Effect of 3-Aminobenzamide on the Ultrastructure of Astrocytes and Microvessels After Focal Cerebral Ischemia in Rats.

The disruption of blood-brain barrier (BBB) is a critical event in the formation of brain edema during early phases of ischemic brain injury. Poly(ADP-ribose) polymerase (PARP) activation, which contributes to BBB damage, has been reported in ischemia-reperfusion and traumatic brain injury. Here, we investigated the effect of 3-aminobenzamide (3-AB), a PARP-1 inhibitor, on the ultrastructure of BBB. Male Sprague Dawley rats were suffered from 90 minutes of middle cerebral artery occlusion, followed by 4.5 hours or 22.5 hours of reperfusion (R). The vehicle or 3-AB (10 mg/kg) was administered intraperitoneally (ip) 60 minutes after lacking of blood. Tissue Evans Blue (EB) levels, ultrastructures of astrocytes and microvessels, and areas of perivascular edema were examined in penumbra and core, at I 1.5 hours /R 4.5 hours and I 1.5 hours /R 22.5 hours, respectively. The severity of ultrastructural changes was graded with a scoring system in each group. We showed that 3-AB treatment significantly decreased tissue EB levels and ultrastructural scores, attenuated damages in astrocytes and microvessels, and reduced areas of perivascular edema. In conclusion, PARP inhibition may provide a novel therapeutic approach to ischemic brain injury.

Introduction

Despite recent advances in understanding cerebral ischemic stroke, no therapies can effectively and widely cure this condition that causes adult disability and cognitive impairment. Stroke is the third most common cause of death worldwide with tissue plasminogen activator as the only Food and Drug Administration–approved treatment for the ischemic type. The blood–brain barrier (BBB), which comprises the cerebral microvascular endothelium together with astrocytes, pericytes, microglia, neurons, and the extracellular matrix, is the regulated interface between the central neural system and the peripheral cycle.[1-4] Disruption of the BBB is a critical event in the formation of brain edema during the early phase of ischemic brain injury.[5] In experimental studies, BBB opening is biphasic: the initial breakdown is most likely caused by oxidative stress and is followed by partial BBB recovery. The second increase in BBB permeability leads to neutrophil infiltration through tight junction (TJ) redistribution.[6] During oxidative stress, the release of oxidants, proteolytic enzymes, and pro-inflammatory cytokines alters BBB permeability, leading to brain edema.

Recent researches have shown that acute oxidative and inflammatory mechanisms are determining factors in the development of the secondary brain-induced damage like BBB disruption.[7,8] The prognostic value of different circulating lectin pathway (LP) of complement promoters in acute stroke was evaluated by Zangari and coworkers.[7] They identified ficolin-1 as a sensitive prognostic marker for stroke for the first time and their results showed that LP was involved to a different extent and with distinctive functions over time in ischemic and hemorrhagic stroke. That is, ficolin-1 has higher sensitivity toward stroke outcome compared to other LP activators. In order to determine whether the peripheral leukocyte count and neutrophil-to-lymphocyte ratio predicted neurological deterioration (ND) after intracerebral hemorrhage (ICH), Lattanzi et al identified consecutive patients with ICH who had blood sampling performed within 24 hours from symptom’s onset for systematic research and analysis.[7] Their results proved that the prediction of ND could do benefits to clinical management design and assist patient prognostication. Poly(ADP-ribose) polymerase-1 (PARP-1) is a ubiquitously expressed enzyme that catalyzes the poly(ADP-ribosyl)ation of acceptor proteins by using nicotinamide adenine dinucleotide as a substrate and plays a key role in the regulation of DNA repair and gene transcription.[9] Multiple studies have shown that excessive PARP-1 activation mediates oxidative stress- and excitotoxicity-induced cell death, and increased PARP activities have been found in animal models of cerebral ischemia.[10] The contribution of PARP activation to BBB damage has been shown in ischemia–reperfusion and traumatic brain injury.[11,12] The neuroprotective effects of 3-aminobenzamide (3-AB) by inhibiting PARP-1 has also been studied. Wu et al investigated the protection of 3-AB on BBB and dopaminergic neurons from functional damage in a lipopolysaccharide-induced Parkinson disease rat model.[13] They found the administration of 3-AB could significantly inhibit the expression of the TJ-associated proteins claudin-5, occludin, and ZO-1. Furthermore, the p-ERK1/2 expression was reduced, while the expression of p-p38MAPK was increased, according to the Western blot analysis. These results supported that 3-AB could protect BBB from functional damage in the rat model and dopaminergic neurons were protected from degeneration by upregulation of TJ-associated proteins. In order to get further results on this topic, we aimed to investigate the effect of 3-AB, a PARP-1 inhibitor, on the ultrastructure of BBB after the focal cerebral ischemia of rats in this study.

