Purinergic Signalling (2016) 12:453–463 DOI 10.1007/s11302-016-9511-x
ORIGINAL ARTICLE
Lack of functional P2X7 receptor aggravates brain edema development after middle cerebral artery occlusion Melanie Kaiser 1 & Anja Penk 2 & Heike Franke 1 & Ute Krügel 1 & Wolfgang Nörenberg 1 & Daniel Huster 2 & Michael Schaefer 1
Received: 26 November 2015 / Accepted: 23 March 2016 / Published online: 5 April 2016 # Springer Science+Business Media Dordrecht 2016
Abstract Effective therapeutic measures against the development of brain edema, a life-threatening complication of cerebral ischemia, are necessary to improve the functional outcome for the patient. Here, we identified a beneficial role of purinergic receptor P2X7 activation in acute ischemic stroke. Involvement of P2X7 in the development of neurological deficits, infarct size, brain edema, and glial responses after ischemic cerebral infarction has been analyzed. Neurologic evaluation, magnetic resonance imaging, and immunofluorescence assays were used to characterize the receptor’s effect on the disease progress during 72 h after transient middle cerebral artery occlusion (tMCAO). Sham-operated animals were included in all experiments for control purposes. We found P2X7-deficient mice to develop a more prominent brain edema with a trend towards more severe neurological deficits 24 h after tMCAO. Infarct sizes, T2 times, and apparent diffusion coefficients did not differ significantly between wild-type and P2X7−/− animals. Our results show a characteristic spatial distribution of
Electronic supplementary material The online version of this article (doi:10.1007/s11302-016-9511-x) contains supplementary material, which is available to authorized users.
reactive glia cells with strongly attenuated microglia activation in P2X7−/− mice 72 h after tMCAO. Our data indicate that P2X7 exerts a role in limiting the early edema formation, possibly by modulating glial responses, and supports later microglia activation. Keywords Brain ischemia . MCAO . Microglia . P2X7 . Purinergic receptor
Abbreviations ADC Apparent diffusion coefficient AQP4 Aquaporin 4 CCA Common carotid artery DWI Diffusion weighted image ECA External carotid artery ICA Internal carotid artery MCA Middle cerebral artery MCAO Middle cerebral artery occlusion P2X7−/− P2X7 receptor knockout NMR Nuclear magnetic resonance pMCAO Permanent middle cerebral artery occlusion RARE Rapid acquisition with relaxation enhancement ROI Region of interest tMCAO Transient middle cerebral artery occlusion VOI Volume of interest WT Wild-type
* Michael Schaefer [email protected]
Introduction 1
2
Rudolf Boehm Institute of Pharmacology and Toxicology, Medical Faculty, Universität Leipzig, Härtelstr. 16-18, 04107 Leipzig, Germany Institute of Medical Physics and Biophysics, Medical Faculty, Universität Leipzig, Leipzig, Germany
Ischemic stroke is one of the most important vascular diseases, which causes loss of nervous tissue and may lead to severe disability or even death. A major complication is the development of brain edema, which
454
exerts pressure on neighboring tissues and further aggravates the disease. Despite the clinical importance of the disease, knowledge gaps concerning the pathophysiology of cerebral ischemia continue to complicate the search for an effective therapy. The purinergic receptor P2X7, an ATP-gated cation channel belonging to the purinergic P2 receptor family, has been shown to affect pathophysiological events following brain ischemia. P2X7 is a low-affinity ATP receptor mainly expressed on immune and glial cells. Regarding the central nervous system, P2X7 is known to be present on microglia, astrocytes, oligodendrocytes, and some neuronal populations [1]. Inflammation and tissue trauma may entail high extracellular ATP concentrations [2] and trigger the activation of P2X7 [3]. Subsequent Ca2+ entry may cause neuronal and oligodendrocyte cell death [4, 5], but activation of P2X7 has also been shown to induce microglia proliferation [6]. Consistently, the receptor is involved in pathophysiological events following ischemic brain damage, but studies to clarify its exact role yielded inconsistent results. On the one hand, a detrimental pro-inflammatory role is attributed to P2X7 in experimental brain injury, which was reversible by antagonists or diminished by gene knockout [7, 8]. These results are in agreement with the fact that the maturation and release of IL-1β as well as the generation of reactive oxygen species, both pathologic stimuli in experimental and human stroke [9, 10], are triggered upon the activation of the receptor [11]. On the other hand, results obtained by pharmacological modulation of P2X7 point towards a beneficial function of the receptor, presumably conveyed by activating neuroprotective responses in glial cells [12]. In fact, P2X7-bearing microglia and astrocytes may contribute to the reduction of neuronal cell death [13, 14]. A primary in vivo study did not argue for an influence of P2X7 on the infarct volume 24 h after ischemia [15]; however, no further infarct parameters or time points were investigated. In this context, we evaluated the contribution of P2X7 to the outcome after middle cerebral artery occlusion (MCAO), an established model of ischemic stroke, in mice after surgery for the subsequent 72 h. We applied clinical, nuclear magnetic resonance (NMR) imaging, and immunofluorescence methods to obtain data regarding neurological deficits, infarct size, brain edema, and glial cell activation. Our results are compatible with a protective role of P2X7 during initial phases after the onset of ischemia. At later time points, microglia activation was supported by P2X7, but no significant differences were observed with respect to the clinical outcome.
