Brain Redox Imaging Using Blood—Brain Barrier-Permeable Nitroxide MRI Contrast Agent

Abstract

Reactive oxygen species (ROS) and compromised antioxidant defense may contribute to brain disorders such as stroke, amyotrophic lateral sclerosis, etc. Nitroxides are redox-sensitive paramagnetic contrast agents and antioxidants. The ability of a blood—brain barrier (BBB)-permeable nitroxide, methoxycarbonyl-2,2,5,5-tetramethylpyrrolidine-1-oxyl (MC-P), as a magnetic resonance-imaging (MRI) contrast agent for brain tissue redox imaging was tested. MC-P relaxation in rodent brain was quantified by MRI using a fast Look-Locker T₁-mapping sequence. In the cerebral cortex and thalamus, the MRI signal intensity increased up to 50% after MC-P injection, but increased only by 2.7% when a BBB-impermeable nitroxide, 3CxP (3-carboxy-2,2,5,5,5-tetramethylpyrrolidine-1-oxyl) was used. The maximum concentrations in the thalamus and cerebral cortex after MC-P injection were calculated to be 1.9±0.35 and 3.0±0.50 mmol/L, respectively. These values were consistent with the ex vivo data of brain tissue and blood concentration obtained by electron paramagnetic resonance (EPR) spectroscopy. Also, reduction rates of MC-P were significantly decreased after reperfusion following transient MCAO (middle cerebral artery occlusion), a condition associated with changes in redox status resulting from oxidative damage. These results show the use of BBB-permeable nitroxides as MRI contrast agents and antioxidants to evaluate the role of ROS in neurologic diseases.

Keywords

blood—brain-permeable contrast agent free radical MRI nitroxide redox

Introduction

Increased reactive oxygen species (ROS) and decreased antioxidant defense may contribute to numerous brain disorders such as stroke, amyotrophic lateral sclerosis, Parkinson's disease, and Alzheimer's disease (Halliwell, 2006; Huang et al, 2001; Mariani et al, 2005; Muir et al, 2006; Raichle and Mintun 2006; Zimmermann et al, 2004). Noninvasive assessment of oxidative stress is therefore useful to monitor the role of ROS in brain disorders. For image studies of ROS in vivo, nitroxide compounds have been used as redox-sensitive contrast agents by electron paramagnetic resonance (EPR) imaging (He et al, 2002; Kuppusamy et al, 2002; Yamada et al, 2002).

Nitroxides are nontoxic, stable, organic free radicals having a single unpaired electron and therefore capable of providing MRI contrast by shortening the longitudinal relaxation time (T₁). Although nitroxides have a lower relaxivity compared with conventional T₁ contrast agents such as Gd³⁺ complexes, the volume distribution of nitroxides is significantly greater because of better cell permeability (Hahn et al, 1992; Samuni et al, 2001). The rapid bioreduction from paramagnetic nitroxides to the corresponding diamagnetic hydroxylamines has compromised their utility as contrast agents (Keana and Pou, 1985). The combination of relatively low relaxivity and short life spans precluded the use of nitroxides as MRI contrast agents. However, with improved sensitivity of MRI and shift to higher magnetic fields, recent work has begun to investigate whether the reduction of nitroxides might be useful in obtaining information pertaining to the redox status of tissue. (Hyodo et al, 2006; Matsumoto et al, 2006). Nitroxides can be reduced to the corresponding hydroxylamines by reductants such as ascorbate (Figure 1). Nitroxides were investigated to be used as antioxidants and radioprotectors in preclinical cancer research (Erker et al, 2005; Hahn et al, 1997; Krishna et al, 1996a; Liaw et al, 2005; Metz et al, 2004). In in vivo studies, the ratio of nitroxide/hydroxylamine has been shown to be dependent on the redox status of the tissue (Hahn et al, 2000; Kuppusamy et al, 1996, 2002). These observations suggest that nitroxides participate in cellular redox reactions, and that their relative levels may be indicative of the cellular redox status.

Figure 1

Reversible one-electron reduction/oxidation showing the interconversion and the molecular structure of MC-P. The structure on the left is the radical form of MC-P. The structure on the right is the reduced form (hydroxylamine) of MC-P.

