Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Recurrent mild cerebral ischemia: enhanced brain injury following acute compared to subacute recurrence in the rat

Recurrent mild cerebral ischemia: enhanced brain injury following acute compared to subacute... Background: In the current study, a transient cerebral ischemia producing selective cell death was designated a mild ischemic insult. A comparable insult in humans is a transient ischemic attack ( TIA) that is associated with functional recovery but can have imaging evidence of minor ischemic damage including cerebral atrophy. A TIA also predicts a high risk for early recurrence of a stroke or TIA and thus multiple ischemic insults are not uncommon. Not well under- stood is what the effect of differing recovery times between mild ischemic insults has on their pathophysiology. We investigated whether cumulative brain damage would differ if recurrence of a mild ischemic insult occurred at 1 or 3 days after a first insult. Results: A transient episode of middle cerebral artery occlusion via microclip was produced to elicit mild ischemic changes—predominantly scattered necrosis. This was followed 1 or 3 days later by a repeat of the same insult. Brain damage assessed histologically 7 days later was substantially greater in the 1 day recurrent group than the 3 days recurrent group, with areas of damage consisting predominantly of regions of incomplete infarction and pannecrosis in the 1 day group but predominantly regions of selective necrosis and smaller areas of incomplete infarction in the 3 days group (P < 0.05). Enhanced injury was reflected by greater number of cells staining for macrophages/microglia with ED1 and greater alterations in GFAP staining of reactive astrocytes in the 1 day than 3 days recurrent groups. The differential susceptibility to injury did not correspond to higher levels of injurious factors present at the time of the second insult such as BBB disruption or increased cytokines (tumor necrosis factor). Microglial activation, with poten- tial for some beneficial effects, appeared greater at 3 days than 1 day. Also blood analysis demonstrated changes that included an acute increase in granulocytes and decrease in platelets at 1 day compared to 3 days post transient ischemia. Conclusions: Dynamic changes in multiple inflammatory responses likely contribute to the time dependence of the extent of damage produced by recurrent mild ischemic insults. The time of mild stroke recurrence is crucial with early recurrence producing greater damage than subacute recurrence and this supports urgency for determining and implementing optimal stroke management directly after a TIA. Keywords: Cerebral ischemia, Transient ischemic attack, Stroke, Recurrence, Inflammation, Granulocytes social and economic burden [1]. Often considered a warn- Background ing sign for stroke, a transient ischemic attack (TIA) Ischemic stroke remains a leading cause of death and a results from a cerebral ischemic episode of short duration major cause of adult disability resulting in a huge personal, associated with a temporary blockage of a cerebral artery resulting in transient (less than 24 h) functional deficits [ 2, *Correspondence: utuor@ucalgary.ca 3]. Although TIA’s are associated with substantial or full Department of Clinical Neurosciences and Hotchkiss Brain Institute, functional recovery, there is both experimental and clini- Cumming School of Medicine, University of Calgary, Calgary, AB T2N 4N1, cal evidence that reperfused brain may often suffer some Canada Full list of author information is available at the end of the article © 2016 The Author(s). This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/ publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Tuor et al. BMC Neurosci (2016) 17:28 Page 2 of 14 permanent damage such as scattered cell death or selec- to one of three groups involving double surgical proce- tive neuronal loss following a transient ischemic event dures: a control group with sham surgery  +  transient [4–7]. Understanding better the pathophysiology of such middle cerebral artery occlusion (MCAO) at 1  day and mild transient ischemic injury is of some urgency consid- experimental groups with MCAO + recurrent MCAO at ering an anticipated higher incidence of TIA as the popu- 1 day and MCAO + recurrent MCAO at 3 days (n =  6/ lation becomes increasingly elderly, and, an anticipated group). Sample size was selected to minimize num- increasing number of transient cerebral artery occlusions bers of animals used yet provide power as estimated to will be successfully recanalized with tissue plasminogen detect a 30  % difference between the anticipated means activator or endovascular therapy [8–10]. of groups assuming a standard deviation of 15  %. Ani- In addition to being key to treatment approaches, mals were euthanized 7  days after the last surgery and improved knowledge regarding the pathophysiology of a brains were processed for histology to determine the relatively short transient ischemic insult resulting in recov- extent of ischemic brain damage. In additional experi- ery is also of importance for understanding the interactions ments performed to investigate longitudinal cerebral of multiple events considering that a TIA is associated with ischemic changes, animals were randomized to MCAO an increased risk of a second or recurrent stroke [2, 3, 11]. and euthanasia at 1 or 3 days after the transient ischemia A majority of TIA’s are associated with cortical ischemic (n = 8/group). Peripheral systemic inflammatory changes events and large vessel disease and recurrence is most com- were also investigated in additional animals from blood monly associated with a second ipsilateral cortical ischemic sampled at either 1 or 3 days following transient MCAO event [12, 13]. However, surprisingly little is known of the (n = 5/group). interaction of the ischemic injury processes that occur with multiple mild ischemic insults where the insults are Transient mild focal ischemic insult sufficiently severe to cause selective cell death within the Surgical procedures were performed aseptically under ischemic territory. Duration between insults is likely a cru- isoflurane anesthesia and included analgesic measures to cial factor considering that ischemic injury and brain recov- alleviate suffering. On the day of the first surgery, a tran - ery evolve over time. Furthermore, the pathophysiology of sient mild ischemic insult (mild transient MCAO) was early recurrence is important to understand considering produced in the laboratory by temporarily occluding the recurrence is frequently observed by 1 day after a TIA [14]. distal MCA using a microsurgical approach as described u Th s the objective of the current study was to deter - previously [7, 15]. Briefly, a microaneurysm clip was mine whether cumulative damage from multiple mild placed on the MCA through a small burr hole in the tem- transient ischemic insults, each of sufficient severity poral bone above the MCA where it crosses the rhinal to cause selective necrosis, is influenced by the recov - fissure. Rectal temperature was maintained using a servo- ery time between insults. A secondary objective was to controlled heating lamp. A tail artery was cannulated for determine whether the evolution of cellular changes after obtaining arterial blood samples and a small burr hole the first insult were associated with the total damage over the ipsilateral parietal cortex was used to measure produced by the recurrent insult. We hypothesized that cerebral perfusion with laser Doppler flowmetry. Con - damage with an early recurrent stroke would be greater current to the MCAO, both common carotid arteries than with a subacute recurrent stroke. An animal model were transiently occluded using suture thread. At the end was considered key because ischemic severity could of the 30 min occlusion, the microclip and carotid artery be varied to produce mild ischemic cerebral changes ligatures were removed. A dura substitute (Gore Pre- as can be observed clinically. Furthermore the timing clude MVP, Better Hospital Supplies Corp., Miami, FL) between insults could be well controlled. Thus we used was positioned on closing to facilitate the production of a rat model of transient focal ischemia [7, 15] adjusted to a subsequent second MCAO. Topical anesthetic in addi- produce selective cell death, which in young Wistar rats tion to the administration of buprenorphine (0.03  mg/ required 30 min of occlusion; this was designated to be a kg, s.c. every 12 h as needed) provided analgesia. During mild transient cerebral ischemic insult. We also produced a sham operation, animals underwent the same surgical a recurrence of the insult at 1 or 3  days with the results procedures with the exception of vascular occlusion. Fol- demonstrating that brain damage is greater when recur- lowing surgery, rats were housed in separate cages with rence is acute (1 day) rather than subacute (3 days). free access to soft and hard food, water and environ- mental enrichment. Animals were monitored twice daily Methods for the first 2  days and then daily to ensure good recov - Male Wistar rats (Charles River, Montreal, Canada) ery. Animals were subjected to a recurrent mild focal were acclimatized to a 12  h light/dark cycle with free ischemic insult by repeating the 30  min MCAO proce- access to food and water. Animals were randomized dure at either 1 or 3 days of recovery following the initial Tuor et al. BMC Neurosci (2016) 17:28 Page 3 of 14 mild MCAO. The location of the second clip placement AB_2307440, Vector Laboratories Burlingame, CA) that on the MCA was immediately distal to the original posi- stains both microglia and vessels [19]. The same sec - tion to minimize surgical complications. tions stained for lectin were also stained using a primary mouse anti-rat endothelial barrier antigen (EBA) IgM Histology to detect brain injury or repair responses (AB_10120605, SMI 71, 1:400, Covance) that stained For histological analysis, pentobarbital anesthetized vessels [20] to assess blood–brain barrier dysfunction. rats were perfusion fixed with formalin and brains were Comparing EBA and lectin stained sections provided embedded in paraffin. Sections (6 μm) were stained with an additional detection of morphological and intensity standard and immunohistochemical methods and altered staining changes in microglia. Finally, integrity of the staining or ischemic injury were assessed blinded to the blood–brain barrier was assessed using an antibody to animal’s identification or surgical group. detect extravasation of large plasma proteins (1:200, goat anti rat immunoglobulin G (IgG), Jackson ImmunoRe- Standard staining with hematoxylin and eosin search Lab, West Grove, PA). In sections stained with hematoxylin and eosin, the For all the immunohistochemical stains, methods extent of cortical damage was identified by classifying were generally as described previously [7]. Briefly, par - ischemic injury in the ischemic middle cerebral artery affin sections were first processed for antigen retrieval territory as selective necrosis, incomplete infarction or using citrate buffer at pH  =  6, an exception being sec- pannecrosis as described in previous studies [5, 16]. In tions stained with Iba1. The sections were incubated addition, a score for brain injury was determined [7] by with either 10  % goat or donkey serum, then primary dividing the cortex into 4 regions of interest and a score antibody, followed by the appropriate biotin-conju- was assigned to each region graded as: 0 for normal, 1 for gated IgG (Jackson ImmunoResearch Lab, West Grove, <10 % of cellular injury, 2 for 10–50 % of cellular injury, 3 PA) or fluorescent conjugated secondary antibody. For for >50 % cellular injury and, 4 for confluent areas of pan - colorimetric staining, horseradish peroxidase conju- necrosis. A cumulative score was obtained by summing gated streptavidin (1:400, Dako, Burlington, ON) and the scores for each region. diaminobenzidine (Sigma) were applied. For double staining of lectin and EBA, antigen retrieval was omit- Immunohistochemical staining for cellular ischemic ted and after blocking with goat serum sections were changes incubated with FITC-lectin at room temperature for Macrophages/microglia and astrocytes are two glial cell 1 h followed by incubation with EBA at room tempera- types participating in the CNS inflammatory response ture overnight. to cerebral ischemia. The ischemic changes in these cells observed following 7  days of recovery after multi- Histological analysis of altered immunostaining ple ischemic insults also reflect the severity of ischemic Sections were visualized and analyzed blinded to the injury. Reactive astrocytes were assessed using a primary experimental group using an Olympus BX61 micro- antibody to glial fibrillary acidic protein (GFAP) (rabbit scope and Microbrightfield Stereo Investigator (MBF anti-GFAP, 1:10,000, AB_10013482, Dako, Burlington, Bioscience, Williston, VT). Immunohistochemical ON). Increased activation of macrophages/microglia were changes were semi-quantified according to the stain assessed using anti-ED1 (AB_2291300, mouse anti-rat used and the types of cellular changes observed in the CD68 clone ED1, 1:100, AbD Serotec, Raleigh, NC) [17]. MCA territory of the parietal cortex ipsilateral and con- The effect of a single mild ischemic insult on the pro - tralateral to the transient occlusion. For ED1 and TNF, gression of inflammatory changes between 1 and 3  days regions of parietal cortex within the MCA territory were assessed first by staining for reactive astrocytes were inspected and cells within at least three fields of and macrophages/microglia as above. Changes in the view (0.263 mm each) with positive staining for ED1 or cytokine, tumor necrosis factor (TNF) was assessed using TNF were counted and used to provide a mean value. goat anti rat TNF-alpha (1:100, AB_354511, R&D system, For EBA, fluorescent images of parietal cortex were Minneapolis, MN) [17]. In addition, early activation of captured digitally (200× magnification) and those with microglia was assessed using an antibody against Ion- technically good vessel staining in the contralateral ized Calcium-Binding Adapter Molecule 1 (Iba1, 1:1000, hemisphere were analyzed using Image J for counting AB_839504, Wako Chemicals, USA) which is specifically of the number of EBA stained vessels in representa- expressed in microglia and is upregulated following cer- tive ipsilateral and contralateral fields of view. GFAP ebral ischemia [18]. stained sections were assigned cumulative scores simi- Vascular changes were investigated using a fluo - lar to the hematoxylin and eosin stained sections, with rescein isothiocyanate-labeled tomato lectin (1:200, altered levels of staining being scored in each of the four Tuor et al. BMC Neurosci (2016) 17:28 Page 4 of 14 different cortical areas as: 0 for normal, 1 for scattered Results reactive cells labelled or <10  % of area, 2 for 10–50  % Brain damage varies with timing between multiple mild of the area labelled, 3 for >50  % of the area labelled insults and 4 for confluent areas of pannecrosis with total loss For animals receiving multiple surgical interventions, eight of GFAP stain. A lectin staining score was attained by animals were lost to analysis due to mortality post recurrent assessing the extent of increased tomato lectin staining stroke (n = 1), inadequate flow reduction during clip place - of activated microglia and their processes within the ment (n  =  2), surgical complications (n  =  2) and a prob- cortical parenchyma (i.e. in excess of vessels labelled lem with tissue embedding (n  =  2). A total of 18 animals with EBA) and scored according to: 0—none, 1—sparse (n  =  6/group) were analyzed in the sham  +  mild MCAO, numbers and 2—substantial numbers of parenchymal 1  day recurrent MCAO and 3  days recurrent MCAO cells with positive lectin staining. In regions of infarct groups. Body weights in these 3 groups were similar at base- a score of 3 was assigned if there were substantial num- line and time of the second surgery (290 ± 65 and 282 ± 50; bers of positive stained parenchymal cells in addition to 319  ±  56 and 314  ±  65; 296  ±  45 and 296  ±  40, respec- a diffuse increase in signal from within the parenchyma. tively). In all animals, ischemic changes were observed in the For Iba1 stained sections, those with the best positive cerebral cortex ipsilateral to the transient middle cerebral staining of processes were selected for counting both artery occlusion (Fig.  1b, d, e–h) but not in the contralat- the total number of microglial cells stained per field and eral