interactions (e.g., protein–protein, protein–peptide, protein–DNA, protein–RNA, protein–inhibitor [small molecule], and any variation thereof) at a variety of binding strengths; it is especially well suited to detect weak and even very weak interactions (μM–mM).

3.3.1 NMR Sample Preparation

  1. 1.

    Identification of the optimal sample buffer: Perform a high-throughput buffer screen to identify the buffer, pH, and salt concentration at which the protein is most stable (see Notes 11 and 12 ).

  2. 2.

    Identification of the optimal temperature for NMR measurements: The faster a protein tumbles in solution, the shorter its overall correlation time (τ c) and the better the spectral quality due to longer transverse relation times (R 2). Thus, once the optimal buffer has been identified, determine the melting temperature of the protein and set the measurement temperature at ~30 °C below the melting temperature (see Note 13 ).

  3. 3.

    A small amount of D2O must be added to the sample (1–10 % depending on the spectrometer and probe used (see Note 14 ).

3.3.2 NMR Experiments, Sequence Specific Backbone Assignment

  1. 1.

    Sequence specific backbone assignment: 2D [1H,15N]-TROSY, 3D HNCA [40], 3D HNCACB, 3D HN(CO)CA, 3D HN(CO)CACB, 3D HNCO, 3D HN(CA)CO, and a 3D 15N-edited [1H,1H] NOESY spectrum are the most common NMR spectra used to achieve the sequence-specific backbone assignment for a protein (see Notes 15 and 16 ).

  2. 2.

    Verification of assignment accuracy, single amino acid labeling: Due to the significantly reduced number of experiments (see Note 15 ) and the potential reduced completion percentage of the sequence-specific backbone assignment in large proteins due to, for example, the lack of H/D back-exchange, it is critical to verify the accuracy of the assignment. One of the primary methods by which this is achieved is by using single amino acid labeled samples (see examples in [41, 42]), which can be rapidly produced for 15N-Phe, 15N-Ile, 15N-Leu, and 15N-Tyr. Here it is best to use a specially prepared medium that has all amino acids in unlabeled form except for the ones that should be labeled, which are added as 15N-labeled amino acids. As these samples are produced using H2O and not D2O, sequence-specific backbone assignment problems arising from D2O/H2O back-exchange can be assayed.

  3. 3.

    Verification of assignment accuracy, CSPs: The sequence-specific backbone assignment can be further verified using ligand and/or inhibitor binding. Here, only residues known to interact with the ligand/inhibitor should show significant chemical shift perturbations (CSP) [43]. For example, we used C3′-(2,2,5,5-tetramethyl-3-pyrroline-1-oxyl-3-carboxylic acid ester) adenosine triphosphate (SL-ATP , a generous gift from Dr. Pia Vogel, Southern Methodist University) [44] to verify the assignment of p38α . AMP-PNP and/or specific p38α inhibitors can be used to test for CSP changes. This allows for the rapid detection of potentially mis-assigned residues.

  4. 4.

    Data deposition: All NMR chemical shift information must be deposited in the BioMagResBank (www.bmrb.wisc.edu; p38α accession number is 17471) so it can be downloaded and used by other research scientists for additional analysis and/or for the design of novel experiments.

3.3.3 NMR Experiments, Protein–Protein Interactions

  1. 1.

    Identification of residues that mediate a protein–protein interaction: The formation of a protein–protein complex influences the local environment of chemical shifts, which can be readily measured using a concentration dependent recording of a series of 2D [1H,15N] HSQC spectra [43]. Specifically, one protein is 15N-labeled (NMR-active) and the second protein is un-labeled (NMR-inactive). The bound, unlabeled protein will change the local environment of the amino acids that mediate binding, which can be readily detected in the NMR spectrum.

  2. 2.

    Measuring the K d of the interaction: A concentration-dependent measurement is used to determine an amino acid specific K d. Protein–protein binding events typically lead to a concentration dependent linear change in the chemical shift value. If a deviation from this linear change is detected, this means an additional conformational change is occurring during the binding event (see Note 17 ).

  3. 3.

    Reverse titrations: A reverse titration can be performed in which the previous unlabeled protein is now NMR active, while the previous NMR-active protein is now unlabeled. In this way, the complementary binding site can be determined.

  4. 4.