Material and Methods

Rat Model of Focal Cerebral Ischemia

All animal experiments were carried out according to the ARRIVE (Animal Research: Reporting of In Vivo Experiments) guidelines.[14] Male Sprague Dawley rats, weighing 300-330 g (Beijing Vital River Experimental Animal Science and Technology Co, Ltd, Beijing, China), were anesthetized by inhaling 4% isoflurane in the O2 mixture on a warming pad at 37°C. Anesthesia was maintained with 2.5% isoflurane. The right middle cerebral artery occlusion (MCAO) was established by inserting a silicone coated nylon filament (φ = 0.028 cm) as previously described.[15] Generally, under the operating microscope, the right external carotid artery (ECA) and internal carotid artery (ICA) were exposed through a midline incision. Nylon filaments were inserted from the ECA into the ICA and gently advanced approximately 20 mm until resistance indicating that the tip of the filament had passed through and blocked the origin of the middle cerebral artery. Reperfusion was accomplished by removing the filament at 1.5 hours after MCAO. Sham-operated animals also underwent the same process, but sutures did not exceed the ICA bifurcation.

Animal Grouping and Drug Administration

Rats were randomly divided into 3 groups: (1) the sham-operation group (2 mL/kg); (2) the vehicle-treated group (normal saline), and (3) the 3-AB group (3-AB; 10 mg/kg). The drug was given intraperitoneally (ip) for 1 hour after ischemia.

Sample Collection and Preparation

At 1.5 hours of ischemia, 4.5 hours of reperfusion (I 1.5 h/R 4.5 h) and at 1.5 hours of ischemia, and 22.5 hours of reperfusion (I 1.5 h/R 22.5 h), MCAO rats were ip injected with 10% chloral hydrate 400 mg/kg and perfused with normal saline through the heart. Evans Blue (EB) was used as a tracer to assess the BBB disruption. Animals were intravenously injected with 2 mL/kg 3% EB in saline solution via the femoral vein, 2 hours prior to perfusion, under anesthesia. After perfusion, the tissues of core and penumbra were dissected at 4°C, according to the method used before.[16] Briefly, the brain sections were divided into 3 parts, starting with 3 mm in the prefrontal lobe and 3 mm in the posterior occipital lobe.

Regions from the right hemisphere of the middle section, corresponding to the ischemic core and penumbra, were dissected. The midline initially was identified between the 2 hemispheres. A longitudinal cut (top-down) was performed from the midline to the right (ischemic) hemispheres about 2 mm. Then, a horizontal diagonal cut at 2 o’clock was done to separate the core from the penumbra. For transmission electron microscopic examination, excised brain was sectioned into 1-mm-thick coronal blocks with a rat brain matrix, and the third coronal blocks from the frontal pole were used. Tissue blocks of 1 mm3 in the core and penumbra were collected.

Blood–Brain Barrier Permeability Assay

After perfusion, brain tissues of core and penumbra were weighed and added with formamide (Sigma, Shanghai, China), incubated at room temperature for 72 hours, the supernatant was centrifuged at 4°C for 15 minutes (12 000g) and transferred to the 96-well blackboard. The concentration of EB was measured at 620 nm by using a fluorescent multi-well plate reader (Infinite M200OTECAN, Switzerland). The tissue is oven-dried at 120°C for 24 hours and then weighed. The EB content of the tissue was quantified by the linear standard curve derived from the known dye content and then normalized to the tissue weight (μg/g).