cortex (Fig.  1a, c). Severity of injury varied between the number of activated microglial cells with ramified or groups (see also data in Additional file  1). Substantial areas thickened bushy processes. Potential increased colori- of incomplete infarct and pannecrosis in addition to some metric staining in cortex for plasma proteins with IgG areas of selective necrosis were apparent within the middle were assessed by measuring gray levels in ipsilateral and cerebral artery territory in all 6 animals subjected to a recur- contralateral cortex along with levels on the blank slide rent mild ischemic insult separated by 1  day (e.g. Fig.  1b, adjacent to the region of interest in order to analyze d). Substantial changes in ED1 and GFAP staining also left–right differences. occurred with recurrence at 1 day post first insult (Fig.  1j, l and o, q, respectively). In contrast, in animals subjected to Systemic circulatory response a recurrent mild ischemic insult separated by 3  days there Hematology analysis of blood samples was also per- were only small regions of incomplete infarction with more formed in additional rats to assess for signs of immuno- extensive regions of selective necrosis in four of the six ani- suppression or infection. Venous blood samples (0.2  ml) mals and only regions of selective necrosis in 2 animals (e.g. were collected in EDTA coated tubes and analyzed Fig.  1e, g). Positive staining of macrophages/microglia with immediately for concentrations of platelets and red and ED1 or for reactive astrocytes with GFAP was also modest white blood cells (HemaTrue Hematology Analyzer, (Fig.  1k, m, p, r). Brains of animals that underwent a sham Heska Corp., Loveland, CO, USA). Relative proportions procedure followed by transient ischemia 1 day later had the of the different types of white cells were also assessed by least extensive ischemic changes consisting of only selective measuring the numbers of monocytes, granulocytes and necrosis (6/6 animals) (e.g. Fig.  1f, h) and minimal staining lymphocytes. for ED1 or GFAP (3/3) (Additional file 1 ). Semi-quantitative analysis of the stained sections sup- Statistical analysis ported these observations (Fig. 1i). All groups had signifi - Data were analyzed using SigmaPlot 13 software (Sys- cantly greater damage (median cumulative score ≥3) in tat Software Inc, San Jose, CA). Data are reported as the ischemic cortex than in the contralateral hemisphere mean  ±  SD for groups with continuous values or as the which was normal (median score of 0). Animals with an median and first and third quartiles for nonparametric early recurrent stroke 1 day after the first insult had sig - data. Statistical comparisons were considered significant nificantly greater damage assessed in hematoxylin and at P < 0.05. Prior to a comparison of means, samples were eosin stained sections than a sham procedure followed first tested for normality and equal variance followed by by a mild MCAO. Total damage with early (1 day) recur- selection of an appropriate analysis of variance and post rence also exceeded that produced by recurrent stroke hoc test (e.g. Bonferroni). Differences between ipsilateral separated by 3  days (P  <  0.05). The differences in sever - and contralateral hemispheres were compared using a ity of damage and their dependence on the recovery time paired Student’s t-test or a Rank Sum Test for categorical between the first and recurrent insult were also reflected values. in an increased number of cells staining with ED1 Tuor et al. BMC Neurosci (2016) 17:28 Page 5 of 14 MCAOx2 b c Contralateral d 1 d ** Contralateral MCAOx2 1d †† ** Sham+MCAO MCAOx2 g 3d h e f †† ** 1d 1d 3d Recovery between Procedures MCAOx2 3d 600 Sham+MCAO n ** l MCAOx2 1d m MCAOx2 3d 1d 3d Recovery Before Recurrent MCAO MCAOx2 1d MCAOx2 3d MCAOx2 1d q s o MCAOx2 3d p * s 4 25 µm 1d 3d MCAOx2 1d MCAOx2 3d Recovery Before Recurrent MCAO Fig. 1 Cerebral injury dependence on recovery time between recurrence of mild transient ischemic insults. Representative cerebral cortical micro- graphs of sections at low and high magnification (scale bar of 25 µm) stained with: hematoxylin and eosin (a–h), ED1 for activated macrophages/ microglia (j–m), and GFAP for reactive astrocytes (o–r). Brain was perfusion fixed and processed 7 days following the last insult. Cerebral cortex con- tralateral to the transient middle cerebral artery occlusion (MCAO) is normal (A,C). Within cerebral cortex (b, d) after a mild transient MCAO followed by a recurrent MCAO at 1 day there is extensive pannecrosis (p). Selective necrosis (s) is common following a mild transient MCAO followed by a recurrent MCAO at 3 days (e, g). Selective necrosis also occurs in rats with a sham surgery followed 1 day later by transient MCAO (f, h). Semiquanti- tative assessment of injury scores indicated greater damage with 1 day compared to 3 days recurrence in the H&E sections (i). Immunohistochemi- cal staining with ED1 antibody demonstrated an increase in activated macrophages/microglia following recurrent MCAO produced at either 1 or 3 days after an initial MCAO (j–m). The mean number of cells per field with ED1 staining demonstrating a difference between 1 and 3 days recurrent MCAO groups (n). GFAP staining of astrocytes was increased with recurrent MCAO at 1 or 3 days (o–r). The median score for altered GFAP staining reflected greater injury and astrocytic changes in brains of animals with recurrent MCAO at 1 day than 3 days (s). n = 6/group. *P < 0.05; **P < 0.006, † †† ‡ Ipsilateral different from contralateral; P < 0.05; P < 0.006, different from 1 day Recurrent MCAO. P < 0.01, different from Sham (Fig. 1n) and the severity score for differences in staining blood gases were similar between 1 and 3 days recurrent for reactive astrocytes with GFAP (Fig. 1s). groups. Mean values were 86 ± 11 and 88 ± 11 mm Hg for mean arterial blood pressure, 102  ±  7 and 109  ±  5 Similar physiological measures in recurrent stroke groups for PO2, 44.2  ±  2 and 38.6  ±  2  mm Hg for PCO2, Potential differences in severity of ischemia or other 7.35  ±  0.01 and 7.36  ±  0.04 for PH and 13.6  ±  3 and measured physiological parameters that are well known 11.6 ± 2 mmol/L for blood glucose, respectively. See also to affect ischemic injury were comparable between data in Additional file 2. groups. Cortical perfusion was less than 10  % baseline during MCAO and reperfusion levels were similar dur- Inflammatory changes following a single mild transient ing both the first and second ischemic insult irrespective ischemia of the recovery time between them (Fig.  2a). There was In order to investigate whether differential responses to also excellent control of body temperature during MCAO brain injury and recovery from ischemia could explain and during early reperfusion, irrespective of the experi- the ensuing damage produced by a recurrent MCAO, mental group (Fig. 2b). Other physiological variables and longitudinal tissue changes to a single mild MCAO were Ipsilateral H&E (Score) Ipsilateral ED1 Ipsilateral GFAP (Cells/Field) (Score) Tuor et al. BMC Neurosci (2016) 17:28 Page 6 of 14 which had minimal TNF staining (Fig.  4a), increased numbers of positive TNF stained cells were observed 1 3 d ipsilaterally at 1 and 3  days post-insult (e.g. Fig.  4b, c). The mean increased number of TNF stained cells were not different statistically at 1 and 3  days post insult (Fig. 4d). See also data in Additional file 4. Microglial changes, which are usually considered pro- inflammatory acutely but can also contribute later to 0 repair [21], were altered following a single mild insult. During Post During Post Microglial cells stained with Iba1 (Fig.  4e–h) were simi- st nd 1 MCAO 2 MCAO lar in the 1 and 3  days groups contralaterally (e.g. mean of 23.0 ± 6.0 cells per field). Total numbers of cells ipsi - laterally increased to 32.4  ±  9 and 64.9  ±  25 in the 1 1 3 d and 3  days groups respectively (P  <  0.05) and included cells with ramified or bushy morphology or round cells (Fig.  4f, g). Those with ramified or bushy morphology were also different between groups (Fig.  4h). See also data in Additional file  4. Using a tomato lectin stain of vessels and microglia, examination of staining in the non-vascular parenchyma contralaterally indicated a During Post During Post lack of appreciable staining for microglial cells with lec- st nd 1 MCAO 2 MCAO tin [22] (e.g. Fig.  4i). At 1  day post MCAO there was a Fig. 2 Mean cortical perfusion or rectal temperature during or post modest increase in lectin stained microglia and their pro- middle cerebral artery occlusion (MCAO). a Cortical perfusion meas- cesses, associated with microglial activation (e.g. Fig. 4j). ured using Doppler flowmetry presented as a percent of baseline. At 3  days post-insult, substantial numbers of microglia Mean perfusion decreased to <8 % baseline during MCAO subse- quently returning toward baseline within the first 5 min following stained for lectin (e.g. Fig.  4k) and the median scores reperfusion irrespective of MCAO recurrence at either 1 or 3 days post for lectin staining differed significantly between groups the first MCAO (n = 6/group). b Mean core body temperature was (P < 0.05) (Fig. 4l). well controlled resulting in similar means for the recurrent groups either during or after the first or second MCAO BBB injury following a mild transient ischemia In sections stained with lectin, we also performed immunostaining for endothelial barrier antigen (EBA) investigated in additional randomized animals. Several whose loss is considered to provide a highly sensitive of these were lost to analysis due to mortality (n  =  1), indicator of dysfunction of the blood–brain barrier [20]. surgical complications (n  =  2) and a problem with tis- There was reduced vascular EBA immunoreactivity sue embedding (n  =  1). Sections from all animals ana- observed following a mild ischemic insult. Compared to lyzed following a single transient MCAO demonstrated the contralateral hemisphere (Fig. 4m), there were fewer mild damage in hematoxylin and eosin stained sections EBA stained vessels (Fig. 4n–o) and analysis of sections (Fig. 3a, b). Damage score was similar in the 1 and 3 days determined a reduced number of EBA stained vessels recurrent MCAO groups (n = 8/group) (Fig. 3c). In adja- both at 1 and 3 days post a mild MCAO (Fig. 4p). There cent stained sections, there were increased numbers of was no significant difference between groups. Despite ED1 stained macrophages/microglia ipsilaterally in both a reduction in EBA staining there was no evidence of groups (Fig. 3d, e). There was also altered GFAP staining major BBB damage. Following a single mild ischemic of reactive astrocytes (Fig.  3g, h) compared to contralat- insult, there was a lack of appreciable positive staining eral cortex. These inflammatory changes scored in the 1 for IgG within parietal cortex (not shown), indicating and 3  days recurrent groups were not different statisti - that the blood–brain barrier was intact to passage of cally (Fig. 3f, i). See also data in Additional file 3. large plasma proteins detectable with IgG. Quantita- In sections from these animals we also investigated tive analysis of these sections demonstrated there was whether TNF-alpha, often considered to be a pro-inflam - no difference in the darkness of staining in ipsilateral matory cytokine, is substantially increased 1 day post a versus contralateral cortex for the 1 and 3  days groups mild ischemic insult and thereby a potential contributor (−0.7  ±  1.2 and −0.1  ±  1.9 gray levels, respectively) to increased ischemic damage observed with multiple (Additional file 4 ). insults at this time. Relative to the contralateral cortex, Cortical perfusion Rectal Temperature ( C) (% Baseline) Tuor et al. BMC Neurosci (2016) 17:28 Page 7 of 14 MCAO+1d MCAO+3d ** ab c ** 1d 3d Time post mild MCAO de f 600 1d 3d Time post mild MCAO gh i 1d 3d Time post mild MCAO Fig. 3 Acute and subacute cortical responses to a single mild ischemic insult. a, b Hematoxylin and eosin (H&E) stained sections from brains perfusion fixed at 1 or 3 days following a mild transient middle cerebral artery occlusion (MCAO). c The median and first and third quartiles of the damage score was not different statistically between groups. d, e ED1 stained sections from brains 1 and 3 days post MCAO. f Mean numbers of ED1 cells stained per field were not different statistically. g, h GFAP stained sections from animals 1 and 3 days post MCAO. i Median and first and third quartiles of the GFAP score were similar between groups (P > 0.05). n = 8/group *P < 0.05; **P < 0.005, ipsilateral different from contralateral. (Mann–Whitney Rank Sum test) Systemic inflammation following a mild transient ischemia compared to levels either in naïve animals or in blood Considering the evidence that systemic inflammation samples collected 3  days after MCAO (Fig.  5c). Altered or suppression of the immune system can play a role in lymphocyte and granulocyte concentrations but no influencing ischemic damage [21, 23, 24], we also investi- change in platelets were also observed 1 day following the gated the potential contribution of systemic immune cell stress of a sham surgery. However, the increase in granu- changes to the enhancement of ischemic damage by per- locytes alone was insufficient to enhance damage; sham forming a hematology analysis of blood samples collected surgery with increased granulocytes and a mild ischemic at 1 or 3 days following a mild MCAO or a sham surgery. insult 1  day later produced only mild ischemic changes Blood samples from a naïve group of rats provided an (Fig. 1c). additional control (n  =  5/group). Groups had a simi- lar hematocrit (Fig.  5d and Additional file  5). Following Discussion a mild MCAO there was a decrease in the lymphocyte The current results demonstrate the major importance percentage of white blood cells and the platelet concen- of early recurrence of a mild transient ischemic insult tration at 1  day but not 3  days relative to naïve controls following an initial TIA. Distinctive in our study, com- (Fig.  5a, b). In contrast, there was a transient increase in pared to studies of preconditioning, is the investigation the granulocyte percentage of white blood cells at 1  day of multiple transient ischemic insults—both of sufficient Ipsilateral ED1 Ipsilateral H&E Ipsilateral GFAP (Cells/Field) (Score) (Score) Tuor et al. BMC Neurosci (2016) 17:28 Page 8 of 14 1d 3 d Contralateral Ipsilateral a bc 1d 3d 25 M Time post mild MCAO 80 0 60 0 40 0 20 0 25 M 1d 1d 3d 3d Time post mild MCAO i j 4 k 2 ** 1d 3d 25 M Time post mild MCAO ** ** 1d 3d 25 M Time post mild MCAO Fig. 4 Cerebral Inflammatory and blood–brain barrier marker changes following a single mild transient MCAO. a Representative sections stained from contralateral cortex for the cytokine, tumor necrosis factor ( TNF). b, c TNF immunostained sections from ipsilateral cortex at 1 or 3 days post MCAO, respectively. d Mean number of TNF stained cells per field is similarly increased at 1 or 3 days post MCAO (n = 8/group). Representative positive staining of microglia with Iba1 in contralateral cortex e and ipsilateral cortex at 1 and 3 days post MCAO (f, g). Activated microglia were observed at 1 day and these were more numerous at 3 days post MCAO (h, n = 8/group). Double staining of vessels and microglia with tomato lectin (i–l) and vessels with endothelial barrier antigen (EBA) (m–p). Staining of microglia with lectin was minimal in the contralateral cortex (e.g. i) and increased (e.g. arrows) in ipsilateral cortex at 1 day (j) or 3 days (k) post transient MCAO. l Median scores for lectin staining (n = 8/group). m Representative staining of vessels for endothelial barrier antigen (EBA) in contralateral cortex (12 animals with good contralateral positive staining quantified). n, o EBA staining in ipsilateral cortex from animals at 1 and 3 days post MCAO. p Reduced percentage of vessels stained with EBA ipsilat - erally relative to contralaterally (n = 6/group). *P < 0.05; **P < 0.005, ipsilateral different from contralateral. P < 0.01 lectin stained microglia differ between 1 and 3 days post MCAO groups severity to cause mild ischemic damage. We found that in influencing brain damage and supports urgency for a recurrent mild insult following a TIA in rats produced determining and implementing optimal stroke preven- substantially more brain damage when recurrence is tion management early after TIA to avoid a second acute (i.e. 1  day) than with a subacute recovery time of ischemic event. 