Purinergic Signalling (2016) 12:453–463
Material and methods Animals Mouse breeding and experiments were performed in the animal facility of the Faculty of Medicine, University of Leipzig according to the ARRIVE guidelines as well as the European (Council Directive 2010/63/EU) and German guidelines for the welfare of experimental animals and were approved by the local authorities (Landesdirektion Leipzig; ID: TVV 64/ 13). All possible efforts were made to replace, reduce, and refine experiments including laboratory animals. Male mice (25–35 g) with P2X7−/− or wild-type (WT) genotype of C57BL/6 J background originally purchased from the Jackson Laboratory (Bar Harbor, ME, USA) were used [16]. Mice were housed under standard conditions with free access to food and water and a 12-h light–dark cycle with lights on from 7:00 a.m. From first NMR session on, animals were housed individually. Surgical procedures Anesthesia was induced with 2.5 % of isoflurane (Baxter, Unterschleissheim, Germany) in oxygen and maintained with 1.5 % of the anesthetic. Analgesia was assured by subcutaneous injection of 200 mg/kg metamizol (WDT, Garbsen, Germany), a non-opioid analgesic with only weak antiinflammatory effects. Metamizol was administered before surgery and every 6 hours afterwards. MCAO was essentially achieved as described before with minor modifications [17]. Briefly, after a midline neck incision, the left common carotid artery (CCA) and its bifurcation into the external (ECA) and internal carotid artery (ICA) were exposed and carefully separated from the adjacent vagus nerve. The CCA as well as the ECA were ligated (Suprama Silk 7–0, Feuerstein, Berlin, Germany) while blood flow through the ICA was impeded by a microvascular clip. A polysiloxane-coated filament (Suprama Nylon 8–0, Feuerstein) coated with Xantopren M mucosa (Heraeus Kulzer GmbH, Hanau, Germany) was then inserted through a small opening cut into the CCA distal to the ligature. Upon removal of the microvascular clip, the filament was advanced approximately 11 mm along the ICA into the circle of Willis, where it blocked the origin of the middle cerebral artery (MCA). The filament was immediately removed in sham-operated animals or after 60 min in mice undergoing transient MCAO (tMCAO). It remained in place in animals belonging to the permanent MCAO (pMCAO) group. The ICA was then ligated (Suprama Silk 5–0, Feuerstein) and the neck incision closed. Neurological score Neurological deficits were evaluated daily, and a neurological score based on a scale described before [18] was assigned to
Purinergic Signalling (2016) 12:453–463
each animal. Initially, mice were suspended by their tail above the surface of a table. Animals able to extend both forelimbs towards the surface were scored 0. If the right forelimb remained flexed and subsequent tests did not reveal further neurological deficits, the score was set to 1. Mice given a score of 2 showed decreased loading of the right forelimb when moving without the support of the hind limbs as by elevating them by the tail. If fixation of the animal by the tail caused them to move in circles, they were scored 3. Circling behavior when allowed to move freely resulted in a score of 4. Severely reduced general condition and inability to move resulted in a score of 5 and immediate euthanasia. Magnetic resonance imaging protocol and image analysis One and 3 days after MCAO, mice were anesthetized with 75 mg/kg ketamine (WDT) and 6 mg/kg xylazine (Bayer Health Care, Leverkusen, Germany). Subsequently, each mouse underwent MRI measurements, which were performed on a Bruker DRX 300 (7.05 T) wide bore NMR spectrometer equipped with a microimaging unit including a mouse bed (MicWB40 Probe using a single channel birdcage resonator with 30 mm inner diameter and a Micro2.5 Gradient System, Bruker BioSpin GmbH, Rheinstetten, Germany) providing a maximal gradient strength of 730 mT/m. The cerebral brain region was mapped using nine coronal slices with a slice thickness of 1 mm (field of view 20 mm × 20 mm) and no gap in between (Online Resource 1A). Three different scanning protocols were used. High resolution, intermediate weighted images were acquired using a fast spin-echo sequence (rapid acquisition with relaxation enhancement, RARE) with a RARE factor of 8 (repetition time 1200 ms; echo time 9.2 ms; effective echo time 32.4 ms; matrix 256 × 256; averages 2; total scanning time 76 s). A multi slice multi echo (MSME) sequence was used to measure the T2 time (repetition time 2500 ms; echo time 10.2 ms; number of echo images 12; matrix 256 × 128; total scanning time 320 s), and five diffusion weighted images (DWI) for each slice were acquired using a pulsed gradient spin echo to measure the apparent diffusion coefficient (ADC) along the rostral-caudal direction (repetition time 1200 ms; echo time 23.7 ms; time between gradient pulses (Δ) 12.2 ms; gradient duration (δ) 6 ms; matrix 128 × 64; b values of 52.2, 157.1, 355.8, 638.6, and 1110.3 s/mm2 (diffusion gradient 35–200 mT/m); total scanning time 384 s). Image analysis was performed in a randomized, blinded fashion using custom MATLAB scripts. First, RARE images were used to create a mask for each of the nine slices including only the brain to speed up analysis. These masks were then downsampled for further analysis of T2 and DWI image trains to the desired resolution (256 × 128 for T2 and 128 × 64 for DWI), including only pixels in the low resolution mask if at least half of the corresponding pixels formed part of the high-
455
resolution mask from the RARE. T2 and ADC maps were calculated using a mono-exponential fitting function. However, as the images show a significant noise floor due to limited scanning time, the fitting functions for T2 and ADC were corrected for an offset term (determined as mean value of an area including pure noise). T2 maps were then adjusted in brightness and contrast to define a region of interest (ROI) with altered T2 values. All ROIs were then summed up to a volume of interest (VOI) representing the size of the infarct. A corresponding VOI on the contralateral site was created using mirroring (slightly adjusted to not include physiologically hyperintense neuroanatomic regions) (Online Resource 1C). For these two VOIs (ipsi- and contralateral), mean values for T2 times and ADC were calculated. Mice with no observable altered region in the whole brain were analyzed in the same manner by drawing VOIs similar to those in animals with cerebral infarction. Second, the position of the midline was determined using neuroanatomic landmarks such as the fissura longitudinalis, lateral, and third ventricles in three consecutive slices (Online Resource 1B). The percentage of the infarcted hemisphere as a proportion of the total brain volume in the middle slice was used to describe oedematous brain swelling. Eventually, swelling of the infarcted hemisphere was put in relation to the size of the ischemic lesion.
