Three-dimensional electron spin resonance imaging of endogenous nitric oxide radicals generated in living plants

Electron spin resonance imaging (ESRI) has been developed to detect the spatial contribution of free radicals in recent years, but all of these studies are used in animal systems and almost of them using exogenous spin probes and only a few to study the endogenous free radicals in vivo. However, there is no report about the endogenous free radical three-dimensional (3D) EPRI in plant because of low concentration of endogenous free radicals and low resolution of the L-band ESRI. Recently, we have developed the imaging resolution better than 200 μm which is higher 10 times than L-band (about 1–3 mm) and the sensitivity is higher about 1000 times than that of L-band ESRI. Using this system and spin trapping technique, we studied spatial contribution of the endogenous nitric oxide (NO) radical generation in wheat leaves and got a clear 3D ESRI picture and CT (computed tomography) of NO of a wheat leaf. This imaging picture shows a clear spatial distribution of NO free radicals in the leaf. This is the first 3D ESRI of endogenous NO free radical generated in plant in the world. We have also studied the distributions of NO generation in different plants from different species with different shapes and it is shown that this is a convenient method to study the endogenous free radicals in plants.


INTRODUCTION
Electron spin resonance (ESR) is the most direct and effective technique to detect free radicals, but it can only give the information about the free radical's kind and concentration. Electron spin resonance imaging (ESRI) is a representation of the spatial distribution of the ESR intensity or other parameters in a heterogeneous sample, and it has been developed in recent years, especial ESR instrument operating at low frequencies has made it possible to detect free radical spatial contribution in animals (Fu et al 2014;Kobayashi et al. 2014;Lin et al. 2014;Morosan et al. 2015;Zou and Zhang 2017). Because of the low concentration and short life time of natural free radical in biological bodies, the extraneous spin probes have been used in most ESRI experiments. Nitroxide stable free radicals exhibit varied chemical and biological properties and their biological applications have been greatly expanded over the past few years (Maeda 2012(Maeda , 2013. They can serve as in vivo functional imaging probes that non-invasively report on the oxygen status and redox properties in different tissues, such as tumors (Berliner et al. 1987;Feng et al. 2016;Hou et al. 2005;Mikuni et al. 2004;Matsumoto et al. 2006), heart (Kuppusamy et al. 1994;Zweier et al. 1998), brain (Hiramatsu et al. 1995;Kuppusamy et al. 1995a, b;Ueda et al. 1997;Velayutham et al. 2003), and liver (Yoshimura et al. 1995). It was also used to detect the physical state of plant (Berliner and Fujii 1985).
There are a few studies on the endogenous free radicals generated in vivo in physiological condition using L-band ESRI (Yokoyama et al. 1996(Yokoyama et al. , 2000(Yokoyama et al. , 2002, but no any report about the endogenous free radicals generated in plant using this technique because of low concentration of endogenous free radicals and low resolution of the L-band ESRI. Nitric oxide (NO) is now recognized to play an important physiological role in animals either as a regulator, involved in signal transduction mechanisms, or as a toxic or protective molecule depending on the concentration and the tissue where it is acting (Feng et al. 2016;Gao et al. 2016;Millar and Day 1997;Wendehenne et al. 2001;Wink et al. 1997;Zhao 2015). In plants, although NO research is more recent than in animal, mounting evidences suggest that NO plays important roles in diverse physiological processes such as growth and development (Laxalt et al. 1997), plant disease resistance (Noritake et al. 1996), abiotic stress (Delledonne et al. 1998), and signal transition (Durner et al. 1998).
We have developed ESR spin trapping technique (Zhang et al. 2001;Zhou et al. 1999) to study NO free radicals generated during development (Zhang et al. 2002(Zhang et al. , 2004, ischemia-reperfusion heart (Shen et al. 1998;Zhao et al. 1996a, b), and brain (Zhang et al. 2004a, b), inflammation Li et al. 2000;Zhao et al. 1996a, b), neurodegenerative diseases Zhao 2005) and protective effects of natural antioxidants (Shen et al. 2000). We have also developed this method Xu et al. 2004Xu et al. , 2005 to study the NO free radical generated from plant (Xu and Zhao 2003) and role of endogenous NO burst in the resistance of wheat to stripe rust . We have developed an L-band ESRI system and got a clear picture of baby mouse by spin probe (Wu et al. 2005(Wu et al. , 2007. In this study, we have developed an X-band ESRI system and studied the endogenous NO distribution in wheat leaves and got a clear three-dimensional (3D) picture of NO and CT (computed tomography) of a wheat leaf. We also studied the distributions of NO generation in different plants with different shapes. This is the first 3D ESRI of endogenous NO free radicals in plant until now and can be used in the future study as a strong and convenient technique.

