Lipid-nanodiscs formed by paramagnetic metal chelated polymer for fast NMR data acquisition

Abstract

Lipid-nanodiscs have been shown to be an exciting innovation as a membrane-mimicking system for studies on membrane proteins by a variety of biophysical techniques, including NMR spectroscopy. Although NMR spectroscopy is unique in enabling the atomic-resolution investigation of dynamic structures of membrane-associated molecules, it, unfortunately, suffers from intrinsically low sensitivity. The long data acquisition often used to enhance the sensitivity is not desirable for sensitive membrane proteins. Instead, paramagnetic relaxation enhancement (PRE) has been used to reduce NMR data acquisition time or to reduce the amount of sample required to acquire an NMR spectra. However, the PRE approach involves the introduction of external paramagnetic probes in the system, which can induce undesired changes in the sample and on the observed NMR spectra. For example, the addition of a paramagnetic ions, as frequently used, can denature the protein via direct interaction and also through sample heating. In this study, we show how the introduction of paramagnetic tags on the outer belt of polymer-nanodiscs can be used to speed-up data acquisition by significantly reducing the spin-lattice relaxation (T_l) times with minimum-tono alteration of the spectral quality. Our results also demonstrate the feasibility of using different types of paramagnetic ions (Eu³⁺, Gd³⁺, Dy³⁺, Er³⁺, Yb³⁺) for NMR studies on lipid-nanodiscs. Experimental results characterizing the formation of lipid-nanodiscs by the metal-chelated polymer, and their increased tolerance toward metal ions are also reported.

Graphical Abstract

Introduction

Membrane proteins are ubiquitous in every cell and perform a plethora of fundamental functions for the living organisms[l], Despite their importance, there is still a large gap in the structural studies of membrane proteins when compared to their soluble protein counterparts. This is mainly because of the difficulties in mimicking native lipid membranes and using the membrane mimetics to stabilize native-like folding and function of membrane proteins[2]. While detergents are still in use in the purification and structural studies of membrane proteins, it is known that detergents are capable of denaturing membrane proteins. Other well-utilized model membranes are bicelles and liposomes, which also have limitations[3]. Lipid nanodiscs, in recent years, have shown great potential in the structural biology of membrane proteins.[1,4–10], Nanodiscs are disc-shaped lipid bilayer patches surrounded by an amphiphilic macromolecule that can be made up of membrane scaffold proteins (MSPs)[1,11–13], peptides[14], or synthetic-polymers [5,9,15–22], Although different types of nanodiscs are used, most MSP and peptide based nanodiscs show several limitations that hinder the characterization of reconstituted membrane proteins by several biophysical techniques [18]. On the other hand, the synthetic polymer-based nanodiscs have been shown to provide a number of flexibilities and choices, enabling a variety of applications. Many different types of polymer belts have been reported in the literature, and they have been used to demonstrate that polymer based nanodisc technology is suitable for the study of biomacromolecules by most biophysical techniques including both solution and solid-state NMR spectroscopy[15,23–28], While NMR is unique in rendering the measurement of atomic-resolution dynamics at different time scales, it, however, has the significant disadvantage of being a relatively lowsensitivity technique[29]. This disadvantage is amplified by the instability or scarce amount of most biologically interesting membrane proteins and also to characterize short-lived intermediates such as amyloid oligomers. This low sensitivity is most commonly overcome by either expensive isotopic labeling and encumberingly long experimental times which is not desirable for heat-sensitive proteins[29–35]. Another solution is to increase the concentration of the sample, but it is often untenable for many protein systems due to the potential instability and aggregation issues of the proteins such as amyloid proteins. The sensitivity of NMR spectroscopy can also be enhanced by dynamic nuclear polarization (DNP) technique. This methodology is increasingly used and has already offered significant contributions in the fields of structural biology and materials [36–42], However, one of the most important ways to increase the sensitivity of NMR is by the addition of paramagnetic dopants to utilize the paramagnetic relaxation enhancement (PRE)[43–50].

