A low-E magic angle spinning probe for biological solid state NMR at 750 MHz

Abstract

Crossed-coil NMR probes are a useful tool for reducing sample heating for biological solid state NMR. In a crossed-coil probe, the higher frequency (1)H field, which is the primary source of sample heating in conventional probes, is produced by a separate low-inductance resonator. Because a smaller driving voltage is required, the electric field across the sample and the resultant heating is reduced. In this work we describe the development of a magic angle spinning (MAS) solid state NMR probe utilizing a dual resonator. This dual resonator approach, referred to as "low-E," was originally developed to reduce heating in samples of mechanically aligned membranes. The study of inherently dilute systems, such as proteins in lipid bilayers, via MAS techniques requires large sample volumes at high field to obtain spectra with adequate signal-to-noise ratio under physiologically relevant conditions. With the low-E approach, we are able to obtain homogeneous and sufficiently strong radiofrequency fields for both (1)H and (13)C frequencies in a 4mm probe with a (1)H frequency of 750 MHz. The performance of the probe using windowless dipolar recoupling sequences is demonstrated on model compounds as well as membrane-embedded peptides.

Fig. 1

(a) Physical dimensions of the coils (in millimeters) and their integration into a MAS stator using a Teflon coil centering platform; (b) photograph of the probe head. The Teflon coil platform and 2 pairs of leads can be seen at the bottom of the stator. The observe solenoid is made using a variable pitch winding which is not reflected in the drawing

Fig. 2

Schematics of the double-tuned rf matching network. L₀–C₀ forms the ¹H loop gap resonator with the detection solenoid inside (L₁). Inductors L₃ and L₄ (5–10 nH each) represent flexible leads connecting sample coils to the ¹H and ¹³C circuits. C₁, C₂, and C₃ are variable capacitors for, respectively, matching, balancing, and tuning the ¹H LGR. In the low frequency channel, C₅ is used for matching and C₆ and C₇ tuning capacitors are connected to a single tuning rod via a gear mechanism. Retuning to different observe nuclei (e.g. ¹⁵N) is done by replacing a tuning chip, C₈, and a balancing chip, C_7A. A low-voltage ¹H rejection trap, L₂–C₄, is placed at the entry of the ¹³C rf cable.

Fig. 3

a. B₁ homogeneity characteristics for the ₁H (black squares) and ¹³C (gray diamonds) channels as a function of sample length The ¹³C solenoidal coil length, 8.3 mm, is indicated by the dashed vertical line. As expected, homogeneity is high and improves when the sample is confined within the coil. b. Nutation profiles for the ¹H (top) and ¹³C (bottom) channels collected using an adamantane sample. Each peak corresponds to the monitored resonance as a function of the pulse length. (a) ¹H spectra in 0.5 μs increments for a sample length of 6.7 mm and (b) ¹³C spectra in 0.5 μs increments for a sample length of 6.7 mm.

Fig. 3

a. B₁ homogeneity characteristics for the ₁H (black squares) and ¹³C (gray diamonds) channels as a function of sample length The ¹³C solenoidal coil length, 8.3 mm, is indicated by the dashed vertical line. As expected, homogeneity is high and improves when the sample is confined within the coil. b. Nutation profiles for the ¹H (top) and ¹³C (bottom) channels collected using an adamantane sample. Each peak corresponds to the monitored resonance as a function of the pulse length. (a) ¹H spectra in 0.5 μs increments for a sample length of 6.7 mm and (b) ¹³C spectra in 0.5 μs increments for a sample length of 6.7 mm.

Fig. 4

a. Average rf sample heating at three different sample lengths and two salt concentrations. Black filled circles: 20 mM TmDOTP^5- with a 3.7 mm sample length. Black filled squares: 20 mM TmDOTP^5- and 150 mM NaCl with a 3.7 mm sample length. Gray filled symbols correspond to the same solutions with a 6.7 mm sample length, and the empty symbols correspond to the same solutions with an 11.7 mm sample length. Lines are linear fits to the data as a visual guide. Note that because of its high ionic strength, small concentration of TmDOTP^5- still contribute significant rf loss. The rf heating remains under 15 K for all samples, and either decreasing the sample length or the salt concentration further reduces the heating. Average power was varied by keeping the duty cycle constant at 3.8% and varying the presaturation pulse power. A delay of 5 ms between the presaturation pulse and the acquire pulse was used and 256 dummy scans (~ 5 minutes) were run before acquiring to make sure the sample had reached equilibrium temperature. b. Frictional heating due to MAS. The sample temperature was monitored using a sample containing 10% lead nitrate (Pb(NO₃)₂) diluted with NaCl to reduce its density. The airlines were kept at room temperature. The line is a quadratic fit through the origin. The initial temperature drop is due to Joule-Thomson cooling. Heating at 13 kHz MAS reaches 20 K, which is enough to be a problem for biological samples, but it can be mitigated by cooling the airlines.

