Original article
Multiharmonic electron paramagnetic resonance imaging as an innovative approach for in vivo studies

https://doi.org/10.1016/j.freeradbiomed.2020.03.018Get rights and content

Highlights

  • Multiharmonic analysis reveals more details in animal EPR imaging.

  • The first report of using multiharmonic for 3D in vivo EPR imaging.

  • Multiharmonic analysis significantly improves S/N ratio for EPR in vivo imaging.

Abstract

This work is the first report when multiharmonic analysis (MHA) was applied for electron paramagnetic resonance imaging (EPRI) for in vivo applications. Phantom studies were performed for established methodology, and in vivo imaging was conduct as a proof-of-concept. Phantom studies showed at least six times improvement of the signal – to – noise (S/N) ratio. Application MHA for 3D EPR in vivo imaging provides images of spin probe distribution in mouse head. The EPRI, in combination with nitroxide and trityl spin probe, was performed to obtained 3D EPR in vivo images using MHA. For both used spin probes, MHA provided images with better S/N ratio, especially in the case of nitroxide, where projections obtained using conventional CW did not allow for reconstructing reliable data. Trityl radical exhibited high resolution and quality of obtained images after MHA. The MHA methodology allows the selection of a second modulation amplitude even 40 times higher than the natural EPR linewidth of the spin probe without line shape distortion, which highly improves the sensitivity of the acquired signal and allowing for imaging mice regardless of their size in a routine animal experiment.

Introduction

In recent years, electron paramagnetic resonance [1,2] has become an increasingly used technique in biological and biomedical research. Since current EPR spectroscopy may be performed in 3D [3,4] and with the right choice of the spin probe, it is possible to obtain three-dimensional maps of unique tissue micro-environment parameters. For example, the broadening of the EPR linewidth of some spin probes is proportional to the local oxygen concentration [5,6]. The decay in amplitude or changes in the EPR spectrum shape of nitroxide spin probes reflects the tissue redox state [[7], [8], [9], [10], [11]] as well as shifts of their EPR line position, which allows for pH mapping [12]. In vivo EPR imaging (EPRI) became possible by the development of low-frequency EPR tomographs, with a frequency range between 250 MHz and 1 GHz. The use of the lower frequency allows for animal applications, but in vivo experiments distinctly limit the acquisition time and decrease the sensitivity by lowering the Q factor of the resonator. Therefore, one of the biggest challenges in EPR in vivo imaging is to obtain an EPR signal that is sufficient for reliable data analysis, especially when the concentration of paramagnetic centres is low and the acquisition time is limited.

The easiest approach to improve the S/N ratio involves averaging, but it considerably lengthens the acquisition time, which is not desirable in in vivo studies. Continuous wave (CW) EPRI is a conventional, commercially available and the most popular mode of this technique, in which the magnetic field is modulated, often at 100 kHz, and the first-derivative of the absorption spectrum is acquired by phase-sensitive detection at the modulation frequency at a series of magnetic fields. Several ways were presented to overcome the S/N ratio limitation. Increasing the magnetic field modulation is one way to improve S/N, however, at the expense of line broadening. In conventional spectroscopy amplitude modulation should not be greater than one-third of EPR peak – to – peak linewidth (ΔHpp); otherwise, the spectrum is broadened [13]. Another way to improve the resolution and sensitivity of the reconstructed spectral-spatial EPR images involves the application of different amplitude modulations for each projection, instead of a constant one. In this case, amplitude modulation was dependent on gradient strength, which increased with it [14]. By using direct detection of resonances without conventional low-frequency modulation and phase sensitive detection, it was possible to speed up measurements based on a rapid scan (RS) strategy [[15], [16], [17], [18]]. With a rapid scan, the measured data needs to be deconvolved to obtain the absorption spectra, where its derivative would result in similar data from the classic CW technique. It was shown that the S/N ratio from spectra obtained from rapid scans was around an order of magnitude higher than that obtained from the conventional CW technique [19].

To improve the S/N ratio in the CW method, the multiharmonic analysis was presented. In most EPR spectrometers and imagers, the first – harmonic spectrum is measured. For a modulation amplitude smaller than the linewidth, the S/N ratio of the n-th harmonic decreases rapidly with n, but by using an amplitude modulation much higher than the linewidth, the harmonics have a much higher S/N compared to a conventional non-overmodulated spectrum [20,21]. The study performed at X-band using 10 kHz of frequency modulation where the maximum of 45 harmonics were analysed [20,21] shown that by using multiharmonic analysis, it is possible to reconstruct the correct shape of the first - derivative spectra, however, the peak-to-peak linewidths are broadened by up to 10%.

In this work, we applied multiharmonic analysis in 600 MHz EPR tomograph, which is suitable for in vivo research, and 1 kHz frequency modulation was used. For the first time, the dependency between the shape of the EPR spectrum reconstructed using multiharmonics analysis for a range from 1 to 100 multiharmonics was explored. Such an analysis allows characterizing the behavior of the reconstructed signal shape depending on the number of used harmonics, which allows us to determine crucial parameters for MHA and obtaining spectrum without broadening. Additionally, we examine the correlation between the applied modulation amplitude and the number of multiharmonics used on the reconstructed line shape. Finally, we present a proof – of – concept of this technique for EPR imaging in in vivo applications and compare the results with those from conventional CW EPR imaging.