PRE effect can be used to enhance NMR sensitivity by shortening the spin-lattice (T₁) relaxation properties of the nuclei[51,52], Briefly, relaxation is a process in which the thermal equilibrium of the nuclear spin states is regained after the perturbation of applied radio-frequency (RF) pulses. The relaxation of multiple-spin systems can be an extremely complex process to describe, which can be further complicated by chemical exchange and local motions; but it can be phenomenologically described by two of the major relaxation mechanisms: longitudinal (or spin-lattice or T₁) and transverse (or spin-spin or T₂) relaxations as defined by the Bloch equations[53], PRE-NMR involves the introduction of a paramagnetic salt into the sample or external paramagnetic-tags in the investigated system. Based on the chosen probe, the PRE effect can alter the relaxation in different ways[52,54–56]. Unfortunately, it is not possible to separately influence T₁ and T₂ values; so, while affecting one of the relaxation parameters, it is inevitable to change the other. This aspect can cause undesired effects on the spectra, such as line-broadening (related to T₂-shortening) and peak-shifts (due to hyperfine shift )[43,50,57–60] that can negatively affect the overall quality of the resultant spectra. Additionally, sample preparation is complicated by a further step of "tagging," which need not always proceed to completion[57].

Since the PRE effect can be used to speed-up the T_l process, it can be utilized to accelerate data acquisition by reducing the recycle delay between successive scans used for signal averaging in an NMR experiments. In addition, the paramagnetic effect induced shift in resonance frequency (known as hyperfine shift and composed of two contributions: the contact and the pseudo-contact shifts or PCS) can be used to obtain structural information such as topological analysis and distance measurements since paramagnetic effects are typically proportional to l/rⁿ (n = 6 for PRE and 3 for PCS), where r is the distance between the paramagnetic tag and the investigated nuclei[61–63].

The versatility and flexibility of paramagnetic methods are benefited from a large pool of paramagnetic tags to choose from and the ways of incorporation to access the sites of interest[64–73]. Indeed, the use of diamagnetic and paramagnetic ions bound to specific positions or solvent-exposed surfaces of macromolecules or supramolecular aggregates both in solution and solid-state can reveal structural and topological information.

We recently reported a nanodisc forming synthetic polymer, SMA-EA-DOTA[74], In addition to the already reported advantages of forming polymer-based lipid-nanodiscs, this copolymer enables the introduction of paramagnetic metal ions to the DOTA units in the polymer located on the outer rim of polymer-nanodiscs. Thus, this DOTA-unit containing polymer avoids the presence of free paramagnetic ions in solution and the use of paramagnetically-tagged lipids[52,75]. As a result, an SMA-EA-DOTA polymer-based nanodisc offers a planar lipid bilayer to reconstitute membrane protein(s) without the interference from the added paramagnetic metal ions. The aim of this study is to reduce the spin-lattice relaxation time (T₁) of the sample with a minimum line broadening due to a reduced spin-spin relaxation time (T₂) by using the paramagnetic metals chelated to the DOTA units of the polymers in nanodiscs. Due to the size flexibility of polymer nanodiscs[76], a wide variety of NMR conditions can be utilized for solution NMR, partial alignment for RDC measurements by solution NMR, and a higher degree of alignment for solid-state NMR experiments[77]. Because of this flexibility, a library of a variety of metal ions for differing conditions of PRE is needed. In addition, the introduction of paramagnetic metals can enable the applications of EPR experiments[78,79]. In this study, we report a systematic investigation of the PRE effects from five different paramagnetic metal ions (Eu^s+, Gd⁵⁺, Dy³⁺, Er³⁺, Yb³⁺) for NMR applications[50,57,60,80–85].

Materials and Methods

1. Reagents and Materials

Poly(Styrene-co-Maleic Anhydride)-Cumene terminated, SMAnh, M_w ~ 1.6 kDa, anhydrous 1Methyl2-pyrrolidone (NMP), 2-Aminoethanol (EA), Triethylamine, Trifluoroacetic Acid (TFA), Diethyl Ether (Et₂0), Hydrochloric Acid (HCI), Sodium Hydroxide (NaOH), 4-(2-Hydroxyethyl)Piperazine-l-ethanesulfonic acid (HEPES), Sodium Chloride (NaCI), Europium(lll) Chloride Hexahydrate (EuCI₃-6H₂0), Gadolinium(lll) Chloride Hexahydrate (GdCI₃-6H₂0), Dysprosium(lll) Chloride Hexahydrate (DyCI₃-6H₂0), Erbium(lll) Chloride Hexahydrate (ErCI₃-6H₂0), Ytterbium(lll) Chloride Hexahydrate (YbCI₃-6H₂0) were purchased from SigmaAldrich^® (St. Louis, Missouri). l,4,7,10-Tetraazacyclododecane-l,4,7-tris(t-butyl-acetate)-10(aminoethylacetamide) was purchased from Macrocyclics^® Inc. (Plano, Texas). l,2-dimyristoyl-sn-glycero-3phosphocholine (DMPC) was purchased from Avanti^® Polar Lipids, Inc. (Alabaster, Alabama).

