Polynitroxyl Albumin Reduces Infarct Size in Transient Focal Cerebral Ischemia in the Rat: Potential Mechanisms Studied by Magnetic Resonance Imaging

Abstract

Nitroxide free radicals are known to protect cells from oxidative damage. Diffusion-weighted and perfusion-weighted magnetic resonance imaging was used to evaluate the effects of polynitroxyl albumin(PNA) in a middle cerebral artery intraluminal suture model of transient focal cerebral ischemia in the rat. Three groups of Sprague-Dawley rats were investigated: (1) PNA(N = 6), (2) human serum albumin (N = 6), and (3) saline (N = 7). The middle cerebral artery was occluded for 2 hours. Treatment was started 30 minutes after induction of ischemia. A total dose of 1% body weight (volume/weight) of PNA (23.5 mg/dL protein and 110 mmol/L nitroxide), albumin (23.5 mg/dL), or saline was injected intravenously at three time points: 0.5% at 0.5 hours, 0.25% at 2 hours (i.e., just before reperfusion), and 0.25% at 4 hours after occlusion. Six sets of diffusion- and perfusion-weighted magnetic resonance images were acquired throughout the 2 hours of ischemia and the 2 hours of reperfusion. The rats were killed at 24 hours, and the brains were stained with 2,3,5-triphenyltetrazolium chloride (TTC). Diffusion-weighted imaging showed that the growth of the ischemic lesion was suppressed in the PNA-treated group. The 4 hours diffusion-weighted imaging-derived hemispheric lesion volume in the PNA-treated group (25% ± 9%) was significantly smaller than that in the saline-treated(43% ± 13%; P = 0.016) or albumintreated groups (38% ± 6%; P = 0.017). A larger difference was observed for the 24-hour TTC-derived lesion volumes in the PNA (8% ± 7%), saline (35% ± 8%; P< 0.001), and albumin (31% ± 6%; P < 0.001) groups. Perfusion-weighted imaging demonstrated a marked improvement in cerebral perfusion in the PNA-treated group during ischemia and reperfusion. In conclusion, treatment with PNA results in an improvement in perfusion and a reduction of infarct volume in a model of transient focal cerebral ischemia in the rat.

Keywords

Cerebral ischemia focal Magnetic resonance imaging Stroke Albumin Nitroxide Rats Diffusion Perfusion

Oxidative damage is implicated in ischemic tissue injury (McCord, 1985; Siesjo et al., 1989; Traystman et al., 1991). Nitroxides are stable free radicals that have been shown to protect cells from oxidative damage initiated by superoxide, hydrogen peroxide, or radiation in vitro and in vivo (Nilsson et al., 1989; Mitchell et al., 1990; Samuni et al., 1991a,1991b; Hahn et al., 1994). One such nitroxide free radical, 4-hydroxy-2,2,6,6-tetramethylpiperidine-1-oxyl (tempol), has reduced reperfusion damage in the isolated perfused heart (Gelvan et al., 1991) and has reduced brain damage in rats subjected to closed head injury (Beit-Yannai et al., 1996). The beneficial mechanisms of nitroxides are unclear, but they may involve catalytic reduction of the levels of superoxide (i.e., acting as a superoxide dismutase mimic), oxidation of reduced transition metal ions, and detoxification of secondary organic radicals (Mitchell et al., 1990; Krishna and Samuni, 1994; Krishna et al., 1996). However, the benefit of nitroxides is limited by their short circulatory half-life in vivo. By covalently bonding the nitroxides to human serum albumin, as in polynitroxyl albumin(PNA), the compound under study in this report, the half-life and the antioxidant activity of the nitroxides can be enhanced (Kuppusamy et al., 1996). Unlike free nitroxides, the nitroxides in PNA are bound to a macromolecule and thus remain in the vascular space. Because of its enhanced redox activity, PNA might prove useful in the treatment of cerebral ischemic-reperfusion injury.

An evaluation of PNA in a suture model of transient middle cerebral artery occlusion(MCAO) in the rat is presented in this communication. This study evaluates the following:(1) the effectiveness of PNA in minimizing ischemic injury, (2) the timing of the actions of PNA during ischemia and reperfusion, and (3) the mechanism of action (i.e., hemodynamic or neuroprotective). The magnetic resonance imaging (MRI) techniques of diffusion-weighted MRI (DW imaging) and perfusion-weighted MRI (PW imaging) were used to follow the progression of the ischemic lesion and the perfusion deficit, respectively, during ischemia and reperfusion (Minematsu et al., 1993; Kozniewska et al., 1995). Diffusion-weighted imaging can delineate regions of ischemia because of a reduction of the apparent diffusion coefficient(ADC) of water in ischemic tissue (Moseley et al., 1990). In fact, the ADC in ischemic tissue can decrease significantly as early as a couple of minutes after induction of ischemia (Röther et al., 1996b). Dynamic contrast-enhanced PW imaging (Villringer et al., 1988) can semiquantitatively assess the perfusion deficit during MCAO and can assess the quality of reperfusion after the occluding suture is withdrawn. In addition to evaluating the efficacy of PNA in the aforementioned model of focal cerebral ischemia in the rat, the ability of MRI to monitor both the underlying hemodynamics and the temporal evolution of the ischemic lesion may yield insight about the mode of action of PNA.

MATERIALS AND METHODS

Animal model and protocol

Nineteen male Sprague-Dawley rats (270 to 370 g) were anesthetized initially with 3% halothane in a mixture of air (80%) supplemented with oxygen (20%). The rats were intubated and ventilated to keep the blood gases within the physiologic range. Anesthesia was maintained with 0.75% to 1.5% halothane throughout surgery and imaging. Body temperature was controlled with circulation of warm air. The femoral artery and vein were catheterized to monitor blood pressure and blood gases, and to deliver drug and contrast agent. The rats underwent 2 hours of MCAO induced by an intraluminal suture. Briefly, the tip of a 4-0 monofilament was coated with silicone to a diameter of ∼0.28 mm and was introduced into the internal carotid artery through the isolated stump of the external carotid artery (Koizumi et al., 1986). The suture then was advanced through the skull base to occlude the origin of the middle cerebral artery (MCA).