Transmission Electron Microscopic Examination

At I 1.5 h/R 4.5 h and I 1.5 h/R 22.5 h, MCAO rats were anesthetized and sacrificed. For electron microscopy observations, sections (1 mm × 1 mm) were obtained from ischemic core and penumbra of each animal (Figure 1A) and fixed in 2.5% glutaraldehyde for 2 hours at 4°C. After they were washed with phosphate buffer (pH 7.2), the tissues were fixed in 1% osmium tetroxide for 2 hours, then dehydrated through an ethanol series embedded in epoxy resin and cut into ultrathin sections (100-nm thickness). The ultrathin sections were prepared according to the observed results of semithin sections, which were stained with methylene-blue for examination under the light microscope. The sections were mounted on copper grids, stained in uranyl acetate and lead citrate, and then observed under a transmission electron microscope (H-7650; Hitachi Ltd, Tokyo, Japan) at 80 kV.[17] All sections were blinded for the investigator. The severity of ultrastructural changes was graded in a scoring system (Table 1).[18,19]

Figure 1.

A, A representative image of triphenyl tetrazolium chloride (TTC) staining of a brain slice from a middle cerebral artery occlusion (MCAO) rat. The squares indicate positions where brain samples were collected for electron microscopic examinations from penumbra (P) and core (C). B, Effects of 3-aminobenzamide (3-AB) on tissue Evans Blue (EB) levels in penumbra and core in ischemic tissue at I 1.5 h/R 4.5 h and I 1.5 h/R 22.5 h after middle cerebral artery occlusion (MCAO) in rats. Vehicle or 3-AB was injected intraperitoneally 1 hour after ischemia. Data are presented as mean ± standard deviation of the mean. ## P < .01 vs sham-operated rats. **P < .01 vs vehicle-treated rats.

Table 1.

Ultrastructural Grading System.

Edema
None	0
Interstitial	1
Perivascular	2
Perivascular with vesicles	3
Massive perivascular and perineuronal	4
Fragmentation of blood-brain barrier	5
Neurons
None	0
Indentation	1
Margination of chromatin	2
Apoptosis, necrosis	3
Astrocytes
None	0
Swelling	1
Margination of chromatin	2
Apoptosis, necrosis	3
Myelin
Normal	0
Separation in configuration	1
Interruption in configuration	2
Honeycomb appearance	3
Endothelial BM detachment
Normal	0
Detached	1
Maximum points	15

Abbreviation: BM, basement membrane.

Measurement of the Perivascular Edema

To quantify the perivascular edema, the area of swollen astrocyte end feet surrounding the capillary basement membrane (BM) was measured from 8 randomly selected micro-vessels in the ischemic core and penumbra, respectively, by using the National Institutes of Health ImageJ 1.48v software. Five rats were analyzed for each group treated with saline or 3-AB.

Statistical Analysis

The data of ultrastructural score were expressed in the form of median value. The other values shown in this article are the mean ± standard deviation of the mean value to indicate the number of animals studied. The differences among groups were analyzed by single factor analysis of variance or by independent sample t test when appropriate. The ultrastructural scores were Kruskal-Wallis test or Mann-Whitney U test (P < .05). P < .05 was regarded important.

Results

3-Aminobenzamide Reduces Tissue EB Levels in Penumbra

In penumbra, compared to the sham group, EB contents were markedly increased in the vehicle group, both at I 1.5 h/R 4.5 h and I 1.5 h/R 22.5 h (Figure 1B; P < .01). 3-Aminobenzamide treatment decreased EB levels, compared to vehicle treatment (P < .01). Similarly, in core, EB levels of vehicle-treated rats were significantly higher than those of sham-operated rats at I 1.5 h/R 4.5 h and I 1.5 h/R 22.5 h (Figure 1B; P < .01), while the 3-AB group showed great reduced EB levels compared to the vehicle group (P < .01).