Immunofluorescence staining Animals were deeply anesthetized and transcardially perfused (2 % paraformaldehyde in sodium acetate buffer containing 10 IU/ml heparin, pH 6.5, followed by 2 % paraformaldehyde in sodium borate buffer containing 0.15 % glutaraldehyde, pH 8.5). Brains were immediately removed and stored in the latter solution. In preparation for immunofluorescence staining, coronal sections (50 μm) were cut using a vibratome (Leica, Typ VT 1200S, Nussloch, Germany) and collected as free-floating slices. Unspecific protein binding was prevented by blocking for 30 min with 5 % fetal calf serum (Sigma-Aldrich, St. Louis, MO, USA) dissolved in 0.05 M Tris-buffered saline (TBS;pH 7.6; Sigma-Aldrich) and 10 % Triton X-100 (Sigma-Aldrich). Slices were then incubated with primary antibodies rabbit anti-ionized calcium binding adaptor molecule 1 (Iba1; 1:1000; Wako Chemicals GmbH, Neuss, Germany), mouse anti-glial fibrillary acidic protein (GFAP; 1:1000; Dako Cytomation, Glostrup, Denmark), and rabbit anti-laminin (1:50; Dako) for 48 h at 4 °C. Following subsequent washing steps with TBS, slices were exposed to secondary antibodies Cy3-conjugated donkey anti-rabbit IgG (1:800; Jackson ImmunoResearch, West Grove, PA, USA) or Cy5-conjugated donkey anti-mouse IgG (1:800; Jackson ImmunoResearch) for 2 h at room temperature. Slices were then mounted on slide glasses, dehydrated in ethanol, cleared
456
in n-butylacetate, embedded in Entellan (Merck, Darmstadt, Germany), and covered. Confocal microscopy and quantification Immunofluorescence was examined using a confocal laser scanning microscope (LSM 510 Meta, Zeiss, Oberkochen, Germany) equipped with two helium/neon lasers emitting at 543 and 633 nm, respectively. Cy5-labeled signals were color coded in green. Images taken with a Plan-Neofluar 10x/0.3 objective allowed visualization of infarct core, penumbra, and surrounding tissues and served as a template to define regions of interest (ROIs). While the infarcted tissue was characterized by loss of cellular integrity and only permitted measurements of unspecific immunofluorescence, the infarct border itself was strongly demarcated by activated microglia, and intense GFAP immunofluorescence could be seen somewhat further from the infarct core. In order to establish spatial profiles of glial responses, we thus positioned ROIs so that they encompassed the infarct border and the surrounding tissue. Immunofluorescence intensity along three such ROIs per animal was measured, corrected for the respective background signals, and normalized to unaffected brain regions to compensate for uneven staining in different slices. Similarly, three ROIs were defined to count microglial cells. C-Apochromat 40x/1.2 and C-Apochromat 63x/1.2 water immersion objectives were used to obtain detailed images of glial cell morphology. Data evaluation and statistical analysis All data are expressed as mean ± SEM, obtained in n animals. One and two-way repeated measures ANOVA, and subsequent pairwise multiple comparison procedures were applied to compare variables in more than two groups. The two-tailed Student’s t test was used to test for statistically significant differences between two experimental groups. Values for P < 0.05 were considered significant.
Results P2X7 reduces rapid development of brain edema after ischemic infarction We evaluated the progression of neurological deficits, infarct size and brain swelling during 72 h after tMCAO. Mice lacking P2X7 tended to initially (6 h) experience more severe neurological problems than those of the WT (P = 0.169), but these findings only constitute a non-significant trend that leveled out until the end of the observation (Fig. 1a). Furthermore, the postischemic interval was characterized by a significantly greater midline shift in the P2X7-deficient
Purinergic Signalling (2016) 12:453–463