Immediately after the advancement of the suture, the rats were transferred to the MRI spectrometer. The rats were removed from the magnet at 2 hours after MCAO for withdrawal of the suture and subsequent restoration of blood flow through the internal carotid artery. The rats then were repositioned carefully in the MRI spectrometer for an additional 2 hours of MRI. Anatomical landmarks were employed to allow for the sequential comparison of MRI data. After the imaging protocol was complete at 4 hours after MCAO, the rats were removed from the spectrometer and allowed to recover. At 24 hours, the rats underwent a neurologic examination and were scored as follows: 0 for no apparent deficit, 1 for contralateral forelimb flexion, 2 for decreased grip of the contralateral forelimb when tail is pulled, 3 for contralateral circling only when pulled by the tail, and 4 for spontaneous contralateral circling (Bederson et al., 1986; Menzies et al., 1992). Rats that died before the 24-hour time point were given the poorest score (i.e., 4). After the neurologic examination, the rats were killed, and fresh coronal brain sections (2-mm thick) were stained with a 2% solution of 2,3,5-triphenyltetrazolium chloride (TTC) in saline to obtain an MRI independent measure of the extent of the ischemic injury.

Three groups were studied: (1) PNA (N = 6) (SynZyme Technologies, Inc., Irvine, CA, U.S.A.), (2) human serum albumin(N = 6) (Baxter Healthcare Corp, Glendale, CA, U.S.A.), and (3) saline (N = 7). The nitroxide concentration in the PNA solution was 110 mmol/L, which corresponds to 42 nitroxide moieties per molecule of albumin. Albumin was used to differentiate between the effects of albumin and the nitroxides in PNA. Treatment was started 30 minutes after induction of ischemia. A total dose of 1% body weight (volume/weight) of PNA (23.5 mg/dL protein and 110 mmol/L nitroxide), albumin (23.5 mg/dL), or saline was injected intravenously at three time points after MCAO, namely, 0.5% at 0.5 hours, 0.25% at 2 hours (i.e., just before reperfusion), and 0.25% at 4 hours after MCAO.

Magnetic resonance imaging

To follow the progression of the ischemic lesion, T₂-, diffusion-, and perfusion-weighted imaging was performed at six time points after MCAO: 0.25 hours (before first drug treatment at 0.5 hours), 1 hour, ∼1.7 hours(just before reperfusion at 2 hours), ∼2.3 hours (just after reperfusion at 2 hours), 3 hours, and 4 hours. All imaging was performed on a 2-T Bruker Omega CSI spectrometer (Karlsruhe, Germany) equipped with shielded gradients capable of producing 20 G/cm. The head of the rat was centered supinely in a 5.5-cm diameter bird cage radiofrequency coil.

Isotropic or trace diffusion-weighted spin-echo imaging(DW imaging) was used for delineation of the ischemic region, since it can depict the region of ischemia more accurately than single-axis diffusion weighting by removing the confounding effects of anisotropic diffusion on signal intensity (van Gelderen et al., 1994; Lythgoe et al., 1997). The isotropic weighting was achieved within a single scan by using diffusion-sensitizing gradient waveforms (Butts et al., 1997) that satisfy the orthogonality requirements detailed by Wong and colleagues (1995). Each set of diffusion-sensitizing gradients before and after the π radiofrequency pulse had a duration of 25 milliseconds. A high gradient factor, b, of 1300 s/mm² was used for a DW image and a low b value of 20 s/mm² was used for a primarily T₂-weighted image. Pixel-by-pixel maps of the ADC of water were calculated from these two images (Le Bihan et al., 1986). The images were acquired with a 128 × 128 matrix, field of view 50 mm, repetition time 2.5 seconds, echo time 80 milliseconds, one average, eight coronal slices, slice thickness 1.5 mm, interslice gap of 0.2 mm, and an acquisition time of ∼5.5 minutes.

Perfusion-weighted imaging was performed using spin-echo echo planar imaging to follow the passage of a bolus injection of 0.3 mmol/kg gadolinium diethylenetriaminepentaacetic acid (Magnevist, Schering, Berlin) (Villringer et al., 1988). The echo planar images were acquired with a 64 × 64 matrix, field of view 40 mm, repetition time 1.5 seconds, echo time 88 milliseconds, one average, four coronal slices, slice thickness 2.5 mm, and an interslice gap of 0.2 mm. Sixteen sets of these four slices were acquired continuously for 24 seconds. After 6 seconds of baseline imaging, a bolus of contrast was injected into the femoral vein, and the passage of the bolus of contrast through the brain was monitored by imaging for a further 18 seconds.

Data analysis

The evolution of the ischemic injury was followed by calculating lesion volumes from the DW images. Ischemic areas were defined as regions with hyperintensity greater than 1 SD (consistently∼18%) of the normal contralateral side on the DW images, which is comparable with the thresholds of 15% and 20% used in earlier DW imaging studies (Müller et al., 1995; Kohno et al., 1995). The areas of DW imaging hyperintensity were summed over the central six MR images, thereby covering 10 mm of the brain, and the lesion volume was expressed as a percentage of the hemispheric volume (%HLV). Lesion volumes at 24 hours after MCAO were measured by summing the unstained areas on five TTC sections.

