Exploring Protein-Glycan Interactions: Advances in Nuclear Magnetic Resonance

Summary

This NMR-based protocol investigates weak protein-glycan interactions using cyanovirin-N and D-mannose. Combining ligand- and protein-detected methods, it maps binding sites, detects allosteric effects, and identifies encounter complexes. The approach outlines sample preparation and data analysis, offering structural and dynamic insights valuable for glycan-specific diagnostics and recognition mechanisms.

Abstract

Protein-glycan interactions are central to many biological processes and are increasingly recognized as promising targets for diagnostic strategies in infectious, inflammatory, and neoplastic diseases. However, characterizing these interactions, especially when they are weak and transient, remains technically challenging. Nuclear magnetic resonance (NMR) spectroscopy offers unique advantages for studying such interactions under near-physiological conditions, providing atomic-resolution insights into both structural and dynamic aspects. In this study, we present an integrated NMR protocol designed to investigate low-affinity protein-glycan interactions using the mannose-specific lectin cyanovirin-N and the monosaccharide D-mannose as a model system. The protocol combines ligand- and protein-based NMR approaches to map binding sites comprehensively and detect subtle binding events, including potential allosteric effects or encounter complexes. The critical steps of the protocol include careful sample preparation, precise control of concentrations, and spectral standardization to ensure data reproducibility and reliability. NMR in solution enables the detection of transient interactions that are often inaccessible by other techniques. Although weak interactions are essential for life, there is a bias in the literature toward higher-affinity complexes, even though low-affinity complexes are relevant in many important biological functions, such as signal transduction and cell-cell communication. This technique provides a powerful platform for investigating protein-glycan recognition and may serve as a valuable tool for identifying novel diagnostic markers on the basis of glycan-specific interactions.

Introduction

Protein-glycan interactions are fundamental to a wide range of biological processes, including cell-cell recognition, immune response modulation, and pathogen-host interactions^1,2. These molecular interactions are highly specific and dynamic and play crucial roles in both physiological and pathological mechanisms^3,4,5. A classic example of such interactions occurs with lectins, proteins that specifically recognize and bind to glycans. Cyanovirin-N (CVN), for example, has been widely studied as an experimental model because of its ability to interact in high-affinity with high mannose oligosaccharides^5,6,7.

In addition to lectins, many glycosylated proteins, such as integrins, also play essential roles in recognition and cellular signaling. Integrins are transmembrane receptors that frequently undergo glycosylation of their subunits, influencing their stability and affinity for ligands, including glycans^5,8,9,10. However, the structural characterization of these interactions remains challenging owing to the intrinsic heterogeneity and flexibility of glycans, which complicates traditional structural biology approaches. In this context, the development of advanced methodologies capable of capturing these interactions in solution is of great value to this field, particularly for glycobiology^11,12.

Nuclear magnetic resonance (NMR) spectroscopy has emerged as a powerful tool for investigating protein-ligand interactions at the atomic level. Unlike other techniques, such as X-ray crystallography and cryo-electron microscopy, NMR allows the study of biomolecular interactions under near-physiological conditions, providing insights into both structure and dynamics. This flexibility enables researchers to modulate environmental parameters such as pH (typically between 4.0-8.0), ionic strength (e.g., 0-500 mM NaCl), and temperature (generally 273-330 K), thereby tailoring conditions to the specificities of protein-glycan complexes. Additionally, NMR is particularly advantageous for analyzing transient and weak interactions, which are often characteristic of protein-glycan recognition events^13,14.

Several NMR techniques have been employed to investigate protein-glycan interactions, utilizing both protein- and ligand-based approaches. From the protein perspective, heteronuclear single quantum coherence (HSQC) titration is widely used to map binding sites by monitoring chemical shift perturbations upon ligand addition. Relaxation dispersion experiments enable the detection of conformational and chemical exchange processes occurring on the micro- to millisecond timescale, providing insights into the dynamic aspects of glycan recognition^15,16. These protein-based techniques typically require sample concentrations in the range of 50 µM to 2 mM, depending on protein stability and labeling efficiency^17,18.