3  days between insults. The cause of this difference is likely multifactorial and potentially includes an augmen- Interaction between two Mild Ischemic Insults tation of damage by systemic inflammatory changes. The principal finding of the current study was that the Irrespective of the mechanisms, the results demonstrate combined damage produced by multiple mild insults that the time of mild stroke recurrence can be crucial was influenced by their timing. In contrast to that in Ipsilateral EBA Ipsilateral TNF Ipsilateral Lectin Ipsilateral Iba1 (Vessels % Contral) (Cells/Field) (Score) (Cells/Field) Tuor et al. BMC Neurosci (2016) 17:28 Page 9 of 14 b † ‡ ‡ Sham MCAO MCAO Sham MCAO MCAO Naive Naive +1d +1d +3d +1d +1d +3d c d ‡‡ ‡‡ Sham MCAO MCAO Sham MCAO MCAO Naive Naive +1d +1d +3d +1d +1d +3d Fig. 5 Systemic changes following a Single Mild Transient MCAO. A complete blood count of blood samples from naïve animals or animals at 1 or 3 days post a mild transient MCAO. a The concentration of lymphocytes, presented as a % of total white blood cells, was decreased at 1 day post transient MCAO when compared to naïve animals. b The mean concentration of platelets was also decreased at 1 day post MCAO. c The mean number of granulocytes, normalized to the total number of white blood cells, was transiently increased at 1 day. d Mean hematocrit was similar in ‡ ‡‡ † all groups. N = 5/group. P < 0.05; P < 0.005, different from Naive. P < 0.05, different from 1 day post MCAO (ANOVA and Bonferroni t-test) the current study, previous studies of multiple ischemic also observed an increase in the damage produced by a insults have generally focussed on studies where the recurrent compared to a single MCAO, but differences first or second insult is a very short preconditioning or were most marked for an acute (1 day) rather than a suba- postconditioning ischemia that alone produces no per- cute (3 days) recurrence. Extensive infarction rather than manent cellular damage; such a non-damaging ischemia partial infarction and/or selective necrosis was produced has protective effects [25–28]. Indeed in the current when stroke recurrence was acute rather than subacute. study the group with a sham procedure prior to transient In addition to this finding, we observed several differen - MCAO had reduced damage compared to that of a sin- tial systemic and cerebral inflammatory changes at acute gle MCAO assessed 3  days later. Using an initial insult and subacute times following a mild transient ischemic that is more injurious, one laboratory has examined the insult providing novel insights into the pathophysiology effects of repeated ischemia in the gerbil using only ani - of TIA and potential contributions to recurrent stroke mals that displayed neurological signs of stroke after a damage. These results are of potential relevance clinically first temporary carotid occlusion and subjecting them considering an anticipated rise in the incidence of TIA, to additional episodes of the same duration of ischemia both due to our ageing population and an escalation in [29, 30]. Although small numbers (3–5) per group were number of ischemic stroke patients that will be treated investigated, their results suggested a transition from with early reperfusion therapy [8–10]. Understanding mild to more severe injury with increasing numbers of better the pathophysiology of early recurrence is also insults and greater damage when second insults were at essential considering that the median time for recurrence 3 or 5 versus 48 h apart. Recently, we used a model of two of functional deterioration related to a vascular event fol- relatively short transient ischemic episodes, produced by lowing a TIA is 1 day [14]. Furthermore, although debili- microclip occlusion of the MCA and separated by 3 days tating strokes are relatively infrequent following a TIA to demonstrate an increased damage with recurrent alone, when there is a recurrent event, 53  % have been stroke compared to a single insult [15]. The present study found to be debilitating [31]. Lymphocytes (%) Granuloyctes(%) Hematocrit (%) Platelets (x10 /L) Tuor et al. BMC Neurosci (2016) 17:28 Page 10 of 14 Physiological and vascular changes was increased within areas of selective necrosis at 2 and Our physiological measures determined that various 4 weeks post insult followed by a decline in OX42 stain- factors known to influence infarct size such as body ing thereafter [5, 6, 36]. We investigated acute microglial/ temperature, blood pressure and blood gases were simi- macrophage activation using Iba1, tomato lectin and ED1 lar between the groups and thus are unlikely to account staining and observed a consistent early (1 day) increase for the effect of different timing of recurrent insults on in Iba1 stained microglia but sparse positive ED-1 stain- brain damage. Although preconditioning ischemia has ing in 3/8 animals. Staining for activated microglia/mac- been observed by others to improve blood flow during rophages using Iba1 and lectin staining was increased ischemia compared to a single severe ischemic insult at 3  days compared to 1  day post a mild ischemic insult [32, 33], the levels of cerebral blood flow reduction were and ED-1 staining was present in 6/8 animals. This sup - equivalent during both insults and for both groups. With ports that microglial activation occurs as early as 1  day respect to reperfusion, there were no significant statisti - post a mild ischemic insult, consistent with previous cal differences in perfusion within the first 5 min of rep - observations of early microglial activation in peri-infarct erfusion. However, there was a variable return of flow to regions i.e. regions which likely experienced mild tran- baseline, with some animals demonstrating hyperperfu- sient ischemia; and, this is followed by the appearance of sion and this variance was associated with suboptimal macrophages within the core and peri-infarct region at statistical power requiring caution for deciding the group 3–5 days post-insult [37]. There is a complex and incom - means were the same. Also a limitation presently was pletely understood progression of activation of different that only acute reperfusion was monitored and the possi- microglial phenotypes (e.g. macrophages that are of an bility remains that at a delayed hypoperfusion could have M1 or M2 phenotype), however, there is agreement in adversely influenced damage. general that activated microglia appear to be involved in Regarding vascular changes, there was little evidence both necrotic and repair responses [21, 37–39]. Thus it for differences in disruption of the blood brain barrier at is possible that in the present study, increased microglial early versus subacute times. There was a lack of detect - activation observed at 3  days provides some neuropro- able extravasation of IgG at either time point in agree- tection and acts to help reduce damage during the sec- ment with previous reports of a lack of vasogenic edema ond insult via expression of anti-inflammatory cytokines detectable in T magnetic resonance images following a such as IL-10 or TGF-β and factors such as IGF-1 [21, mild transient ischemic insult [5, 7]. The reduced stain - 37, 39]. Future elucidation of the pathophysiology of the ing for EBA did suggest some endothelial injury at both 1 polarity changes of the various immune cells activated and 3 days post a single MCAO. Early ischemic endothe- following ischemia should consider identifying the spe- lial injury is associated with an acute activation of the cific mediators within the microenvironment in response endothelium resulting in an increased expression of to differing severities of cerebral ischemia—including endothelial adhesion molecules associated with endothe- a mild transient ischemia. Presently, evidence was lack- lial injury [21, 24, 34]. However, since reductions in EBA ing for a potential involvement of a differential release were similar at both time points, there was no direct evi- of pro-inflammatory neurotoxic cytokines enhancing dence for greater dysfunction in the blood–brain barrier acute recurrent ischemic damage considering staining for at 1 than 3  days to help account for greater enhanced TNF-alpha tended to be greater rather than less at 3 days. damage of a recurrent 1 day insult. Systemic inflammation Cerebral inflammatory changes Examination of peripheral blood following a short tran- We also investigated whether the first mild ischemic sient ischemia indicated that even a mild ischemic insult insult could produce factors in the tissue that could either evoked alterations in inflammatory cells within the blood. augment ischemic damage acutely (i.e. at 1  day) or pro- Although the complex and dynamic immune responses vide some later (3 days) neuroprotection. The literature is post ischemia are still being characterized in detail and sparse regarding acute glial inflammatory changes in the their roles in affecting ischemic outcomes remain con - brain following a mild ischemic insult mimicking a TIA. troversial, numerous studies indicate that ischemic At relatively chronic times (1–4  weeks) following a mild alterations in peripheral inflammatory cells can modify focal transient ischemia increased astroglial reactivity brain damage [21, 23, 24]. Following the mild transient using GFAP has been reported previously [5, 7, 35, 36]. ischemia produced currently, we found alterations in the Similarly, at chronic times post-insult, microglial activa- concentrations of lymphocytes, granulocytes and plate- tion detected as positive immunostaining for OX42 (an lets in the blood. immune marker for CD11b which also stains activated Regarding changes in lymphocytes, both our sham neutrophils and microglial derived macrophages) [37] animals and animals 1  day post a mild ischemic insult Tuor et al. BMC Neurosci (2016) 17:28 Page 11 of 14 demonstrated a decrease in peripheral lymphocytes changes in various molecular signals and their interac- compared to our naïve controls. This is consistent with tions on multiple cell types with potential differences the lymphopenia associated with immune suppression according to species. and an increased susceptibility to infection that is well Currently, the neurovascular unit is considered a key recognized to occur post stroke in humans and experi- site of neutrophil action at delayed times following tran- mental animals [40–43]. The reduction in lymphocytes sient cerebral ischemia. Following transient ischemia, with the current mild transient ischemia produced by Ly6G positive neutrophil infiltration into the brain 30  min clip occlusion was rather modest (10–15  %) and parenchyma was minimal or occurred within the vicin- short lasting compared to the >50  % reduction in lym- ity of vessels [54, 55]. Also following transient ischemia phocytes observed for at least 2  weeks following 60  min in the mouse, homogenized brain samples analyzed of transient MCAO in mice using a transient intralumi- with flow cytometry demonstrated a delayed (2–3  days) nal thread occlusion [40]. Presently, lymphocyte numbers increase in neutrophils not present at 1 day [53, 56]. Note returned towards naïve control levels already by 3  days that neutrophil migration appears to occur sooner and is post-insult. more intense without reperfusion; it has been observed The white blood cell analysis also provided a count of early (e.g. at 1  day) following permanent ischemia [47, polymorphonuclear granulocytes which consists of pre- 57]. Although there may be a lack of migration into brain dominantly neutrophils along with eosinophils and baso- acutely following a single transient ischemia, granulo- phils. The increase in granulocytes observed at 1  day cyte recruitment, activation and infiltration [34, 37] on post-insult is consistent with reports of increased gran- primed cerebral endothelium may be greatly acceler- ulocytes following stroke in humans and experimental ated by the subsequent flow disruption produced with a animals [41, 43–46]. The origin of the additional granulo - recurrent ischemic insult. A contribution from platelets cytes can vary with the type of injury but includes mobi- is also possible as supported by their decrease at 1  day lization from the spleen and bone marrow and potentially compared to 3  days post-insult possibly reflecting con - increased production and decreased apoptosis [34, 37]. tinued activation/interaction with cerebral ischemic The mechanism mediating the increase likely includes a vasculature. Reduced platelet counts have also been stress response [23, 41] which is consistent with previous observed in stroke patients and there is evidence for their reports and our observation of elevated numbers of gran- interaction with peripheral inflammatory responses and ulocytes and decreased numbers of lymphocytes, also in ischemic injury [58–60]. Additional research is required animals subjected to sham surgery [40, 47, 48]. With a to clarify the importance of granulocyte and platelet mild transient ischemic insult, we observed an increase interactions and the dynamic changes in injured cerebro- in granulocytes that was temporary with a normaliza- vascular endothelium following an initial mild ischemic tion by 3  days post MCAO. A similar return towards insult prior to a second TIA. The possibility for different baseline was observed in mice with either sham surgery peripheral immune responses occurring with a first TIA or MCAO using a transient ligature [48]. Important to versus a recurrent stroke/TIA should also be considered note is that the stress induced neutrophilia would appear [61]. insufficient to enhance damage because a sham surgery 1  day prior to mild transient MCAO reduced ischemic Conclusion damage indicating that a combination of both transient To conclude, the production of multiple mild ischemic cerebral ischemia and neutrophilia would be involved in insults with different recovery times between them exacerbating damage. demonstrated an important dependence on the tim- Indeed, the interaction of granulocytes or neutrophils ing between insults. The damage observed with a short with cerebral ischemia is a complex operation of dynamic recovery of 1 day following a mild transient ischemia sub- changes in neutrophil activation at various sites (e.g. vas- stantially exceeded that with a subacute recovery time of cular and parenchymal) underlying multiple functions 3  days. The results might help explain clinical reports of (e.g. blood–brain barrier disruption, thrombus/clot for- increased risk of recurrent stroke in the first day follow - mation or neurotoxicity) [34]. Both clinical and experi- ing a TIA/minor stroke or the sudden clinical deteriora- mental studies have reported an increased accumulation tion observed in some cases soon after stroke [62, 63]. The of neutrophils in the brain after an ischemic stroke [34, mechanisms involved in the response are likely complex 37, 49]; and, there is evidence for their role in enhancing involving a progression of systemic and cerebral/vascular brain damage but also evidence for their lack of an effect inflammatory changes after the first mild ischemic insult or for producing beneficial effects via neuroprotection that affect the second insult. Irrespective, the results indi - or an involvement in brain repair e.g. [34, 37, 50–53]. cate that in the absence of clinically available therapeutic This variability may reflect the array of time dependent therapy to attenuate injury following a transient ischemic Tuor et al. BMC Neurosci (2016) 17:28 Page 12 of 14 Ethics attack or its recurrence, patients should receive immedi- This experimental animal study was carried out in accordance with the guide- ate care to best manage and prevent stroke recurrence lines and policies of the Canadian Council on Animal Care for the ethical care e.g. with early investigation of thromboembolic sources, and handling of animals. The current study was approved by the University of Calgary Health Sciences Animal Care Committee (Protocols M11017 and intensive antiplatelet therapies, lipid lowering, and/or AC15-0043). anticoagulation for cardioembolic stroke. Funding This study was supported by funding from the Canadian Institutes for Health Additional files Research (Grant MOP111037). This funding body had no role in the design of the study nor in the collection, analysis and interpretation of data nor in writ- ing of the manuscript. Additional file 1. Histology assessments for Figure 1. Shown are the data for each animal for the 1d and 3d recurrent (Recur.) groups and the Sham Received: 17 November 2015 Accepted: 11 May 2016 (sh) plus 1d mild (mld) stroke group. The descriptive assessments, the H&E scores, the ED1 counts and the GFAP scores are presented. Additional file 2. Physiological measures for each animal in the 1d and 3d recurrent mild ischemic insult groups. Shown are the doppler flow measures (% baseline), rectal temperature, mean arterial blood pressure (MABP) and blood gas values. References 1. Black M, Wang W, Wang W. Ischemic stroke: From next gen- Additional file 3. Histological assessments for Figure 3. Shown are the eration sequencing and GWAS to community genomics? OMICS. data for each animal at either 1d or 3d post a single mild ischemic insult. 