As a semiquantitative measure of cerebral perfusion in the ischemic hemisphere, the time-to-peak, and the maximum signal loss in the bolus-tracking curves were measured in the basal ganglia and the basal and lateral part of the cortex, and were normalized relative to the corresponding regions in the nonischemic hemisphere (Minematsu et al., 1993; Röther et al., 1996a). The bolus delay in the stroked ipsilateral hemisphere compared with the contralateral side is given by

\begin{array}{l} Bolus delay & = & (time to max . signal loss, ipsi) \\ - (time to max . signal loss, contra) \end{array}

and the bolus peak ratio (BPR) is given by

\begin{array}{l} Bolus peak ratio & = & (max . signal loss, ipsi) / \\ (max . signal loss, contra) . \end{array}

Perturbations in cerebral perfusion can be characterized by a delay in the arrival of the bolus of contrast to the ipsilateral hemisphere of the brain compared with the contralateral hemisphere. In addition, the maximum signal loss yields an estimate of the amount of contrast, and by implication the amount of blood, passing through the brain. Since the signal loss in the ipsilateral hemisphere must be normalized with respect to the contralateral hemisphere, which presumably has normal perfusion, a BPR of 1 implies normal perfusion in the ipsilateral hemisphere and 0 indicates no perfusion. Perfusion changes were analyzed on the central slice at the level of the caudoputamen. In addition to these measurements, the area of tissue with a bolus delay of 2 seconds or more on the ipsilateral hemisphere of the central slice was calculated to demarcate the tissue affected by the MCAO.

Comparisons between the various groups were made with the Student's t test and the level of significance was defined as P ⩽ 0.05 (represented as * in the figures). The mean and SD are presented for the various groups.

RESULTS

Physiology

Body temperature, pH, PaO₂, PaCO₂, and mean arterial blood pressure were within physiologic range for all groups (Table 1). Both albumin and PNA reduced the hematocrit at 2 and 4 hours after MCAO relative to saline (Table 1). None of the animals experienced subarachnoid hemorrhage. Fourteen of 19 animals survived until the 24-hour time point, and hence infarct volumes at 24 hours could be calculated from TTC-stained sections for these animals. Therefore, TTC data were obtained on six of six PNA-treated, four of six albumin-treated, and four of seven saline-treated animals. One saline-treated animal had a hemorrhagic transformation, and one albumintreated animal had a parenchymal hemorrhage within infarcted tissue. The mean neurologic scores at 24 hours were 3.3 ± 1.1, 2.8 ± 1.2, and 2.2 ± 0.8 for the saline-, albumin-, and PNA-treated rats, respectively. The PNA-treated rats had a better neurologic score than the saline-treated rats, but the difference did not reach significance (P = 0.06).

Table 1

Parameters of general physiology before and after occlusion of the middle cerebral artery

			Time after MCAO
Parameter	Treatment group	Before MCAO	1 h	2 h	3 h	4 h
All values are mean ± SD. The start of treatment was 30 minutes after MCA occlusion. N = 7 for saline, N = 6 for albumin, and N = 6 for PNA. MCAO, middle cerebral artery occlusion; PNA, polynitroxyl albumin.
Mean arterial blood pressure (mm Hg)	Saline	91 ± 7	98 ± 5	101 ± 5	96 ± 4	97 ± 7
	Albumin	93 ± 8	96 ± 5	98 ± 5	94 ± 6	96 ± 7
	PNA	94 ± 10	99 ± 4	103 ± 9	99 ± 5	102 ± 4
Body temperature (°C)	Saline	36.5 ± 0.4	37.6 ± 0.4	37.4 ± 0.6	37.3 ± 0.4	36.9 ± 0.7
	Albumin	36.9 ± 0.4	37.4 ± 0.3	37.1 ± 0.4	37.1 ± 0.6	37.1 ± 0.2
	PNA	36.9 ± 0.5	37.3 ± 0.6	37.0 ± 0.5	36.9 ± 0.4	37.4 ± 0.6
Arterial Po₂ (mm Hg)	Saline	134 ± 27	121 ± 24	128 ± 34	125 ± 16	134 ± 26
	Albumin	135 ± 21	114 ± 19	113 ± 17	113 ± 11	122 ± 18
	PNA	140 ± 32	128 ± 25	134 ± 29	138 ± 20	142 ± 25
Arterial Pco₂ (mm Hg)	Saline	41 ± 5	41 ± 7	43 ± 8	45 ± 18	40 ± 14
	Albumin	38 ± 8	44 ± 4	43 ± 11	39 ± 5	36 ± 9
	PNA	37 ± 6	44 ± 9	40 ± 6	35 ± 7	38 ± 6
Arterial pH	Saline	7.40 ± 0.05	7.38 ± 0.08	7.36 ± 0.06	7.37 ± 0.15	7.40 ± 0.14
	Albumin	7.43 ± 0.05	7.36 ± 0.05	7.36 ± 0.10	7.41 ± 0.05	7.44 ± 0.06
	PNA	7.41 ± 0.06	7.36 ± 0.07	7.40 ± 0.05	7.47 ± 0.06	7.44 ± 0.07
Hematocrit (%)	Saline	42 ± 2∗		41 ± 4∗		40 ± 4∗
	Albumin	42 ± 2		35 ± 2		34 ± 2
	PNA	41 ± 0^†		36 ± 1^†		34 ± 0^†

∗

N = 5

†

N = 2

TTC-derived lesion volume

The final%HLV on the TTC-stained sections at 24 hours after MCAO was significantly lower in the PNA-treated group (8% ± 7%) than in the saline-treated (35% ± 8%;P < 0.001) or albumin-treated (31% ± 6%; P < 0.001) groups (Table 2). The final infarcts at 24 hours after MCAO were not significantly different between the saline- and albumin-treated groups(P = 0.42).

Table 2

Lesion volumes at 4 and 24 hours after middle cerebral artery occlusion

	4 h DWI∗	24 h TTC^†
The mean ± SD is presented. DWI, diffusion-weighted imaging; PNA, polynitroxyl albumin; TTC, 2,3,5-triphenyltetrazolium chloride staining.
% Hemispheric lesion volume
Saline	43.3 ± 13	35.1 ± 8
Albumin	38.4 ± 6	30.6 ± 6
PNA	25.3 ± 9	8.0 ± 7
P values
Saline vs PNA	0.016	<0.001
Albumin vs PNA	0.017	<0.001
Saline vs Albumin	0.405	0.415

∗

N = 7, 6, and 6 for saline, albumin, and PNA, respectively.