Ligand-based NMR techniques offer complementary information by focusing on glycan changes during binding. Saturation transfer difference (STD-NMR) is particularly useful for identifying glycan epitopes involved in recognition, as it selectively saturates the NMR signals of ligand regions in close contact with the protein^19,20. Water-Ligands Observed via Gradient Spectroscopy (WaterLOGSY)^21,22 and Transferred Nuclear Overhauser Effect Spectroscopy (Tr-NOESY) provide additional means to assess ligand binding and conformational changes during interactions²³. Furthermore, Carr-Purcell-Meiboom-Gill (CPMG) relaxation experiments aid in identifying weak and transient interactions that are difficult to detect with conventional methods²⁴. Ligand-based experiments are often performed at protein concentrations of 10-50 µM and may require optimization of mixing times and saturation parameters²⁵. Notably, these methods can be limited by the low solubility of glycans or the poor signal-to-noise ratio when working with small ligands.

Together, these NMR methods provide a comprehensive framework for elucidating the structural and dynamic properties of protein-glycan interactions. With ongoing advancements in sensitivity and isotopic labeling strategies, this technique is becoming increasingly essential in glycobiology, offering new perspectives on molecular recognition mechanisms with potential applications in drug development and biomarker discovery. Nonetheless, the success of these approaches depends on factors such as sample homogeneity, glycan complexity, and the presence of flexible or disordered regions that may broaden NMR signals or hinder interpretation²⁶. NMR is a powerful method for studying transient events, which is a major challenge in structural biology.

CVN recognizes α(1,2)-linked mannosyl residues present in high-mannose oligosaccharides with nanomolar affinities^27,28,29. However, it poorly recognizes the monosaccharide D-mannose. In the present work, we studied the interaction of CVN with D-mannose as a way to illustrate how NMR is powerful in understanding transient interactions.

Protocol

NOTE: Perform all steps involving bacterial cultures and buffer preparations in accordance with institutional biosafety protocols. Use personal protective equipment (PPE) and, when handling open cultures or chemicals, work in a biosafety cabinet or chemical fume hood if required.

1. Expression of Cyanovirin-N

1. Transform E. coli BL21 cells with a plasmid encoding the CVN gene inserted into the pET26a vector.
2. Cultivate the cells in 1 L of Super Broth (32 g of tryptone, 20 g of yeast extract, 5 g of NaCl per L), supplemented with kanamycin (50 µg/mL), 0.5% (w/v) glucose, and 1.6 mM MgSO₄. Incubate under continuous agitation at ~250 rpm at 37 °C.
3. For production of ¹⁵N-labeled CVN, grow the cells in modified M9 minimal medium using ¹⁵NH₄Cl as the sole nitrogen source.
   1. Prepare 1 L of M9 minimal medium by mixing 200 mL of sterile 5x M9 salts solution containing 64 g/L of Na₂HPO₄, 15 g/L of KH₂PO₄, 2.5 g/L of NaCl, 5 g/L of NH₄Cl.
   2. Add 2 mL of 1 M MgSO₄ (0.22 µm filter-sterilized), 0.1 mL of 1 M CaCl₂ (0.22 µm filter-sterilized), and 4 g of glucose (final concentration 0.4%, dissolved in sterile distilled water and 0.22 µm filter-sterilized).
   3. Bring the final volume to 1 L with sterile distilled water.
      NOTE: Supplements such as amino acids or vitamins can be added as needed.
4. Monitor the culture until it reaches an optical density (OD₆₀₀) of approximately 1.2. Then, add isopropyl-beta-D-thiogalactopyranoside (IPTG) to a final concentration of 1.0 mM to induce protein expression.
5. Incubate for 20 h post-induction at 37 °C.
6. Harvest the cells by centrifugation at 7,000 × g for 10 min at 4 °C.
7. Isolate the periplasmic fraction as follows:
   1. Resuspend the cell pellets in 0.4 volumes of 30 mM Tris-HCl buffer (pH 8.0) containing 20% sucrose and 1 mM EDTA and incubate at room temperature with shaking for 10 min.
   2. Centrifuge at 10,000 × g for 10 min at 4 °C.
   3. Resuspend the resulting pellets in 0.4 volumes of ice-cold distilled water and incubate on ice with shaking for 10 min.
   4. Centrifuge again at 10,000 × g for 10 min at 4 °C.
   5. Collect the supernatant, which corresponds to the periplasmic fraction, for protein extraction.
      NOTE: All buffers must contain a protease inhibitor cocktail to prevent protein degradation.