3-Aminobenzamide Improves Ultrastructural Score in Penumbra and Core

Semiquantitative evaluations (Figure 2) related to the effects of 3-AB on ultrastructure were performed in the penumbra and core of each group (5 rats/group). In penumbra, compared to the sham group, ultrastructural scores were markedly increased in the vehicle group, both at I 1.5 h/R 4.5 h (Figure 2; P < .001) and I 1.5 h/R 22.5 h (Figure 2; P < .01). 3-Aminobenzamide treatment reduced the scores contrast with vehicle treatment (P < .05). What is same as this, in core, the ultrastructural scores of vehicle-treated rats were necessarily higher than those of sham-operated rats at I 1.5 h/R 4.5 h (Figure 2; P < .01) and I 1.5 h/R 22.5 h (Figure 2; P < .01), while the 3-AB group exhibited lower scores than the score of the vehicle group (P < .05).

Figure 2.

Effects of 3-aminobenzamide (3-AB) on ultrastructural score in penumbra and core in ischemic tissue at I 1.5 h/R 4.5 h and I 1.5 h/R 22.5 h after middle cerebral artery occlusion (MCAO) in rats. Vehicle or 3-AB was injected intraperitoneally 1 hour after ischemia. Data are presented as scatterplots, with bar as the median (n = 5). ## P < .01 and ### P < .001 vs sham-operated rats in each group. *P < .05 vs vehicle-treated rats in each group.

3-Aminobenzamide Attenuated Astrocyte Ultrastructure Damage by Ischemia

Astrocyte ultrastructures were analyzed in penumbra and core. The images showed that 3-AB definitely attenuated the damages on astrocyte ultrastructure caused by ischemia. In penumbra, the astrocytic cytoplasm was clear and showed a large quantity of organelles in the sham group which meant astrocyte cellular components were not damaged with minor edema. At I 1.5 h/R 4.5 h, significant cell swelling was found in astrocytes of the vehicle group, with swollen nuclei and chromatin margination. At I 1.5 h/R 22.5 h, the vehicle group showed astrocyte proliferation. However, astrocyte swelling was improved at both times in the 3-AB group (Figure 3).

Figure 3.

Representative images of ultrastructure of astrocyte in penumbra and core in ischemic tissue at I 1.5 h/R 4.5 h and I 1.5 h/R 22.5 h after middle cerebral artery occlusion (MCAO) in rats. Vehicle or 3-aminobenzamide (3-AB) was injected intraperitoneally 1 hours after ischemia. Images in penumbra are shown in A-E. A, sham-operated group; B, vehicle-treated group at I 1.5 h/R 4.5 h; C, 3-AB-treated group at I 1.5 h/R 4.5 h; D, vehicle-treated group at I 1.5 h/R 22.5 h; E, 3-AB-treated group at I 1.5 h/R 22.5 h. Images in core are shown in F-J. F, sham-operated group; G, vehicle-treated group at I 1.5 h/R 4.5 h; H, 3-AB-treated group at I 1.5 h/R 4.5 h; I, vehicle-treated group at I 1.5 h/R 22.5 h; J, 3-AB-treated group at I 1.5 h/R 22.5 h (original magnification 7500×).

In core, similar results were observed. Samples in the sham group appeared minor edema, while at I 1.5 h/R 4.5 h, astrocyte swelling was severe along with nuclear membrane cracking and chromatin leakage in the vehicle group. At I 1.5 h/R 22.5 h, much more severe damages were observed, such as difficult to identify astrocytes, or only residual structures could be observed. After given 3-AB, astrocytes maintained intact structure and cell swelling was significantly improved.