†

N = 4, 4, and 6 for saline, albumin, and PNA, respectively.

Diffusion-weighted imaging

Diffusion-weighted MR images demonstrating the ischemic lesion in representative saline- and PNA-treated rats over the 4 hours during and after MCAO are shown in Fig. 1. Isotropic diffusion gradients permitted the measurement of ischemic lesions without the complicating factor of signal intensity variations from anisotropic diffusion (van Gelderen et al., 1994; Lythgoe et al., 1997). The evolution of the mean lesion volume in the three treatment groups is shown in Fig. 2. The %HLV for the saline- and PNA-treated groups do not differ before the first treatment at 0.5 hours (P = 0.26), but then the lesion volumes diverge and rapidly become different as soon as 1 hour after MCAO (Fig. 2). The %HLV values in the albumin- and PNA-treated rats are similar before treatment (P = 0.30), but the albumin group has a significantly larger lesion volume than the PNA group at the latest time point of 4 hours after MCAO (P = 0.017). The growth of the lesion is suppressed markedly in the PNA group in contrast to the saline and albumin groups.

FIG. 1.

Diffusion-weighted images of four slices are shown for representative rats in the saline-treated (A) and polynitroxyl albumin (PNA)-treated (B) groups during the initial 2 hours of ischemia and the 2 hours after reperfusion. The initial regions of ischemia at 0.25 hours after middle cerebral artery occlusion (MCAO) are similar between these two rats, yet the final outcomes are different. As is most easily seen on slice 2, the cortex is primarily spared from infarction in the PNA-treated rat. Treatment began at 0.5 hours. The ischemic lesions are well demarcated at 4 hours after MCAO. All images are scaled globally.

FIG. 2.

The temporal evolution of the percent hemispheric lesion volume (mean ± SD), as measured by diffusion-weighted imaging, is presented for saline-, albumin-, and PNA-treated rats. A hyperintensity greater than 1 SD of the contralateral signal intensity was used to define the ischemic lesion. The lesion volumes before the initial treatment at 0.5 hours are not significantly different (P = 0.26) between the saline and PNA groups, but as soon as 1 hour after MCAO, the lesion volumes deviate significantly. The lesion grows markedly over the initial 2 hours in the saline group, whereas the growth of the lesion in the PNA group is markedly suppressed. The lesion in the albumin-treated rats also grows within the first 2 hours after MCAO, but it is not significantly different in size from the PNA group until the 4-hour time point. *P < 0.05 between saline or albumin versus PNA.

Table 2 summarizes the %HLV at 4 hours after MCAO. The lesion volumes in the PNA-treated animals are significantly smaller than in the saline- or albumin-treated rats. The lesion volumes at 4 hours did not differ significantly between the saline and albumin groups. Interestingly, the final mean %HLV derived from the DW imaging measurements at 4 hours after MCAO agreed well with that obtained by TTC at 24 hours after MCAO for the saline- and albumin-treated groups (Table 2), which is in agreement with the good correlation between TTC and DW imaging observed by Minematsu and others (1992) in control animals with 3 hours of ischemia. On the other hand, the %HLV from the TTC of the PNA-treated group is smaller than that measured from DW imaging and demonstrates a marked decrease in the volume of ischemic-reperfusion injury between 4 hours and 24 hours after MCAO in the PNA-treated rats.

The ADC values for the normal contralateral hemisphere were similar for all three treatment groups, as were the ADC values in the ischemic regions. The ADC values were (0.70± 0.07) × 10⁻³, (0.73 ± 0.02) × 10⁻³, and (0.71 ± 0.03) × 10⁻³ mm²/s at 15 minutes after MCAO and (0.70 ± 0.05) × 10⁻³, (0.72 ± 0.02) × 10⁻³, and (0.73 ± 0.02) × 10⁻³ mm²/s at 4 hours after MCAO in the contralateral hemisphere of the saline-, albumin-, and PNA-treated rats, respectively. Since the ADC is a function of temperature, this result implies that brain temperature was not influenced by treatment (Hasegawa et al., 1994). In ipsilateral areas, the mean ADC values in ischemic tissue were reduced to (0.48 ± 0.06) × 10⁻³, (0.52 ± 0.02) × 10⁻³, and (0.49 ± 0.04) × 10⁻³ mm²/s at 15 minutes after MCAO and (0.45 ± 0.07) × 10⁻³, (0.46± 0.05) × 10⁻³, and (0.49 ± 0.04) × 10⁻³ mm²/s at 4 hours after MCAO in the saline-, albumin-, and PNA-treated rats, respectively.

Unlike the DW images, the T2-weighted images (not shown) do not delineate the area of the ischemic lesion at the early time points after MCAO. Few abnormal hyperintensities were observed on the T2-weighted images before reperfusion. However, most animals did show significant hyperintensities on T2-weighted images immediately after reperfusion. The T2-weighted intensity, when present within the lesion, increased throughout the remainder of the reperfusion period in parallel with the development of vasogenic edema. In agreement with other reports (Minematsu et al., 1992), the areas of abnormality on the T2-weighted images at 4 hours were less conspicuous and were underestimating the size of the ischemic lesion relative to DW imaging.

Perfusion-weighted imaging

To demarcate the area of brain that is affected by MCAO, the ipsilateral area experiencing a bolus delay of 2 seconds or more was measured on a single slice over the 4 hours after MCAO (Fig. 3). The affected region comprises about 55% to 65% of the hemisphere and is similar for all three groups. After the suture is withdrawn from the origin of the MCA, the area with a bolus delay decreases to about 10% to 20% of the hemisphere. The similarity of the area of bolus delay within the three treatment groups confirms the reproducibility of the suture model of MCAO presented in this study. In addition, the bolus delays within these areas were similar and averaged 5.8 ± 1.0, 5.4 ± 1.0, and 5.1 ± 1.1 seconds for the saline-, albumin-, and PNA-treated rats, respectively, during the initial 2 hours after MCAO. The lack of a significant bolus delay after withdrawal of the suture signified that the origin of the MCA had been reopened properly.