2. NMR experiments

NOTE: All experiments were acquired at 14.1 T using a ¹H/¹⁵N/¹³C triple resonance inverse-detection probe (see Table of Materials) at 298K.

1. Ligand-based experiments
   1. Sample preparation
      1. Prepare two samples: one containing CVN and one without CVN (ligand-only control). Each sample contains 600 µL of 80 µM CVN in sodium phosphate buffer (pH 7.4) containing 50 mM NaCl.
      2. Add D-mannose to a final concentration of 4 mM.
      3. Add 5% (v/v) D₂O to each sample.
      4. Transfer the sample to a 5 mm NMR tube.
      5. Use a 50-fold molar excess of the ligand relative to the protein.
         NOTE: A ligand excess in the range of 10-fold to 1000-fold is typically recommended, with lower excesses preferred for small proteins.
   2. Spectral acquisition for D-mannose assignment
      1. Acquire 2D TOCSY and 2D ¹H-¹³C HMBC spectra using standard pulse sequences:
         2D TOCSY: Phase-sensitive total correlation spectroscopy (TOCSY) with excitation sculpting for water suppression (DIPSI2ESGPPH).
         2D HMBC: ¹H-¹³C Heteronuclear Multiple-Bond Correlation with gradient selection and non-decoupled acquisition (HMBCGPNDQF).
         NOTE: These experiments were performed for the purpose of D-mannose assignment but are not shown, as chemical shift assignment is not within the scope of this study.
   3. Mapping ligand binding through chemical shift perturbation (CSP) of the ligand
      1. For the sample preparation and acquisition of ¹H-¹³C HSQC spectra, prepare a 4 mM solution of D-mannose in sodium phosphate buffer (pH 7.4), with 50 mM NaCl and 5% D₂O. Divide the solution into two samples: one with 80 µM CVN and one without CVN (ligand-only reference).
         NOTE: These spectra will be used to determine the CSP of D-mannose upon interaction with CVN.
      2. For the setup of ¹H-¹³C HSQC acquisition parameters, use the standard pulse sequence HSQCETGPSI (gradient selection and sensitivity-enhanced ¹H-¹³C HSQC). Set the following parameters for acquisition:
         Time Domain (TD): 1024 × 128 complex points (¹H × ¹³C dimensions)
         Spectral Width (SW): 10.0171 ppm (6009.615 Hz) for ¹H, 80 ppm (12069.106 Hz) for ¹³C
         Carrier Frequencies: 4.7 ppm for ¹H, 75 ppm for ¹³C
         - Number of Scans: 448.
      3. Calculate the CSP using the following equation:
         