2015;19:451–60. The H&E scores, the ED1 counts and the GFAP scores are presented. 2. Easton JD, Saver JL, Albers GW, Alberts MJ, Chaturvedi S, Feldmann E, Additional file 4. Histological assessments for Figure 4. Shown are the et al. Definition and evaluation of transient ischemic attack: a scientific data for each animal at either 1d or 3d post a single mild ischemic insult. statement for healthcare professionals from the American Heart Associa- Positive staining counts for TNF, Iba1 and EBA in addition to the Lectin tion/American Stroke Association Stroke Council; Council on Cardiovas- scores and IgG gray level measures are presented. cular Surgery and Anesthesia; Council on Cardiovascular Radiology and Intervention; Council on Cardiovascular Nursing; and the Interdisciplinary Additional file 5. Complete blood analysis values for each of the animals Council on Peripheral Vascular Disease. The American Academy of in each group—the naïve control group, the control group with sham Neurology affirms the value of this statement as an educational tool for middle cerebral artery (MCA) surgery, the 1d post MCA occlusion and 3d neurologists. Stroke. 2009;40:2276–93. post MCA occlusion groups. Presented are the numbers of white blood 3. Kernan WN, Ovbiagele B, Black HR, Bravata DM, Chimowitz MI, Ezekowitz cells, lymphocytes, monocytes, granulocytes (x109/L) and the numbers MD, et al. Guidelines for the prevention of stroke in patients with stroke of lymphocytes, monocytes and granuloctyes (% total white blood cells). and transient ischemic attack: a guideline for healthcare profession- Hematocrit and numbers of platelets are also presented. als from the American Heart Association/American Stroke Association. Stroke. 2014;45:2160–236. 4. Baron JC, Yamauchi H, Fujioka M, Endres M. Selective neuronal loss in Abbreviations ischemic stroke and cerebrovascular disease. J Cereb Blood Flow Metab. EBA: endothelial barrier antigen; GFAP: glial fibrillary acidic protein; IgG: immu- 2014;34:2–18. noglobulin G; MCAO: middle cerebral artery occlusion; TIA: transient ischemic 5. Ejaz S, Williamson DJ, Ahmed T, Sitnikov S, Hong Y T, Sawiak SJ, et al. attack; TNF: tumor necrosis factor. Characterizing infarction and selective neuronal loss following temporary focal cerebral ischemia in the rat: a multi-modality imaging study. Neuro- Authors’ contributions biol Dis. 2013;51:120–32. All authors helped to draft the manuscript. In addition, UIT participated 6. Ejaz S, Emmrich JV, Sawiak SJ, Williamson DJ, Baron JC. Cortical selective in study conception, design and data interpretation and contributed to neuronal loss, impaired behavior, and normal magnetic resonance histological assessment and statistical analysis of the data. ZZ carried out the imaging in a new rat model of true transient ischemic attacks. Stroke. recurrent stroke surgeries and participated in coordination of the study and 2015;46:1084–92. data interpretation. PAB participated in study conception and design and 7. Qiao M, Zhao Z, Barber PA, Foniok T, Sun S, Tuor UI. Development interpretation of the study. MQ participated in the recurrent stroke surgeries of a model of recurrent stroke consisting of a mild transient stroke and carried out the histological procedures and assessment of histological followed by a second moderate stroke in rats. J Neurosci Methods. injury. All authors read and approved the final manuscript. 2009;184:244–50. 8. Berkhemer OA, Fransen PS, Beumer D, van den Berg LA, Lingsma HF, Yoo Author details AJ, et al. A randomized trial of intraarterial treatment for acute ischemic Department of Clinical Neurosciences and Hotchkiss Brain Institute, Cum- stroke. N Engl J Med. 2015;372:11–20. ming School of Medicine, University of Calgary, Calgary, AB T2N 4N1, Canada. 9. Campbell BC, Mitchell PJ, Kleinig TJ, Dewey HM, Churilov L, Yassi N, et al. Department of Physiology and Pharmacology, University of Calgary, Calgary, Endovascular therapy for ischemic stroke with perfusion-imaging selec- AB T2N 4N1, Canada. tion. N Engl J Med. 2015;372:1009–18. 10. Goyal M, Demchuk AM, Menon BK, Eesa M, Rempel JL, Thornton J, et al. Acknowledgements Randomized assessment of rapid endovascular treatment of ischemic We gratefully acknowledge the technical assistance during the surgical experi- stroke. N Engl J Med. 2015;372:1019–30. ments provided by Melissa Morgunov. The animal resources centre performed 11. Bal S, Patel SK, Almekhlafi M, Modi J, Demchuk AM, Coutts SB. High the hematology analysis of the blood samples. rate of magnetic resonance imaging stroke recurrence in crypto- genic transient ischemic attack and minor stroke patients. Stroke. Availability of supporting data 2012;43:3387–8. The data sets supporting the results of this article are included within the 12. Kappelle LJ, Van Latum JC, Van Swieten JC, Algra A, Koudstaal PJ, van Gijn article and its additional files. J. Recurrent stroke after transient ischaemic attack or minor ischaemic stroke: Does the distinction between small and large vessel disease Competing interests remain true to type? Dutch TIA Trial Study Group. J Neurol Neurosurg The authors declare that they have no competing interests. Psychiatry. 1995;59:127–31. Tuor et al. BMC Neurosci (2016) 17:28 Page 13 of 14 13. Johansson E, Cuadrado-Godia E, Hayden D, Bjellerup J, Ois A, Roquer J, 36. Hughes JL, Beech JS, Jones PS, Wang D, Menon DK, Baron JC. Mapping et al. Recurrent stroke in symptomatic carotid stenosis awaiting revascu- selective neuronal loss and microglial activation in the salvaged neocorti- larization: a pooled analysis. Neurology. 2016;86:498–504. cal penumbra in the rat. Neuroimage. 2010;49:19–31. 14. Coutts SB, Modi J, Patel SK, Demchuk AM, Goyal M, Hill MD. CT/CT 37. Benakis C, Garcia-Bonilla L, Iadecola C, Anrather J. The role of microglia angiography and MRI findings predict recurrent stroke after transient and myeloid immune cells in acute cerebral ischemia. Front Cell Neurosci. ischemic attack and minor stroke: results of the prospective CATCH study. 2014;8:461. Stroke. 2012;43:1013–7. 38. Perego C, Fumagalli S, De Simoni MG. Temporal pattern of expression and 15. Clark D, Tuor UI, Thompson R, Institoris A, Kulynych A, Zhang X, et al. Pro- colocalization of microglia/macrophage phenotype markers following tection against recurrent stroke with resveratrol: endothelial protection. brain ischemic injury in mice. J Neuroinflammation. 2011;8:174. PLoS ONE. 2012;7:e47792. 39. Patel AR, Ritzel R, McCullough LD, Liu F. Microglia and ischemic stroke: a 16. Garcia JH, Liu KF, Ye ZR, Gutierrez JA. Incomplete infarct and delayed neu- double-edged sword. Int J Physiol Pathophysiol Pharmacol. 2013;5:73–90. ronal death after transient middle cerebral artery occlusion in rats. Stroke. 40. Prass K, Meisel C, Hoflich C, Braun J, Halle E, Wolf T, et al. Stroke-induced 1997;28:2303–9. immunodeficiency promotes spontaneous bacterial infections and is 17. Qiao M, Meng S, Foniok T, Tuor UI. Mild cerebral hypoxia-ischemia mediated by sympathetic activation reversal by poststroke T helper cell produces a sub-acute transient inflammatory response that is less type 1-like immunostimulation. J Exp Med. 2003;198:725–36. selective and prolonged after a substantial insult. Int J Dev Neurosci. 41. Dirnagl U, Klehmet J, Braun JS, Harms H, Meisel C, Ziemssen T, et al. 2009;27:691–700. Stroke-induced immunodepression: experimental evidence and clinical 18. Ito D, Tanaka K, Suzuki S, Dembo T, Fukuuchi Y. Enhanced expression of relevance. Stroke. 2007;38:770–3. Iba1, ionized calcium-binding adapter molecule 1, after transient focal 42. Urra X, Cervera A, Villamor N, Planas AM, Chamorro A. Harms and benefits cerebral ischemia in rat brain. Stroke. 2001;32:1208–15. of lymphocyte subpopulations in patients with acute stroke. Neurosci- 19. Acarin L, Vela JM, Gonzalez B, Castellano B. Demonstration of poly- ence. 2009;158:1174–83. N-acetyl lactosamine residues in ameboid and ramified microglial 43. Vogelgesang A, Grunwald U, Langner S, Jack R, Broker BM, Kessler C, et al. cells in rat brain by tomato lectin binding. J Histochem Cytochem. Analysis of lymphocyte subsets in patients with stroke and their influence 1994;42:1033–41. on infection after stroke. Stroke. 2008;39:237–41. 20. Lin B, Ginsberg MD. Quantitative assessment of the normal cerebral micro- 44. Buck BH, Liebeskind DS, Saver JL, Bang OY, Yun SW, Starkman S, et al. Early vasculature by endothelial barrier antigen (EBA) immunohistochemistry: neutrophilia is associated with volume of ischemic tissue in acute stroke. application to focal cerebral ischemia. Brain Res. 2000;865:237–44. Stroke. 2008;39:355–60. 21. Iadecola C, Anrather J. The immunology of stroke: from mechanisms to 45. Ross AM, Hurn P, Perrin N, Wood L, Carlini W, Potempa K. Evidence of the translation. Nat Med. 2011;17:796–808. peripheral inflammatory response in patients with transient ischemic 22. Acarin L, Gonzalez B, Castro AJ, Castellano B. Primary cortical glial reaction attack. J Stroke Cerebrovasc Dis. 2007;16:203–7. versus secondary thalamic glial response in the excitotoxically injured 46. Ito U, Hakamata Y, Kawakami E, Oyanagi K. Temporary [corrected] cerebral young brain: microglial/macrophage response and major histocompat- ischemia results in swollen astrocytic end-feet that compress microves- ibility complex class I and II expression. Neuroscience. 1999;89:549–65. sels and lead to delayed [corrected] focal cortical infarction. J Cereb 23. Chamorro A, Meisel A, Planas AM, Urra X, van de Beek D, Veltkamp R. The Blood Flow Metab. 2011;31:328–38. immunology of acute stroke. Nat Rev Neurol. 2012;8:401–10. 47. Moller K, Boltze J, Posel C, Seeger J, Stahl T, Wagner DC. Sterile inflamma- 24. Denes A, Thornton P, Rothwell NJ, Allan SM. Inflammation and brain tion after permanent distal MCA occlusion in hypertensive rats. J Cereb injury: acute cerebral ischaemia, peripheral and central inflammation. Blood Flow Metab. 2014;34:307–15. Brain Behav Immun. 2010;24:708–23. 48. Denes A, Pradillo JM, Drake C, Buggey H, Rothwell NJ, Allan SM. Surgical 25. Koch S, Della-Morte D, Dave KR, Sacco RL, Perez-Pinzon MA. Biomarkers manipulation compromises leukocyte mobilization responses and for ischemic preconditioning: finding the responders. J Cereb Blood Flow inflammation after experimental cerebral ischemia in mice. Front Neuro - Metab. 2014;34:933–41. sci. 2014;7:271. 26. Wang Y, Reis C, Applegate R, Stier G, Martin R, Zhang JH. Ischemic 49. Price CJ, Menon DK, Peters AM, Ballinger JR, Barber RW, Balan KK, et al. conditioning-induced endogenous brain protection: applications pre-, Cerebral neutrophil recruitment, histology, and outcome in acute per- or post-stroke. Exp Neurol. 2015;272:26–40. ischemic stroke: an imaging-based study. Stroke. 2004;35:1659–64. 27. Garcia-Bonilla L, Benakis C, Moore J, Iadecola C, Anrather J. Immune 50. Krams M, Lees KR, Hacke W, Grieve AP, Orgogozo JM, Ford GA. Acute mechanisms in cerebral ischemic tolerance. Front Neurosci. 2014;8:44. Stroke Therapy by Inhibition of Neutrophils (ASTIN): an adaptive 28. Esposito E, Hayakawa K, Maki T, Arai K, Lo EH. Eec ff ts of postconditioning dose-response study of UK-279,276 in acute ischemic stroke. Stroke. on neurogenesis and angiogenesis during the recovery phase after focal 2003;34:2543–8. cerebral ischemia. Stroke. 2015;46:2691–4. 51. Allen C, Thornton P, Denes A, McColl BW, Pierozynski A, Monestier M, et al. 29. Hanyu S, Ito U, Hakamata Y, Yoshida M. Transition from ischemic neuronal Neutrophil cerebrovascular transmigration triggers rapid neurotoxicity necrosis to infarction in repeated ischemia. Brain Res. 1995;686:44–8. through release of proteases associated with decondensed DNA. J Immu- 30. Hanyu S, Ito U, Hakamata Y, Nakano I. Topographical analysis of cortical nol. 2012;189:381–92. neuronal loss associated with disseminated selective neuronal necrosis 52. Posel C, Scheibe J, Kranz A, Bothe V, Quente E, Frohlich W, et al. Bone and infarction after repeated ischemia. Brain Res. 1997;767:154–7. marrow cell transplantation time-dependently abolishes efficacy of 31. Coutts SB, Modi J, Patel SK, Aram H, Demchuk AM, Goyal M, et al. What granulocyte colony-stimulating factor after stroke in hypertensive rats. causes disability after transient ischemic attack and minor stroke? Results Stroke. 2014;45:2431–7. from the CT and MRI in the Triage of TIA and minor Cerebrovascular Events 53. Garcia-Bonilla L, Moore JM, Racchumi G, Zhou P, Butler JM, Iadecola C, to Identify High Risk Patients (CATCH) Study. Stroke. 2012;43:3018–22. et al. Inducible nitric oxide synthase in neutrophils and endothelium 32. Zhao L, Nowak TS Jr. CBF changes associated with focal ischemic pre- contributes to ischemic brain injury in mice. J Immunol. 2014;193:2531–7. conditioning in the spontaneously hypertensive rat. J Cereb Blood Flow 54. Enzmann G, Mysiorek C, Gorina R, Cheng YJ, Ghavampour S, Hannocks Metab. 2006;26:1128–40. MJ, et al. The neurovascular unit as a selective barrier to polymorphonu- 33. Hoyte LC, Papadakis M, Barber PA, Buchan AM. Improved regional cer- clear granulocyte (PMN) infiltration into the brain after ischemic injury. ebral blood flow is important for the protection seen in a mouse model Acta Neuropathol. 2013;125:395–412. of late phase ischemic preconditioning. Brain Res. 2006;1121:231–7. 55. Ullrich N, Strecker JK, Minnerup J, Schilling M. The temporo-spatial locali- 34. Jickling GC, Liu D, Ander BP, Stamova B, Zhan X, Sharp FR. Targeting zation of polymorphonuclear cells related to the neurovascular unit after neutrophils in ischemic stroke: translational insights from experimental transient focal cerebral ischemia. Brain Res. 2014;1586:184–92. studies. J Cereb Blood Flow Metab. 2015;35:888–901. 56. Gelderblom M, Leypoldt F, Steinbach K, Behrens D, Choe CU, Siler DA, 35. Arsava EM, Gurer G, Gursoy-Ozdemir Y, Karatas H, Dalkara T. A new model et al. Temporal and spatial dynamics of cerebral immune cell accumula- of transient focal cerebral ischemia for inducing selective neuronal necro- tion in stroke. Stroke. 2009;40:1849–57. sis. Brain Res Bull. 2009;78:226–31. Tuor et al. BMC Neurosci (2016) 17:28 Page 14 of 14 57. Chu HX, Kim HA, Lee S, Moore JP, Chan CT, Vinh A, et al. Immune cell infil- 61. Ross AM, Lee CS. Description and identification of the peripheral immune tration in malignant middle cerebral artery infarction: comparison with response trajectories over time in first-time and recurrent stroke/transient transient cerebral ischemia. J Cereb Blood Flow Metab. 2014;34:450–9. ischemic attack. J Neurosci Nurs. 2015;47(5):256–62. 58. Tohgi H, Suzuki H, Tamura K, Kimura B. Platelet volume, aggregation, 62. Awadh M, MacDougall N, Santosh C, Teasdale E, Baird T, Muir KW. Early and adenosine triphosphate release in cerebral thrombosis. Stroke. recurrent ischemic stroke complicating intravenous thrombolysis 1991;22:17–21. for stroke: incidence and association with atrial fibrillation. Stroke. 59. D’Erasmo E, Aliberti G, Celi FS, Romagnoli E, Vecci E, Mazzuoli GF. Platelet 2010;41:1990–5. count, mean platelet volume and their relation to prognosis in cerebral 63. Kennedy J, Hill MD, Ryckborst KJ, Eliasziw M, Demchuk AM, Buchan AM. infarction. J Intern Med. 1990;227:11–4. Fast assessment of stroke and transient ischaemic attack to prevent early 60. Denes A, Pradillo JM, Drake C, Sharp A, Warn P, Murray KN, et al. Strep- recurrence (FASTER): a randomised controlled pilot trial. Lancet Neurol. tococcus pneumoniae worsens cerebral ischemia via interleukin 1 and 2007;6:961–9. platelet glycoprotein Ibalpha. Ann Neurol. 2014;75:670–83. Submit your next manuscript to BioMed Central and we will help you at every step: • We accept pre-submission inquiries • Our selector tool helps you to find the most relevant journal • We provide round the clock customer support • Convenient online submission • Thorough peer review • Inclusion in PubMed and all major indexing services • Maximum visibility for your research Submit your manuscript at www.biomedcentral.com/submit http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png BMC Neuroscience Springer Journals

Recurrent mild cerebral ischemia: enhanced brain injury following acute compared to subacute recurrence in the rat

Download PDF
14 pages

Loading next page...