One of the nitroxides, methoxycarbonyl-2,2,5,5-tetramethylpyrrolidine-1-oxyl (MC-P; Figure 1), has been used as a contrast agent for brain imaging in EPR-imaging experiments (Anzai et al, 2006; Lee et al, 2004; Sano et al, 1997; Yokoyama et al, 2002). Anzai et al (2003) confirmed that the nitroxide, MC-P, after intravenous injection, accumulated in the brain tissue as evidenced by autoradiography. These findings suggest that MC-P can pass through the blood—brain barrier (BBB) and give T₁ contrast in brain tissue. Previously, we showed that nitroxides are useful contrast agents for monitoring the redox status of tissue using MRI with improved spatial and temporal resolutions compared to EPR imaging (Hyodo et al, 2006; Matsumoto et al, 2006). Therefore, MC-P can be used as a BBB-permeable MRI contrast agent that can participate in redox reactions. Middle cerebral artery occlusion (MCAO) in rats is generally accepted as a standard model of focal cerebral ischemia (Koizumi et al, 1986; Longa et al, 1989; Nagasawa and Kogure, 1989). After periods of ischemia, on reperfusion with oxygenated buffer solutions, a burst of ROS has been shown to be generated with a propensity to inflict oxidative damage, alter the redox status of the tissue, and shift the equilibrium shown in Figure 1 toward left. To show the possibility of using MC-P in brain redox imaging in neurologic disease model, rat transient MCAO model was employed. Furthermore, T₁ mapping has the potential to quantify the tissue concentration of the contrast agent with the knowledge of the relaxivity of the agent used. Chuang and Koretsky (2006) reported fast T₁ mapping based on the Look-Locker (LL) sequence that can significantly reduce scanning time without compromising on accuracy. This has been shown to have almost the same efficiency as the spin echo inversion recovery method in terms of the signal-to-noise ratio per unit time (Crawley and Henkelman, 1988). Using the LL sequence, we used MRI to study the pharmacokinetics of MC-P concentration in the brain by monitoring T₁ changes before and after MC-P injection. T₁ changes were readily detected throughout the brain at a well-tolerated dose of MC-P. Maximum T₁ reduction occurred at about 3 mins after injection and returned to control by 15 mins. A BBB-impermeable nitroxide, 3CxP, did not change T₁ in the brain. Estimation of the concentration of MC-P using in vitro relaxivities agreed with EPR results from extracted tissue. The results indicate that MC-P goes into the brain and causes reduction of T₁. In addition, the reduction rate of MC-P was significantly decreased in the infarct region of brain after ischemia followed by reperfusion compared with the same region of the sham group. This unfolds the possibility of using this agent for MRI study of the redox status of brain and the changes in it under pathologic conditions.

Materials and methods

Chemicals

MC-P was purchased from Columbia Advanced Science, LC (Baltimore, MD, USA). 3CxP was purchased from Sigma-Aldrich Chemical Co. (St Louis, MO, USA). Deionized water was used for all experiments. Potassium chloride (KCl) was purchased from Hospira Inc. (Lake Forest, IL, USA). Other materials used were of analytical grade.

Animals

Male Wistar rats were supplied by the Frederick Cancer Research Center, Animal Production (Frederick, MD, USA). The rats were 6 to 7 weeks of age at the time of experimentation and were housed three per cage in climate-controlled, circadian rhythm-adjusted rooms and were allowed food and water ad libitum. All in vivo experiments were carried out in compliance with the Guide for the Care and Use of Laboratory Animal Resources (1996), National Research Council, and approved by the National Cancer Institute Animal Care and Use Committee.

Magnetic Resonance Imaging and Pulse Sequence

Magnetic resonance-imaging measurements were performed at a 4.7 T scanner controlled with ParaVision 3.0.1 (Bruker BioSpin MRI GmbH, Rheinstetten, Germany). A series of T₁-weighted spoiled gradient-echo (SPGR; repetition time (TR)=75 ms, echo time (TE)=3 ms, flip-angle (FA)=45°, and N_EX=2) was employed to observe T₁ effects. The scan time for an image set (which included six slices) by the SPGR sequence was 20 s. Other imaging parameters were as follows: image resolution was 256 × 128 zero-filled to 256 × 256 (0.125 mm resolution), field of view (FOV)=3.2 × 3.2 cm, slice thickness=2.0 mm and 0.2 mm gap. The number of slices was six.

For quantification of the dynamic change of T₁ and to estimate the concentration of the contrast agent, three MC-P-injected rats were scanned by a fast LL T₁-mapping sequence (Chuang and Koretsky, 2006). After adjusting higher order shimming, echo spacing, echo symmetry, and B₀ compensation, multislice LL data were acquired by segmented gradient-echo EPI (echo planar imaging) with TR=10 secs, TE 6.7 ms, FA=20°, acquisition interval=400 ms, and number of LL time points=20. To avoid resolution blurring because of the T₂∗ (an intrinsic MRI parameter) of brain, a bandwidth of 250 kHz and 32 echoes in one segment was used to keep the echo train length short. With a matrix size of 64 × 64 (0.4 mm in-plane resolution), a set of 13 1.5-mm slices could be obtained in 20 secs.

Phantoms

Phantoms comprising five or seven tubes, with each tube (4.7 mm inner diameter) containing 0 to 50 mmol/L MC-P in phosphate-buffered saline (PBS) as shown in Figure 2A, were used for measuring the in vitro T₁ relaxivity. The experimental conditions of SPGR and LL sequence were SPGR: TR=75 ms, TE=3 ms, FA=45°, N_EX=8, scan time=160 secs, FOV=3.2 × 3.2 cm, reconstructed image resolution=256 × 256, slice thickness=2.0 mm, and number of slices=6. The LL T₁-mapping data were acquired by conventional gradient-echo MRI with TR=10 secs, TE=2.4 ms, FA=20°, acquisition interval=400 ms, number of time points=20, and matrix size=96 × 96.