2. Polymer synthesis and characterization

2.1.

SMA-EA copolymer was synthesized, purified, and characterized according to the procedure described in the literature[76].

2.1.

SMA-EA-DOTA copolymer was synthesized, as described previously[74]. In short, 1 g of SMAnh, 0.435 g of 1,4,7,10-Tetraazacyclododecane-l,4,7-tris(t-butyl-acetate)-10-(aminoethylacetamide) (amino-DOTA), and 1 mL of Triethylamine were dissolved in 60 ml of anhydrous NMP and heated to 70 °C under continuous stirring (Step-1). The addition of an excess of ethanolamine in the presence of an extra 1 mL of triethylamine for 2 more hours at the same temperature in the same round bottom flask completed the ǹucleophilic ring-opening reaction (Step-2). The final product was precipitated using 1 M HCI. To deprotect the chelator units, the polymer was dissolved in TFA and reacted for 2 hours at room temperature under gentle stirring (Step-3). Finally, the product was precipitated using diethyl ether and washed multiple times with deionized water prior to lyophilization. The 3-step reaction yielded about 600 mg. The reaction scheme is presented in Figure S1.

2.3.

The newly synthesized polymer was characterized by FT-IR. The shift of the carbonyl stretching frequency from 1770 cm⁻¹ to 1702 cm⁻¹ indicates the conversion of the anhydride to the amide, confirming the success of the reaction. FT-IR spectra are reported in Figure S2.

3. Polymer nanodiscs preparation

Stock solutions of each copolymer (SMA-EA and SMA-EA-DOTA) were obtained by dissolving the desired amount of powder in a 0.1 M NaOH solution. The pH was then adjusted to 7.4 using 1 M HCI. Polymer-based lipid nanodiscs were prepared by mixing the desired quantity of DMPC lipids and polymers in the ratio 1:1 by weight from 20 mg/mL stock solutions. Each sample was incubated overnight at room temperature prior to its use.

4. Polymer nanodiscs characterization

4.1. Size Exclusion Chromatography (SEC)

Polymer-based phospholipid nanodiscs were purified by SEC using a self-packed Sephadex 200 16/600 column operated on a GE Healthcare^® AKTA purifier. Samples were eluted at room temperature and at a flow rate of 1 mL/min. The elution was monitored using a UVdetector at λ = 254 nm. Chromatograms are shown in Figure S3.

4.2. Dynamic Light Scattering (DLS)

All the DLS experiments were performed on a Wyatt Technology^® DynaPro NanoStar^® using a 1 μL quartz MicroCuvette. The size distribution profiles for both DMPC: SMA-EA 1:1 w/w and DMPC: SMA-EA-DOTA 1:1 w/w polymer-lipid nanodiscs used in this study are reported in Figure 1d.

Figure 1. Characterization of polymers and polymer nanodiscs.

a) Molecular structures of SMA-EA and SMA-EADOTA. b) FT-IR spectra of the starting material (SMAnh) and synthetic polymers. FT-IR results confirm the functionalization of the starting material (SMAnh in dark gray) and similarities among SMA-EA (in red) and SMA-EADOTA (in blue). Full spectra are included in the Supporting Information (Figure S2). c) Dissolution of multilamellar vesicles (MLVs) by SMA-EA (red) and by SMA-EA-DOTA (blue) for a 1:1 lipid:polymer weight ratio, d) Tolerance of DMPC:SMA-EA 1:1 w/w and DMPC: SMA-EA-DOTA 1:1 w/w macro-nanodiscs against different metal ions. Size exclusion chromatography (e) and DLS (f) profiles for 1:1 w/w ratio of DMPC:SMA-EA and DMPC:SMA-EA-DOTA samples. The DLS profiles were obtained after SEC purification for the nanodiscs fraction highlighted in (e).