3-Aminobenzamide Improves Microvessel Ultrastructure After Ischemia

In the sham group, capillaries consisted of a single layer of endothelial cells (ECs) surrounded by a layer of BM, sometimes enclosed by additional pericytes and another layer of BM. Organelles in all cells were well preserved and mitochondria showed a normal cristae pattern. In addition, mild perivascular edema was visible in the sham group.

In penumbra (Figure 4), at I 1.5 h/R 4.5 h, perivascular edema was found in the vehicle group, with disruption of mitochondrial cristae and loose BM. At I 1.5 h/R 22.5 h, the vehicle group showed significant perivascular edema, EC proliferation, and unclear BM. In 3-AB group, ultrastructural abnormalities were improved at both times.

Figure 4.

Effects of 3-aminobenzamide (3-AB) on ultrastructure of microvessels and basement membrane in penumbra and core in ischemic tissue at I 1.5 h/R 4.5 h and I 1.5 h/R 22.5 h after middle cerebral artery occlusion (MCAO) in rats. Vehicle or 3-AB was injected intraperitoneally 1 hours after ischemia. Images in penumbra are shown in A-E. A, sham-operated group; B, vehicle-treated group at I 1.5 h/R 4.5 h; C, 3-AB-treated group at I 1.5 h/R 4.5 h; D, vehicle-treated group at I 1.5 h/R 22.5 h; E, 3-AB-treated group at I 1.5 h/R 22.5 h. Images in core are shown in F-J. F, sham-operated group; G, vehicle-treated group at I 1.5 h/R 4.5 h; H, 3-AB-treated group at I 1.5 h/R 4.5 h; I, vehicle-treated group at I 1.5 h/R 22.5 h; J, 3-AB-treated group at I 1.5 h/R 22.5 h (original magnification 7500×). Arrows indicate the basement membrane (BM) of the microvessels.

In core, the vehicle group showed massive perivascular edema and narrow capillary lumen at both times. Endothelial cells revealed numerous cytoplasmic vacuoles. Fragments of microvilli were observed free-floating in the capillary lumen, in addition to the obvious disruption of the mitochondrial cristae in the ECs. The BM appeared loose and unclear, while the thickness of the BM was reduced compared to the sham group. In the 3-AB group, ultrastructural abnormalities were significantly improved.

The results of the perivascular edema area are shown in Figure 5. In the sham group, the area of perivascular edema was 6.418 ± 0.764 μm[2] in penumbra and 5.622 ± 0.951 μm[2] in core. In penumbra, the vehicle group demonstrated a significantly larger edema area, compared to the sham group, at I 1.5 h/R 4.5 h and I 1.5 h/R 22.5 h (Figure 5; P < .01 and P < .001, respectively). 3-Aminobenzamide treatment reduced the edema area, contrast with vehicle treatment (Figure 5; P < .01 and P < .001, respectively). In core, the edema area in the vehicle group was higher than that of the sham group at I 1.5 h/R 4.5 h and I 1.5 h/R 22.5 h (Figure 5; P < .001, respectively). Moreover, 3-AB treatment reduced the area of perivascular edema, contrast with vehicle treatment, at both times (Figure 5; P < .01 and P < .001, respectively).

Figure 5.

Effects of 3-aminobenzamide (3-AB) on areas of perivascular edema in penumbra and core in ischemic tissue at I 1.5 h/R 4.5 h and I 1.5 h/R 22.5 h after middle cerebral artery occlusion (MCAO) in rats. Vehicle or 3-AB was injected intraperitoneally 1 hour after ischemia. Data are presented as mean ± standard deviation of the mean (n =8). ## P < .01 and ### P < .001 versus sham-operated rats in each group. **P < .01 and ***P < .001 versus vehicle-treated rats in each group.

The TJ was observed in the sham group, both in penumbra and core are (Figure 6). In the vehicle group, the TJ was significantly reduced or even disappeared, compared to sham group. With the 3-AB treatment in I 1.5 h/R 4.5 h group, the TJ was largely recovered. However, with the increase in the ischemia time (I 1.5 h/R 22.5 h), the recovery of TJ is not as significant as the I 1.5 h/R 4.5 h group, though it is improved compared to vehicle group.