Figure 2

MR images of MC-P phantom by SPGR and LL sequences. (A) Schematics of the MC-P phantom (left column), T₁-weighted SPGR images (middle column), and T₁ maps (right column) calculated from LL sequence. The center phantom was filled with PBS. MC-P was dissolved in saline. (B) Change in the MR signal intensity of MC-P phantom as a function of concentration from 0.2 mmol/L to 50 mmol/L. Signal intensity change (%) was calculated on the basis of PBS phantom intensity. (C) Relaxation rate (1/T₁) of MC-P as a function of concentration. The MRI parameters of SPGR and LL sequence are SPGR: TR=75 ms, TE=3 ms, FA=45°, N_EX=8, scan time=160 secs, FOV=3.2 × 3.2 cm, reconstructed image resolution=256 × 256, slice thickness=2.0 mm, and number of slices=6; LL sequence: the LL T₁-mapping data were acquired by conventional gradient-echo with TR=10 secs, TE=2.4 ms, FA=20°, acquisition interval=400 ms, number of time points=20, and matrix size=96 × 96.

Animal Experimental Design

Rats were anesthetized by isoflurane (2.0%) in medical air (700 mL/min) and secured on a special holder by an adhesive skin tape. A breathing sensor (SA Instruments Inc., NY, USA) was placed on the dorsal side of the rat. A nonmagnetic temperature probe (FISO, QC, Canada) was used to monitor the rectal temperature of the rats. The tail vein was cannulated for the injection of MC-P or 3CxP. The rat was then placed in the resonator, which was previously warmed up using a hot-water-cycling pad. The resonator unit including the rat was placed in the magnet bore. The MR measurements were started after the rat's core body temperature stabilized at 37°C. The rat's body temperature was maintained at 37±1°C during experiments. Before making the measurements with MC-P or 3CxP, images of the rat were taken with the parameters mentioned above. T₁-weighted images or T₁ maps were acquired continuously for 13 mins, using SPGR or LL sequence. A solution of MC-P or CxP (1.5 μmol/g body weight) was injected by tail vein cannulation, 2.0 mins after starting the scan. To stop the blood flow immediately, 2 mL of KCl solution (2 mequiv/mL) was injected 40 secs after the administration of MC-P and then continuously measured by SPGR till 13 mins.

Surgical Procedure of Ischemia and Reperfusion in Rat

The intraluminal suture model was used for induction of focal cerebral ischemia, as formerly described by published procedure (Nagasawa and Kogure, 1989). Briefly, rats were anesthetized with isoflurane using a nose cone. Anesthesia induction started with 4% isoflurane carried by medical air and maintained by 2% isoflurane with medical air. Animal core body temperature was maintained at approximately 37°C using a water blanket. A 2 cm midline incision was made on the anterior neck. The right CCA (common carotid artery), ECA (extra carotid artery), and ICA (internal carotid artery) were exposed with careful conservation of the vagus nerve. The CCA and ECA were ligated and a suture was placed at the ICA. To block the origin of the middle cerebral artery, a 4-0 nylon surgical thread was advanced into the ICA via a small incision. The left CCA was ligated with a suture to complete the MCAO. After 1 h of MCAO, normal blood flow was restored in the middle cerebral artery by withdrawing the suture tip and the suture of the left CCA region. The rats were moved to the MR magnet 3 h after reperfusion. Then MRI procedure was performed as described above.

Image Analysis

The MRI data were analyzed using the ImageJ software package (http://rsb.info.nih.gov/ij/). The T₁ of each pixel was calculated in two steps using a custom-written program running in Matlab (The MathWorks Inc., Natick, MA, USA). First, the signal recovery of each pixel was fitted by the Levenburg—Marquardt nonlinear three-parameter curve-fitting algorithm. Then, the obtained parameters were used to resolve the dependency of the fitted T₁ on the FA and acquisition interval to derive the actual T₁ (Chuang and Koretsky, 2006). Semi-logarithmic plots of the time course of MRI signal change in the ROI (region of interest) were used for reduction rate calculation. Region-of-interest analysis was performed on multislice LL data using ImageJ. The concentration of MC-P was estimated by the LL T₁ maps using the relaxivity measured in the phantom study using the following equation:

[MC-P]=(R₁−R₁₀)/Relaxivity.

Statistical analysis was performed using the StatView software Stat View 5.0 (SAS Institute Inc., Cary, NC, USA). All data are presented as mean±s.d.

