| BACKGROUND:Intracerebral hemorrhage (ICH) is a cerebrovascular disease with high incidence, morbidity and mortality. In China, the annual incidence is about (60-80)/100000, accounting for10-15%of strokes. The30day mortality of ICH has been reported to be40-50%. Brain edema formation may be a major cause of neurological deficit or cerebral hernia in patients with ICH. Vasogenic cerebral edema is the main type of brain edema after ICH.The BBB is a crucial anatomic location in the brain, playing an important role in maintaining the homeostasis of brain. Its disruption is believed to be a major cause of brain edema after ICH. The BBB is composed of four main cellular elements: endothelial cells (ECs), astrocyte end-feet, extracellular matrix, and perycites. The tight junction is believed to be the key structure of BBB.Reports have shown that inflammatory mediators, thrombin, hemoglobin breakdown products, oxidative stress, complement, and matrix metalloproteinases are all participated in the BBB disruption.As a major component of hematoma, erythrocyte is mainly composed of Hb. After erythrocyte lysis, large quantities of Hb are introduced into the brain parenchyma. Recently, more attention has focused on the role of hemoglobin in the BBB disruption after ICH. Intracerebral injection of packed RBCs did not produce a dramatic edema until3days, but lysed erythrocytes induced an increase in brain water content and an increase of BBB permeability at24hours after injection. In a rabbit ICH model, Hb and heme were observed in the perihemotoma zone as early as24hours. Similarly, intracerebral injection of hemoglobin also resulted in an increase of brain water content within24hours. In a guinea pig exchange transfusion model, Hb triggered reduced ZO-1expression and increased IgG extravasation. Those results indicated that Hb was released at early stage of ICH and may have a potential role in BBB disruption.Nitric oxide (NO), an important bioregulatory molecular, is continuously synthesized by nitric oxide synthase (NOS), which has three isoforms:inducible NOS (iNOS), endothelial NOS (eNOS) and neuronal NOS (nNOS). Under physiological conditions, it plays a beneficial role in vasculature and central nervous system. However, it may be a double-edged sword. In the brain, disruption of BBB induced by a large amount of NO release has been reported in acute hypertension, infections, bacterial meningitis and focal cerebral ischemia. In addition, despite of the cytotoxicity of NO itself, its derivatives, especially peroxynitrites may lead to a disastrous effect on cell damage, excitotoxicity and BBB dysfunction.OBJECT:Hemoglobin was infused with stereotactic guidance into the right caudate nucleus of male Sprague Dawley rats. Then we investigated the effect of Hb on the BBB permeability, changes of TJ proteins (claudin-5, occludin, ZO-1and JAM-1), iron deposition, expression of inducible NOS and endothelial NOS, as well as NO production. The role of NOS and NO in this process was also further examined.METHODS:1. Experimental groups and surgical procedureMale normal Sprague Dawley rats, weighing approximately250-300g, were randomly divided into three groups:Hb-injected group, sham-operated group and normal control group. According to the time points after operation, Hb group and sham-operated group were divided into6subgroups (6h,12h,24h,48h,3d and7d,12rats for each subgroup). After inserted stereotactically into the right caudate nucleus, 20μl hemoglobin at a concentration of150mg/ml was injected. As a control experiment, same volume of saline was injected in sham-operated rats in the same way.2. Preparation of paraffin sections and HE stainingAfter anaesthetized, rats were perfused transcardially with saline followed by4%paraformaldehyde solution. Brain tissues were then removed and fixed by immersion in the same solution for24hours. After dehydrated and vitrified, they were embedded in paraffin, and5-um sections were prepared. Sections were de-waxed in xylene and rehydrated in graded ethanol and deionized water and then stained by hematoxylin for2.5min and eosin for2.5min. Images were captured with a microscope.3. Quantitative analysis of blood-brain barrier permeability2%EB solution was injected (4mL/kg, i.v.) via femoral vein before2hours of killing. Rats were transcardially perfused with normal saline until the drainage was colorless. Ipsilateral hemisphere was removed and carefully weighted. Samples were then incubated in trichloroacetic acid solution. Following homogenization and centrifugation (15,000rpm for20minutes), the supernatants was diluted with ethanol (1:3), and its fluorescence intensity was determined (excitation at620nm and emission at680nm) with an automatic microplate reader. The EB leakage was represented asμg/g brain weight.4. Real-time quantitative polymerase chain reactionRats were anesthetized and decapitated. Brain tissues (about30mg) around the lesion sites were obtained and total RNA was extracted from the tissue with GeneJETTM RNA Purification Kit. First strand cDNA was synthesized from RNA using a Maxima(?) First Strand cDNA Synthesis Kit. PCR reactions were run using Power SYBR(?) Green PCR Master Mix. PCR incubation and cycling parameters were95℃for10minutes, and a two-step temperature cycle consisting a denaturing step at95℃for15seconds and an annealing cycle of60℃for1minutes, repeated40times. The expression levels of target genes were calculated by using the standard curve method. 5. ImmunofluorescenceSections were de-waxed and rehydrated and then performed by heat treatment in a microwave oven in Tris-EDTA buffer solution (PH8.5) for antigen retrieval. After incubated by serum, sections were incubated overnight at4℃with primary antibody. After washed with PBS, sections were then incubated with the second antibody for1hours at37℃. For double staining experiments, primary antibodies were mixed and incubated overnight at4℃. Images were obtained with fluorescence microscope.6. Immunohistochemical Studies for iNOS, eNOSFor iNOS and eNOS antigen retrieval, slides were boiled in a microwave oven for30minutes in0.01mol/L citrate buffer solution (PH6.0). Sections were incubated in0.3%H2O2, goat serum, primary antibody, biotinylated secondary antibody and HRP-Streptavidin reagent. At last, immunoreactivity was detected using DAB. Images were obtained with the use of Leica Application Suite.7. NO concentrationAfter anesthetized, rats were perfused transcardially with saline. Brain tissues around the lesion site were obtained. The examination for the concentration of NO adopted nitrate reductase method. NO concentration was represented as μmol per gram protein.8. Enhanced Perls’ stainingSections were exposed to Perls’ solution (5%potassium ferrocyanide and5%hydrochloric acid,1:1) for30minutes at room temperature. After washed in deionized water, sections were incubated in1%DAB for15minutes, followed by0.03%H2O2and0.5%DAB for10minutes. At last, sections were washed in deionized water and lightly counterstained with haematoxylin and eosin. A total of five random images of each rat were assessed with the mean optical density and performed for semiquantitative analysis.9. Statistical analysisAll data in the text were presented as mean±standard error (SE). Statistic analysis was performed using SPSS19.0statistical programs. Comparison between groups were determined by Student’s T test or one-way analysis of variance (ANOVA) and followed by LSD test for two group’s comparison within the multiple groups. Correlation analysis between NO production and EB leakage was performed by Pearson test. Differences were considered significant at probability values less than0.05.RESULTS:1. Histopathologic changes after Hb injection were observed through HE staining. Around Hb,there was evident edema and nercrosis which showed degeneration, pyknosis or karyopyknosis. The number of normal neurons is reduced with a great deal of inflammatory cells effused. At7days after injection, Hb was partly absorbed, and the edema stripe around it becomes less distinct. The residual Hb was mainly around vessels.2. Damage to the BBB was assessed by Evans blue dye extravasation test. The EB leakage of sham groups were low and remained relatively constant, which had no statistical difference compared with normal control rats. There was also no significant difference between the subgroups of sham group. In contrast, EB leakage acutely increased after Hb injection, indicating a strong dysfunction of BBB. There were significant differences between Hb groups and normal controls or sham groups at the same time points (P<0.05). The permeability of BBB increased five times and reached the plateau at12hours after Hb injection, which had significant differences compared with other subgroups of Hb group (P<0.05).3. To examine the effect of Hb on endothelial tight junctions of BBB, we examined the expression of claudin-5, occludin, ZO-1and JAM-1in mRNA level. The change trends of each TJ proteins were not identical after Hb injection. In Hb group, the expression of claudin-5mRNA dramatically reduced as early as6hours, reached minimal level at12hours and remained low levels for at least7days (Fig. lb). There were significant differences between Hb groups and sham groups (P<0.05) except for24hours. Statistical differences were also observed between Hb