         where and represent the chemical shift differences between D-mannose in the presence of CVN (Figure 1A, labeled in red) and free D-mannose (Figure 1B).
   4. Mapping ligand binding through the saturation transfer difference (STD)
      1. Create a new dataset and configure parameters for STD-NMR experiments. Two experimental strategies may be applied: (i) screening of different saturation frequencies to optimize signal-to-noise ratio; (ii) fixing the saturation frequency and varying the saturation time to evaluate build-up.
      2. Set up 1D ¹H NMR experiments using the zgpr pulse sequence. Tune the spectrometer for ¹H, perform shimming, and measure a hard 90° pulse (e.g., ~10 µs at -10.6 dB attenuation, equivalent to ~11.482 W).
      3. Center the ¹H carrier frequency at approximately 4.7 ppm, corresponding to the water signal.
      4. Select the STDDIFFESGP.3 pulse sequence and configure the acquisition parameters as follows:
         Set the spectral width according to sample requirements.
         Set the interscan delay (d1) to 4 s.
         Define the saturation time (d20) based on the desired experiment.
         Load the FQ2LIST containing the off-resonance saturation frequency (-40 ppm) and the on-resonance frequencies to saturate the protein signal only.
         ​NOTE: On-resonance frequencies must correspond to protein protons and must be at least 1000 Hz away from any ligand signal to avoid direct saturation.
      5. Test the following on-resonance saturation frequencies to determine optimal conditions: -0.59 ppm, 0.73 ppm, and 8.1 ppm (see Figure 2A).
         NOTE: 0.73 ppm was selected as an on-resonance saturation frequency, as it provided the best signal-to-noise ratio. Using this frequency, the STD amplification factor (A_STD) was quantified as a function of saturation time. To this end, STD spectra were acquired at different saturation times (0.5, 1, 1.5, 2, 2.5, 3, and 4 s) to construct the STD build-up curve (Figure 2B). The corresponding A_STD values were then plotted as a function of saturation time (Figure 2C), allowing the evaluation of the efficiency of the saturation transfer, which is proportional to the time and proximity of the ligand hydrogen to the protein. A_STD factor is defined as:
         
         I_STD is the difference in signal intensity between the on- and off-resonance spectra, I₀ is the signal intensity in the off-resonance spectrum, [L]_T is the total ligand concentration, and [P] is the protein concentration. This normalization accounts for concentration effects and allows for comparison across different experimental conditions. STD relies on spin diffusion and slow molecular tumbling to efficiently transfer saturation from the protein to the bound ligand. Larger proteins (slow tumbling) allow more efficient saturation and spread of magnetization within the protein^19,30,31,32.
      6. Set the number of scans (ns) to 64 and average the experiment l4 (loop 4) times.
      7. Calculate the total number of scans as: Total scans = ns × l4. Use 32,768 complex points in the direct dimension (TD) with a spectral width (SW) of 10.0171 ppm (6,009.615 Hz).
      8. Set the interscan delay (d1) to 4 s and acquisition time (AQ) to 2.7262976 s. Total relaxation delay equals d1 + AQ.
      9. Configure the saturation pulse as follows: duration 50 ms; shape Gaussian (p42); controlled via shaped program 9 (SP9).
      10. Use an STD pulse sequence variant that includes a T₁ρ filter to suppress residual protein signals. Set the spin-lock time (d29) based on protein molecular weight. For larger proteins, use smaller spin-lock times (~20 ms) and for smaller proteins, larger (> 40 ms).
          NOTE: All STD spectra were acquired at 599.93 MHz (BF1). Adjustment of d29 is critical to minimize protein background while preserving ligand signal integrity. Optimize this parameter empirically if necessary.
   5. Mapping of ligand binding via ¹H-R₂
      1. Select the CPMG_ESGP2D pulse program from the Bruker standard library. This sequence is based on the Carr-Purcell-Meiboom-Gill (CPMG) experiment, incorporating excitation sculpting for water suppression³³.
      2. Set the experiment as a 2D acquisition.
         NOTE: This is a pseudo-2D experiment in which only the direct dimension is used for chemical shift information.
      3. Configure the acquisition parameters as follows:
         Time domain (TD): 32,768 complex points in the direct ¹H dimension.
         (TD1): 2 points.
         Spectral width (SW): 7.9932 ppm (4,795.396 Hz).
         Interscan delay (d1): 4 s.
         Carrier frequency (o1): 4.7 ppm (centered on water signal).
         Number of scans (ns): 8.
         Number of dummy scans (ds): 4.
         CPMG delay (d20): ≤1 ms (i.e., shorter than 1⁄3 J_HH).
         ​Number of averages (TDav): 200.
      4. Run the experiment for 200 cycles of 8 scans each, resulting in a total accumulation of 1,600 scans. Adjust the variable counter list (vclist) according to the desired total time of the CPMG cycle (T_CPMG). Here, 2 cycles were chosen for a T_CPMG=8 ms and 200 cycles were chosen for a T_CPMG=800 ms.
      5. Run two ¹H-CPMG experiments, one for the sample containing the protein (4 mM D-mannose and 80 µM CVN) and the other for the free D-mannose (4 mM D-mannose).
      6. Plot the ¹H spectra for each T_CPMGvia the command efp (exponential multiplication followed by Fourier transform and phase correction) or sinm (sine multiplication shifted by 90°, SSB=2) followed by fp (Fourier transform and phase correction). Adjust the window function according to the best processing strategy.
         NOTE: In this experiment sinm and fp were used. The 1^st serial will be the spectra with a T_CPMG of 8 ms (first in the vclist), and the 2^nd serial will be 800 ms (second in the vclist). The processed ¹H-spectra are plotted in Figure 3.
      7. Calculate the CPMG quotient (Q) via equation 3.
         ​
2. Protein-based experiments
   1. Sample preparation
      1. Prepare 600 µL of a sample containing 436 µM uniformly ¹⁵N-labeled CVN in sodium phosphate buffer (pH 7.4) supplemented with 50 mM NaCl and 5% deuterium oxide (D₂O). Titrate D-mannose into the sample to a final concentration of 60 mM. For the titration, use a 5 mm NMR tube.
   2. Resonance Assignment (Optional Validation Step)
      1. Confirm the assignment of free CVN in solution using previously reported chemical shift values³⁴. Acquire the following triple resonance spectra: HNCO, HNCA, HNCACB, and CBCACONH^35,36,37.
         NOTE: These experiments serve only to validate the assignment and are not shown, as chemical shift assignment is not the primary objective of this protocol.
   3. Mapping the binding of D-mannose on CVN by chemical shift perturbation of the protein
      1. Acquire ¹H-¹⁵N HSQC spectra of ¹⁵N-labeled CVN at 298 K. Perform a titration by adding D-mannose to reach the following final concentrations: 0, 1, 2, 5, 10, 20, 40, and 60 mM.
      2. Set up the ¹H-¹⁵N Fast-HSQC experiment using the phase-sensitive FHSQCF3GPPH pulse sequence from the Bruker standard library. Use the following parameters:
         Time domain (TD): 1024 × 140 complex points in the ¹H and ¹⁵N dimensions.
         Spectral width (SW): 16.0274 ppm (9615.385 Hz) in the ¹H dimension and 34 ppm (2067.119 Hz) in the ¹⁵N dimension.
         Carrier frequency: 4.7 ppm (¹H) and 119 ppm (¹⁵N).
         Number of scans: 128.
      3. Calculate the chemical shift perturbation (Δδ) using the following equation:
         