Quantification of MC-P in Brain and Blood by EPR Spectroscopy

MC-P dissolved in saline was injected in the tail vein of rats at a dose of 0.75 μmol/g body weight (5 μL/g). The concentration in the bolus solution was 150 mmol/L. Blood was collected from the retro-orbital sinus using heparinized tubes. Brain tissue was extracted 2, 5, and 8 mins after injection of MC-P and wet weight was determined. Blood or tissue samples were diluted or homogenized with fourfold volume PBS for blood or threefold volume PBS for brain. The homogenate solution was mixed with ferricyanide solution (2 mmol/L). Ferricyanide quantitatively oxidizes the hydroxylamine produced as a result of in vivo reduction back to the oxidized form (Krishna et al, 1992, 1996a, 1996b). The signal intensities of the 100 μL samples were immediately measured using a Varian E-9 X-band EPR spectrometer. The operating conditions for the EPR spectrometer were as follows: modulation frequency, 100 kHz; microwave power, 1 mW. Because there is linearity up to 2 mmol/L between concentration and EPR signal height of MC-P (data not shown), EPR signal heights of homogenate mixtures were converted to the concentration using previously obtained standard curves with or without ferricyanide solution (100 μmol/L to 2 mmol/L). Finally, MC-P concentrations in the brain tissue and blood were calculated based on the dilution factor.

Results

Figure 2A (middle column) shows T₁-weighted images of phantom with various concentrations of MC-P acquired using the SPGR sequence. The MR signal intensity increased with concentration of MC-P up to 50 mmol/L (Figure 2B), with a linear response up to 5 mmol/L. The intensity change (%) in the 5 mmol/L MC-P tube was about 100% compared with PBS (center tube). The T₁ maps of the phantom by the LL sequence are shown in Figure 2A (right column). The T₁ values decreased depending on the MC-P concentration from 2,246 ms (0.5 mmol/L) to 122 ms (50 mmol/L), and the relaxation rates (1/T₁) increased linearly up to 50 mmol/L (Figure 2). The relaxivity of MC-P in PBS solution was measured to be 0.16 sec⁻¹ mmol/L⁻¹ by the LL sequence.

To study the dynamics of nitroxides in brain, a T₁-weighted SPGR sequence was used. Using this technique, six slices (2 mm slice thickness) were obtained every 20 secs with an image resolution of 125 μm. After the injection of MC-P solution, the intensity in the rat head region immediately increased (green color). The increased signal intensity in the brain tissue was large compared with the facial muscle (Figure 3A). The maximum change in the signal intensity in the brain tissue after injection of MC-P was about 50% compared with the pre-injection image (Figure 3C). The signal intensity derived from MC-P (oxidized form, MRI visible) reached a maximum at 30 secs after injection, and the signal intensity decreased during the subsequent 6 mins. On the other hand, 3CxP (cell-impermeable) didn't cause significant signal changes in the brain region although it has a similar in vitro relaxivity (0.17 sec⁻¹ mmol/L⁻¹) as MC-P (Figure 3B). The signal intensity in the cerebral cortex and thalamus increased by only 2.7% during the measurement time, and the increased signal intensity plateaued for up to 10 mins after injection of 3CxP (Figure 3D). To check if blood flow affected MC-P reduction, blood flow was stopped by KCl injection 40 secs after administration of MC-P. Respiration stopped (rats died) 15 to 20 secs after KCl injection. The MR signal intensity in brain decreased after enhancement by MC-P in a manner similar to that found for brain but slower than in animals with normal cerebral perfusion (Figure 3E). The MC-P reduction rate (k=0.30±0.066 min⁻¹) in brain without flow was slower than in living rat brain (k=0.50±0.039 min⁻¹), indicating that a fraction of the reduction of intensity enhancement might have been because of washout from the brain. Alternatively, a change in the redox state of the tissue on death could explain these differences.

Figure 3

Time-course SPGR MR images of rat head region after injection of (A) MC-P (cell-permeable) and (B) 3CxP (cell-impermeable). Contrast agents were injected 2 mins after the MR scan was started. Sixty serial images were obtained in 20 mins. T₂-weighted spin echo images (MSME image) were obtained before the SPGR scan. The SPGR MR parameters were as follows: image resolution was 256 × 128 zero-filled to 256 × 256 (0.125 mm resolution), FOV=3.2 × 3.2 cm, slice thickness=2.0 mm and 0.2 mm gap. The number of slices was six. The green color shows MR signal intensity increased percentage (%) from the pixel of the pre-injection image. The time courses of intensity change of (C) MC-P and (D) 3CxP in the ROI of cerebral cortex (red color) and thalamus (blue color) are shown. (E) Intensity change of MC-P without blood flow is shown. KCl (2 mL) was injected 40 secs after MC-P injection and rats died within 20 secs.

In the ischemia—reperfusion (IR)-treated group, all rats showed typical symptoms during ischemia, and all rats were alive after IR. The T₂-weighted MR images in the IR group did not show any differences at this time (Figure 4A). However, significantly high intensity regions were observed 24 h after IR (data not shown). The reduction rate of MC-P in the right hemisphere was significantly decreased compared with the same region of sham rat brain (Figure 4C).