groups and normal controls at each time points (P<0.05). Significant decrease of occludin mRNA was observed at6hours (P<0.05compared with normal control) and12hours (P<0.05compared with normal control or sham groups) after Hb injection (Fig.1c). Expression of occludin mRNA recovered to its normal level from24hours after Hb injection, showed no significant difference at24hours,48hours and3days in contrast with that of sham groups. However, a notable fivefold increase in mRNA level was measured at7days after Hb injection (P<0.001). For ZO-1(Fig.1d), a gradual reduction was detected at6h after Hb injection. It significantly decreased at12h and reached its nadir at7days (P<0.001). For JAM-1(Fig. le), significant reductions were detected at each time points (P<0.05) except for24hours and3days.4. In order to determine the changes in the expression and location of TJ proteins, immunofluorescence for claudin-5, ZO-1and occludin was performed. In sham control rats, intact and intense reactivity for TJ proteins were observed between adjacent endothelial cells in a continuous and linear pattern. Conversely, in Hb injected animals, claudin-5and ZO-1staining became diffuse and discontinuous, especially at medium and small-sized vessels. Occludin reactive intensity decreased at medium and small-sized vessels at early stage after Hb injection.5. By means of enhanced Perls’ staining, non-heme iron accumulation in the injured caudate nucleus was analyzed after Hb injection. No positive staining was visualized in brain sections from normal controls and sham groups. In Hb-injected animals, iron accumulation was detectable around the lesion site as early as48hours. Densitometry analysis showed a significant increase at48hours (P<0.05) compared with normal controls and sham groups. It remained high levels at3days (P<0.05) and7(P<0.05) days after Hb injection. Iron deposition was primarily found in the gliocyte. In some cases, it was observed within endothelial cells, especially at7days.6. A dramatic increase of iNOS mRNA was observed after Hb injection (Fig.4a). Maximal expression occurred at6hours, which100times higher than normal and sham controls (P<0.001). The mRNA of iNOS remained high levels till48hours (P<0.05) and then slowly diminished. Interesting, at3days and7days after Hb injection, iNOS mRNA was significantly decreased compared to normal and sham controls (P<0.05). In contrast to iNOS, eNOS mRNA (Fig.4h) had a gentle increase at6hours and12hours after Hb injection. A fourfold higher expression was detected at24hours which had a significant difference compared with normal and sham controls (P<0.05). Then, the expression significantly decreased at48hours (P<0.05) and7days(P<0.001). 7. To further confirm the expression and distribution of iNOS and eNOS, immunohistochemistry was performed. The iNOS-positive cells were distributed widely in the lesion hemisphere. Immunoreactivity was obviously observed in astrocytes and inflammatory cells. In some cases, iNOS reactivity was detected in the endothelial cells. Positive eNOS staining was most prominent at24hours in the lesion hemisphere but not in sham controls.8. We examined the production of NO by measurement nitrite/nitrate of brain tissues. In Hb-injected rats, NO production began to increase by12hours, peaked at3days, then declined to basal levels by7days. It showed significant changes at12hours (P<0.05),24hours (P<0.001),48hours (P<0.001) and3days (P<0.001) compared to normal controls. NO production was correlated significantly with the increase of BBB permeability (r=0.532, P<0.05).9. To further investigate the role of NOS in BBB disruption, brain sections were double labeled for claudin-5and iNOS or eNOS. In sham control animals, vessels with minimal iNOS or eNOS reactivity exhibited intense and intact claudin-5staining. However, in the Hb-injected animals, enhanced expression of iNOS was detected around lots of blood vessels. Meanwhile, eNOS immunoreactivity was observed in endothelial cells in the lesion site. The blood vessels with enhanced iNOS or eNOS immunoreactivity consistently showed discontinuous or diminished claudin-5staining, suggesting a close correlation between NOS and sites of BBB alterations.CONCLUSION:1. Hb may induce alterations of tight junction proteins and cause an increase of BBB permeability.2. Increased expression of iNOS and eNOS and excessive NO production were also observed after Hb injection. Moreover, enhanced iNOS and eNOS expression co-localized with sites of discontinuous or diminished claudin-5reactivity.3. Hb-induced BBB disruption appeared to be related to the NOS overexpression and excessive NO formation. |
PDF Full Text Request