group at 24 h after reperfusion (P = 0.026), (Fig. 1c). Whereas the WT group expressed a significant enlargement of the midline shift over time (P = 0.046), the midline was not further shifted in the P2X7-deficient animals until 72 h after surgery. Of note, the values represented in Fig. 1c correspond to the overall swelling of the infarcted hemisphere. In order to calculate these values, we determined the line separating both hemispheres, and for this task, we used high resolution RARE images and T2 maps (also see Online Resource 1). Due to resolution, these two types of images were superior to DWI images regarding the recognizability of anatomical structures marking the midline. Since only the position of these anatomical landmarks (and the total brain volume as restricted by the intact mouse skull) affected the results depicted in Fig. 1c, it is not possible to deduct the respective share of cytotoxic and vasogenic brain edema from these measurements. Because the infarct size appeared to be slightly larger in P2X7-deficient mice (Fig. 1b), one could speculate the corresponding pronounced midline shift to be a mere result of an increased volume of infarcted tissue. We therefore established the ratio between swelling of the infarcted hemisphere and the respective lesion volume and obtained strongly increased values for P2X7-deficient animals at 24 h that confirm our aforementioned results (P = 0.011, Fig. 1d). Consistent with the neurological scores after 72 h, differences regarding brain edema were no longer detectable. At this time point in addition to the midline shift, the relative brain swelling of WT mice had significantly increased (P = 0.032). Although T2 relaxation times and ADC did not differ at any time between WT and P2X7−/− mice, both displayed significantly prolonged T2 relaxation times and decreased ADC when compared with sham-operated animals (Fig. 2; Table 1, upper panel), indicating the development of a cytotoxic edema, i.e., the shift of extracellular water to intracellular compartments [19]. Furthermore, T2 relaxation times significantly grew between 24 and 72 h after MCAO (WT: P = 0.009; P2X7−/−: P = 0.005), indicating interstitial water accumulation in all animals that underwent tMCAO (Table 1, upper panel). Sham-operated animals did not show neurological deficits. T2 relaxation times as well as ADC values of the ipsi- or contralateral brain hemispheres in sham-operated mice did not differ from those obtained in the contralateral hemisphere of animals that underwent tMCAO. Hence, contralateral values served as a reference for unaffected tissue in Table 1. Microglia activation is attenuated in animals lacking P2X7 The ionized calcium binding adaptor protein 1 (Iba1), a marker expressed in microglia and macrophages that is upregulated after tissue damage, was used to evaluate microglial activation 72 h after tMCAO [20]. Activated astrocytes were monitored by expression of glial fibrillary acidic protein (GFAP). Both Iba1 and GFAP expression were highly increased after
Purinergic Signalling (2016) 12:453–463
457
Fig. 1 Evaluation of neurological deficits, infarct size, and brain swelling at different time points after tMCAO. a Neurological scores indicating the severity of neurological deficits at various time points after tMCAO in WT (grey bars) and P2X7−/− animals (black bars). b Infarct sizes were determined in vivo by NMR imaging. Consecutive image slices covering the whole brain were evaluated to obtain the respective infarct volumes. c Brain swelling, resulting in a midline shift and compression of the contralateral hemisphere, was assessed by calculating the volume occupied by the infarcted hemisphere as percentage of the total brain volume.d Swelling of the infarcted hemisphere normalized to respective infarct size. Of note, relative brain swelling augmented significantly between 24 and 72 h in WT mice. n = 8–12; asterisks indicate P < 0.05
cerebral ischemia in animals that underwent tMCAO when compared with sham-operated mice and represent the characteristic spatial pattern of glia cells with respect to the infarct site (Fig. 3, left panel). In order to obtain information regarding the distribution of activated glial cells, we defined a cortical ROI that encompasses the penumbra and adjacent brain tissue (Fig. 4a, b, Fig. 5a, b) and measured immunofluorescence intensity along this ROI. Since cellular integrity was almost completely lost in the infarct cores, immunofluorescence signals in these regions were inconsistent and
Fig. 2 Calculated T2 (upper panel) and ADC maps (lower panel) overlaid with RARE images from NMR. Representative examples for tMCAO and sham groups of WT and P2X7-deficient mice 24 h after surgery are shown. Identified ischemic regions are shown grey rimmed. Scale bar 2 mm. The linear color coding bar represents values of 0–100 ms for T2 and 0– 250 × 10−5 mm2/s for the ADC