Figure 4

Redox MR imaging of rat brain after IR. (A) T₂-weighted MR image and T₁-weighted MR image of sham and IR treated rats after MC-P injection. (B) Intensity change of right hemisphere in the T₁-weighted MR images after MC-P injection. (C) Reduction rate of right hemisphere in sham- and IR-treated rats. The data were averaged for four animals. Error bars are ±s.d. ∗P<0.01, the significant difference between sham and IR rats by t-test.

Figure 5A shows the time course of quantitative T₁ maps before and after injection of MC-P. T₁ map data were obtained every 20 secs with 13 slices (1.5 mm slice thickness) by the EPI—LL sequence. The T₁ maps clearly showed the difference of relaxation time T₁ between the cerebral cortex and thalamus. The relaxation times in cerebral cortex and thalamus before injection were 1,577±19.9 and 1,315±11.9 ms, respectively (Figure 5B). The T₁ of these regions rapidly decreased after injection of MC-P, and the minimum T₁ (secs) of the cerebral cortex and thalamus region were 1,034±70.4 and 779±70.2 ms, respectively. The recovery of T₁ in the cerebral cortex and thalamus showed similar trends. From T₁ and in vitro relaxivity of MC-P (0.16 mmol/L⁻¹s⁻¹), the concentration of MC-P at all time points in the ROI was calculated (Figure 5C). The maximum concentration in the cerebral cortex and thalamus 30 secs after injection was 1.9±0.35 and 3.0±0.50 mmol/L, respectively. The concentration was significantly higher in the thalamus compared with the cerebral cortex. This difference in concentration was observed up to 2 mins after injection, after which the concentration in those regions converged.

Figure 5

Dynamic T₁ maps of rat head region after MC-P injection acquired by LL sequence. (A) Time-course T₁ maps of rat head region after MC-P injection. Multislice LL data were acquired every 20 secs by segmented EPI with TR=10 secs, TE 6.7 ms, FA=20°, acquisition interval=400 ms, and number of time points=20. Slice thickness was 1.5 mm. The changes in (B) T₁ and (C) MC-P concentration in the cerebral cortex and thalamus regions were shown. The MC-P concentration was calculated from T₁ using the equation described in the Materials and methods section. Error bar is ±s.d. (the number of experiments was three). ∗P<0.01, the significant difference of maximum concentration of MC-P between cerebral cortex and thalamus by t-test.

To confirm whether the concentration of MC-P determined by the MRI technique is reliable, an ex vivo study using X-band EPR was carried out on the brain tissue for MC-P levels (Figure 6). The oxidized and total MC-P (oxidized+reduced form) levels were determined with and without ferricyanide solution, respectively. The brain tissue MC-P level was 0.84±0.035 mmol/L at 2 mins after MC-P injection (Figure 6A). This was in agreement with the MRI data in the cerebral cortex (0.83±0.107 mmol/L) and in the thalamus (0.94±0.086 mmol/L). On the other hand, total MC-P level (oxidized+reduced form) was 1.57±0.199 mmol/L, suggesting that approximately half of MC-P was in the reduced form in the brain at this time point. At 8 mins, most of MC-P was reduced (0.24±0.107 mmol/L); however, total MC-P (reduced+oxidized) still remained (0.92±0.092 mmol/L).

Figure 6

The quantification of oxidized and total (oxidized+reduced) MC-P in brain tissue and blood using X-band EPR. (A) The concentration change in oxidized (white bar) and total (black bar) MC-P in brain tissue (N=4). (B) Time-course change in the concentration of oxidized Hillered et al (1988) and total MC-P (brown) in the blood. The gray color shows concentration change from the MRI, as in Figure 5C. Blood was collected from eyeground using heparinized capillary (N=6).

The blood concentrations of MC-P at different time points in six rats are shown in Figure 6B. The oxidized form of MC-P followed similar kinetics as the brain, peaking at similar concentration and falling off at a similar rate. Total MC-P levels slowly decreased, but MC-P was still present at 30 mins (level of oxidized MC-P: 0.33±0.084 mmol/L, level of total MC-P: 1.04±0.07 mmol/L), suggesting that the clearance of MC-P from the blood was much slower than the reduction detected in the brain.

Discussion

The results of this study show that the cell-permeable and low-molecular-weight BBB-permeable MRI contrast agent, MC-P, is useful for enhancing MRI intensity in the brain. Furthermore, dynamic contrast-enhancement study with LL sequence allowed us to get rapid T₁ maps from the brain region before and after MC-P injection. This enabled determination of the MC-P concentration in brain regions assuming the in vivo relaxivity of MC-P as the same in vivo.

Previously, we reported that nitroxide radicals are useful in monitoring the redox status of tumor (Hyodo et al, 2006; Matsumoto et al, 2006). In this study, the cell-permeable and highly lipophilic nitroxide, MC-P, was tested in the brain as an imaging probe. The T₁-weighted MR signal intensity of MC-P showed linearity up to 5 mmol/L (Figure 2B), although it increased up to 50 mmol/L. The relaxation rate (1/T₁) was linear up to 50 mmol/L (Figure 2C). Because the MC-P concentration in the thalamus region went up to approximately 3 mmol/L, EPR spectral line broadening effect might have compromised the accuracy of quantification as probes in EPR and Overhauser MRI, the two commonly used techniques to monitor nitroxides. However, MRI is not affected by the EPR spectral line broadening; therefore, it can provide more accurate information on the concentration.