Figure 6.

Effects of 3-aminobenzamide (3-AB) on ultrastructure of microvessels’ tight junction (TJ) in penumbra and core in ischemic tissue at I 1.5 h/R 4.5 h and I 1.5 h/R 22.5 h after middle cerebral artery occlusion (MCAO) in rats. Vehicle or 3-AB was injected intraperitoneally 1 hour after ischemia. Images in penumbra are shown in A-E. A, Sham-operated group; B, vehicle-treated group at I 1.5 h/R 4.5 h; C, 3-AB-treated group at I 1.5 h/R 4.5 h; D, vehicle-treated group at I 1.5 h/R 22.5 h; E, 3-AB-treated group at I 1.5 h/R 22.5 h. Images in core are shown in F-J. F, Sham-operated group; G, vehicle-treated group at I 1.5 h/R 4.5 h; H, 3-AB-treated group at I 1.5 h/R 4.5 h; I, vehicle-treated group at I 1.5 h/R 22.5 h; J, 3-AB-treated group at I 1.5 h/R 22.5 h (original magnification 7500×). Red and blue notes indicate the TJ and basement membrane (BM) of the microvessels, respectively.

Discussion

In this study, we provided ultrastructural evidence for the protective effects of 3-AB on BBB disruption induced by brain ischemic insult. We focused on the pathological morphology of astrocytes and microvessels in maintenance of BBB integrity. The main findings were that: (1) astrocyte swelling occurred before neuronal pyknosis, resulting in perivascular edema. According to the results of electron microscopy images, the morphology of neurons in the ischemic group was generally normal, but there was obvious edema in astrocyte. However, the scope of electron microscopy was limited, and it was difficult to get complete neurons and astrocytes simultaneous in the same vision field; (2) intact TJs confirmed cytotoxic edema in ischemic rats (Figure 6); (3) 3-AB protected astrocytes against cytotoxic edema, thus protecting neurons.

After the onset of cerebral ischemia, brain cells in core area are basically dead, while those in penumbra are in the state of edema. Irreversible damage will occur as time goes on. Currently, the primary window period for brain protection is 6 hours. We need to protect the brain as soon as possible because the protective effect of drugs and the brain tissue itself will be extremely weak beyond this time window. In this study, the pathological morphology in penumbra and core appeared to be largely similar; changes in core were more severe. At I 1.5 h/R 4.5 h and I 1.5 h/R 22.5 h, BBB alterations included glial cell adhesion, swollen astrocytes, and ECs with enlarged mitochondria. Consistent with the morphological changes in astrocytes and ECs, permeability of the BBB to EB extravasation increased after ischemia. Kwon et al reported that the intact areas of BM-astrocyte contacts were significantly reduced in a time-dependent manner after MCAO.[20] Moreover, disruption of mitochondrial cristae and cytoplasmic vacuoles indicated EC dysfunction. Thus, in our opinion, swollen astrocytes, which can’t encircle ECs, and EC dysfunction, which may impact vascular leakage, lead to the disruption of the BBB. In addition, we found that the TJs remained nearly intact and inferred that cytotoxic edema was the main type of brain edema in ischemic rats. The capillary endothelial TJs are the main mediators, restricting the transfer of vascular-derived substances.[21] The most important membrane-associated TJ proteins include occludins, claudins, and junctional adhesion molecules, which are connected to the cytoskeleton by the zonula occludens.[22,23] Although the morphology of TJs remained normal, associated proteins may change in expression, and further research is needed. Importantly, the role of pericytes in maintenance of BBB was unclear in this study.

Conclusion

In our study, we observed early pathological changes at I 1.5 h/R 4.5 h; we found swollen astrocytes and normal neurons. Subsequently, at I 1.5 h/R 22.5 h, both neuronal pyknosis and astrocyte swelling can be observed. Thus, the significant inhibitory effects of 3-AB on astrocyte swelling may indicate a therapeutic approach for ischemic brain injury.