4.3. Static Light Scattering (SLS)

All the SLS experiments were performed using a 4 mL cuvette (1 cm optical path) under continuous stirring at 25°C on a FluoroMax-4^® Spectrofluorometer from Horiba Scientific^®. The excitation wavelength was set at 400 nm while the emission wavelength was set at 404 nm, and the slit was set to 2 nm.

a). Solubilization Experiments

The time-dependent solubilization of the DMPC suspension in 10 mM HEPES buffer 50 mM NaCI was monitored by the intensity of scattered light at a 90° angle. The solubilization power of two different polymers SMA-EA and SMA-EA-DOTA was tested on a 1 mg/mL DMPC MLVs solution. The amount of polymer added was equivalent (1:1 w/w ratio) for all of them. Data are shown in Figure 1c.

b). Metal ion titrations

Polymer-based lipid nanodiscs tolerance toward the investigated cations was tested by titrating a 1 mg/mL solution of DMPC: copolymer 1:1 w/w nanodiscs in 10 mM HEPES buffer at pH = 7.40 with a 4 M solution of each metal. The results are shown in Figure 1b.

4.4. NMR Spectroscopy

Solution NMR experiments were performed at 11.75 T on a 500 MHz Bruker Avance III HD spectrometer. NMR samples were prepared according to the procedure described in the “Nanodisc Formation” section and adding the desired concentration of metal ions (0, 5 × 10⁻⁴, 2.5 × 10⁻², 5 × 10⁻², 0.125, 0.250, 0.500, 1.250, and 2.500 mM), then lyophilized for 24 hours prior to resuspension in 600μL of D₂O and then transferred to 5 mm Norrell^® Sample Vault Series™ glass tubes and placed in a commercial 5 mm quadruple resonance ²H/¹H/¹⁵N/¹³C Bruker round-coil TXI™ 500 SB probe. The experiments were performed in D₂O at neutral pH at three different temperatures, 15, 25, and 35°C. Each sample was made using 4 mg of lipids, and an equal amount (by weight) of SMA-EA-DOTA polymer, titrated at pH 7.4 to obtain a DMPC: SMA-EA-DOTA (1:1 w/w) system. Each concentration of each paramagnetic metal investigated was prepared individually and tested on three sequential different experiments.

a). ¹H-NMR

¹H spectra were recorded by collecting 32 scans with a spectral width of 25 ppm. The transmitter frequency offset was set at 4.7 ppm. Acquisition time and relaxation delay were respectively set at 0.8 s and 1.0 s. ¹H-NMR spectra for DMPC:SMA-EA 1:1 (w/w) and DMPC:SMA-EADOTA 1:1 (w/w) are shown in Figure 3c.

Figure 3. NMR spectra of nanodiscs in the presence of Gd³⁺ ions.

a) Schematic representation of a macro-nanodisc, b) Molecular structure of l,2-dimyristoyl-sn-glycero-3-phosphocholine (DMPC). c) ¹H NMR spectra of 1:1 w/w DMPC:SMA-EA (red) and DMPC:SMA-EA-DOTA (blue) macro-nanodiscs, d) ¹H NMR spectra of 1:1 w/w DMPC: SMA-EA (red) and SMA-EA-DOTA (blue) nanodiscs titrated with the indicated amount of Gd³⁺ ions. All NMR were obtained using 4 mg of lipids in 10 mM phosphate buffer (pH = 7.4) in 100% D₂O at 35°C.

b). Inversion recovery experiment

To measure T₁, an inversion recovery experiment was performed. Fifteen data points from 0.001 s to 10.0 s were collected by acquiring 8 scans with a spectral width of 10 ppm. The transmitter frequency offset was set at 4.7 ppm. Acquisition time and relaxation delay were respectively set at 0.001 s and 10 s. Data are shown in Figures S6–S11.

c). Data Processing

Data have been processed using both Bruker Topspin™ 3.2 and Mestrelab Research S.L. MestReNova™ software was used to integrate the peaks of interest.