Stable of the ESR signal of NO spin trapped by (MGD) 2 -Fe 21 and (DETC) 2 -Fe 21
Complex Figure 1A shows a standard curve of (MGD) 2 -Fe 2? -NO complex obtained from the reaction of NO solution with (MGD) 2 -Fe 2? . Inset in Fig. 1A is a three-line ESR signal (g = 2.035, aN = 1.27 mT) of (MGD) 2 -Fe 2? -NO obtained after MGD and Fe mixed with authority NO (certain concentrations of NO gas). Figure 1B is a series of ESR signal of (MGD) 2 -Fe 2? -NO with good S/N threelines obtained from a wheat leaf after wheat is cultured in a MGD and Fe 2? solution and the signal is increased continuously until 2 h, then keeps on a plateau, after 4 h, the signal continuously decreased (Fig. 1C) and it can be detected after 24 h. When the wheat leaf is cut off from the wheat, the signal will continuously decrease but be stable during 30 min (Fig. 1D, E) (it takes about 30 min to collect the data for ESRI).
In the presence of NO inhibitors, PTIO (2-phenyl-4, 4,5,5-tetramethylimidazoline-1-oxyl) or NMMA (N Gmonomethyl-L-arginine), the signals decreased (result not shown), indicating that this signal was generated, at least part, by the endogenous NO free radicals from the pathway of L-arginine. Figure 1E shows a decay curve of (DETC) 2 -Fe 2? -NO complex obtained from the wheat leaf absorbed DETC and Fe 2? solution respectively. Inset in Fig. 1E is a ESR spectrum of (DETC) 2 -Fe 2? -NO complex. It is found that the signal of (MGD) 2 -Fe 2? -NO is better and more stable than that of (DETC) 2 -Fe 2? -NO. So (MGD) 2 -Fe was used to spin trap the endogenous NO generated in the plants in this paper.

Imaging resolution
The imaging resolution achieved by this X-band ESRI system was determined by two capillaries containing carbon powder as shown in Fig. 2. The two samples spaced apart at a distance of 500 lm, center to center, and 125 lm, edge to edge. Figure 2A shows the schematic diagram of position and sizes of the detected samples. Figure 2B shows the partial slice images of ESR-CT; Fig. 2C shows the stacked 3D ESR-CT imaging, and Fig. 2D shows the 3D spatial image of the phantom. The image clearly identifies the two samples. This means that the imaging resolution obtained with our developed ESRI system is better than 200 lm.
ESRI of NO free radical generated from the wheat leaf and auricle Figure 3 shows a typical set of ESRI images in the wheat leaf after the absorption of a solution containing Fe 2? and MGD. Figure 3A shows two-dimensional (2D) ESRI images of spin density in the cross section (ZX-plane) of endogenous NO free radical generated in the wheat leaf. Figure 3D shows a 3D spatial image of endogenous NO free radical in a wheat leaf. Figure 3B, C shows partial slice and slice stacked images of ESR-CT from Fig. 3D, respectively. It can be found that the strongest signal of NO free radical is located in the leaf vein. Figure 4 shows a typical set ESRI images in the wheat auricle after the absorption of a solution containing Fe 2? and MGD. Figure 4A shows 2D ESRI images of spin density in the cross section (ZX-plane) of endogenous NO free radical generated in the wheat auricle. Figure 4D shows a 3D spatial image of the NO free radical. Figure 4B, C show partial slice and slice stacked images of ESR-CT from Fig. 4D, respectively. It can be found that the shape of the auricle is different from the leaf but the strongest signal of NO is also located in the leaf vein.
ESRI of NO free radical generated from millet grain and bud Figure 5 shows the 3D ESRI images at different visual angles of the millet grain and bud after absorption of a solution containing Fe 2? and MGD ( Fig. 5A-C), and a partial slice images of ESR-CT (Fig. 5D). 3D ESRI images were successfully reconstructed only for those grain and bud with (MGD) 2 -Fe 2? -NO. There is no signal can be found in the dry grain and the grain incubated in water before sprouting. The NO signal can only be detected in the sprouting grain with spin trapping. It can be found that the strongest signal of NO is located in the bud and the signal of NO is over whole grain.
ESRI imaging of the NO free radicals generated from forsythia young stem Figure 6 shows a typical set ESRI images in the forsythia young stem after the absorption of a solution containing Fe 2? and MGD. Figure 6A shows the 2D ESRI images of spin density in the cross section (ZX-plane) of endogenous NO free radical generated in the forsythia young stem. Figure 6D shows a 3D spatial image of the NO free radical in the forsythia young stem. Figure 6B, C show partial slice and slice stacked images of ESR-CT from Fig. 6D, respectively. It can be found that the shape of the forsythia young stem is different with the wheat leaf and auricle and the strongest signal of NO is located in the best. Inset is the ESR spectrum that represents NO concentration at 500 nmol/L. B ESR spectra of (MGD) 2 -Fe 2? -NO complex generated in live wheat leave. C The accumulation of (MGD) 2 -Fe 2? -NO complex generated in live wheat leave with time. D The decay curve of ESR signal of (MGD) 2 -Fe 2? -NO complex generated in live wheat leave with time. Inset is the ESR spectra of (MGD) 2 -Fe 2? -NO complex generated in live wheat leave with time. E The decay of ESR signal of (DETC) 2 -Fe 2? -NO complex generated in live wheat leave with time, the inset is an ESR spectrum of (DETC) 2 -Fe 2? -NO complex generated in live wheat leave. The ESR conditions: X-band, microwave power 5 mW, Modulation 4G, room temperature Three-dimensional electron spin resonance imaging