FIG. 3.

The percent hemispheric area of brain that exhibits a contrast bolus delay of 2 seconds or more in the perfusion-weighted images of a single slice corresponding to the level of the caudoputamen in the saline-, albumin-, and PNA-treated rats. Occlusion of the middle cerebral artery perturbs the cerebral hemodynamics over a similar area in all three treatment groups. As expected, the area of bolus delay decreases markedly after the withdrawal of the occluding suture.

Table 3 presents a single-slice comparison of the hemispheric area that has perturbed hemodynamics, as measured by a bolus delay of 2 seconds or more on the PW images, with the hemispheric lesion area on DW imaging during the initial 2 hours while the suture is occluding the origin of the MCA. Interestingly, the area with perturbations in perfusion is larger than the lesion area on DW imaging at 0.25 hours after MCAO in all three treatment groups, but this mismatch between perfusion and diffusion decreases as the ischemic lesion evolves over time. By the 1.7-hour time point, the DW imaging lesion area has matched the PW imaging lesion area in both the saline- and albumin-treated groups whereas, there is still a mismatch in the PNA-treated group. The diffusion-perfusion mismatch at the early time point reflects the fact that a portion of the original area with perturbed hemodynamics remains above the perfusion threshold for changes in water diffusion. As time progresses, some of these hemodynamically challenged areas of brain are no longer able to sustain their energy stores and succumb to cellular swelling, which is subsequently seen as hyperintensity on the DW images. Transient depolarizations in the periinfarct zone are an additional stress that is likely to contribute to the evolution of the infarct (Busch et al., 1996). Mismatches in PW and DW images have been observed in stroke patients (Sorenson et al., 1996) and in a rat MCAO model (Quast et al., 1993).

Table 3

A comparison of the area that shows a perturbation of hemodynamics on perfusion-weighted imaging with the lesion area measured by diffusion-weighted imaging during the initial 2 hours after middle cerebral artery occlusion on a single slice

Group and time after MCAO	Area with bolus delay ≥2 s∗ (% hemisphere)	Area with lesion on DWI^† (% hemisphere)	Percent difference in area (PWI-DWI)
The mean ± SD is presented. DWI, diffusion-weighted imaging; PWI, perfusion-weighted imaging; PNA, polynitroxyl albumin; MCAO, middle cerebral artery occlusion.
Saline
0.25 h	60 ± 12	47 ± 23	13
1 h	67 ± 8	60 ± 12	7
1.7 h	68 ± 9	67 ± 8	1
Albumin
0.25 h	58 ± 9	33 ± 12	25
1 h	58 ± 18	53 ± 7	5
1.7 h	56 ± 24	63 ± 12	−7
PNA
0.25 h	66 ± 11	37 ± 19	29
1 h	61 ± 12	43 ± 20	18
1.7 h	60 ± 16	50 ± 16	10

∗

Data was obtained on a single slice (slice thickness = 2.5 mm) corresponding to the level of the caudoputamen.

†

Data was obtained on the slice (slice thickness = 1.5 mm) that corresponded most closely to the slice used for the perfusion data.

Even though the area of the brain that has a bolus delay is similar for all three treatment groups (Fig. 3), the actual amount of blood passing through the affected region may not necessarily be the same. The BPR is a semiquantitative measure of the perfusion deficit in the stroked ipsilateral hemisphere relative to the presumably normal contralateral hemisphere. The time course of the BPR during the 4 hours after MCAO is shown in Fig. 4 for the ipsilateral basal ganglia and cortex of a single slice in the three treatment groups. The BPR immediately after insertion of the suture, but before any treatment (i.e., at 0.25 hours), is reduced and is not significantly different between the treatment groups for either the basal ganglia or the cortex. However, by 1.7 hours after MCAO, the BPR has decreased to a larger extent in the saline group than in the PNA group. The perfusion status of the albumin group is not significantly different than the PNA group during the initial 2 hours after MCAO. As expected during the reperfusion phase, the BPR increases immediately after withdrawal of the occluding suture in all cases. The PNA-treated animals demonstrate BPR values closer to 1 after reperfusion, reflecting an improved status of perfusion in the ipsilateral hemisphere that is similar to that found in the normal contralateral hemisphere. On the other hand, the saline-treated rats have BPR values well below 1, which implies that perfusion is poor during reperfusion. The mean BPR values for the 2 hours before and 2 hours after withdrawal of the suture are summarized in Table 4. The PNA-treated group has significantly better perfusion than the saline-treated group in both the cortex and the basal ganglia during ischemia and reperfusion. In comparison with the albumin-treated group, the PNA group has significantly better perfusion in the basal ganglia only during the 2 hours after withdrawal of the suture. The albumin group displays higher overall perfusion than the saline group in the cortex but not in the basal ganglia (Table 4).

Table 4

The bolus peak ratios∗ averaged over either the initial 2 hours of middle cerebral artery occlusion or the 2 hours after the withdrawal of the suture from the origin of the middle cerebral artery†

	Cortex		Basal Ganglia
	Pre suture withdrawal	Post suture withdrawal	Pre suture withdrawal	Post suture withdrawal
The mean ± SD is presented, PNA, polynitroxyl albumin.
Bolus Peak Ratio
Saline	0.38 ± 0.21	0.57 ± 0.26	0.37 ± 0.15	0.65 ± 0.20
Albumin	0.53 ± 0.25	0.78 ± 0.19	0.43 ± 0.18	0.71 ± 0.10
PNA	0.57 ± 0.19	0.87 ± 0.12	0.53 ± 0.15	0.86 ± 0.14
P values
Saline vs PNA	0.005	<0.001	0.003	<0.001
Albumin vs PNA	0.60	0.11	0.12	<0.001
Saline vs Albumin	0.047	0.007	0.24	0.22

∗

Bolus peak ratio is defined in Materials and Methods section.