         where and are the chemical shift differences between the CVN in the presence and absence of D-mannose (Figure 4A).
         NOTE: The factor 10 accounts for the difference in spectral widths between the proton and nitrogen dimensions.
      4. Calculate the dissociation constant (K_D) by plotting the CSP values for each residue as a function of D-mannose concentration. Fit the resulting data to a single-site binding isotherm using the following equation (5).
           equation 5
         where [P_T] is the protein concentration used in the titration (CVN), [P_T] is the ligand concentration (D-mannose), and CSPmax is the CSP at the saturation concentration of the ligand and CSPmin in the absence of the ligand.
   4. Mapping the binding of D-mannose to CVN via the ¹⁵N-R₂ position of the protein.
      1. Measure the ¹⁵N-R₂ values of the individual residues of CVN in the absence and presence of 60 mM D-mannose. Perform all experiments at 298 K.
      2. Set up the ¹⁵N-R₂ experiment using the phase-sensitive CPMG-based pulse sequence HSQCT2ETF3GPSI3D from the Bruker standard library. Use the following parameters:
         Time domain (TD): 1024 × 128 complex points in the ¹H and ¹⁵N dimensions, with 8 points in the pseudo dimension.
         Spectral width (SW): 16.0274 ppm (¹H; 9615.385 Hz) and 34 ppm (¹⁵N; 2067.119 Hz).
         Carrier frequency: 4.7 ppm (¹H) and 119 ppm (¹⁵N).
         ​Number of scans: 32. (v) Variable counter list: 1, 8, 2, 6, 3, 5, 4, and 7, corresponding to relaxation delays (T_relax) of 16.96, 135.68, 33.92, 101.76, 50.88, 84.96, 67.84, and 118.72 ms, respectively.
      3. Process the pseudo3D spectra as a series of HSQC-like ¹H/¹⁵N correlation relaxation spectra using NMRPipe³⁸, following the software tutorial (see Table of Materials for a link to the tutorial).
      4. Import the processed spectra into a suitable NMR analysis platform (e.g., CCPNMR³⁹). Select all spectra and apply the "Follow Intensity Changes" tool. Plot the intensity of each cross-peak as a function of T_relax and fit the decay to a mono-exponential function to extract ¹⁵N-R₂ values for each residue.
      5. Calculate the experimental error from the signal-to-noise ratio of the HSQC-like spectra at 67.84 ms. Process a region of the spectrum containing only noise and convert it to a text file using the following command:
         pipe2txt.tcl ./PROCs/ft/R2_67_84.ft2 > noise67_84.txt"
      6. Determine the standard deviation of the noise region using statistical software (e.g., Origin(Pro), Version 2021). Define this value as the noise intensity I_noise.
      7. Calculate the experimental error for each residue using the following equation:
         