Results and Discussion

Formation of SMA-EA-DOTA lipid-nanodiscs

SMA-EA-DOTA copolymer was synthesized according to the procedure as briefly described above[74]. Figure 1a shows the chemical structures of SMA-EA and SMA-EA-DOTA co-polymers (la). FT-IR characterization (Figure 1b) confirms the functionalization of SMA-EA and DOTA, as reported previously.[74] The ability to solubilize l,2-dimyristoyl-sn-glycero-3-phosphocholine (DMPC) lipid vesicles (or aggregates) was tested at room temperature (~25 °C). For this study, about 20–25 nm size polymer-nanodiscs formed by the addition of a 1:1 lipids-to-polymer weight ratio (1:1 w/w) was used (Figure 1(e and f)). Data shown in Figure 1c confirms similar solubilization capabilities of SMA-EA and SMA-EA-DOTA copolymers. Since only a small number of DOTA units per polymer chain were introduced, both SMA-EA and SMA-EA-DOTA have comparable nanodiscs forming capabilities as shown by the experimental results in Figure 1. Additionally, dynamic light scattering (DLS) profiles (Figure 1f) after size-exclusion chromatography (SEC) purification (Figure 1e) confirm the formation of nanodiscs and their isolation. These large-size nanodiscs (>20 nm) are also known as "macro-nanodiscs" which can be aligned in the presence of an external magnetic field as demonstrated in the previous studies and are useful for solid-state NMR studies.[76]

SMA-EA-DOTA nanodiscs are more stable than SMA-EA nanodiscs in the presence of metals

The tolerance of the polymer nanodiscs toward metal ions was examined using static light scattering (SLS) experiments as described above. The SLS experimental results shown in Figure 1d demonstrate the enhanced-tolerance of DOTA containing polymer nanodiscs in the presence of metal ions. In addition to the above-mentioned FT-IR data (Figure 1), the SLS results confirm the successful functionalization of SMA-EA with DOTA. Because of the presence of DOTA groups in SMA-EA-DOTA, the high binding affinity for various lanthanides renders a near complete chelation and thus the high tolerance observed in the presence of various metal ions as shown in Figure 1d[86,87].

Figure 2 shows a schematic representation of the chelation of paramagnetic cations to SMA-EADOTA co-polymer nanodiscs and its use for rapid NMR data acquisition. A first successful application of Cu²⁺ based PRE-NMR with this polymer system was recently reported from our research group which also probed the interaction between polymer nanodiscs and G-quadruplex[74], In this study, the use of paramagnetic properties of five additional trivalent cations from elements of the f-block of the periodic table (Eu³⁺, Gd³⁺, Dy³⁺, Er³⁺, Yb³⁺) is investigated by measuring the spin-lattice (T_l) relaxation times of protons for various concentrations of the metal ions.

Figure 2. Use of paramagnetic metal-chelated polymers to speed-up NMR data acquisition.

A schematic representation of how SMA-EA-DOTA copolymer-based nanodiscs chelated with paramagnetic metals can be used to reduce T₁ relaxation times and to shorten the recycle delay between the successive scans in NMR data acquisition.

Based on the assignment of NMR peaks as reported previously[74,88], it is possible to assign the ¹H peaks observed from lipid and polymer components of nanodiscs: 1 ppm for protons from the lipid acyl (-CH₃), 1.4 ppm for protons from the lipid acyl (-CH₂-), 3.3 ppm for protons from the lipid head quaternary ammonium (-CH₃, γ), and 7.3 ppm for protons from the aromatic styrene group of the polymer. The structure of DMPC and peak assignment of ¹H NMR spectra for both DMPC:SMA-EA and DMPC:SMA-EADOTA polymer-based nanodiscs are shown in Figure 3c. The various paramagnetic metals used in the present study (Eu³⁺, Gd³⁺, Dy³⁺, Er³⁺, Yb³⁺) were chosen among the 14 f-block elements to be representative of the magnetic differences reported in the literature[50,82,84,88,89].