    NMR spectroscopy has a size limit; however, these CSP studies can be performed even with very high molecular weight proteins (depending on the labeling scheme up to 1 MDa). A further advantage of this technique is the ability to use peptides and full length proteins, which allows, for example, a small part of an interaction to be tested via a peptide [17, 45].

3.4 Global Structures

NMR spectroscopy has a limited set of experiments for measuring long-distance interactions. Nuclear Overhauser Effect (nOe) measurements (1H–1H dipole–dipole) are limited to ~5–6 Å (depending on labeling and the size of the B 0 field used for the measurements; NOE scales with B 0 3/2) [46]. Paramagnetic spin relaxation measurements (1H–e− dipole–dipole) allow for the measurements of longer distances (up to ~20 Å), but require routine modification of the protein (e.g., covalently attached via a cysteine or a N/C-terminal residue) using a paramagnetic spin label (most commonly used spin-label is (1-Oxyl-2,2,5,5-tetramethylpyrroline-3-methyl)methanethiosulfonate [MTSL]) [47]. As the relaxation effects depend on the internal dynamics of the spin label as well as the proteins themselves, all quantitative distance interpretation of spin-label data must be performed with great care. Other possibilities to define tertiary and quaternary structures are the use of residual dipolar couplings (RDCs) [48]. Rapid isotropic overall tumbling of the proteins in solution usually cancels out dipolar couplings. However, introducing a slight preferred alignment relative to the external field, e.g., through changes in the solution environment or high magnetic field strengths, restores a portion of these couplings. The measurement of these RDC data makes it possible to derive the orientation of a large number of local NH vectors in relationship to the overall alignment tensor, and thus can define the relative orientation of distant parts of for example a multi-domain protein.

Another method to obtain long-distance information is the incorporation of overall distance and shape constrains from independent small angle X-ray scattering measurements (SAXS; this method reports on the isotropic scattering of X-ray photons on proteins in solution, which enables the measurements of the radius of gyration (R g) among other parameters; see ref. 49 for a review). For example, to determine the structure of p38α in complex with HePTP , multiple experiments were performed. X-ray crystallography and NMR spectroscopy were used to achieve a high resolution local structure while SAXS allowed for the description of the global structure [17]. In recent years, a variety of computational methods have been introduced that allow data from multiple techniques to be optimally combined in order to determine the structure and/or the ensemble of structures that best fit the experimental results (HADDOCK [50, 51], EROS [18, 52], EROS-NMR [17], among other software [53]).

SAXS measurements can be performed either at a home source or at a synchrotron. Here we summarize steps we routinely perform when measuring this data.

  1. 1.

    Obtaining non-aggregated samples: SAXS is exceptionally sensitive to aggregation , as soluble aggregates, even if they represent less than 1 % of the sample, are significantly larger and thus will have a major impact on the overall measured scattering. Thus, it is optimal to prepare the protein sample immediately (or as shortly as possible before) the SAXS measurements; to facilitate optimal sample preparation, some synchrotron beamlines offer in-line SEC purification prior to SAXS measurements. For some samples, it is possible to filter aggregates prior to measurement. For this, 0.02 μm syringe filters (GE Healthcare Anotop 10) are suitable (see Note 10 ).

  2. 2.

    Sample concentration: The optimal concentration for SAXS measurements depends on the X-ray source, the size and assembly of the cell (flow-through or static), among other parameters. For example, on our home source (BioSAXS, FRE+ anode, Rigaku), we measure samples at concentrations as low as 1.5 mg/mL. Typically, we initiate our measurements using samples with the lowest concentration possible. We then concentrate the sample and re-collect the SAXS data from the sample at as high of concentration as can be achieved before aggregation is detected (sometimes this can be as high as 30 mg/mL).

  3. 3.

    Aggregation detection: Some samples are not amenable to SAXS measurements as they exhibit aggregation even at low concentrations. Aggregation is readily detected by observing a concentration dependent change in the R g as well as a change in the I/c 0 plot. In general, all experimental SAXS curves should be parallel at multiple concentrations; discrepancies are a clear indication of aggregation .

  4. 4.

    Radiation damage: Exposure to high energy photons can induce radiation damage. This is less of an issue at home sources, but can occur when using synchrotron radiation. Here, the use of flow cells can alleviate the problem [54].