†

N = 7, 6, and 6 for saline, albumin, and PNA, respectively.

FIG. 4.

The bolus peak ratio (mean ± SD) in the ipsilateral cortex (saline and PNA [A], albumin and PNA [C]) and basal ganglia (saline and PNA [B], albumin and PNA [D]) relative to the contralateral nonischemic hemisphere is presented during middle cerebral artery occlusion (MCAO) and after withdrawal of the suture. A bolus peak ratio of 1 implies that the passage of contrast agent (related to perfusion) in the ipsilateral side is equal to that in the contralateral side, whereas a bolus peak ratio of 0 implies no perfusion in the affected territory. The perfusion deficit is similar in all three groups before treatment (i.e., at 0.25 hours after MCAO). As expected, the bolus peak ratio increases in all cases after withdrawal of the occluding suture. (A, B) The PNA-treated group has significantly better perfusion than the saline-treated group after the first drug treatment and before withdrawal of the occluding suture, as well as after withdrawal of the suture. (C, D) In contrast, the albumin and PNA groups are significantly different only for one time point in the basal ganglia. The bolus peak ratio decreases significantly over the initial 2 hours after MCAO in the cortex of the saline-treated(P = 0.035) and albumin-treated (P = 0.008) groups, but it does not worsen significantly (P = 0.22) in the PNA-treated case. *P < 0.05.

The BPR over the 4 hours after MCAO in the basal ganglia and cortex were combined and averaged to get an overall parameter for the degree of perfusion during the MRI measurements. Fig. 5 illustrates the good correlation between the infarct volume on the 4-hour DW images and the overall perfusion during the 4 hours after MCAO, and demonstrates the importance of the level of perfusion in determining the final lesion volume in all three treatment groups.

FIG. 5.

Correlation of the percent hemispheric lesion volume assessed by diffusion-weighted imaging at 4 hours after MCAO with the bolus peak ratio averaged over the entire 4 hours after MCAO. Poorer perfusion (i.e., lower average bolus peak ratios) correlates well with larger infarcts.

DISCUSSION

Perfusion and diffusion-weighted MRI are able to monitor the underlying hemodynamics and the evolution of an ischemic lesion during ischemia and reperfusion. In our study of transient focal cerebral ischemia in the rat, the administration of PNA was shown to reduce the size of the ischemic lesion relative to saline- and albumin-treated rats, as measured by DW imaging at 4 hours after MCAO and by TTC staining at 24 hours after MCAO (Fig. 2; Table 2). By using DW imaging to noninvasively follow the progression of the ischemic lesion, it became clear that the amelioration in the PNA-treated group was occurring before removal of the occluding suture. Both the saline- and albumin-treated groups had significantly larger lesion volumes at 1.7 hours than at 0.25 hours after MCAO, whereas the growth of the lesion in the PNA group was markedly suppressed over this time period. In addition, there was a substantial decrease in %HLV between the 4-hour DW imaging-derived lesion volume and the 24-hour TTC-derived lesion volume in the PNA group (Table 2). The cortex was spared to a greater extent than the subcortical structures of the brain (Fig. 1).

The lesion volumes, as measured on the DW images, at the first time point after MCAO (i.e., 0.25 hours) were not significantly different between the three treatment groups, but part of the improved outcome of the PNA group, relative to the saline group, could result from its smaller ischemic lesion immediately after MCAO. However, this argument does not hold when comparing the PNA group to the albumin group, since the albumin group started off with a slightly smaller lesion. Furthermore, the area with a bolus delay on the PW images was similar for all three groups (Fig. 3) and the initial BPR were not significantly different among the groups (Fig. 4), thereby demonstrating that the perfusion deficits were alike among all three groups used in this study.

Perfusion-weighted imaging performed before and after withdrawal of the occluding suture provides insight into the marked decrease of the ischemic lesion volume in the PNA-treated group. The PNA group demonstrated larger BPR and, hence, better perfusion than the saline-treated group (Fig. 4; Table 4), even during MCAO. In particular, the BPR in the saline- and albumin-treated rats decreased to a large extent between 0.25 hours and 1.7 hours after MCAO, possibly resulting from plugging up of the microvascular bed, in contrast to the PNA-treated rats. This observation suggests that perfusion was improved as a result of PNA and that this improved perfusion may be directly responsible, at least in part, for the reduced size of the ensuing infarct (Fig. 5). Even after removal of the occluding suture, the perfusion in the ischemic territory does not rebound to that found on the normal contralateral side, and thus the BPR values remain below 1. This is particularly evident for the saline-treated group, which has a BPR of only ∼0.6 during the “reperfusion” phase, whereas the BPR of ∼0.9 in the PNA-treated group demonstrates an almost normal degree of perfusion (Table 4). Similarly, others have shown that treatment with albumin-bound superoxide dismutase results in an improved recovery of CBF after reperfusion in a model of global brain ischemia in rats (Takeda et al., 1993). Since the perfusion deficits were not as severe in the PNA group before reperfusion, it is difficult in this study to assess the degree of protection invoked by PNA against reperfusion injury specifically. Further studies with PNA given first at reperfusion or at later time points are needed to see if delayed treatment with PNA could improve outcome such as that observed with the spin trap α-phenyl-N-tert-butyl nitrone (Zhao et al., 1994; Folbergrová et al., 1995).

When following the BPR over time, it is assumed implicitly that the perfusion in the contralateral side remains relatively constant, independent of either stroke evolution in the ipsilateral hemisphere or drug treatment. This assumption is supported by the fact that the actual minimum signal intensity during the passage of the bolus of contrast in the contralateral hemisphere is within error for all six time points during ischemia-reperfusion (data not shown) and is within error for all three treatment groups; namely, absolute minimum signal intensities of 59 ± 10%, 56 ± 9%, and 59 ± 8% for the saline-, albumin-, and PNA-treated groups are measured in the contralateral basal ganglia throughout the 4 hours of ischemia-reperfusion. In addition, others have shown that the CBF, as measured by ¹⁴C iodoantipyrine autoradiography, in the contralateral hemisphere is not significantly different between 30 minutes and 2 hours of ischemia in rats (Kohno et al., 1995).