Figure 3d shows ¹H NMR spectra of SMA-EA and SMA-EA-DOTA nanodiscs containing DMPC lipid; and the indicated concentration of Gd³⁺ ions. Here, using two different types of nanodisc systems (with anc without the DOTA) of comparable sizes (Figure 1e and 1f), one noticeable difference we observed was that significant line-broadening occurred at a much lower Gd³⁺ ions concentration (0.25 mM) for the SMA-E/⁵ polymer nanodiscs as compared to that observed for SMA-EA-DOTA. The severe line-broadening observec for the lipid headgroup’s methyl protons of SMA-EA polymer nanodiscs is most likely due to the presence ol a higher population of free Gd³⁺ ions that directly bind to the zwitterionic lipid headgroup. On the othei hand, the SMA-EA-DOTA polymer containing nanodiscs attract the Gd³⁺ ions from the sample to be chelated to the DOTA unit on the polymer belt of the nanodisc and avoid the line-broadening effects. The inversion-recovery NMR experiments were performed to measure the T₁ values of protons in order tc examine the effect of the added paramagnetic metal ions in the sample (see Materials and Methods). The T₁ values were measured by integrating the observed ¹H NMR peaks of interest and then fitting the experimentally measured data to the following equation:

$${\mathrm{M}}_{\mathrm{z}}(\mathrm{\tau })={\mathrm{M}}_{\mathrm{z},\mathrm{eq}}(1-2{\mathrm{e}}^{-\mathrm{\tau }/{\mathrm{T}}_1})$$ (1)

Where M_z(τ) is the z-component of the magnetization dependent on time τ, M_z,eq is the z-magnetization al thermal equilibrium, τ is the time between the 180° and 90° pulses, and T_l is the spin-lattice relaxatior time.

Figure 4 shows the fitting curves obtained for the 1:1 w/w DMPC:SMA-EA-DOTA nanodiscs sample foi various concentrations of Gd³⁺ ions. These experiments were also carried out for the other paramagnetic metal ions, and the results are included in the Supporting Information (Figures S6–S11).

Figure 4. Measurement of T₁ for protons.

Spin-inversion NMR experimental data obtained from 1:1 w/w DMPC:SMA-EA-DOTA nanodiscs to determine Tl values of protons for varying concentrations of Gd³⁺ as indicated. Equation 1 was used to obtain the best-fitting values given in Table S2.

The resulting T₁ of the four-aforementioned peaks (aromatic, γ, methylene and methyl groups) are plotted against [Gd³⁺] for both DMPC:SMA-EA (1:1 w/w) and DMPC:SMA-EA-DOTA (1:1 w/w) polymer-nanodiscs samples in Figures S6–S7. Figure 5 shows the T₁ values for the four ¹H peaks of interest obtained for the different metal ions. By comparing the different lanthanide trivalent cations ([Ln³⁺]) at the same molar concentration, different effects are noted. When [Ln³⁺] = 0.5 mM, Gd³⁺, followed by Dy³⁺, showed the largest T_rreduction for all of the peaks as compared to a reference diamagnetic system made up of DMPC:SMA-EA-DOTA 1:1 (w/w). The aromatic peak from the styrene fraction of the polymer is the most affected by the paramagnetic ions due to the proximity of the styrene group to the DOTA uniťs chelated metals on the nanodiscs belt. Gd³⁺ and Dy³⁺ ions show T₁ times reduced respectively by 1.9- and 1.3-times. Even though Gd³⁺ ions provided the greatest T₁ relaxation effect at the lowest concentration among all the investigated metals, other metal ions, like Dy³⁴, had little to no significant line-broadening effect up to 1.25 mM (Figure S8 and Table S3). Metal ions such as Er³⁺, Eu³⁺, and Yb³⁺ showed significant line-broadening at the highest concentration investigated (2.5 mM) but no effects on T₁-reduction were observed (Figures S7Sll and Tables S2–S6). Data for [Er³⁺] = 2.5 mM are not shown because the sample showed instability.

Figure 5. Efficiency of paramagnetic metal ions in shortening T₁.

T₁ comparison among the different investigated metal ions at [Ln³⁺] = 0.5 mM. The black bar represents the data obtained from a control sample, 1:1 (w/w) DMPC: SMA-EA-DOTA without paramagnetic metal ions. Data shown for ¹H peaks of: a) styrene/aromatic group, b) y methyl groups in the quaternary ammonium of DMPC, c) the methylene groups from the acyl chains (C4-C13) of DMPC, and d) the terminal acyl-chain methyl groups (C14) of DMPC. Each of the NMR samples used in these measurements consisted of 4 mg of DMPC in 10 mM Phosphate buffer pH = 7.4 in D₂O.