Fig. 4

a. Average rf sample heating at three different sample lengths and two salt concentrations. Black filled circles: 20 mM TmDOTP^5- with a 3.7 mm sample length. Black filled squares: 20 mM TmDOTP^5- and 150 mM NaCl with a 3.7 mm sample length. Gray filled symbols correspond to the same solutions with a 6.7 mm sample length, and the empty symbols correspond to the same solutions with an 11.7 mm sample length. Lines are linear fits to the data as a visual guide. Note that because of its high ionic strength, small concentration of TmDOTP^5- still contribute significant rf loss. The rf heating remains under 15 K for all samples, and either decreasing the sample length or the salt concentration further reduces the heating. Average power was varied by keeping the duty cycle constant at 3.8% and varying the presaturation pulse power. A delay of 5 ms between the presaturation pulse and the acquire pulse was used and 256 dummy scans (~ 5 minutes) were run before acquiring to make sure the sample had reached equilibrium temperature. b. Frictional heating due to MAS. The sample temperature was monitored using a sample containing 10% lead nitrate (Pb(NO₃)₂) diluted with NaCl to reduce its density. The airlines were kept at room temperature. The line is a quadratic fit through the origin. The initial temperature drop is due to Joule-Thomson cooling. Heating at 13 kHz MAS reaches 20 K, which is enough to be a problem for biological samples, but it can be mitigated by cooling the airlines.

Fig. 5

a. ^NAGLY ¹³C CPMAS spectrum after 8 scans with a sample length of 6.9 mm containing 67.3 mg of glycine spinning at 13 kHz. Inset is a magnification of the noise used in the S/N measurement. This particular spectrum has a S/N measurement of 301. b. S/N for ^NAGLY as a function of sample length. Shown are the S/N normalized for the number of scans (solid circles, left axis) as well as the S/N per unit mass (open squares, right axis), also normalized for the number of scans.

Fig. 5

a. ^NAGLY ¹³C CPMAS spectrum after 8 scans with a sample length of 6.9 mm containing 67.3 mg of glycine spinning at 13 kHz. Inset is a magnification of the noise used in the S/N measurement. This particular spectrum has a S/N measurement of 301. b. S/N for ^NAGLY as a function of sample length. Shown are the S/N normalized for the number of scans (solid circles, left axis) as well as the S/N per unit mass (open squares, right axis), also normalized for the number of scans.

Fig. 6

CPMAS matching condition profile for adamantane at 10 kHz MAS using a square spinlock pulse on both rf channels. The ¹H rf field was held constant at (ω₁/2π) ~ 35 kHz while the ¹³C B₁ field was varied; the signal is graphed as a function of the rf mismatch. The solid line corresponds to a sample length of 3.7 mm; the dashed line corresponds to a sample length of 11.7 mm. The X-axis is the average ¹³C B₁ field minus the average ¹H B₁ field. The longer sample (11.7 mm, dotted line) has broader, weaker matching conditions. This is from the wider range of rf energy the sample is exposed to due to the lower homogeneity of the ¹³C B₁ field. The shorter sample (3.7 mm, solid line) shows excellent agreement with theory with maximal signal at mismatch levels equal to integer values of the spinning speed.

Fig. 7

¹³C CPMAS spectrum of nanocrystalline natural abundance lysozyme.

Fig. 8

A 2D-DQCSA spectrum for *G*AV.

Fig. 9

(top) CPMAS and DQ-filtered spectra for the 21 amino acid peptide KL₄ ¹³C′-enriched at positions L9 and L10 and incorporated into DPPC:POPG lipid vesicles at a peptide:lipid molar ratio >1:50; the signals in the aliphatic region are primarily from the surrounding lipids. (middle) 2D-DQ CSA correlation data for the KL₄ sample along with best fit simulations. Closed and open symbols are data collected at 750 and 500 MHz, respectively; solid and dashed lines correspond to the signal trajectories for the best fit (ϕ,ψ) simulations at the two fields. (bottom) χ² evaluation of simulations with varying ψ while holding ϕ at a value obtained from DQ buildup experiments; solid and dashed lines correspond to fitting of data at 750 and 500 MHz, respectively. Note the improved selectivity of the χ² evaluation at 750 MHz due to the increased CSAs at the higher magnetic field.