         NOTE: It is important to use the experimental error rather than the fitting error. The experimental error defines the interpretation limits when comparing the ¹⁵N-R₂ in the absence and presence of D-mannose.

Figure 1: Chemical shift perturbation of D-mannose in the presence of CVN. (A) ¹H-¹³C-HSQC spectrum of D-mannose in the presence of CVN. (B) ¹H-¹³C-HSQC spectrum of free D-mannose. The labels in the spectra show each of the chemical shift assignments and the respective anomeric configuration (α or β). The labels in red are those assigned to the bound conformation, and those in black are those assigned to the free conformation. (C) Superposition of the ¹H-¹³C-HSQC spectra in the presence (green) and absence (blue) of CVN. (D) Chemical structure of α- and β-D-mannose highlighting the observed CSP, which was calculated according to equation (1). The CSP values are depicted below the green circles. The red circle represents the β-anomeric hydrogen that vanished in the presence of CVN. Please click here to view a larger version of this figure.

Figure 2: Saturation transfer difference (STD) spectra of D-mannose in the presence of CVN. (A) STDs acquired at different saturation frequencies: -0.59, 0.73, and 8.1 ppm. (B) STD acquired at different saturation times: 0.5, 1, 1.5, 2, 2.5, 3, and 4 s. (C) STD amplification factor (A_STD) calculated according to equation 2 as a function of the saturation time. Please click here to view a larger version of this figure.

Figure 3: ¹H-R₂ of D-mannose in the presence and absence of CVN. (A) Representation of the expected D-mannose ¹H-CPMG spectra for a binder and a nonbinder acquired with R₂ relaxation times of 8 and 800 ms in the absence (bottom) and presence (top) of CVN. Q was calculated according to equation 3. In the absence of interaction (Q ≈ 1), the signal intensities remain similar at both relaxation times of the ligand with or without the protein. Upon binding (Q < 1), the signal intensities decrease with increasing relaxation for the sample containing the protein. (B) Schematic representation of the chemical exchange between free and bound states of D-mannose. Depending on the relative values of the chemical shift difference (Δω) and the exchange rate constant (k_ex = k_on + k_off), the system falls into slow, intermediate, or fast exchange regimes. (C)¹H NMR spectra of free D-mannose and D-mannose in the presence of CVN, showing differences in signal intensities at 8 ms and 800 ms relaxation times. The calculated Q factors for specific hydrogens (H_6β, H_3β, H_5α, H_4β) are shown, with lower Q values indicating stronger interactions with CVN and identifying the regions of the sugar involved in binding. Please click here to view a larger version of this figure.

