Miho C Emoto1, Hideo Sato-Akaba2, and Hirotada G Fujii1
1Sapporo Medical University, Sapporo, Japan, 2Osaka University, Osaka, Japan
Synopsis
Electron paramagnetic
resonance (EPR) imaging is a noninvasive imaging method for visualizing the brain
redox status using nitroxide compounds as imaging probes. A digital imaging system for three-dimensional
continuous-wave EPR imaging of small animals was developed using a high-speed
analog-to-digital converter, digital-to-analog
converters and
field programmable gate array integrated circuits to improve the signal-to-noise
ratio (SNR) and visualize the precise brain redox status in mice. Compared to
an analog EPR imager, the digital EPR imager obtains a higher SNR of a phantom
and produces images with sufficient quality to create a more accurate brain
redox map of the mouse head.
INTRODUCTION
Electron paramagnetic
resonance (EPR) imaging is a noninvasive imaging method for visualizing brain
redox status using nitroxide compounds as imaging probes. Nitroxides have unique antioxidant properties, and
function as redox-sensitive probes because of redox change from paramagnetic nitroxides
to diamagnetic hydroxylamine that occur in the presence of reactants in the
cellular environment. Nitroxides, such as Tempol or Tempone, can be used as
imaging probes in three-dimensional (3D) EPR imaging studies (1,2), but only a few of
them have been imaged in live mice because of their short life-times in vivo.
Their short life-times result in EPR images with low temporal and spatial resolution.
To overcome these problems and visualize the precise brain redox status in
mice, a digital imaging system for 3D continuous-wave (CW) EPR imaging of small
animals was developed using a high-speed analog-to-digital converter (ADC) and
field programmable gate array (FPGA) integrated circuits (ICs). Compared to an
analog EPR imager, the digital EPR imager is less expensive, has a higher
signal-to-noise ratio (SNR) and captures higher quality images that can be used
to generate more accurate redox maps of a mouse brain.MATERIALS AND
METHODS
EPR
imaging system: A magnetic circuit (X-5253, Hitachi Metals Neomaterial, Ltd., Osaka, Japan) and a multi-coil
parallel-gap resonator for a 750-MHz CW-EPR imager were used with a developed
CW-EPR digital console. This console mainly consisted of a 500-Msps ADC
(AD9434), eight 2-Msps digital-to-analog converters (DAC8811) and inexpensive
FPGAs (Cyclone V). The reflected radio frequency (RF) signal, which
contained the EPR signal, was directly sampled from the resonator using a
band-pass subsampling method. The reflected RF
signal was used to produce control signals that were fed to the resonator for
automatic tuning and matching. To extract the EPR signal from the
reflected RF signal, a field modulation method was applied, since the impedance
mismatch due to slow EPR absorption was cancelled. Data
acquisition through a USB port was controlled using a LabVIEW-based program on a
computer running Mac OS 10.9.5. A filtered back-projection method was used to
reconstruct images from the EPR spectra. Paramagnetic
nitroxide imaging probe:
3-methoxycarbonyl-PROXYL (MCP)
was obtained from NARD Chemicals, Ltd. (Osaka, Japan). Animals: Male C57BL/6 mice aged 5–7
weeks with body weights of 20–25 g were used. Nitroxide probes in saline were
injected by tail vein cannulation into the mice under isoflurane anesthesia.RESULTS AND
DISCUSSION
The developed digital CW-EPR imaging system (FPGA
system) is depicted in Fig. 1. The EPR spectrum of a
nitroxide phantom sample was acquired using both the FPGA system and an analog EPR
system. The SNR of the FPGA system was 1.8 times larger than that of the analog
system. To evaluate the effect of the improved SNR on EPR imaging, EPR images
of the same phantom used for SNR evaluation were taken using both EPR imagers. Figures
2 A and B show the EPR images of the phantom obtained using the analog and FPGA
system, respectively. The intensity of each EPR image was normalized by the
maximum intensity in the respective image. The quality of the image in Fig. 2 B
obtained using the FPGA system is better than that in Fig. 2 A obtained using
the analog system. The image in Fig. 2 A contains more background noise than
the image in Fig. 2 B. Figure 3 shows the EPR images of a mouse head in the sagittal
plane injected with MCP using the FPGA system. The co-registration of the EPR
image and the anatomical image obtained by MRI indicates that MCP is present inside
the brain (Fig. 3 A). The reduction rate map of MCP in the mouse head was also co-registered
to the MRI (Fig. 3 B). The reduction rate of MCP in the mouse brain was
found to be higher than that in other areas of the mouse head. These results
suggest that the developed FPGA system obtains EPR images of mouse heads, which
will allow researchers to obtain more accurate redox maps of the mouse brain.Acknowledgements
This work was supported by a grant from the Japan Society for the Promotion of Science (16K10291).References
1. Sato-Akaba H, Fujii H, et al. Development and testing of a CW-EPR apparatus for imaging of short-lifetime nitroxyl radicals in mouse head. J Magn Reson. 2008;193(2):191-198
2. Emoto MC, Sato-Akaba H, Hirata H, Fujii HG. Dynamic changes in the distribution and time course of blood-brain barrier-permeative nitroxides in the mouse head with EPR imaging: visualization of blood flow in a mouse model of ischemia. Free Radic Biol Med. 2014;74:222-228