 
/lp/springer-journals/recurrent-mild-cerebral-ischemia-enhanced-brain-injury-following-acute-nHQAoOxv2z

References (66)

Publisher
Springer Journals
Copyright
Copyright © 2016 by The Author(s)
Subject
Biomedicine; Neurosciences; Neurobiology; Animal Models
eISSN
1471-2202
DOI
10.1186/s12868-016-0263-x
pmid
27230275
Publisher site
See Article on Publisher Site

Abstract

Background: In the current study, a transient cerebral ischemia producing selective cell death was designated a mild ischemic insult. A comparable insult in humans is a transient ischemic attack ( TIA) that is associated with functional recovery but can have imaging evidence of minor ischemic damage including cerebral atrophy. A TIA also predicts a high risk for early recurrence of a stroke or TIA and thus multiple ischemic insults are not uncommon. Not well under- stood is what the effect of differing recovery times between mild ischemic insults has on their pathophysiology. We investigated whether cumulative brain damage would differ if recurrence of a mild ischemic insult occurred at 1 or 3 days after a first insult. Results: A transient episode of middle cerebral artery occlusion via microclip was produced to elicit mild ischemic changes—predominantly scattered necrosis. This was followed 1 or 3 days later by a repeat of the same insult. Brain damage assessed histologically 7 days later was substantially greater in the 1 day recurrent group than the 3 days recurrent group, with areas of damage consisting predominantly of regions of incomplete infarction and pannecrosis in the 1 day group but predominantly regions of selective necrosis and smaller areas of incomplete infarction in the 3 days group (P < 0.05). Enhanced injury was reflected by greater number of cells staining for macrophages/microglia with ED1 and greater alterations in GFAP staining of reactive astrocytes in the 1 day than 3 days recurrent groups. The differential susceptibility to injury did not correspond to higher levels of injurious factors present at the time of the second insult such as BBB disruption or increased cytokines (tumor necrosis factor). Microglial activation, with poten- tial for some beneficial effects, appeared greater at 3 days than 1 day. Also blood analysis demonstrated changes that included an acute increase in granulocytes and decrease in platelets at 1 day compared to 3 days post transient ischemia. Conclusions: Dynamic changes in multiple inflammatory responses likely contribute to the time dependence of the extent of damage produced by recurrent mild ischemic insults. The time of mild stroke recurrence is crucial with early recurrence producing greater damage than subacute recurrence and this supports urgency for determining and implementing optimal stroke management directly after a TIA. Keywords: Cerebral ischemia, Transient ischemic attack, Stroke, Recurrence, Inflammation, Granulocytes social and economic burden [1]. Often considered a warn- Background ing sign for stroke, a transient ischemic attack (TIA) Ischemic stroke remains a leading cause of death and a results from a cerebral ischemic episode of short duration major cause of adult disability resulting in a huge personal, associated with a temporary blockage of a cerebral artery resulting in transient (less than 24 h) functional deficits [ 2, *Correspondence: utuor@ucalgary.ca 3]. Although TIA’s are associated with substantial or full Department of Clinical Neurosciences and Hotchkiss Brain Institute, functional recovery, there is both experimental and clini- Cumming School of Medicine, University of Calgary, Calgary, AB T2N 4N1, cal evidence that reperfused brain may often suffer some Canada Full list of author information is available at the end of the article © 2016 The Author(s). This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/ publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Tuor et al. BMC Neurosci (2016) 17:28 Page 2 of 14 permanent damage such as scattered cell death or selec- to one of three groups involving double surgical proce- tive neuronal loss following a transient ischemic event dures: a control group with sham surgery  +  transient [4–7]. Understanding better the pathophysiology of such middle cerebral artery occlusion (MCAO) at 1  day and mild transient ischemic injury is of some urgency consid- experimental groups with MCAO + recurrent MCAO at ering an anticipated higher incidence of TIA as the popu- 1 day and MCAO + recurrent MCAO at 3 days (n =  6/ lation becomes increasingly elderly, and, an anticipated group). Sample size was selected to minimize num- increasing number of transient cerebral artery occlusions bers of animals used yet provide power as estimated to will be successfully recanalized with tissue plasminogen detect a 30  % difference between the anticipated means activator or endovascular therapy [8–10]. of groups assuming a standard deviation of 15  %. Ani- In addition to being key to treatment approaches, mals were euthanized 7  days after the last surgery and improved knowledge regarding the pathophysiology of a brains were processed for histology to determine the relatively short transient ischemic insult resulting in recov- extent of ischemic brain damage. In additional experi- ery is also of importance for understanding the interactions ments performed to investigate longitudinal cerebral of multiple events considering that a TIA is associated with ischemic changes, animals were randomized to MCAO an increased risk of a second or recurrent stroke [2, 3, 11]. and euthanasia at 1 or 3 days after the transient ischemia A majority of TIA’s are associated with cortical ischemic (n = 8/group). Peripheral systemic inflammatory changes events and large vessel disease and recurrence is most com- were also investigated in additional animals from blood monly associated with a second ipsilateral cortical ischemic sampled at either 1 or 3 days following transient MCAO event [12, 13]. However, surprisingly little is known of the (n = 5/group). interaction of the ischemic injury processes that occur with multiple mild ischemic insults where the insults are Transient mild focal ischemic insult sufficiently severe to cause selective cell death within the Surgical procedures were performed aseptically under ischemic territory. Duration between insults is likely a cru- isoflurane anesthesia and included analgesic measures to cial factor considering that ischemic injury and brain recov- alleviate suffering. On the day of the first surgery, a tran - ery evolve over time. Furthermore, the pathophysiology of sient mild ischemic insult (mild transient MCAO) was early recurrence is important to understand considering produced in the laboratory by temporarily occluding the recurrence is frequently observed by 1 day after a TIA [14]. distal MCA using a microsurgical approach as described u Th s the objective of the current study was to deter - previously [7, 15]. Briefly, a microaneurysm clip was mine whether cumulative damage from multiple mild placed on the MCA through a small burr hole in the tem- transient ischemic insults, each of sufficient severity poral bone above the MCA where it crosses the rhinal to cause selective necrosis, is influenced by the recov - fissure. Rectal temperature was maintained using a servo- ery time between insults. A secondary objective was to controlled heating lamp. A tail artery was cannulated for determine whether the evolution of cellular changes after obtaining arterial blood samples and a small burr hole the first insult were associated with the total damage over the ipsilateral parietal cortex was used to measure produced by the recurrent insult. We hypothesized that cerebral perfusion with laser Doppler flowmetry. Con - damage with an early recurrent stroke would be greater current to the MCAO, both common carotid arteries than with a subacute recurrent stroke. An animal model were transiently occluded using suture thread. At the end was considered key because ischemic severity could of the 30 min occlusion, the microclip and carotid artery be varied to produce mild ischemic cerebral changes ligatures were removed. A dura substitute (Gore Pre- as can be observed clinically. Furthermore the timing clude MVP, Better Hospital Supplies Corp., Miami, FL) between insults could be well controlled. Thus we used was positioned on closing to facilitate the production of a rat model of transient focal ischemia [7, 15] adjusted to a subsequent second MCAO. Topical anesthetic in addi- produce selective cell death, which in young Wistar rats tion to the administration of buprenorphine (0.03  mg/ required 30 min of occlusion; this was designated to be a kg, s.c. every 12 h as needed) provided analgesia. During mild transient cerebral ischemic insult. We also produced a sham operation, animals underwent the same surgical a recurrence of the insult at 1 or 3  days with the results procedures with the exception of vascular occlusion. Fol- demonstrating that brain damage is greater when recur- lowing surgery, rats were housed in separate cages with rence is acute (1 day) rather than subacute (3 days). free access to soft and hard food, water and environ- mental enrichment. Animals were monitored twice daily Methods for the first 2  days and then daily to ensure good recov - Male Wistar rats (Charles River, Montreal, Canada) ery. Animals were subjected to a recurrent mild focal were acclimatized to a 12  h light/dark cycle with free ischemic insult by repeating the 30  min MCAO proce- access to food and water. Animals were randomized dure at either 1 or 3 days of recovery following the initial Tuor et al. BMC Neurosci (2016) 17:28 Page 3 of 14 mild MCAO. The location of the second clip placement AB_2307440, Vector Laboratories Burlingame, CA) that on the MCA was immediately distal to the original posi- stains both microglia and vessels [19]. The same sec - tion to minimize surgical complications. tions stained for lectin were also stained using a primary mouse anti-rat endothelial barrier antigen (EBA) IgM Histology to detect brain injury or repair responses (AB_10120605, SMI 71, 1:400, Covance) that stained For histological analysis, pentobarbital anesthetized vessels [20] to assess blood–brain barrier dysfunction. rats were perfusion fixed with formalin and brains were Comparing EBA and lectin stained sections provided embedded in paraffin. Sections (6 μm) were stained with an additional detection of morphological and intensity standard and immunohistochemical methods and altered staining changes in microglia. Finally, integrity of the staining or ischemic injury were assessed blinded to the blood–brain barrier was assessed using an antibody to animal’s identification or surgical group. detect extravasation of large plasma proteins (1:200, goat anti rat immunoglobulin G (IgG), Jackson ImmunoRe- Standard staining with hematoxylin and eosin search Lab, West Grove, PA). In sections stained with hematoxylin and eosin, the For all the immunohistochemical stains, methods extent of cortical damage was identified by classifying were generally as described previously [7]. Briefly, par - ischemic injury in the ischemic middle cerebral artery affin sections were first processed for antigen retrieval territory as selective necrosis, incomplete infarction or using citrate buffer at pH  =  6, an exception being sec- pannecrosis as described in previous studies [5, 16]. In tions stained with Iba1. The sections were incubated addition, a score for brain injury was determined [7] by with either 10  % goat or donkey serum, then primary dividing the cortex into 4 regions of interest and a score antibody, followed by the appropriate biotin-conju- was assigned to each region graded as: 0 for normal, 1 for gated IgG (Jackson ImmunoResearch Lab, West Grove, <10 % of cellular injury, 2 for 10–50 % of cellular injury, 3 PA) or fluorescent conjugated secondary antibody. For for >50 % cellular injury and, 4 for confluent areas of pan - colorimetric staining, horseradish peroxidase conju- necrosis. A cumulative score was obtained by summing gated streptavidin (1:400, Dako, Burlington, ON) and the scores for each region. diaminobenzidine (Sigma) were applied. For double staining of lectin and EBA, antigen retrieval was omit- Immunohistochemical staining for cellular ischemic ted and after blocking with goat serum sections were changes incubated with FITC-lectin at room temperature for Macrophages/microglia and astrocytes are two glial cell 1 h followed by incubation with EBA at room tempera- types participating in the CNS inflammatory response ture overnight. to cerebral ischemia. The ischemic changes in these cells observed following 7  days of recovery after multi- Histological analysis of altered immunostaining ple ischemic insults also reflect the severity of ischemic Sections were visualized and analyzed blinded to the injury. Reactive astrocytes were assessed using a primary experimental group using an Olympus BX61 micro- antibody to glial fibrillary acidic protein (GFAP) (rabbit scope and Microbrightfield Stereo Investigator (MBF anti-GFAP, 1:10,000, AB_10013482, Dako, Burlington, Bioscience, Williston, VT). Immunohistochemical ON). Increased activation of macrophages/microglia were changes were semi-quantified according to the stain assessed using anti-ED1 (AB_2291300, mouse anti-rat used and the types of cellular changes observed in the CD68 clone ED1, 1:100, AbD Serotec, Raleigh, NC) [17]. MCA territory of the parietal cortex ipsilateral and con- The effect of a single mild ischemic insult on the pro - tralateral to the transient occlusion. For ED1 and TNF, gression of inflammatory changes between 1 and 3  days regions of parietal cortex within the MCA territory were assessed first by staining for reactive astrocytes were inspected and cells within at least three fields of and macrophages/microglia as above. Changes in the view (0.263 mm each) with positive staining for ED1 or cytokine, tumor necrosis factor (TNF) was assessed using TNF were counted and used to provide a mean value. goat anti rat TNF-alpha (1:100, AB_354511, R&D system, For EBA, fluorescent images of parietal cortex were Minneapolis, MN) [17]. In addition, early activation of captured digitally (200× magnification) and those with microglia was assessed using an antibody against Ion- technically good vessel staining in the contralateral ized Calcium-Binding Adapter Molecule 1 (Iba1, 1:1000, hemisphere were analyzed using Image J for counting AB_839504, Wako Chemicals, USA) which is specifically of the number of EBA stained vessels in representa- expressed in microglia and is upregulated following cer- tive ipsilateral and contralateral fields of view. GFAP ebral ischemia [18]. stained sections were assigned cumulative scores simi- Vascular changes were investigated using a fluo - lar to the hematoxylin and eosin stained sections, with rescein isothiocyanate-labeled tomato lectin (1:200, altered levels of staining being scored in each of the four Tuor et al. BMC Neurosci (2016) 17:28 Page 4 of 14 different cortical areas as: 0 for normal, 1 for scattered Results reactive cells labelled or <10  % of area, 2 for 10–50  % Brain damage varies with timing between multiple mild of the area labelled, 3 for >50  % of the area labelled insults and 4 for confluent areas of pannecrosis with total loss For animals receiving multiple surgical interventions, eight of GFAP stain. A lectin staining score was attained by animals were lost to analysis due to mortality post recurrent assessing the extent of increased tomato lectin staining stroke (n = 1), inadequate flow reduction during clip place - of activated microglia and their processes within the ment (n  =  2), surgical complications (n  =  2) and a prob- cortical parenchyma (i.e. in excess of vessels labelled lem with tissue embedding (n  =  2). A total of 18 animals with EBA) and scored according to: 0—none, 1—sparse (n  =  6/group) were analyzed in the sham  +  mild MCAO, numbers and 2—substantial numbers of parenchymal 1  day recurrent MCAO and 3  days recurrent MCAO cells with positive lectin staining. In regions of infarct groups. Body weights in these 3 groups were similar at base- a score of 3 was assigned if there were substantial num- line and time of the second surgery (290 ± 65 and 282 ± 50; bers of positive stained parenchymal cells in addition to 319  ±  56 and 314  ±  65; 296  ±  45 and 296  ±  40, respec- a diffuse increase in signal from within the parenchyma. tively). In all animals, ischemic changes were observed in the For Iba1 stained sections, those with the best positive cerebral cortex ipsilateral to the transient middle cerebral staining of processes were selected for counting both artery