Abstract: Introduction Parahydrogen induced polarization (PHIP) stands out among various NMR hyperpolarization techniques due to its versatility and cost-effectiveness, offering significant signal enhancement for NMR spectroscopy. Unlike dynamic nuclear polarization (DNP), PHIP utilizes parahydrogen (pH2), a form of H2 with zero nuclear spin, to transfer nonequilibrium spin order to the target substrate magnetization. In the SABRE (Signal Amplification By Reversible Exchange) approach, this conversion occurs during the reversible coordination of pH2 and substrate at specialized organometallic Ir-complexes. Efficient polarization transfer to substrate nuclei can be achieved using radiofrequency (RF) excitation at high magnetic field or low/ultralow (mT/μT) static magnetic field. Despite advancements in high-field SABRE, low and ultralow-field methods were found to be more reliable and efficient, thus preferrable for research. In practice, 13C or 15N nuclei of the substrate molecule are commonly targeted for polarization transfer due to background-free detection and their ability to retain magnetization for minutes, crucial for in vivo studies. Spontaneous polarization transfer from pH2 to these isotopes typically occurs at magnetic field range, significantly lower than Earth's magnetic field (tens of μT). By SHEATH-SABRE (SABRE in shield enables alignment transfer to heteronuclei) it is possible to achieve polarization transfer within a specialized magnetic shield, enabling to control magnetic fields below microtesla levels. For over a decade, SHEATH-SABRE was the most efficient technique1. However, recently it has been shown that polarization transfer by a weak oscillating transverse field resonant with the heteronuclear spin in a static field of around 50-100 microtesla gives similar or even better results than SHEATH-SABRE2. This approach is very similar to the methods of low irradiation at high field or spin-lock induced crossing- (SLIC-)SABRE, but without the adverse effects of the high field, such as singlet-triplet conversion and the necessity of selective irradiation of the bound and free substrate. LF-SLIC-SABRE has only been used for enhancing polarization in 13C nuclei of pyruvate3. Results and Discution We report the first result of the application of LF-SLIC-SABRE method with substantially high levels of 15N polarization at a field of 100 μT without sophisticated magnetic field shielding. Our simple and easy approach utilizes a homemade device that costs less than 100$ comprising magnetic coils for generating static longitudinal field and oscillating transverse field. We tested two approaches for SLIC-SABRE – with resonant and slightly off-resonant excitation of the heteronuclear spin. We tested new approach with pyridine and biocompatible substrates (metronidazole, tinidazole, secnidazole, and ornidazole) for SABRE. SLIC-SABRE approach reproducibly yields two-fold increase in polarization levels compared to SHEATH-SABRE. To perform SABRE experiments at low field we utilized a self-made device, consisting of two Z coils for creating static field (for adjusting and for gradient compensating) and two X,Y coils for applying oscillating transverse field (Fig. 1A). The coils are hand-wound on a 3D-printed plastic frame. The design of this SLIC-device allows generating static magnetic fields around hundreds of microtesla with good homogeneity and apply oscillating transverse fields of the desired configuration. The current generating the transverse field was produced with an integrated sound card of the desktop computer without any further amplification. This approach ensures the high quality of the applied pulses, although at the cost of their power. Thus, the maximum achievable 1 field was equal to around 10 μT. The polarization of the transverse field was circular, which provided with the additional factor of two in the 1amplitude, as compared to linear polarization. To optimize the SLIC-SABRE experiments we fixed the B0 field in the polarization creation region equal to 97 μT (420 Hz of 15N Larmor frequency) and varied the frequency and the amplitude of the SLIC pulse. To transfer the sample between the SLIC coils above the cryomagnet and the NMR probe, we used a motorized sample shuttling and parahydrogen bubbling setup described previously. Fig. 1. (A) Experimental setup for SLIC-SABRE at low field: (1) automated pH2 bubbling unit, controlled by a standard NMR console, (2) stepper motor for sample shuttling, (3) self-made device for performing LF-SLIC-SABRE, (4 ) Helmholtz coil for the static Z magnetic field, (5) “anti-Helmholtz” coil compensating the Z field gradient, (6) saddle-shaped coils generating the oscillating transverse field, (7) NMR sample. (B) Experimental scheme of LF-SLICSABRE (left). Protocols for the “off-resonant” and “on-resonant” variants of SLIC-SABRE (right). The LF-SLIC-SABRE experiment involved: (1) relaxation at high field, (2) sample transfer to a low field area, (3) bubbling pH2 while irradiating with a transverse field, (4) returning to the high field for spectrum detection using a nonselective RF pulse (Fig. 1B left) There are two ways of harnessing SLIC pulse in SABRE experiments (Fig. 1B right). In the first approach, hereafter referred to as "off-resonant", the pulse is applied with a small detuning from the resonance of the target heteronuclei, which leads to the creation of polarization along the effective field, which has non-zero components both along the static field B0 and the oscillating B1 field. In contrast to high-field experiments, at low magnetic fields the Larmor frequencies of the heteronuclei in the bound and free forms of the substrate coincide, which allows accumulating the polarization along the effective field in the free form. In the second approach, called "on-resonant", the SLIC pulse frequency exactly equals the Larmor frequency and the polarization build-up occurs along the B1 field. By varying the amplitude and frequency of the SLIC excitation to leave the resonance condition, this polarization can be adiabatically rotated coaxially with the static field. This allows preserving the entire created polarization when the sample is transferred to the high field, making the "on-resonant" approach more effective than the "off-resonant". However, the experimental procedure for “off-resonant” SLIC-SABRE is more straightforward, as it requires only a CW pulse, without any amplitude or frequency sweep. As a reference for assessing the efficiency of SLIC-SABRE we measured the magnetic field dependence of 15N polarization in SHEATH-SABRE experiments. As a result, the maximum polarization for protonated pyridine was equal to 10%, while for deuterated compound it was only 7%. In “off-resonant” SLIC method with the nutation frequency of the SLIC pulse equal to 15 Hz the maximum 15N polarization of protonated pyridine increases up to 12.4%. For deuterated pyridine “off-resonant” SLIC performs even better, providing 14% polarization with 5 Hz SLIC pulse. The dependence of the resulting polarization on the SLIC frequency has an antisymmetric shape with two maxima, occurring when the effective field fulfils the LAC condition: ,eff = 2 12. However, one should notice that chemical exchange in SABRE method affects the coherent spin dynamics, shifting the optimal LAC conditions. Utilizing “on-resonance” variant of SLIC method with 25 Hz nutation frequency allowed us to obtain 16.2% polarization for protonated pyridine and 12.2% for deuterated one. As expected, the SLIC frequency dependence in that case has a single maximum, corresponding to zero detuning from the 15N resonance. To sum up, for protonated pyridine both variants of SLIC-SABRE provide higher 15N polarization than SHEATH-SABRE, with “on-resonant” SLIC excitation being advantageous to “off-resonant”. Fig. 2. Dependence of 15N polarization on the frequency 1 field in SLIC-SABRE experiment for ornidazole (ORZ), metronidazole (MTZ), secnidazole (SCZ) and tinidazole (TNZ) for “offresonant” (A) and “on-resonant” (B) SLIC. 15N{1H} NMR spectra of the sample after performing “on-resonant” SLIC-SABRE. The thermal NMR spectrum (red) of 500 mM of 15N labelled pyridine, used as a reference (C). The sample contains 10 mM of each of the substrates, 100 mM DMSO-d6 and 2 mM of SABRE catalyst dissolved in 500 μL of methanold4. We also performed experiments on polarization of 15N nuclei with the equimolar (10mM) mixture of four compounds: metronidazole (MTZ), tinidazole (TNZ), secnidazole (SCZ), and ornidazole (ORZ) on the natural abundance of 15N isotope. These compounds are used as antimicrobial drugs, they all have similar structure and are suitable substrates for SABRE experiments. However, the chemical shifts of the 15N nuclei of these molecules differ in both free and bound forms. Performing SLIC-SABRE experiments at 100 μT field we created remarkably high polarization of 15N nuclei for mixture of them. (Fig. 2). The highest polarization was observed for tinidazole, which can be caused by the better matching conditions between the exchange rate and spin coupling of 15N and hydride protons withing the active complex for this compound. As in the experiments with pyridine, the "on-resonant" variant of the SLIC method (plot B in Fig 3) allowed us to achieve noticeably greater polarization than the "off-resonant" (plot A in Fig.3).

Cite: Kiryutin A. , Kozinenko V. , Yurkovskaya A.
SLIC-SABRE at Earth’s Magnetic fields: Simple and Effective Method of getting Strong 15N NMR Polarization
21nd International School-Conference MAGNETIC RESONANCE AND ITS APPLICATIONS 01-05 Apr 2024