NMR peaks from the DMPC lipid are also T₁ enhanced by the presence of the paramagnetic tags on the belt but to a lesser extent, especially for Eu³⁺, Er³⁺, and Yb³⁺ ions (Figure 5). Particularly Gd³⁺ ions shorten T₁ values of protons from γ, methylene (C4-C13), and methyl-groups (C14) by respectively ~48%, ~40%, ~10%, and ~12% respectively. While line-broadening was observed for Gd³⁺ ions, the effects observed for other metals at 0.5 mM are on the following order: Eu³⁺ < Yb³⁺ < Er³⁺ < Dy³⁺. Eu³⁺ and Yb³⁺ do not show any T₁-reduction. Er³⁺ ions showed an important paramagnetic relaxation enhancement on the aromatic peak from styrene groups (T₁-reduction of ~35% at a [Er³⁺] = 0.5 mM) but weak-to-negligible effects for y, methylene (4–13), and methyl (14) peaks. The fact that DMPC’s acyl chains are less affected by the paramagnetic tags may be attributed to the large size of the nanodisc investigated) ~20 nm. We expect the PRE effect to be higher for very large macro-nanodiscs under high viscous conditions as they donť tumble fast and therefore the unaveraged ¹H-¹H dipolar couplings should aid the PRE effect from the metals present on the belt. Such results have been observed for large bicelles that align magnetically[90]. Overall, the effect of paramagnetic metals on shortening Ti can be ranked as Gd³⁺> Dy³⁺> Er³⁺> Eu³⁺> Yb³⁺. Figures S7–S11 show the stacking of ID ¹H NMR spectra and the T₁ fitting curves. Tables S1–S6 report the experimentally measured T₁ values.

Because of the large size of the nanodiscs, within the spectral resolution, the observed proton NMR spectra do not show any shift in the observed resonance frequency when the nanodiscs are loaded with any of the lanthanides even for those that are considered as "shifting-agents" such as Eu³⁺, Dy³⁺, Er³⁺[50], Spectra and full titrations with the fitting of the experimental data are included in the Supporting Information (Figure S12).

Conclusions

SMA-based nanodiscs have shown to be a great innovation in biochemistry and biophysics, and are widely used as a membrane-mimicking system to investigate membrane proteins through several techniques. NMR spectroscopy, as a valuable non-disruptive technique, offers structural and dynamic information. Unfortunately, limitations such as its intrinsically low sensitivity result in long acquisition times to enhance the signal-to-noise ratio of NMR spectra. PRE can be used to reduce the acquisition times effectively but involves the introduction of external probes in the system and may cause undesired line-broadening in the spectra, when compared to the diamagnetic counterpart. To overcome these limitations, we have demonstrated the efficiency of SMA-EA-DOTA copolymer in T₁-reduction. This modification of SMA-EA copolymer allows the use of PRE effects in nanodiscs samples. Particularly, the introduction of chelating units that strongly bind paramagnetic metals on the outer belt of the nanodiscs avoids the addition of paramagnetic dopants directly in buffer solutions, in the membrane protein of interest, and in the bilayer components. SMA-EA-DOTA-nanodiscs represent a much less invasive approach toward the preservation of the integrity of the sample, offering PRE effects in a native-like environment.

As demonstrated in this study, this approach can be used to speed up NMR data acquisition (up to ~50%) with minimum-to-no alteration of the spectral quality due to spin-spin relaxation enhancement. A comparison of the effects of different paramagnetic metals shows that Gd³⁺ and Dy³⁺ can be successfully used to shorten T₁, and so, the recycle delay of NMR experiments. We believe that these results can broaden the applications of polymer-nanodiscs in the investigation of membrane proteins in a native-like environment, using both solution and solid-state NMR spectroscopy. Additionally, paramagnetically-labeled nanodiscs can be used for both dynamic nuclear polarization (DNP) NMR and electron paramagnetic resonance (EPR) studies. Both techniques require the presence of paramagnetic tags and could benefit from this improved version of polymer-nanodiscs.

Highlights.

• SMA-EA-DOTA shows effective chelation of a variety of metal ions.

• Polymer chelated paramagnetic lanthanide ions enable T₁-reduction.

• Eu³⁺, Gd³⁺, Dy³⁺, Er³⁺, and Yb³⁺ were investigated for PRE effects.