  5. 5.

    SAXS data analysis: Numerous software packages are available to analyze SAXS data, with the two most widely used being ATSAS [55] and SCATTER, both of which also allow for the calculation of 3D envelops from the data; this calculations are usually performed with the highest signal/noise dataset (commonly highest measured concentration). This can also be done using Fast-SAXS-pro [56].

4 Notes

  1. 1.

    Warning: HCl must be diluted into H2O or D2O; do not add directly to metals.

  2. 2.

    For p38α expression, we use BL21 (DE3) RIL cells (Agilent) to overcome rare codon usage bias in E. coli.

  3. 3.

    A fresh transformation is essential for high yield expression of deuterated proteins.

  4. 4.

    500 mL is the minimum flask volume to ensure sufficient culture aeration.

  5. 5.

    Generally, after the overnight growth step with 70 % D2O-based medium, the 1 L of D2O-based medium is prepared with 100 mL distributed in a 500 mL Ultra-yield flask and the remaining 900 mL distributed equally into four 2 L Ultra-yield flasks.

  6. 6.

    For proteins that are expressed at 18 °C, cultures must be cooled to 18 °C prior to induction.

  7. 7.

    One typically waits until ~10–20 L of used D2O have accumulated before initiating D2O recycling.

  8. 8.

    Flasks with two inlets are optimal because they can be readily refilled.

  9. 9.

    We routinely recycle D2O twice. The doubly recycled D2O is only used for test expression, initial protein characterization and protein–protein interaction studies and sequence specific backbone assignment experiments.

  10. 10.

    Typically, we prepare 4–8 L of the appropriate ITC buffer that is then used for all subsequent experiments in order eliminate any heat of dilution from a buffer mismatch. The same approach is used for SAXS measurements, where accurate buffer subtraction is equally important.

  11. 11.

    We use the high-throughput thermofluor assay to identify optimal buffers for NMR samples [57].

  12. 12.

    Modern NMR spectrometers are most commonly equipped with cryogenically cooled NMR probes. Here, cold (~20 K) helium gas is used to cool the coil and the built-in signal preamplifier of the probe, which reduces thermal noise and increases the signal-to-noise ratio by an factor of 3–4. Because cryoprobes are very sensitive to salt concentrations, samples with lower ionic strength (≤50 mM) are preferred. In addition, HEPES should be preferably used over phosphatase buffers.

  13. 13.

    We determine the melting temperature by following the transition of the 222 nM circular dichroism [CD] signal during heating from 25 to 90 °C.

  14. 14.

    The deuterium NMR signal (nuclear spin = 1) is used as an internal reference and allows for rapid compensation for changes of the magnetic field due to environmental changes (cars on nearby roads, etc.) or the stability of the superconductive magnet.

  15. 15.

    Due to their large gyromagnetic ratio, protons (42.58 MHz/T; ratio of magnetic dipole moment/angular momentum of proton) are most commonly used for excitation and detection of NMR experiments. Thus, in fully deuterated samples, where the only H2O exchangeable protons are available for excitation and detection (specifically 1HN), the number of useable NMR experiments are significantly reduced.

  16. 16.

    The sequence-specific backbone assignment allows for the identification (assignment) of each peak in an NMR spectrum specifically to its amino acid and atom in the protein under investigation. It is based on probability matching of measured chemical shifts—thus, the more the data points, the higher the likelihood of a correct assignment. As a consequence, the higher the completion percentage of the sequence-specific backbone assignment is, the smaller the chance of errors in the sequence-specific backbone assignment.

  17. 17.

    Depending on the K d, the interaction will be in the fast, intermediate or slow exchange regime. A fast and slow exchange regime titration can be most readily analyzed. In the fast exchange regime, the switching between chemical states is so rapid that one average chemical shift is reported and thus the change can be fully traced (K d μM range). Here the experimenter needs to use ratios of 1:1, 1:2, 1:3, and higher until full saturation is achieved, i.e., until the peaks do not shift further upon the addition of more protein. An interaction in the slow exchange regime will lead to two peaks, whose intensities reflect their populations. An interaction in the intermediate exchange regime often leads to a broadening of the peaks beyond detectability. Here titration ratios of 1:0.1, 1:0.25, 1:0.5, etc. can be useful to be able to trace changes.