Section snippets

Phantoms

The first phantom (PH1) – a glass tube contained TCNQ salts with N-methyl-pyridine base as an electron donor [N–Me-Py-(TCNQ)2] with a molecular mass of 502.5 g/mol was used [22,23]. PH1 was filled with 0.32 mg of [N–Me-Py-(TCNQ)2] radical salt powder, which corresponds to around 1.2 × 1021 spins/gram. A water solution of 2 ml of 20 μM of nitroxide 3-Carbamoyl-PROXYL (3-Carbamoyl-2,2,5,5-tetramethyl-1-pyrrolidineoxy) (3-CP) purchased from Sigma-Aldrich (St Louis, MI; USA) was used as the second

Spectroscopy study

As a reference for the EPR linewidth and S/N ratio, traditional detection of EPR 1st derivatives was measured. For the [N–Me-Py-(TCNQ)2] sample, the modulation amplitude was 0.05 G, which is less than 25% of ΔHpp. For the 3-CP sample, the modulation amplitude was 0.4 G, which is about 1/3 of the linewidth. For the [N–Me-Py-(TCNQ)2] samples, modulation amplitudes were increased from 0.5 G up to 8 G. For the 3-CP samples, amplitude modulations up to 10 G were used. After the multiharmonic

Discussion

In vivo EPR imaging experiments are usually performed using conventional CW or pulse EPR tomographs. This work is the first report of the application of the EPR tomograph to perform imaging on a live animal using multiharmonic analysis, which so far was used only in the phantom study.

In a conventional CW EPR experiment, whether it is spectroscopy or imaging, only the 1 st derivative EPR signal is measured. In this study, the practical aspects of the application of multiharmonic analysis were

Conclusions

In conclusion, the MHA methodology developed by our team, allows the selection of a second modulation amplitude even 40 times higher than the natural linewidth of the probe without spectral distortion, which highly improves the sensitivity of the acquired signal. The spectroscopy study confirmed that MHA significantly improved the S/N ratio without distorting the signal shape. The improvement depends on the overmodulation factor. A greater effect in the improvement in the S/N of EPR spectra was

Funding

This research was funded by Polish National Science Centre, grant number: UMO-2014/15/B/NZ5/01488 and UMO-2014/15/B/ST4/04946 as well as National Centre for Research and Development - Smart Growth Operation Programme under project POIR.01.01.01-00-0025/15.

Declaration of competing interest

M.B. T.C., P.S. are all current employees of Novilet. Novilet is a spin-off company specializing in the discovery, development, manufacturing of EPR tomographs. M.B. T.C. and P.S. provided technical support to adapt the ERI TM600 scanner to the scope of this work and provided engineering expertise. This does not alter the authors’ adherence to all the policies on sharing data and materials.

References (30)

  • Z. Yu et al.

    Multiharmonic electron paramagnetic resonance for extended samples with both narrow and broad lines

    J. Magn. Reson.

    (2015)
  • S. Stamenković et al.

    In vivo EPR pharmacokinetic evaluation of the redox status and the blood brain barrier permeability in the SOD1G93AALS rat model

    Free Radic. Biol. Med.

    (2017)
  • K. Anzai et al.

    Assessment of ESR-CT imaging by comparison with autoradiography for the distribution of a blood-brain-barrier permeable spin probe

    MC-PROXYL , to rodent brain

    (2003)
  • L.J. Berliner

    In Vivo EPR (ESR)

    (2003)
  • L.J. Berliner

    The evolution of biomedical EPR (ESR)

    Biomed. Spectrosc. Imag.

    (2017)
  • Cited by (11)

    • Noninvasive detection of the endogenous free radical melanin in human skin melanomas using electron paramagnetic resonance (EPR)

      2022, Free Radical Biology and Medicine
      Citation Excerpt :

      We believe that this seminal contribution will inspire scientists in the field. Regarding technological developments, there is clear room for boosting the sensitivity of the method through improvement in instrumentation, for example by developing S-Band (2–3 GHz) bridges compatible with clinical EPR spectrometers, allowing a penetration depth of 2 mm in the skin [41] and/or use of new acquisition modes (i.e., using multiharmonics or rapid-scan), allowing the use of higher modulation amplitude and power [42,43]. Flexible resonators adapted for skin measurements could also be advantageous for minimizing motion artifacts [44].

    • Benchtop X-band electron paramagnetic resonance detection of melanin and Nitroxyl spin probe in zebrafish

      2022, Free Radical Biology and Medicine
      Citation Excerpt :

      Multi-harmonic analysis allows the selection of a modulation amplitude even 40 times higher than the natural linewidth of the probe, without distorting the signal shape, which highly improves the S/N ratio of the acquired signal and thus sensitivity. The improvement depends on the overmodulation factor, and for large overmodulation multiplicity of the EPR linewidth could be up to 7 times better compared to conventional phase-sensitive detection for the same matching and tuning conditions [20]. EPR analyses were performed at different stages of the zebrafish embryo development, starting 2 h from the onset of pigmentation (i.e., at 29 hpf) till 4 days post fertilization (dpf).

    • Low-cost, programmable infusion pump with bolus mode for in-vivo imaging

      2021, HardwareX
      Citation Excerpt :

      ERI is based on the EPR phenomenon and it allows for the mapping of tissue microenvironment parameters such as the level of oxygen (pO2) [7,8], pH [9,10], inorganic phosphate [11] and redox state [12]. The development of the rapid scan ERI makes it possible to acquire images with a very high temporal resolution, in the order of a few seconds [13,14]. The precision and the accuracy of the injected spin probe during the ERI is an important factor for in-vivo studies.

    • Electron paramagnetic (spin) resonance spectroscopy for biomedical nanotechnology

      2023, Analytical Techniques for Biomedical Nanotechnology
    View all citing articles on Scopus
    1

    Both authors contributed equally to this manuscript.

    View full text