In the animal studies, MRI signal intensities in brain regions clearly increased up to 50% after injection of MC-P, although there was no enhancement (∼3%) in the case of the cell- and BBB-impermeable nitroxide, 3CxP. Both agents have one unpaired electron, their molecular structures are very similar, and their relaxivities are also similar (0.16 mmol/L⁻¹ sec⁻¹ for MC-P; 0.17 mmol/L⁻¹ sec⁻¹ for 3CxP). Thus, the difference detected is most likely because of the distribution of MC-P in the brain. MC-P is a highly lipophilic substance with an octanol—water partition coefficient (Po/w) of 14.4 (Miura et al, 1997). In addition to lipophilicity, its low molecular weight (200) allows it to pass through the BBB. Therefore, SPGR signal intensity in the brain region increased after MC-P injection (Figure 3A). Conversely, Po/w of 3CxP is 0.002, and it is not cell-permeable except by active anion transport (Ichikawa et al, 2006). As a result, the MR signal intensity in the brain increased by only 2.7%. The ratio of maximum intensity change between MC-P and 3CxP was 18.5. This ratio was very similar to the blood volume in the brain. The brain blood value was about 5 to 10% in the whole brain and its ratio to tissue was about 10 to 20 (Ladurner et al, 1976; Mathew et al, 1972; Rostrup et al, 2005). These data suggest that MC-P got distributed in the whole-brain tissue region after passing through the BBB.

The levels of 3CxP were stable in the blood for up to 13 mins. Our previous paper showed that 3CxP displays slow pharmacokinetics compared with cell-permeable contrast agents and its reduction rate was attributed to its membrane permeability (Hyodo et al, 2006). In addition, blood flow may contribute to nitroxide reduction in vivo. Therefore, MC-P reduction rate without blood flow was tested (Figure 3E). Even though blood flow was stopped after injecting KCl solution, MC-P reduced in the brain. The reduction rate of MC-P without blood flow was 60% compared with blood flow, suggesting that MC-P reduction represents reaction with intracellular reductants such as GSH (glutathione), AsA (ascorbic acid), and possibly free radicals in the intracellular compartment. It is also possible that, when an animal died, the redox status of the brain changed in a manner that showed the reduction of MC-P.

Monitoring of T₁ (secs) before and after injection of MC-P using LL sequence allowed the calculation of MC-P concentration in the brain (Figure 5). Thirteen slices of brain T₁ maps with 1.5 mm slice thickness were obtained every 20 secs and monitored continuously up to 13 mins (a representative slice is shown in Figure 5A). The T₁ values in the ROI were converted to MC-P concentration (Figure 5C). The concentration determined by MRI was in agreement with the results measured from brain tissues and the blood concentration of the oxidized form of MC-P as determined ex vivo from EPR (Figure 6). Although there was about 0.5 mmol/L of the oxidized form of MC-P at 10 mins after injection, the total MC-P levels (nitroxide+hydroxylamine) were still around 1.2 mmol/L, suggesting that the reduction from MC-P (oxidized, paramagnetic) to hydroxylamine (reduced, diamagnetic) was predominant compared with clearance and elimination from the brain tissue or blood.

Lowered reduction rate in the infarction area after ischemia and reperfusion were readily seen in the right hemisphere (Figure 4) suggesting that the postischemic reperfusion damage course induced elevated oxidative conditions in infarct region resulting in oxidative damage. Major antioxidants, such as AsA and GSH, are efficient reducing substances in the brain and are found in millimolar concentrations (Rice and Russo-Menna, 1998; Rice et al, 2002). Low levels of these antioxidants have been implicated in neurologic damage and stroke. Under normal physiologic conditions, neurologic damages mediated by ROS can be prevented by the cellular antioxidant network. However, AsA levels decrease in the brain and rise sharply in extracellular fluid of the brain after severe ischemic hypoxia (Hillered et al, 1988; Landolt et al, 1992). The nitroxide radical by itself is also an antioxidant and can be used as a superoxide dismutase mimic (Krishna et al, 1996a). Tempol, which is another cell-permeable nitroxide, has been used in a phase I clinical trial for the prevention of alopecia induced by whole-brain radiotherapy (Metz et al, 2004). Interestingly, in a tumor model, Kuppusamy et al (2002) showed an increased rate of EPR signal decay of 3CxP compared with normal tissue. This decay was decreased by pre-injection of a glutathione synthesis inhibitor, l-buthionine-S,R-sulfoximine (Kuppusamy et al, 2002). These data suggest that the effect of MC-P on MRI signal intensity might enable monitoring of brain damage following alterations in the levels of GSH. As a consequence, the BBB-permeable nitroxide, MC-P, may be useful as an MRI contrast agent to indicate a change in the redox status of the brain. Furthermore, MC-P might provide protection of brain because of tissue antioxidant properties.