occlusion (Fig.  1b, d, e–h) but not in the contralat- the total number of microglial cells stained per field and eral cortex (Fig.  1a, c). Severity of injury varied between the number of activated microglial cells with ramified or groups (see also data in Additional file  1). Substantial areas thickened bushy processes. Potential increased colori- of incomplete infarct and pannecrosis in addition to some metric staining in cortex for plasma proteins with IgG areas of selective necrosis were apparent within the middle were assessed by measuring gray levels in ipsilateral and cerebral artery territory in all 6 animals subjected to a recur- contralateral cortex along with levels on the blank slide rent mild ischemic insult separated by 1  day (e.g. Fig.  1b, adjacent to the region of interest in order to analyze d). Substantial changes in ED1 and GFAP staining also left–right differences. occurred with recurrence at 1 day post first insult (Fig.  1j, l and o, q, respectively). In contrast, in animals subjected to Systemic circulatory response a recurrent mild ischemic insult separated by 3  days there Hematology analysis of blood samples was also per- were only small regions of incomplete infarction with more formed in additional rats to assess for signs of immuno- extensive regions of selective necrosis in four of the six ani- suppression or infection. Venous blood samples (0.2  ml) mals and only regions of selective necrosis in 2 animals (e.g. were collected in EDTA coated tubes and analyzed Fig.  1e, g). Positive staining of macrophages/microglia with immediately for concentrations of platelets and red and ED1 or for reactive astrocytes with GFAP was also modest white blood cells (HemaTrue Hematology Analyzer, (Fig.  1k, m, p, r). Brains of animals that underwent a sham Heska Corp., Loveland, CO, USA). Relative proportions procedure followed by transient ischemia 1 day later had the of the different types of white cells were also assessed by least extensive ischemic changes consisting of only selective measuring the numbers of monocytes, granulocytes and necrosis (6/6 animals) (e.g. Fig.  1f, h) and minimal staining lymphocytes. for ED1 or GFAP (3/3) (Additional file 1 ). Semi-quantitative analysis of the stained sections sup- Statistical analysis ported these observations (Fig. 1i). All groups had signifi - Data were analyzed using SigmaPlot 13 software (Sys- cantly greater damage (median cumulative score ≥3) in tat Software Inc, San Jose, CA). Data are reported as the ischemic cortex than in the contralateral hemisphere mean  ±  SD for groups with continuous values or as the which was normal (median score of 0). Animals with an median and first and third quartiles for nonparametric early recurrent stroke 1 day after the first insult had sig - data. Statistical comparisons were considered significant nificantly greater damage assessed in hematoxylin and at P < 0.05. Prior to a comparison of means, samples were eosin stained sections than a sham procedure followed first tested for normality and equal variance followed by by a mild MCAO. Total damage with early (1 day) recur- selection of an appropriate analysis of variance and post rence also exceeded that produced by recurrent stroke hoc test (e.g. Bonferroni). Differences between ipsilateral separated by 3  days (P  <  0.05). The differences in sever - and contralateral hemispheres were compared using a ity of damage and their dependence on the recovery time paired Student’s t-test or a Rank Sum Test for categorical between the first and recurrent insult were also reflected values. in an increased number of cells staining with ED1 Tuor et al. BMC Neurosci (2016) 17:28 Page 5 of 14 MCAOx2 b c Contralateral d 1 d ** Contralateral MCAOx2 1d †† ** Sham+MCAO MCAOx2 g 3d h e f †† ** 1d 1d 3d Recovery between Procedures MCAOx2 3d 600 Sham+MCAO n ** l MCAOx2 1d m MCAOx2 3d 1d 3d Recovery Before Recurrent MCAO MCAOx2 1d MCAOx2 3d MCAOx2 1d q s o MCAOx2 3d p * s 4 25 µm 1d 3d MCAOx2 1d MCAOx2 3d Recovery Before Recurrent MCAO Fig. 1 Cerebral injury dependence on recovery time between recurrence of mild transient ischemic insults. Representative cerebral cortical micro- graphs of sections at low and high magnification (scale bar of 25 µm) stained with: hematoxylin and eosin (a–h), ED1 for activated macrophages/ microglia (j–m), and GFAP for reactive astrocytes (o–r). Brain was perfusion fixed and processed 7 days following the last insult. Cerebral cortex con- tralateral to the transient middle cerebral artery occlusion (MCAO) is normal (A,C). Within cerebral cortex (b, d) after a mild transient MCAO followed by a recurrent MCAO at 1 day there is extensive pannecrosis (p). Selective necrosis (s) is common following a mild transient MCAO followed by a recurrent MCAO at 3 days (e, g). Selective necrosis also occurs in rats with a sham surgery followed 1 day later by transient MCAO (f, h). Semiquanti- tative assessment of injury scores indicated greater damage with 1 day compared to 3 days recurrence in the H&E sections (i). Immunohistochemi- cal staining with ED1 antibody demonstrated an increase in activated macrophages/microglia following recurrent MCAO produced at either 1 or 3 days after an initial MCAO (j–m). The mean number of cells per field with ED1 staining demonstrating a difference between 1 and 3 days recurrent MCAO groups (n). GFAP staining of astrocytes was increased with recurrent MCAO at 1 or 3 days (o–r). The median score for altered GFAP staining reflected greater injury and astrocytic changes in brains of animals with recurrent MCAO at 1 day than 3 days (s). n = 6/group. *P < 0.05; **P < 0.006, † †† ‡ Ipsilateral different from contralateral; P < 0.05; P < 0.006, different from 1 day Recurrent MCAO. P < 0.01, different from Sham (Fig. 1n) and the severity score for differences in staining blood gases were similar between 1 and 3 days recurrent for reactive astrocytes with GFAP (Fig. 1s). groups. Mean values were 86 ± 11 and 88 ± 11 mm Hg for mean arterial blood pressure, 102  ±  7 and 109  ±  5 Similar physiological measures in recurrent stroke groups for PO2, 44.2  ±  2 and 38.6  ±  2  mm Hg for PCO2, Potential differences in severity of ischemia or other 7.35  ±  0.01 and 7.36  ±  0.04 for PH and 13.6  ±  3 and measured physiological parameters that are well known 11.6 ± 2 mmol/L for blood glucose, respectively. See also to affect ischemic injury were comparable between data in Additional file 2. groups. Cortical perfusion was less than 10  % baseline during MCAO and reperfusion levels were similar dur- Inflammatory changes following a single mild transient ing both the first and second ischemic insult irrespective ischemia of the recovery time between them (Fig.  2a). There was In order to investigate whether differential responses to also excellent control of body temperature during MCAO brain injury and recovery from ischemia could explain and during early reperfusion, irrespective of the experi- the ensuing damage produced by a recurrent MCAO, mental group (Fig. 2b). Other physiological variables and longitudinal tissue changes to a single mild MCAO were Ipsilateral H&E (Score) Ipsilateral ED1 Ipsilateral GFAP (Cells/Field) (Score) Tuor et al. BMC Neurosci (2016) 17:28 Page 6 of 14 which had minimal TNF staining (Fig.  4a), increased numbers of positive TNF stained cells were observed 1 3 d ipsilaterally at 1 and 3  days post-insult (e.g. Fig.  4b, c). The mean increased number of TNF stained cells were not different statistically at 1 and 3  days post insult (Fig. 4d). See also data in Additional file 4. Microglial changes, which are usually considered pro- inflammatory acutely but can also contribute later to 0 repair [21], were altered following a single mild insult. During Post During Post Microglial cells stained with Iba1 (Fig.  4e–h) were simi- st nd 1 MCAO 2 MCAO lar in the 1 and 3  days groups contralaterally (e.g. mean of 23.0 ± 6.0 cells per field). Total numbers of cells ipsi - laterally increased to 32.4  ±  9 and 64.9  ±  25 in the 1 1 3 d and 3  days groups respectively (P  <  0.05) and included cells with ramified or bushy morphology or round cells (Fig.  4f, g). Those with ramified or bushy morphology were also different between groups (Fig.  4h). See also data in Additional file  4. Using a tomato lectin stain of vessels and microglia, examination of staining in the non-vascular parenchyma contralaterally indicated a During Post During Post lack of appreciable staining for microglial cells with lec- st nd 1 MCAO 2 MCAO tin [22] (e.g. Fig.  4i). At 1  day post MCAO there was a Fig. 2 Mean cortical perfusion or rectal temperature during or post modest increase in lectin stained microglia and their pro- middle cerebral artery occlusion (MCAO). a Cortical perfusion meas- cesses, associated with microglial activation (e.g. Fig. 4j). ured using Doppler flowmetry presented as a percent of baseline. At 3  days post-insult, substantial numbers of microglia Mean perfusion decreased to <8 % baseline during MCAO subse- quently returning toward baseline within the first 5 min following stained for lectin (e.g. Fig.  4k) and the median scores reperfusion irrespective of MCAO recurrence at either 1 or 3 days post for lectin staining differed significantly between groups the first MCAO (n = 6/group). b Mean core body temperature was (P < 0.05) (Fig. 4l). well controlled resulting in similar means for the recurrent groups either during or after the first or second MCAO BBB injury following a mild transient ischemia In sections stained with lectin, we also performed immunostaining for endothelial barrier antigen (EBA) investigated in additional randomized animals. Several whose loss is considered to provide a highly sensitive of these were lost to analysis due to mortality (n  =  1), indicator of dysfunction of the blood–brain barrier [20]. surgical complications (n  =  2) and a problem with tis- There was reduced vascular EBA immunoreactivity sue embedding (n  =  1). Sections from all animals ana- observed following a mild ischemic insult. Compared to lyzed following a single transient MCAO demonstrated the contralateral hemisphere (Fig. 4m), there were fewer mild damage in hematoxylin and eosin stained sections EBA stained vessels (Fig. 4n–o) and analysis of sections (Fig. 3a, b). Damage score was similar in the 1 and 3 days determined a reduced number of EBA stained vessels recurrent MCAO groups (n = 8/group) (Fig. 3c). In adja- both at 1 and 3 days post a mild MCAO (Fig. 4p). There cent stained sections, there were increased numbers of was no significant difference between groups. Despite ED1 stained macrophages/microglia ipsilaterally in both a reduction in EBA staining there was no evidence of groups (Fig. 3d, e). There was also altered GFAP staining major BBB damage. Following a single mild ischemic of reactive astrocytes (Fig.  3g, h) compared to contralat- insult, there was a lack of appreciable positive staining eral cortex. These inflammatory changes scored in the 1 for IgG within parietal cortex (not shown), indicating and 3  days recurrent groups were not different statisti - that the blood–brain barrier was intact to passage of cally (Fig. 3f, i). See also data in Additional file 3. large plasma proteins detectable with IgG. Quantita- In sections from these animals we also investigated tive analysis of these sections demonstrated there was whether TNF-alpha, often considered to be a pro-inflam - no difference in the darkness of staining in ipsilateral matory cytokine, is substantially increased 1 day post a versus contralateral cortex for the 1 and 3  days groups mild ischemic insult and thereby a potential contributor (−0.7  ±  1.2 and −0.1  ±  1.9 gray levels, respectively) to increased ischemic damage observed with multiple (Additional file 4 ). insults at this time. Relative to the contralateral cortex, Cortical perfusion Rectal Temperature ( C) (% Baseline) Tuor et al. BMC Neurosci (2016) 17:28 Page 7 of 14 MCAO+1d MCAO+3d ** ab c ** 1d 3d Time post mild MCAO de f 600 1d 3d Time post mild MCAO gh i 1d 3d Time post mild MCAO Fig. 3 Acute and subacute cortical responses to a single mild ischemic insult. a, b Hematoxylin and eosin (H&E) stained sections from brains perfusion fixed at 1 or 3 days following a mild transient middle cerebral artery occlusion (MCAO). c The median and first and third quartiles of the damage score was not different statistically between groups. d, e ED1 stained sections from brains 1 and 3 days post MCAO. f Mean numbers of ED1 cells stained per field were not different statistically. g, h GFAP stained sections from animals 1 and 3 days post MCAO. i Median and first and third quartiles of the GFAP score were similar between groups (P > 0.05). n = 8/group *P < 0.05; **P < 0.005, ipsilateral different from contralateral. (Mann–Whitney Rank Sum test) Systemic inflammation following a mild transient ischemia compared to levels either in naïve animals or in blood Considering the evidence that systemic inflammation samples collected 3  days after MCAO (Fig.  5c). Altered or suppression of the immune system can play a role in lymphocyte and granulocyte concentrations but no influencing ischemic damage [21, 23, 24], we also investi- change in platelets were also observed 1 day following the gated the potential contribution of systemic immune cell stress of a sham surgery. However, the increase in granu- changes to the enhancement of ischemic damage by per- locytes alone was insufficient to enhance damage; sham forming a hematology analysis of blood samples collected surgery with increased granulocytes and a mild ischemic at 1 or 3 days following a mild MCAO or a sham surgery. insult 1  day later produced only mild ischemic changes Blood samples from a naïve group of rats provided an (Fig. 1c). additional control (n  =  5/group). Groups had a simi- lar hematocrit (Fig.  5d and Additional file  5). Following Discussion a mild MCAO there was a decrease in the lymphocyte The current results demonstrate the major importance percentage of white blood cells and the platelet concen- of early recurrence of a mild transient ischemic insult tration at 1  day but not 3  days relative to naïve controls following an initial TIA. Distinctive in our study, com- (Fig.  5a, b). In contrast, there was a transient increase in pared to studies of preconditioning, is the investigation the granulocyte percentage of white blood cells at 1  day of multiple transient ischemic insults—both of sufficient Ipsilateral ED1 Ipsilateral H&E Ipsilateral GFAP (Cells/Field) (Score) (Score) Tuor et al. BMC Neurosci (2016) 17:28 Page 8 of 14 1d 3 d Contralateral Ipsilateral a bc 1d 3d 25 M Time post mild MCAO 80 0 60 0 40 0 20 0 25 M 1d 1d 3d 3d Time post mild MCAO i j 4 k 2 ** 1d 3d 25 M Time post mild MCAO ** ** 1d 3d 25 M Time post mild MCAO Fig. 4 Cerebral Inflammatory and blood–brain barrier marker changes following a single mild transient MCAO. a Representative sections stained from contralateral cortex for the cytokine, tumor necrosis factor ( TNF). b, c TNF immunostained sections from ipsilateral cortex at 1 or 3 days post MCAO, respectively. d Mean number of TNF stained cells per field is similarly increased at 1 or 3 days post MCAO (n = 8/group). Representative positive staining of microglia with Iba1 in contralateral cortex e and ipsilateral cortex at 1 and 3 days post MCAO (f, g). Activated microglia were observed at 1 day and these were more numerous at 3 days post MCAO (h, n = 8/group). Double staining of vessels and microglia with tomato lectin (i–l) and vessels with endothelial barrier antigen (EBA) (m–p). Staining of microglia with lectin was minimal in the contralateral cortex (e.g. i) and increased (e.g. arrows) in ipsilateral cortex at 1 day (j) or 3 days (k) post transient MCAO. l Median scores for lectin staining (n = 8/group). m Representative staining of vessels for endothelial barrier antigen (EBA) in contralateral cortex (12 animals with good contralateral positive staining quantified). n, o EBA staining in ipsilateral cortex from animals at 1 and 3 days post MCAO. p Reduced percentage of vessels stained with EBA ipsilat - erally relative to contralaterally (n = 6/group). *P < 0.05; **P < 0.005, ipsilateral different from contralateral. P < 0.01 lectin stained microglia differ between 1 and 3 days post MCAO groups severity to cause mild ischemic damage. We found that in influencing brain damage and supports urgency for a recurrent mild insult following a TIA in rats produced determining and implementing optimal stroke preven- substantially more brain damage when recurrence is tion management early after TIA to avoid a second acute (i.e. 1  day) than with a subacute recovery time of ischemic event. 