In agreement with our observations of improved hemodynamics in the PNA-treated group, nitroxides have been shown to have vasodilatory actions in the isolated rat heart (Konorev et al., 1995). One potential mechanism is the dismutation of superoxide by the nitroxides of PNA, which may modulate vasoreactivity by influencing the balance between superoxide and nitric oxide (Katusic, 1996). Another possibility is that if PNA could reduce the levels of superoxide and prevent the deterioration of nitric oxide, then microvascular hypoperfusion may be prevented, since nitric oxide can inhibit platelet aggregation and leukocyte adhesion (Iadecola, 1997). In addition, the albumin-treated group has perfusion parameters that are significantly better than the saline-treated group in the cortex (Table 4), and thus some of the hemodynamic improvements of PNA may be related to its albumin component (Matsui et al., 1993). Therefore, the improvement in perfusion may result, in part, from hypervolemic hemodilution. Since PNA remains in the vascular space, a mechanism involving improved hemodynamics is expected. In fact, endothelial cells in the brain can produce hydroxyl free radicals after anoxic injury, and this suggests that the cerebral microvasculature may be an important source of deleterious free radicals (Kumar et al., 1996). Our results support a hemodynamic protective effect of PNA, but we cannot rule out a contribution by neuroprotective mechanisms, since free radicals may diffuse freely throughout the intravascular and extravascular space.

The use of MRI to follow differences in the progression of an ischemic lesion with therapeutic intervention has several advantages over traditional techniques such as histologic examination. Because MRI is noninvasive, the evolution of the ischemic lesion can be monitored in a single animal from the immediate onset of the lesion and onward. This is in contrast to traditional histologic methods, which need a separate group of animals for each time point. In addition, MRI can investigate perfusion throughout the brain with good spatial resolution. Therefore, the degree of the perfusion deficit and recovery can be monitored throughout the ischemiareperfusion episodes, and this can be directly related to the evolving infarct, as viewed with DW imaging.

In conclusion, the combined use of DW and PW MRI permits a thorough investigation of therapeutic drugs for stroke. In this study of transient focal cerebral ischemia in the rat, PNA was shown to reduce the ischemic lesion volume observed on DW imaging at 4 hours after MCAO (P = 0.016) and on TTC-stained slices at 24 hours after MCAO (P < 0.001). Diffusion-weighted imaging demonstrated that there was an improvement in lesion volume as early as intraischemia, that is, before removal of the occluding suture. Perfusion measurements with bolus tracking suggest that an enhancement of perfusion in the stroked hemisphere was one contributing mechanism to the improved outcome of rats treated with PNA.

Footnotes

Acknowledgements

The authors thank Maj Hedehus and Kim Butts for the isotropic diffusion pulse sequence.

References

Bederson

Pitts

Tsuji

Nishimura

Davis

Barkowski

(1986) Rat middle cerebral artery occlusion: evaluation of the model and development of a neurological examination. Stroke 17: 472–476

Beit-Yannai

Zhang

Trembovler

Samuni

Shohami

(1996) Cerebroprotective effect of stable nitroxide radicals in closed head injury in the rat. Brain Res 717:22–28

Busch

Gyngell

Eis

Hoehn-Berlage

Hossman

(1996) Potassium-induced cortical spreading depressions during focal cerebral ischemia in rats: contribution to lesion growth assessed by diffusion-weighted NMR and biochemical imaging. J Cereb Blood Flow Metab 16:1090–1099

Butts

Pauly

de Crespigny

Moseley

(1997) Isotropic diffusion-weighted and spiral-navigated interleaved EPI for routine imaging of acute stroke. Magn Resort Med 38:741–749

Folbergrová

Zhao

Katsura

K-I

Siesjö

(1995) N-tert-butyl-α-phenylnitrone improves recovery of brain energy state in rats following transient focal ischemia. Proc Natl Acad Sci USA 92:5057–5061

Gelvan

Saltman

Powell

(1991) Cardiac reperfusion damage prevented by a nitroxide free radical. Proc Natl Acad Sci 88:4680–4684

Hahn

Krishna

Samuni

DeGraff

Cuscela

Johnstone

Mitchell

(1994) Potential use of nitroxides in radiation oncology. Cancer Res 54:2006s–2010s

Hasegawa

Latour

Sotak

Dardzinski

Fisher

(1994) Temperature dependent change of apparent diffusion coefficient of water in normal and ischemic brain of rats. J Cereb Blood Flow Metab 14:383–390

Iadecola

(1997) Bright and dark sides of nitric oxide in ischemic brain injury. Trends Neurosci 20:132–139

10.

Katusic

(1996) Superoxide anion and endothelial regulation of arterial tone. Free Rad Biol Med 20:443–448

11.

Kohno

Hoehn-Berlage

Mies

Back

Hossman

(1995) Relationship between diffusion-weighted MR images, cerebral blood flow, and energy state in experimental brain infarction. Magn Reson Imaging 13:73–80

12.

Koizumi

Yoshida

Nakazawa

Ooneda

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

13.

Konorev

Tarpey

Joseph

Baker

Kalyanaraman

(1995) Nitronyl nitroxides as probes to study the mechanism of vasodilatory action of nitrovasodilators, nitrone spin traps, and nitroxides: role of nitric oxide. Free Rad Biol Med 18:169–177

14.

Kozniewska

Roberts

TPL

Tsuura

Mintorovitch

Moseley

Kucharczyk

(1995) N^G-nitro-l-arginine delays the development of brain injury during focal ischemia in rats. Stroke 26:282–289

15.