Conclusion

The BBB-permeable MC-P is a potent redox-sensitive MRI contrast agent in brain. The MRI intensity in brain was clearly enhanced after injection of MC-P. Furthermore, the real-time T₁ mapping by LL sequence allowed accurate monitoring of pharmacokinetic MC-P concentration changes with whole-brain coverage. Such studies should be useful to evaluate the redox status of brain in oxidative diseases of the brain, such as stroke, Alzheimer's disease, and other neurologic disorders.

References

Anzai

Saito

Takeshita

Takahashi

Miyazaki

Shoji

Lee

Masumizu

Ozawa

(2003) 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. Magn Reson Imaging 21:765–72.

Anzai

Ueno

Yoshida

Furuse

Aung

Nakanishi

Moritake

Takeshita

Ikota

(2006) Comparison of stable nitroxide, 3-substituted 2,2,5,5-tetramethylpyrrolidine-N-oxyls, with respect to protection from radiation, prevention of DNA damage, and distribution in mice. Free Radic Biol Med 40:1170–8.

Chuang

Koretsky

(2006) Improved neuronal tract tracing using manganese enhanced magnetic resonance imaging with fast T(1) mapping. Magn Reson Med 55:604–11.

Crawley

Henkelman

(1988) A comparison of one-shot and recovery methods in T1 imaging. Magn Reson Med 7:23–34.

Erker

Schubert

Yakushiji

Barlow

Larson

Mitchell

Wynshaw-Boris

(2005) Cancer chemo-prevention by the antioxidant tempol acts partially via the p53 tumor suppressor. Hum Mol Genet 14:1699–708.

Hahn

Krishna

DeLuca

Coffin

Mitchell

(2000) Evaluation of the hydroxylamine Tempol-H as an in vivo radioprotector. Free Radic Biol Med 28:953–8.

Hahn

Sullivan

DeLuca

Krishna

Wersto

Venzon

Russo

Mitchell

(1997) Evaluation of tempol radioprotection in a murine tumor model. Free Radic Biol Med 22:1211–6.

Hahn

Wilson

Krishna

Liebmann

DeGraff

Gamson

Samuni

Venzon

Mitchell

(1992) Identification of nitroxide radioprotectors. Radiat Res 132:87–93.

Halliwell

(2006) Oxidative stress and neurodegeneration: where are we now? J Neurochem 97:1634–58.

10.

Samouilov

Kuppusamy

Zweier

(2002) In vivo imaging of free radicals: applications from mouse to man. Mol Cell Biochem 234–235:359–67.

11.

Hillered

Persson

Bolander

Hallstrom

Ungerstedt

(1988) Increased extracellular levels of ascorbate in the striatum after middle cerebral artery occlusion in the rat monitored by intracerebral microdialysis. Neurosci Lett 95:286–90.

12.

Huang

Agus

Winfree

Kiss

Mack

McTaggart

Choudhri

Kim

Mocco

Pinsky

Fox

Israel

Boyd

Golde

Connolly

Jr (2001) Dehydroascorbic acid, a blood–brain barrier transportable form of vitamin C, mediates potent cerebroprotection in experimental stroke. Proc Natl Acad Sci USA 98:11720–4.

13.

Hyodo

Matsumoto

Mitchell

Krishna

(2006) Probing the intracellular redox status of tumors with magnetic resonance imaging and redox-sensitive contrast agents. Cancer Res 66:9921–8.

14.

Ichikawa

Sato

Kondo

Utsumi

(2006) An ESR contrast agent is transported to rat liver through organic anion transporter. Free Radic Res 40:403–8.

15.

Keana

Pou

(1985) Nitroxide-doped liposomes containing entrapped oxidant: an approach to the ‘reduction problem’ of nitroxides as MRI contrast agents. Physiol Chem Phys Med NMR 17:235–40.

16.

Koizumi

Yoshida

Nakazawa

Ooneda

(1986) Experimental studies of ischemic brain edema. I A new experimental model of cerebral embolism in rats in which recirculation can be introduced in the ischemic area, Jpn J Stroke 8:1–8.

17.

Krishna

Grahame

Samuni

Mitchell

Russo

(1992) Oxoammonium cation intermediate in the nitroxide-catalyzed dismutation of superoxide. Proc Natl Acad Sci USA 89:5537–41.

18.

Krishna

Russo

Mitchell

Goldstein

Dafni

Samuni

(1996a) Do nitroxide antioxidants act as scavengers of $ or as SOD mimics? J Biol Chem 271:26026–31.

19.

Krishna

Samuni

Taira

Goldstein

Mitchell

Russo

(1996b) Stimulation by nitroxides of catalase-like activity of hemeproteins. Kinetics and mechanism. J Biol Chem 271:26018–25.

20.