METHOD DISCUSSION
In plants, NO serves as a signal in hormonal and defense responses (Dangl 1998;Lancaster 1994;Neill et al. 2003;Stohr and Ullrich 2002). NO also plays an important role in plant growth elongation (Gouvea et al. 1997;Leshem and Haramaty 1996), and activation of some enzymes for plant cell differentiation (Ferrer and Ros Bareclo 1999). There are evidences about the regulation of NO on plant maturation and senescence (Leshem et al. 1998). NO treatment inhibits the production of ethylene which inducing plant ripening and aging (Leshem et al. 1998). NO induces the production of phytoalexins which promote plant growth and inhibit senescence (Millar and Day 1997). NO also plays an important role in activities of pathogen-related (PR) protein-phenylalanine ammonia-lyase (PAL) and restriction of multiple pathogens. NO functions as a signal in plant disease resistance for hypersensitive necrotic reaction (Van Camp et al. 1998), production of phytoalexins (Chandok et al. 2004), expression of defence gene (Zeidler et al. 2004), increase of chlorophyll (Leshem and Haramaty 1996). To understand the roles and mechanisms of NO in the plant, it is first important to know where NO is produced and contributed in plant.
To examine the location and contribution of NO in biological system, the available techniques without invasiveness used in practice are EPR imaging and fluorescence and luminal-chemiluminescence imaging (Leshem and Pinchasov 2000). Nagano and Yoshimura reviewed the background and current status of imaging techniques for approach on biological NO, i.e., bioimaging of NO, with emphasis on the imaging techniques that utilize EPR spectrometry and fluorometry (Nagano and Yoshimura 2002).
EPR imaging is considered to be the most effective technique available for non-invasive observation of the spatial distribution of free radicals. Over the past decade, EPR spectroscopic methods combined with NOtrapping techniques have emerged as important tools and have provided much important information about the roles of NO in biological system. The organs or tissues with a locally high, pathological level of NO can be targets, so that this method was used to detect the NO generation in pathophysiological conditions involving the overproduction of NO, such as from the head and abdomen of LPAs stimulated mouse (Suzuki et al. 1999), Fig. 2 Images of phantom for resolution detection. A Schematic diagram of crosssectional view and sizes of the detected samples. B Partial slice images of ESR-CT. C Stacked ESR-CT 3D image. D 3D spatial image of the phantom. Sample: carbon power. Data acquisition parameters: projections 9 9 9, scan time 40 s, sweep width 50 G, microwave power 5 mW, modulation amplitude 2.5 G, gradient field 82 G/cm, room temperature rat brain during ischemia-hypoxia. EPR images from the frozen resected brain were obtained by employing an (DETC) 2 -Fe 2? trap and an EPR imaging system with a microwave frequency of 1.2 GHz (Kuppusamy et al. 1995a(Kuppusamy et al. , b, 1996, the endogenously produced NO in the abdominal region in a LPS-induced sepsis model mouse by using an (DTCS) 2 -Fe 2? complex as an NO trap (Kuppusamy et al. 1994); an imaging of nitric oxide generation during ischaemia was performed with the use of 15 N isotope labeling; it was possible to map the metabolic pathway of this nitric oxide generation (Zweier et al. 1998). However, in vivo EPR imaging is not applicable to physiological levels of NO in plant because of the low concentration of NO generated in the biological system and limitations of currently available L-band EPR instrumentation (low resolution).
There is no any report about the EPRI of animal using X-band instrument, since at conventional X-band (9-10 GHz) frequencies water-rich samples have a high dielectric loss which reduces the Q-factor of the resonant cavity and small size of the cavity, it is difficult to be used in EPRI for animal tissues. In the other hand, these two characters X-band EPR are suitable for plant with lower concentration of water and smaller size of plant tissues.
Fluorescent indicators for bioimaging of NO have been applied to various biological samples to examine production of NO. The most widely used fluorescent indicators for detecting intracellular NO are DAFs and DARs, which can permit the measurement of NO production in various living cells, tissues, and specimen. Leaves and callus of Kalanchoe daigremontiana and Taxus brevifolia were used to investigate NO induced apoptosis in plant cells (Pedroso et al. 2000a). The NO burst preceded a significant increase in nuclear DNA fragmentation and cell death. L-NMMA significantly decreased NO production and apoptosis in both species. Pedroso et al. (2000a) concluded that NO is involved in DNA damage leading to cell death and proposed a potential role of NO as a signal molecule in these plants. Foissner et al. (Pedroso et al. 2000a, b) used 5-diaminofluorescein diacetate (DAF-2 DA), in conjunction with confocal laser scanning microscopy, for in vivo real-time imaging of an elicitor-induced NO burst in tobacco. A growing body of evidence suggests that NO, an important signaling and defense molecule in mammals, plays a key role in activating disease resistance in plants, acting as a signaling molecule and possibly also as a direct antimicrobial agent. The results revealed additional similarities between the plant and the animal hosts responsing to infection. Endogenous NO production Three-dimensional electron spin resonance imaging METHOD was examined in roots by comparing wild-type and mutant seedlings loaded with the permeable NOsensitive dye fluorophore DAF-2 DA. It was found that NO production in plants, which is the strongest at the root tip of young seedlings when roots were treated with 50 lmol/L abscisic acid (ABA), could be inhibited by L-NAME. To address the role of AtNOS1 in stomates, NO production and stomatal closure in epidermal peels from wild-type and Atnos1 plants were compared. ABAinduced NO production (measured as increased fluorescence of the NO-sensitive dye DAF-2 DA) was severely inhibited in Atnos1 guard cells compared with wild-type and rescued (expressing AtNOS1) mutant lines . This technique was also successfully used to study the NO free radicals in endothelial cell (Kojima et al. 1999), smooth muscle cell Kojima et al. 1998a, b), bone marrow stromal cell (Gorbunov et al. 2000) and the regulation of ion channel (Ahern et al. 2000) and process of inflammation (Lopez-Figueroa et al. 2000a), even in mitochondria (Lopez-Figueroa et al. 2000b), ritina (Blute et al. 2000) and bran slices (Kojima et al. 1998a, b) and small animal drosophila (Yermolaieva et al. 2000) and plant cell of leaves and callus. The concentration and resolution limitation of fluorescence imaging for NO is about nmol/L and nm, which is suitable for the study of NO imaging in cell and sub-cell. However, it is difficult to image the special distribution of NO in larger scale about lm-mm such as in the leaves of plant by fluorescence confocal laser microscopy. The X-band EPRI is just suitable to image the NO generated in such scale of plant tissues.
We have studied the generation and spatial contribution of NO free radicals in different plants with different shapes from different species, wheat leaves and auricle, grain and bud, forsythia young stem.