3  days between insults. The cause of this difference is likely multifactorial and potentially includes an augmen- Interaction between two Mild Ischemic Insults tation of damage by systemic inflammatory changes. The principal finding of the current study was that the Irrespective of the mechanisms, the results demonstrate combined damage produced by multiple mild insults that the time of mild stroke recurrence can be crucial was influenced by their timing. In contrast to that in Ipsilateral EBA Ipsilateral TNF Ipsilateral Lectin Ipsilateral Iba1 (Vessels % Contral) (Cells/Field) (Score) (Cells/Field) Tuor et al. BMC Neurosci (2016) 17:28 Page 9 of 14 b † ‡ ‡ Sham MCAO MCAO Sham MCAO MCAO Naive Naive +1d +1d +3d +1d +1d +3d c d ‡‡ ‡‡ Sham MCAO MCAO Sham MCAO MCAO Naive Naive +1d +1d +3d +1d +1d +3d Fig. 5 Systemic changes following a Single Mild Transient MCAO. A complete blood count of blood samples from naïve animals or animals at 1 or 3 days post a mild transient MCAO. a The concentration of lymphocytes, presented as a % of total white blood cells, was decreased at 1 day post transient MCAO when compared to naïve animals. b The mean concentration of platelets was also decreased at 1 day post MCAO. c The mean number of granulocytes, normalized to the total number of white blood cells, was transiently increased at 1 day. d Mean hematocrit was similar in ‡ ‡‡ † all groups. N = 5/group. P < 0.05; P < 0.005, different from Naive. P < 0.05, different from 1 day post MCAO (ANOVA and Bonferroni t-test) the current study, previous studies of multiple ischemic also observed an increase in the damage produced by a insults have generally focussed on studies where the recurrent compared to a single MCAO, but differences first or second insult is a very short preconditioning or were most marked for an acute (1 day) rather than a suba- postconditioning ischemia that alone produces no per- cute (3 days) recurrence. Extensive infarction rather than manent cellular damage; such a non-damaging ischemia partial infarction and/or selective necrosis was produced has protective effects [25–28]. Indeed in the current when stroke recurrence was acute rather than subacute. study the group with a sham procedure prior to transient In addition to this finding, we observed several differen - MCAO had reduced damage compared to that of a sin- tial systemic and cerebral inflammatory changes at acute gle MCAO assessed 3  days later. Using an initial insult and subacute times following a mild transient ischemic that is more injurious, one laboratory has examined the insult providing novel insights into the pathophysiology effects of repeated ischemia in the gerbil using only ani - of TIA and potential contributions to recurrent stroke mals that displayed neurological signs of stroke after a damage. These results are of potential relevance clinically first temporary carotid occlusion and subjecting them considering an anticipated rise in the incidence of TIA, to additional episodes of the same duration of ischemia both due to our ageing population and an escalation in [29, 30]. Although small numbers (3–5) per group were number of ischemic stroke patients that will be treated investigated, their results suggested a transition from with early reperfusion therapy [8–10]. Understanding mild to more severe injury with increasing numbers of better the pathophysiology of early recurrence is also insults and greater damage when second insults were at essential considering that the median time for recurrence 3 or 5 versus 48 h apart. Recently, we used a model of two of functional deterioration related to a vascular event fol- relatively short transient ischemic episodes, produced by lowing a TIA is 1 day [14]. Furthermore, although debili- microclip occlusion of the MCA and separated by 3 days tating strokes are relatively infrequent following a TIA to demonstrate an increased damage with recurrent alone, when there is a recurrent event, 53  % have been stroke compared to a single insult [15]. The present study found to be debilitating [31]. Lymphocytes (%) Granuloyctes(%) Hematocrit (%) Platelets (x10 /L) Tuor et al. BMC Neurosci (2016) 17:28 Page 10 of 14 Physiological and vascular changes was increased within areas of selective necrosis at 2 and Our physiological measures determined that various 4 weeks post insult followed by a decline in OX42 stain- factors known to influence infarct size such as body ing thereafter [5, 6, 36]. We investigated acute microglial/ temperature, blood pressure and blood gases were simi- macrophage activation using Iba1, tomato lectin and ED1 lar between the groups and thus are unlikely to account staining and observed a consistent early (1 day) increase for the effect of different timing of recurrent insults on in Iba1 stained microglia but sparse positive ED-1 stain- brain damage. Although preconditioning ischemia has ing in 3/8 animals. Staining for activated microglia/mac- been observed by others to improve blood flow during rophages using Iba1 and lectin staining was increased ischemia compared to a single severe ischemic insult at 3  days compared to 1  day post a mild ischemic insult [32, 33], the levels of cerebral blood flow reduction were and ED-1 staining was present in 6/8 animals. This sup - equivalent during both insults and for both groups. With ports that microglial activation occurs as early as 1  day respect to reperfusion, there were no significant statisti - post a mild ischemic insult, consistent with previous cal differences in perfusion within the first 5 min of rep - observations of early microglial activation in peri-infarct erfusion. However, there was a variable return of flow to regions i.e. regions which likely experienced mild tran- baseline, with some animals demonstrating hyperperfu- sient ischemia; and, this is followed by the appearance of sion and this variance was associated with suboptimal macrophages within the core and peri-infarct region at statistical power requiring caution for deciding the group 3–5 days post-insult [37]. There is a complex and incom - means were the same. Also a limitation presently was pletely understood progression of activation of different that only acute reperfusion was monitored and the possi- microglial phenotypes (e.g. macrophages that are of an bility remains that at a delayed hypoperfusion could have M1 or M2 phenotype), however, there is agreement in adversely influenced damage. general that activated microglia appear to be involved in Regarding vascular changes, there was little evidence both necrotic and repair responses [21, 37–39]. Thus it for differences in disruption of the blood brain barrier at is possible that in the present study, increased microglial early versus subacute times. There was a lack of detect - activation observed at 3  days provides some neuropro- able extravasation of IgG at either time point in agree- tection and acts to help reduce damage during the sec- ment with previous reports of a lack of vasogenic edema ond insult via expression of anti-inflammatory cytokines detectable in T magnetic resonance images following a such as IL-10 or TGF-β and factors such as IGF-1 [21, mild transient ischemic insult [5, 7]. The reduced stain - 37, 39]. Future elucidation of the pathophysiology of the ing for EBA did suggest some endothelial injury at both 1 polarity changes of the various immune cells activated and 3 days post a single MCAO. Early ischemic endothe- following ischemia should consider identifying the spe- lial injury is associated with an acute activation of the cific mediators within the microenvironment in response endothelium resulting in an increased expression of to differing severities of cerebral ischemia—including endothelial adhesion molecules associated with endothe- a mild transient ischemia. Presently, evidence was lack- lial injury [21, 24, 34]. However, since reductions in EBA ing for a potential involvement of a differential release were similar at both time points, there was no direct evi- of pro-inflammatory neurotoxic cytokines enhancing dence for greater dysfunction in the blood–brain barrier acute recurrent ischemic damage considering staining for at 1 than 3  days to help account for greater enhanced TNF-alpha tended to be greater rather than less at 3 days. damage of a recurrent 1 day insult. Systemic inflammation Cerebral inflammatory changes Examination of peripheral blood following a short tran- We also investigated whether the first mild ischemic sient ischemia indicated that even a mild ischemic insult insult could produce factors in the tissue that could either evoked alterations in inflammatory cells within the blood. augment ischemic damage acutely (i.e. at 1  day) or pro- Although the complex and dynamic immune responses vide some later (3 days) neuroprotection. The literature is post ischemia are still being characterized in detail and sparse regarding acute glial inflammatory changes in the their roles in affecting ischemic outcomes remain con - brain following a mild ischemic insult mimicking a TIA. troversial, numerous studies indicate that ischemic At relatively chronic times (1–4  weeks) following a mild alterations in peripheral inflammatory cells can modify focal transient ischemia increased astroglial reactivity brain damage [21, 23, 24]. Following the mild transient using GFAP has been reported previously [5, 7, 35, 36]. ischemia produced currently, we found alterations in the Similarly, at chronic times post-insult, microglial activa- concentrations of lymphocytes, granulocytes and plate- tion detected as positive immunostaining for OX42 (an lets in the blood. immune marker for CD11b which also stains activated Regarding changes in lymphocytes, both our sham neutrophils and microglial derived macrophages) [37] animals and animals 1  day post a mild ischemic insult Tuor et al. BMC Neurosci (2016) 17:28 Page 11 of 14 demonstrated a decrease in peripheral lymphocytes changes in various molecular signals and their interac- compared to our naïve controls. This is consistent with tions on multiple cell types with potential differences the lymphopenia associated with immune suppression according to species. and an increased susceptibility to infection that is well Currently, the neurovascular unit is considered a key recognized to occur post stroke in humans and experi- site of neutrophil action at delayed times following tran- mental animals [40–43]. The reduction in lymphocytes sient cerebral ischemia. Following transient ischemia, with the current mild transient ischemia produced by Ly6G positive neutrophil infiltration into the brain 30  min clip occlusion was rather modest (10–15  %) and parenchyma was minimal or occurred within the vicin- short lasting compared to the >50  % reduction in lym- ity of vessels [54, 55]. Also following transient ischemia phocytes observed for at least 2  weeks following 60  min in the mouse, homogenized brain samples analyzed of transient MCAO in mice using a transient intralumi- with flow cytometry demonstrated a delayed (2–3  days) nal thread occlusion [40]. Presently, lymphocyte numbers increase in neutrophils not present at 1 day [53, 56]. Note returned towards naïve control levels already by 3  days that neutrophil migration appears to occur sooner and is post-insult. more intense without reperfusion; it has been observed The white blood cell analysis also provided a count of early (e.g. at 1  day) following permanent ischemia [47, polymorphonuclear granulocytes which consists of pre- 57]. Although there may be a lack of migration into brain dominantly neutrophils along with eosinophils and baso- acutely following a single transient ischemia, granulo- phils. The increase in granulocytes observed at 1  day cyte recruitment, activation and infiltration [34, 37] on post-insult is consistent with reports of increased gran- primed cerebral endothelium may be greatly acceler- ulocytes following stroke in humans and experimental ated by the subsequent flow disruption produced with a animals [41, 43–46]. The origin of the additional granulo - recurrent ischemic insult. A contribution from platelets cytes can vary with the type of injury but includes mobi- is also possible as supported by their decrease at 1  day lization from the spleen and bone marrow and potentially compared to 3  days post-insult possibly reflecting con - increased production and decreased apoptosis [34, 37]. tinued activation/interaction with cerebral ischemic The mechanism mediating the increase likely includes a vasculature. Reduced platelet counts have also been stress response [23, 41] which is consistent with previous observed in stroke patients and there is evidence for their reports and our observation of elevated numbers of gran- interaction with peripheral inflammatory responses and ulocytes and decreased numbers of lymphocytes, also in ischemic injury [58–60]. Additional research is required animals subjected to sham surgery [40, 47, 48]. With a to clarify the importance of granulocyte and platelet mild transient ischemic insult, we observed an increase interactions and the dynamic changes in injured cerebro- in granulocytes that was temporary with a normaliza- vascular endothelium following an initial mild ischemic tion by 3  days post MCAO. A similar return towards insult prior to a second TIA. The possibility for different baseline was observed in mice with either sham surgery peripheral immune responses occurring with a first TIA or MCAO using a transient ligature [48]. Important to versus a recurrent stroke/TIA should also be considered note is that the stress induced neutrophilia would appear [61]. insufficient to enhance damage because a sham surgery 1  day prior to mild transient MCAO reduced ischemic Conclusion damage indicating that a combination of both transient To conclude, the production of multiple mild ischemic cerebral ischemia and neutrophilia would be involved in insults with different recovery times between them exacerbating damage. demonstrated an important dependence on the tim- Indeed, the interaction of granulocytes or neutrophils ing between insults. The damage observed with a short with cerebral ischemia is a complex operation of dynamic recovery of 1 day following a mild transient ischemia sub- changes in neutrophil activation at various sites (e.g. vas- stantially exceeded that with a subacute recovery time of cular and parenchymal) underlying multiple functions 3  days. The results might help explain clinical reports of (e.g. blood–brain barrier disruption, thrombus/clot for- increased risk of recurrent stroke in the first day follow - mation or neurotoxicity) [34]. Both clinical and experi- ing a TIA/minor stroke or the sudden clinical deteriora- mental studies have reported an increased accumulation tion observed in some cases soon after stroke [62, 63]. The of neutrophils in the brain after an ischemic stroke [34, mechanisms involved in the response are likely complex 37, 49]; and, there is evidence for their role in enhancing involving a progression of systemic and cerebral/vascular brain damage but also evidence for their lack of an effect inflammatory changes after the first mild ischemic insult or for producing beneficial effects via neuroprotection that affect the second insult. Irrespective, the results indi - or an involvement in brain repair e.g. [34, 37, 50–53]. cate that in the absence of clinically available therapeutic This variability may reflect the array of time dependent therapy to attenuate injury following a transient ischemic Tuor et al. BMC Neurosci (2016) 17:28 Page 12 of 14 Ethics attack or its recurrence, patients should receive immedi- This experimental animal study was carried out in accordance with the guide- ate care to best manage and prevent stroke recurrence lines and policies of the Canadian Council on Animal Care for the ethical care e.g. with early investigation of thromboembolic sources, and handling of animals. The current study was approved by the University of Calgary Health Sciences Animal Care Committee (Protocols M11017 and intensive antiplatelet therapies, lipid lowering, and/or AC15-0043). anticoagulation for cardioembolic stroke. Funding This study was supported by funding from the Canadian Institutes for Health Additional files Research (Grant MOP111037). This funding body had no role in the design of the study nor in the collection, analysis and interpretation of data nor in writ- ing of the manuscript. Additional file 1. Histology assessments for Figure 1. Shown are the data for each animal for the 1d and 3d recurrent (Recur.) groups and the Sham Received: 17 November 2015 Accepted: 11 May 2016 (sh) plus 1d mild (mld) stroke group. The descriptive assessments, the H&E scores, the ED1 counts and the GFAP scores are presented. Additional file 2. Physiological measures for each animal in the 1d and 3d recurrent mild ischemic insult groups. Shown are the doppler flow measures (% baseline), rectal temperature, mean arterial blood pressure (MABP) and blood gas values. References 1. Black M, Wang W, Wang W. Ischemic stroke: From next gen- Additional file 3. Histological assessments for Figure 3. Shown are the eration sequencing and GWAS to community genomics? OMICS. data for each animal at either 1d or 3d post a single mild ischemic insult. 