Kuppusamy

Chzhan

Wang

Zweier

(1996) Three-dimensional gated EPR imaging of the beating heart: time-resolved measurements of free radical distribution during the cardiac contractile cycle. Magn Reson Med 35:323–8.

21.

Kuppusamy

Ilangovan

Cardounel

Zweier

Yamada

Krishna

Mitchell

(2002) Noninvasive imaging of tumor redox status and its modification by tissue glutathione levels. Cancer Res 62:307–12.

22.

Ladurner

Zilkha

Iliff

du Boulay

Marshall

(1976) Measurement of regional cerebral blood volume by computerized axial tomography. J Neurol Neurosurg Psychiatry 39:152–8.

23.

Landolt

Lutz

Langemann

Stauble

Mendelo-witsch

Gratzl

Honegger

(1992) Extracellular antioxidants and amino acids in the cortex of the rat:. monitoring by microdialysis of early ischemic changes. J Cereb Blood Flow Metab 12:96–102.

24.

Lee

Shoji

Miyazaki

Yoshino

Hori

Toyoda

Ikeda

Anzai

Ikota

Ozawa

(2004) Assessment of oxidative stress in the spontaneously hypertensive rat brain using electron spin resonance (ESR) imaging and in vivo L-Band ESR. Hypertens Res 27:485–92.

25.

Liaw

Chen

Lai

Chen

Tzao

(2005) Effects of a membrane-permeable radical scavenger, Tempol, on intraperitoneal sepsis-induced organ injury in rats. Shock 23:88–96.

26.

Longa

Weinstein

Carlson

Cummins

(1989) Reversible middle cerebral artery occlusion without craniectomy in rats. Stroke 20:84–91.

27.

Mariani

Polidori

Cherubini

Mecocci

(2005) Oxidative stress in brain aging, neurodegenerative and vascular diseases: an overview. J Chromatogr B Analyt Technol Biomed Life Sci 827:65–75.

28.

Mathew

Meyer

Bell

Johnson

Neblett

(1972) Regional cerebral blood flow and blood volume measured with the gamma camera. Neuroradiology 4:133–40.

29.

Matsumoto

Hyodo

Matsumoto

Koretsky

Sowers

Mitchell

Krishna

(2006) High-resolution mapping of tumor redox status by magnetic resonance imaging using nitroxides as redox-sensitive contrast agents. Clin Cancer Res 12:2455–62.

30.

Metz

Smith

Mick

Lustig

Mitchell

Cherakuri

Glatstein

Hahn

(2004) A phase I study of topical Tempol for the prevention of alopecia induced by whole brain radiotherapy. Clin Cancer Res 10:6411–7.

31.

Miura

Anzai

Takahashi

Ozawa

(1997) A novel lipophilic spin probe for the measurement of radiation damage in mouse brain using in vivo electron spin resonance (ESR). FEBS Lett 419:99–102.

32.

Muir

Buchan

von Kummer

Rother

Baron

(2006) Imaging of acute stroke. Lancet Neurol 5:755–68.

33.

Nagasawa

Kogure

(1989) Correlation between cerebral blood flow and histologic changes in a new rat model of middle cerebral artery occlusion. Stroke 20:1037–43.

34.

Raichle

Mintun

(2006) Brain work and brain imaging. Annu Rev Neurosci 29:449–76.

35.

Rice

Forman

Chen

Avshalumov

Cragg

Drew

(2002) Brain antioxidant regulation in mammals and anoxia-tolerant reptiles: balanced for neuroprotection and neuromodulation. Comp Biochem Physiol C Toxicol Pharmacol 133:515–25.

36.

Rice

Russo-Menna

(1998) Differential compartmentalization of brain ascorbate and glutathione between neurons and glia. Neuroscience 82:1213–23.

37.

Rostrup

Knudsen

Law

Holm

Larsson

Paulson

(2005) The relationship between cerebral blood flow and volume in humans. Neuroimage 24:1–11.

38.

Samuni

DeGraff

Krishna

Mitchell

(2001) Cellular sites of H₂O₂-induced damage and their protection by nitroxides. Biochim Biophys Acta 1525:70–6.

39.

Sano

Matsumoto

Utsumi

(1997) Synthesis and imaging of blood–brain-barrier permeable nitroxyl-probes for free radical reactions in brain of living mice. Biochem Mol Biol Int 42:641–7.

40.

Yamada

Kuppusamy

English

Yoo

Subramanian

Mitchell

Krishna

(2002) Feasibility and assessment of non-invasive in vivo redox status using electron paramagnetic resonance imaging. Acta Radiol 43:433–40.

41.

Yokoyama

Itoh

Aoyama

Obara

Ohya

Kamada

(2002) In vivo temporal EPR imaging of the brain of rats by using two types of blood–brain barrier-permeable nitroxide radicals. Magn Reson Imaging 20:277–84.

42.

Zimmermann

Winnefeld

Streck

Roskos

Haberl

(2004) Antioxidant status in acute stroke patients and patients at stroke risk. Eur Neurol 51:157–61.