Materials and apparatus
Wheat seeds (Triticum aestivum L. cv. Hanxuan 10) and grain obtained from Northwest Science and Technology University of Agriculture and Forestry were sown in Fig. 4 ESRI images in the wheat auricle after the absorption of Fe 2? and MGD, respectively. A Contour and distribution images of spin density. B Partial slice images of ESR-CT. C Stacked image of ESR-CT. D 3D spatial image of endogenous nitric oxide free radical generated in a wheat auricle. The ESR conditions: X-band, microwave power 2 mW, modulation amplitude 1 G, gradient field 90 G/cm, and room temperature plastic pots and irrigated with Hoagland's solution. The seeds were allowed to germinate and develop in the greenhouse. The growth chamber was set at a relative humidity of 60%-80%, day/night temperatures of 25/20°C, 14-h photoperiod (600 mol quanta/m 2 Ás PAR) supplied by fluorescent lamps and 10-h dark. Sixday-old seedlings and germinating grains were used as experimental materials. ESR experiments were performed on a BRUKER 200D-SRC spectrometer (Germany). The (DETC) 2 -Fe 2? -NO and (MGD) 2 -Fe 2? -NO complex both in solution and in intact leaves was measured, respectively, in a 2.5-mm internal diameter quartz tube at 25°C. The conditions for measurement were X-band, 100 kHz modulation with 3.2 G amplitude, microwave power 20 mW, central magnetic field 3385 G, scan width 400 G, time constant 0.3 s, scan time 4 min. The whole height of triplet hyperfine structure, namely the peak at g = 2.035 and C Stacked image of ESR-CT. D 3D spatial image of endogenous nitric oxide free radical generated in a forsythia young stem. The ESR conditions: X-band frequency, microwave power 2 mW, modulation amplitude 1 G, gradient field 90 G/cm, room temperature Three-dimensional electron spin resonance imaging METHOD its two neighbor peaks, was taken as relative intensity of the ESR signal.