Krishna

Russo

Mitchell

Goldstein

Dafni

Samuni

(1996) Do nitroxide antioxidants act as scavengers of O₂·⁻ or as SOD mimics? J Biol Chem 271:26026–26031

16.

Krishna

Samuni

(1994) Nitroxides as antioxidants. Methods Enzymol 234:580–589

17.

Kumar

Liu

Floyd

Grammas

(1996) Anoxic injury of endothelial cells increases production of nitric oxide and hydroxyl radicals. Biochem Biophys Res Commun 219:497–501

18.

Kuppusamy

Wang

Zweier

Krishna

Mitchell

Trimble

Hsia

CJC

(1996) Electron paramagnetic resonance imaging of rat heart with nitroxide and polynitroxyl albumin. Biochemistry 35:7051–7057

19.

Le Bihan

Breton

Lallemand

Grenier

Cabanis

Laval-Jeantet

(1986) MR imaging of intravoxel incoherent motions: application to diffusion and perfusion in neurologic disorders. Radiology 161:401–407

20.

Lythgoe

Busza

Calamante

Sotak

King

Bingham

Williams

Gadian

(1997) Effect of diffusion anisot-ropy on lesion delineation in a rat model of cerebral ischemia. Magn Reson Med 38:662–668

21.

Matsui

Sinyama

Asano

(1993) Beneficial effect of prolonged administration of albumin on ischemic cerebral edema and infarction after occlusion of middle cerebral artery in rats. Neurosurgery 33:293–300

22.

McCord

(1985) Oxygen-derived free radicals in postischemic tissue injury. N Engl J Med 312:159–163

23.

Menzies

Hoff

Betz

(1992) Middle cerebral artery occlusion in rats: a neurological and pathological evaluation of a reproducible model. Neurosurgery 31:100–107

24.

Minematsu

Fisher

Sotak

(1993) Diffusion and perfusion magnetic resonance imaging studies to evaluate a noncompetitive N-methyl-d-aspartate antagonist and reperfusion in experimental stroke in rats. Stroke 24:2074–2081

25.

Minematsu

Fisher

Sotak

Davis

Fiandaca

(1992) Diffusion-weighted magnetic resonance imaging: rapid and quantitative detection of focal brain ischemia. Neurology 42:235–240

26.

Mitchell

Samuni

Krishna

DeGraff

Ahn

Samuni

Russo

(1990) Biologically active metal-independent superoxide dismutase mimics. Biochemistry 29:2802–2807

27.

Moseley

Kucharczyk

Mintorovitch

Cohen

Kurhanewicz

Derugin

Asgari

Norman

(1990) Diffusion-weighted MR imaging of acute stroke: correlation with T₂-weighted and magnetic susceptibility-enhanced MR imaging in cats. Am J Neuroradiol 11:423–429

28.

Müller

Haraldseth

Jones

Sebastiani

Lindboe

Un-sgård

Øksendal

(1995) Perfusion and diffusion-weighted MR imaging for in vivo evaluation of treatment with U74389G in a rat stroke model. Stroke 26:1453–1458

29.

Nilsson

Olsson

Carlin

Bylund-Fellenius

A-C

(1989) Inhibition of lipid peroxidation by spin labels: relationships between structure and function. J Biol Chem 264:11131–11135

30.

Quast

Huang

Hillman

Kent

(1993) The evolution of acute stroke recorded by multimodal magnetic resonance imaging. Magn Reson Imaging 11:465–471

31.

Röther

de Crespigny

D'Arceuil

Iwai

Moseley

(1996a) Recovery of apparent diffusion coefficient after ischemia-induced spreading depression relates to cerebral perfusion gradient. Stroke 27:980–987

32.

Röther

de Crespigny

D'Arceuil

Moseley

(1996b) MR detection of cortical spreading depression immediately after focal ischemia in the rat. J Cereb Blood Flow Metab 16:214–220

33.

Samuni

Godinger

Aronovitch

Russo

Mitchell

(1991a) Nitroxides block DNA scission and protect cells from oxidative damage. Biochemistry 30:555–561

34.

Samuni

Winkelsberg

Pinson

Hahn

Mitchell

Russo

(1991b) Nitroxide stable radicals protect beating cardiomyocytes against oxidative damage. J Clin Invest 87:1526–1530

35.

Siesjo

Agardth

Bengtsson

(1989) Free radicals and brain damage. Cerebrovasc Brain Metab Rev 1:165–211

36.

Sorenson

Buonanno

Gonzalez

(1996) Hyperacute stroke: evaluation with combined multisection diffusion-weighted and hemodynamically weighted echo planar MR imaging. Radiology 199:391–401

37.

Takeda

Hashimoto

Kosaka

Hirakawa

Inoue

(1993) Albumin-binding superoxide dismutase with a prolonged half-life reduces reperfusion injury. Am J Physiol 264:H1708–H1715

38.

Traystman

Kirsch

Koehler

(1991) Oxygen radical mechanisms of brain injury following ischemia and reperfusion. J Appl Physiol 71:1185–1195

39.

van Gelderen

de Vleeschouwer

MHM

DesPres

Pekar

van Zijl

PCM

Moonen

CTW

(1994) Water diffusion and acute stroke. Magn Reson Med 31:154–163

40.

Villringer

Rosen

Belliveau

Ackerman

Lauffer

Buxton

Chao

Wedeen

Brady

(1988) Dynamic imaging with lanthanide chelates in normal brain: contrast due to magnetic susceptibility effects. Magn Reson Med 6:164–174

41.

Wong

Cox

Song

(1995) Optimized isotropic diffusion weighting. Magn Reson Med 34:139–143

42.

Zhao

Pahlmark

Smith

M-L

Siesjö

(1994) Delayed treatment with the spin trap α-phenyl-N-tert-butyl nitrone (PBN) reduces infarct size following transient middle cerebral artery occlusion in rats. Acta Physiol Scand 152:349–350