2015;19:451–60. The H&E scores, the ED1 counts and the GFAP scores are presented. 2. Easton JD, Saver JL, Albers GW, Alberts MJ, Chaturvedi S, Feldmann E, Additional file 4. Histological assessments for Figure 4. Shown are the et al. Definition and evaluation of transient ischemic attack: a scientific data for each animal at either 1d or 3d post a single mild ischemic insult. statement for healthcare professionals from the American Heart Associa- Positive staining counts for TNF, Iba1 and EBA in addition to the Lectin tion/American Stroke Association Stroke Council; Council on Cardiovas- scores and IgG gray level measures are presented. cular Surgery and Anesthesia; Council on Cardiovascular Radiology and Intervention; Council on Cardiovascular Nursing; and the Interdisciplinary Additional file 5. Complete blood analysis values for each of the animals Council on Peripheral Vascular Disease. The American Academy of in each group—the naïve control group, the control group with sham Neurology affirms the value of this statement as an educational tool for middle cerebral artery (MCA) surgery, the 1d post MCA occlusion and 3d neurologists. Stroke. 2009;40:2276–93. post MCA occlusion groups. Presented are the numbers of white blood 3. Kernan WN, Ovbiagele B, Black HR, Bravata DM, Chimowitz MI, Ezekowitz cells, lymphocytes, monocytes, granulocytes (x109/L) and the numbers MD, et al. Guidelines for the prevention of stroke in patients with stroke of lymphocytes, monocytes and granuloctyes (% total white blood cells). and transient ischemic attack: a guideline for healthcare profession- Hematocrit and numbers of platelets are also presented. als from the American Heart Association/American Stroke Association. Stroke. 2014;45:2160–236. 4. Baron JC, Yamauchi H, Fujioka M, Endres M. Selective neuronal loss in Abbreviations ischemic stroke and cerebrovascular disease. J Cereb Blood Flow Metab. EBA: endothelial barrier antigen; GFAP: glial fibrillary acidic protein; IgG: immu- 2014;34:2–18. noglobulin G; MCAO: middle cerebral artery occlusion; TIA: transient ischemic 5. Ejaz S, Williamson DJ, Ahmed T, Sitnikov S, Hong Y T, Sawiak SJ, et al. attack; TNF: tumor necrosis factor. Characterizing infarction and selective neuronal loss following temporary focal cerebral ischemia in the rat: a multi-modality imaging study. Neuro- Authors’ contributions biol Dis. 2013;51:120–32. All authors helped to draft the manuscript. In addition, UIT participated 6. Ejaz S, Emmrich JV, Sawiak SJ, Williamson DJ, Baron JC. Cortical selective in study conception, design and data interpretation and contributed to neuronal loss, impaired behavior, and normal magnetic resonance histological assessment and statistical analysis of the data. ZZ carried out the imaging in a new rat model of true transient ischemic attacks. Stroke. recurrent stroke surgeries and participated in coordination of the study and 2015;46:1084–92. data interpretation. PAB participated in study conception and design and 7. Qiao M, Zhao Z, Barber PA, Foniok T, Sun S, Tuor UI. Development interpretation of the study. MQ participated in the recurrent stroke surgeries of a model of recurrent stroke consisting of a mild transient stroke and carried out the histological procedures and assessment of histological followed by a second moderate stroke in rats. J Neurosci Methods. injury. All authors read and approved the final manuscript. 2009;184:244–50. 8. Berkhemer OA, Fransen PS, Beumer D, van den Berg LA, Lingsma HF, Yoo Author details AJ, et al. A randomized trial of intraarterial treatment for acute ischemic Department of Clinical Neurosciences and Hotchkiss Brain Institute, Cum- stroke. N Engl J Med. 2015;372:11–20. ming School of Medicine, University of Calgary, Calgary, AB T2N 4N1, Canada. 9. Campbell BC, Mitchell PJ, Kleinig TJ, Dewey HM, Churilov L, Yassi N, et al. Department of Physiology and Pharmacology, University of Calgary, Calgary, Endovascular therapy for ischemic stroke with perfusion-imaging selec- AB T2N 4N1, Canada. tion. N Engl J Med. 2015;372:1009–18. 10. Goyal M, Demchuk AM, Menon BK, Eesa M, Rempel JL, Thornton J, et al. Acknowledgements Randomized assessment of rapid endovascular treatment of ischemic We gratefully acknowledge the technical assistance during the surgical experi- stroke. N Engl J Med. 2015;372:1019–30. ments provided by Melissa Morgunov. The animal resources centre performed 11. Bal S, Patel SK, Almekhlafi M, Modi J, Demchuk AM, Coutts SB. High the hematology analysis of the blood samples. rate of magnetic resonance imaging stroke recurrence in crypto- genic transient ischemic attack and minor stroke patients. Stroke. Availability of supporting data 2012;43:3387–8. The data sets supporting the results of this article are included within the 12. Kappelle LJ, Van Latum JC, Van Swieten JC, Algra A, Koudstaal PJ, van Gijn article and its additional files. J. Recurrent stroke after transient ischaemic attack or minor ischaemic stroke: Does the distinction between small and large vessel disease Competing interests remain true to type? Dutch TIA Trial Study Group. J Neurol Neurosurg The authors declare that they have no competing interests. Psychiatry. 1995;59:127–31. Tuor et al. BMC Neurosci (2016) 17:28 Page 13 of 14 13. Johansson E, Cuadrado-Godia E, Hayden D, Bjellerup J, Ois A, Roquer J, 36. Hughes JL, Beech JS, Jones PS, Wang D, Menon DK, Baron JC. Mapping et al. Recurrent stroke in symptomatic carotid stenosis awaiting revascu- selective neuronal loss and microglial activation in the salvaged neocorti- larization: a pooled analysis. Neurology. 2016;86:498–504. cal penumbra in the rat. Neuroimage. 2010;49:19–31. 14. Coutts SB, Modi J, Patel SK, Demchuk AM, Goyal M, Hill MD. CT/CT 37. Benakis C, Garcia-Bonilla L, Iadecola C, Anrather J. The role of microglia angiography and MRI findings predict recurrent stroke after transient and myeloid immune cells in acute cerebral ischemia. Front Cell Neurosci. ischemic attack and minor stroke: results of the prospective CATCH study. 2014;8:461. Stroke. 2012;43:1013–7. 38. Perego C, Fumagalli S, De Simoni MG. Temporal pattern of expression and 15. Clark D, Tuor UI, Thompson R, Institoris A, Kulynych A, Zhang X, et al. Pro- colocalization of microglia/macrophage phenotype markers following tection against recurrent stroke with resveratrol: endothelial protection. brain ischemic injury in mice. J Neuroinflammation. 2011;8:174. PLoS ONE. 2012;7:e47792. 39. Patel AR, Ritzel R, McCullough LD, Liu F. Microglia and ischemic stroke: a 16. Garcia JH, Liu KF, Ye ZR, Gutierrez JA. Incomplete infarct and delayed neu- double-edged sword. Int J Physiol Pathophysiol Pharmacol. 2013;5:73–90. ronal death after transient middle cerebral artery occlusion in rats. Stroke. 40. Prass K, Meisel C, Hoflich C, Braun J, Halle E, Wolf T, et al. Stroke-induced 1997;28:2303–9. immunodeficiency promotes spontaneous bacterial infections and is 17. Qiao M, Meng S, Foniok T, Tuor UI. Mild cerebral hypoxia-ischemia mediated by sympathetic activation reversal by poststroke T helper cell produces a sub-acute transient inflammatory response that is less type 1-like immunostimulation. J Exp Med. 2003;198:725–36. selective and prolonged after a substantial insult. Int J Dev Neurosci. 41. Dirnagl U, Klehmet J, Braun JS, Harms H, Meisel C, Ziemssen T, et al. 2009;27:691–700. Stroke-induced immunodepression: experimental evidence and clinical 18. Ito D, Tanaka K, Suzuki S, Dembo T, Fukuuchi Y. Enhanced expression of relevance. Stroke. 2007;38:770–3. Iba1, ionized calcium-binding adapter molecule 1, after transient focal 42. Urra X, Cervera A, Villamor N, Planas AM, Chamorro A. Harms and benefits cerebral ischemia in rat brain. Stroke. 2001;32:1208–15. of lymphocyte subpopulations in patients with acute stroke. Neurosci- 19. Acarin L, Vela JM, Gonzalez B, Castellano B. Demonstration of poly- ence. 2009;158:1174–83. N-acetyl lactosamine residues in ameboid and ramified microglial 43. Vogelgesang A, Grunwald U, Langner S, Jack R, Broker BM, Kessler C, et al. cells in rat brain by tomato lectin binding. J Histochem Cytochem. Analysis of lymphocyte subsets in patients with stroke and their influence 1994;42:1033–41. on infection after stroke. Stroke. 2008;39:237–41. 20. Lin B, Ginsberg MD. Quantitative assessment of the normal cerebral micro- 44. Buck BH, Liebeskind DS, Saver JL, Bang OY, Yun SW, Starkman S, et al. Early vasculature by endothelial barrier antigen (EBA) immunohistochemistry: neutrophilia is associated with volume of ischemic tissue in acute stroke. application to focal cerebral ischemia. Brain Res. 2000;865:237–44. Stroke. 2008;39:355–60. 21. Iadecola C, Anrather J. The immunology of stroke: from mechanisms to 45. Ross AM, Hurn P, Perrin N, Wood L, Carlini W, Potempa K. Evidence of the translation. Nat Med. 2011;17:796–808. peripheral inflammatory response in patients with transient ischemic 22. Acarin L, Gonzalez B, Castro AJ, Castellano B. Primary cortical glial reaction attack. J Stroke Cerebrovasc Dis. 2007;16:203–7. versus secondary thalamic glial response in the excitotoxically injured 46. Ito U, Hakamata Y, Kawakami E, Oyanagi K. Temporary [corrected] cerebral young brain: microglial/macrophage response and major histocompat- ischemia results in swollen astrocytic end-feet that compress microves- ibility complex class I and II expression. Neuroscience. 1999;89:549–65. sels and lead to delayed [corrected] focal cortical infarction. J Cereb 23. Chamorro A, Meisel A, Planas AM, Urra X, van de Beek D, Veltkamp R. The Blood Flow Metab. 2011;31:328–38. immunology of acute stroke. Nat Rev Neurol. 2012;8:401–10. 47. Moller K, Boltze J, Posel C, Seeger J, Stahl T, Wagner DC. Sterile inflamma- 24. Denes A, Thornton P, Rothwell NJ, Allan SM. Inflammation and brain tion after permanent distal MCA occlusion in hypertensive rats. J Cereb injury: acute cerebral ischaemia, peripheral and central inflammation. Blood Flow Metab. 2014;34:307–15. Brain Behav Immun. 2010;24:708–23. 48. Denes A, Pradillo JM, Drake C, Buggey H, Rothwell NJ, Allan SM. Surgical 25. Koch S, Della-Morte D, Dave KR, Sacco RL, Perez-Pinzon MA. Biomarkers manipulation compromises leukocyte mobilization responses and for ischemic preconditioning: finding the responders. J Cereb Blood Flow inflammation after experimental cerebral ischemia in mice. Front Neuro - Metab. 2014;34:933–41. sci. 2014;7:271. 26. Wang Y, Reis C, Applegate R, Stier G, Martin R, Zhang JH. Ischemic 49. Price CJ, Menon DK, Peters AM, Ballinger JR, Barber RW, Balan KK, et al. conditioning-induced endogenous brain protection: applications pre-, Cerebral neutrophil recruitment, histology, and outcome in acute per- or post-stroke. Exp Neurol. 2015;272:26–40. ischemic stroke: an imaging-based study. Stroke. 2004;35:1659–64. 27. Garcia-Bonilla L, Benakis C, Moore J, Iadecola C, Anrather J. Immune 50. Krams M, Lees KR, Hacke W, Grieve AP, Orgogozo JM, Ford GA. Acute mechanisms in cerebral ischemic tolerance. Front Neurosci. 2014;8:44. Stroke Therapy by Inhibition of Neutrophils (ASTIN): an adaptive 28. Esposito E, Hayakawa K, Maki T, Arai K, Lo EH. Eec ff ts of postconditioning dose-response study of UK-279,276 in acute ischemic stroke. Stroke. on neurogenesis and angiogenesis during the recovery phase after focal 2003;34:2543–8. cerebral ischemia. Stroke. 2015;46:2691–4. 51. Allen C, Thornton P, Denes A, McColl BW, Pierozynski A, Monestier M, et al. 29. Hanyu S, Ito U, Hakamata Y, Yoshida M. Transition from ischemic neuronal Neutrophil cerebrovascular transmigration triggers rapid neurotoxicity necrosis to infarction in repeated ischemia. Brain Res. 1995;686:44–8. through release of proteases associated with decondensed DNA. J Immu- 30. Hanyu S, Ito U, Hakamata Y, Nakano I. Topographical analysis of cortical nol. 2012;189:381–92. neuronal loss associated with disseminated selective neuronal necrosis 52. Posel C, Scheibe J, Kranz A, Bothe V, Quente E, Frohlich W, et al. Bone and infarction after repeated ischemia. Brain Res. 1997;767:154–7. marrow cell transplantation time-dependently abolishes efficacy of 31. Coutts SB, Modi J, Patel SK, Aram H, Demchuk AM, Goyal M, et al. What granulocyte colony-stimulating factor after stroke in hypertensive rats. causes disability after transient ischemic attack and minor stroke? Results Stroke. 2014;45:2431–7. from the CT and MRI in the Triage of TIA and minor Cerebrovascular Events 53. Garcia-Bonilla L, Moore JM, Racchumi G, Zhou P, Butler JM, Iadecola C, to Identify High Risk Patients (CATCH) Study. Stroke. 2012;43:3018–22. et al. Inducible nitric oxide synthase in neutrophils and endothelium 32. Zhao L, Nowak TS Jr. CBF changes associated with focal ischemic pre- contributes to ischemic brain injury in mice. J Immunol. 2014;193:2531–7. conditioning in the spontaneously hypertensive rat. J Cereb Blood Flow 54. Enzmann G, Mysiorek C, Gorina R, Cheng YJ, Ghavampour S, Hannocks Metab. 2006;26:1128–40. MJ, et al. The neurovascular unit as a selective barrier to polymorphonu- 33. Hoyte LC, Papadakis M, Barber PA, Buchan AM. Improved regional cer- clear granulocyte (PMN) infiltration into the brain after ischemic injury. ebral blood flow is important for the protection seen in a mouse model Acta Neuropathol. 2013;125:395–412. of late phase ischemic preconditioning. Brain Res. 2006;1121:231–7. 55. Ullrich N, Strecker JK, Minnerup J, Schilling M. The temporo-spatial locali- 34. Jickling GC, Liu D, Ander BP, Stamova B, Zhan X, Sharp FR. Targeting zation of polymorphonuclear cells related to the neurovascular unit after neutrophils in ischemic stroke: translational insights from experimental transient focal cerebral ischemia. Brain Res. 2014;1586:184–92. studies. J Cereb Blood Flow Metab. 2015;35:888–901. 56. Gelderblom M, Leypoldt F, Steinbach K, Behrens D, Choe CU, Siler DA, 35. Arsava EM, Gurer G, Gursoy-Ozdemir Y, Karatas H, Dalkara T. A new model et al. Temporal and spatial dynamics of cerebral immune cell accumula- of transient focal cerebral ischemia for inducing selective neuronal necro- tion in stroke. Stroke. 2009;40:1849–57. sis. Brain Res Bull. 2009;78:226–31. Tuor et al. BMC Neurosci (2016) 17:28 Page 14 of 14 57. Chu HX, Kim HA, Lee S, Moore JP, Chan CT, Vinh A, et al. Immune cell infil- 61. Ross AM, Lee CS. Description and identification of the peripheral immune tration in malignant middle cerebral artery infarction: comparison with response trajectories over time in first-time and recurrent stroke/transient transient cerebral ischemia. J Cereb Blood Flow Metab. 2014;34:450–9. ischemic attack. J Neurosci Nurs. 2015;47(5):256–62. 58. Tohgi H, Suzuki H, Tamura K, Kimura B. Platelet volume, aggregation, 62. Awadh M, MacDougall N, Santosh C, Teasdale E, Baird T, Muir KW. Early and adenosine triphosphate release in cerebral thrombosis. Stroke. recurrent ischemic stroke complicating intravenous thrombolysis 1991;22:17–21. for stroke: incidence and association with atrial fibrillation. Stroke. 59. D’Erasmo E, Aliberti G, Celi FS, Romagnoli E, Vecci E, Mazzuoli GF. Platelet 2010;41:1990–5. count, mean platelet volume and their relation to prognosis in cerebral 63. Kennedy J, Hill MD, Ryckborst KJ, Eliasziw M, Demchuk AM, Buchan AM. infarction. J Intern Med. 1990;227:11–4. Fast assessment of stroke and transient ischaemic attack to prevent early 60. Denes A, Pradillo JM, Drake C, Sharp A, Warn P, Murray KN, et al. Strep- recurrence (FASTER): a randomised controlled pilot trial. Lancet Neurol. tococcus pneumoniae worsens cerebral ischemia via interleukin 1 and 2007;6:961–9. platelet glycoprotein Ibalpha. Ann Neurol. 2014;75:670–83. Submit your next manuscript to BioMed Central and we will help you at every step: • We accept pre-submission inquiries • Our selector tool helps you to find the most relevant journal • We provide round the clock customer support • Convenient online submission • Thorough peer review • Inclusion in PubMed and all major indexing services • Maximum visibility for your research Submit your manuscript at www.biomedcentral.com/submit

Journal

BMC Neuroscience – Springer Journals

Published: May 26, 2016

There are no references for this article.