Spin trapping NO in plants
Six-day-old wheat seedlings were treated according to following procedure; seedlings were transferred to a culture medium containing 15 mmol/L FeSO 4 and 40 mmol/ L MGD, but for DETC, it was brushed on the leaves. 24 h later, the leaf discs or auricle were inserted directly into a quartz tube to determine NO on an ESR spectrometer. The forsythia young stem was inserted directly into a culture medium containing 15 mmol/L FeSO 4 , and 40 mmol/L MGD solution for 5 h then transferred into a quartz tube to determine NO on an ESR spectrometer. In the experiment germinating of grain, the grain was incubated at the temperatures 18°C and then the germinating grains were transferred to a culture medium containing 15 mmol/L FeSO 4 , and 40 mmol/L MGD solution. 7 h later, the grain was inserted directly into a quartz tube to determine NO on an ESR spectrometer.

ESR imaging system
The X-band ESR Imaging (ESRI) system developed at our laboratory includes a set of X-band EPR spectrometer modified Varian type E-109, a pair of three-dimensional gradient magnetic field coils, and a PC microcomputerbased data system. The operation frequency range of the EPR spectrometer is at 8.8-9.6 GHz, using a rectangular TE 102 mode cavity with 100 kHz field modulation. The 3D gradient coils were plate form coils that processed using whole copper plates instead of the wound with copper wires, which made its structure so compact that it was much thinner and smaller comparing to those traditionally used in ESRI. A maximum gradient strength of up to 9 mT/cm could be obtained with driving current of about 60A in each dimension coil. The spatial linearity was better than 3% in all three dimensions within a cubic volume range of 10 mm 9 10 mm 9 10 mm.
Projection data acquisition and subsequent image reconstruction were performed using PC computer equipped a data acquisition card with D/A and A/D converters of 12 bit. A software package of application programs was written with Matlab which handles autoprocessing of spectra data and reconstruction of ESRI spatial imaging by the filtered back-projection method.

In vivo ESR measurement and ESR-imaging
After 40-min absorption by FeSO 4 and MGD, the leaf was positioned in the TE 102 cavity with its leaf axial along to the Y-direction of gradient field. The ESR spectrum with a zero-field gradient was measured first, a group of total 9 9 9 projection spectra were acquisitioned at gradient strength of 2.7 mT/cm with the gradient direction rotating in an angle increment of 20°. Using 40-s scans, the total time for data acquisition was approximately 60 min. Before image reconstruction, spectra were processed as follows: noise filtering, baseline correcting, double integrating, and resolution enhancement by convolution difference technique. The processed spectra were then used as projection for the 3D image reconstruction using two-stage back-projection reconstruction algorithm. To avoid some artifacts in the image, a filtering process to the image was carried out before plotting. In this process, the standard deviation of the image was subtracted from each element of the image.

Compliance with Ethical Standards
Conflict of interest Yuanlin Cao, Yongsheng Chen, Qian Wan, Jungai Hu, and Baolu Zhao declare that they have no conflict of interest.
Human and animal rights and informed consent This article does not contain any studies with human or animal subjects performed by any of the authors.
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http:// creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.