Figure 4: Chemical shift perturbation (CSP) analysis revealing CVN residues involved in D-mannose binding. (A) Bar plots showing the backbone amide (¹⁵N-¹H) CSPs of CVN upon titration with D-mannose at 10 mM (top) and 60 mM (bottom) concentrations. The horizontal lines indicate the average CSP plus one or two standard deviations (av + sd, av + 2 sd), which were used as thresholds to identify significantly perturbed residues. (B) Mapping of significantly perturbed residues onto the structure of CVN complexed with the disaccharide Manα1-2Manα (PDB ID: 1IIY⁴⁰). Residues with CSPs above the av + 2 sd threshold are highlighted in red, and those between av + sd and av + 2 sd are shown in blue. The disaccharide Manα1-2Manα molecules are displayed in orange to indicate their binding location. (C) Selected binding curves representing CSPs as a function of D-mannose concentration. The data were fitted according to equation 5 as a single-site binding model to estimate K_D, revealing a range of affinities, with D44 showing the strongest interaction (K_D = 1.1 ± 0.35 mM) and E41 the weakest (K_D = 35 ± 29 mM). Please click here to view a larger version of this figure.

Figure 5: Analysis of ¹⁵N-R2 for CVN free and bound to D-mannose. (A, B) Transverse relaxation rates (¹⁵N-R₂) measured for each residue of CVN in its (A) free and (B) D-mannose-bound states. (C) Difference in R₂ values (ΔR₂ = R₂^bound− R₂^free) plotted per residue. The blue and red lines represent one and two standard deviations from the average (av ± sd and av ± 2 sd), respectively. Residues with significant ΔR₂ values are labeled. (D) Mapping of the ΔR₂ values for the structure of CVN complexed with the disaccharide Manα1-2Manα (PDB ID: 1IIY⁴⁰). Residues with ΔR₂ above or below av± 2 sd are shown in red; those between av± sd and av± 2 sd are shown in blue. D-mannose molecules are displayed in orange to indicate their binding location. Please click here to view a larger version of this figure.

Results

In the present work, we presented NMR experiments to investigate protein-glycan interactions. Protein-glycan molecular recognition underlies numerous biological processes involving the extracellular matrix (ECM), including the regulation of cell adhesion, migration, proliferation, and various other forms of cellular communication.

Here, we chose to work with the monosaccharide D-mannose, which is described as a weak binder. The goal was to describe methods for measuring transient interactions ...

Discussion

NMR spectroscopy represents a versatile approach for investigating protein-glycan interactions, particularly because of its ability to detect weak and transient interactions, such as those observed between CVN and D-mannose. The protocol described here combines ligand- and protein-based NMR methods, allowing for a more comprehensive characterization of the interaction interface.

The critical steps of the protocol include careful sample preparation with precise control of protein and ligand con...

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
13C Glucose	Cambrige Isotopic Lab.	CLM-481
15N Ammonium Chloride	Cambrige Isotopic Lab.	NLM-467
Calcium Chloride	Sigma-Aldrich	C4901
Coquetel inhibit protease	Sigma-Aldrich	4693159001
D2O	Cambrige Isotopic Lab.	DLM-4
D-Mannose	Sigma-Aldrich	M2069-5G
Magnesium Chloride	Sigma-Aldrich	M8266
NMRPipe			Link to the software tutorial: https://www.ibbr.umd.edu/nmrpipe/demo.html
Potassium Chloride (K2HPO4)	Sigma-Aldrich	P3786
Potassium Chloride (KCl)	Sigma-Aldrich	P3911
Potassium Chloride (KH2PO4)	Sigma-Aldrich	P0662
Shigemi tube	Bruker	Z1Z10684A
Sodium chloride (NaCl)	Sigma-Aldrich	S9625
Sodium phosphate (Na2HPO4)	Sigma-Aldrich	567545
Sodium phosphate (NaH2PO4)	Sigma-Aldrich	S97263
Spectrometer Bruker 600 MHz	Bruker
Spectrometer Bruker 800 MHz	Bruker
Tiamine	Sigma-Aldrich	T1270
Triple resonance TXI probe	Bruker
Tryptone	Sigma-Aldrich	93657
Yeast extract	Sigma-Aldrich	Y1625