considered non-specific. Consistent with microglial activation, immunofluorescence of Iba1 was highly enhanced at the infarct border in both WT and P2X7−/− mice, where microglia literally formed a wall between the infarct core and the peri-infarct area (Fig. 4c). This microglial response was significantly attenuated in the absence of P2X7 (P = 0.016, Fig. 4d). Notably, this effect was not caused by a decreased microglial cell number but reflected a lower per-cell Iba1 immunoreactivity in P2X7−/− mice compared with their WT (302.4 ± 15.7 vs. 300.0 ± 7.8 microglial cells per ROI in WT
458 Table 1
Purinergic Signalling (2016) 12:453–463 Clinical and NMR parameters obtained in WT and P2X7−/− mice after tMCAO and pMCAO
T2 relaxation time, ms Apparent diffusion coefficient, mm2 s−1 10−5
Time after tMCAO (h) 24 72 24 72
WT infarct 46.51 ± 0.94 54.66 ± 2.08* 49.07 ± 2.60 47.55 ± 2.75
Time after pMCAO (h) WT infarcte Neurological score Infarct size, mm3 T2 relaxation time, ms Apparent diffusion coefficient, mm2 s−1 10−5
P2X7−/− infarct 45.53 ± 0.60 54.61 ± 3.76* 51.36 ± 1.82 49.14 ± 3.51
WT contralat. P2X7−/− contralat. n 32.00 ± 0.28 31.74 ± 0.18 64.36 ± 0.93 64.69 ± 3.19
31.89 ± 0.25 31.23 ± 0.30 64.03 ± 2.36 66.42 ± 1.14
8 5–7
P2X7−/− infarct WT contralat. P2X7−/− contralat. n
6 24 24
4.0 ± 0.32 4.4 ± 0.24 3.6 ± 0.51 4.2 ± 0.37 114.43 ± 17.33 91.97 ± 9.85
– – –
– – –
4
24 24
54.75 ± 0.98 39.34 ± 0.79
32.13 ± 0.56 69.21 ± 7.88
31.30 ± 0.19 76.63 ± 3.46
4 3
54.91 ± 0.84 45.80 ± 5.66
5
*P < 0.05 different from mean values at 24 h
and P2X7−/− mice, respectively; 220.1 ± 4.3 and 204.3 ± 11.5 microglial cells per ROI in sham-operated WT and P2X7−/− mice, respectively). This finding indicates no strong effect on microglia recruitment or local proliferation, but a limitation of microglial activation when the function of P2X7 is missing. Under resting conditions, microglia possess a ramified morphology with small somata and extensive processes that facilitate surveillance. While this phenotype was visible in sham-
operated animals, transient cerebral ischemia caused morphologic changes, and microglia adopted a reactive, in part even amoeboid-like phenotype (Fig. 3, magnified images of Iba1). Of note, there were no discernible morphological differences between reactive microglia and astrocytes of WT and P2X7−/− mice (Fig. 3, magnified images of Iba1 and GFAP). Reactive astrocytes were seen in a broader region outside the infarct core (Fig. 5c). Although both activation of microglia and
Fig. 3 Confocal images of double immunofluorescence to characterize microglia and astroglia distribution in peri-infarct tissue and corresponding brain sections of sham-operated mice 72 h after surgery. The upper rows, corresponding to WT and P2X7−/− mice after tMCAO, illustrate how Iba1-labeled, reactive microglia (arrows) delimit the ischemic core
(asterisks) while increased astroglial GFAP-expression (arrowheads) prevails in the surrounding tissue over a wider brain area. Sham-operated mice, as represented in the lower rows, depict a homogeneous distribution of glial cells in their resting state. Merged images are included for better spatial visualisation. Scale bars 100 μm
Purinergic Signalling (2016) 12:453–463
459
Fig. 4 P2X7-dependent microglia activation in ischemic lesion and surrounding brain tissue 72 h after tMCAO or sham-operation. a Representative confocal image of Iba1 immunofluorescence in the infarcted cortex of a WT mouse 3 days after tMCAO. The core region is located towards the left of the image, the peri-infarct tissue towards the right side. The white rimmed rectangle corresponds to a ROI defined for further analysis. b Representative image of a P2X7 −/− mouse. c Quantification of Iba1 immunofluorescence. Values shown are based on
measurements along the complete ROI and were normalized to immunofluorescence intensities measured in a distant and apparently unaffected region b. Grey traces represent the WT mice and black traces represent P2X7−/− mice. Continuous traces describe Iba1 immunofluorescence after tMCAO, and dotted traces derive from sham-operated animals. d Microglia activation at infarct border depicted as ratio of Iba1 immunofluorescence between infarct-delimiting region Ba^ and region Bb^. n = 8– 10 for tMCAO groups, n = 4 for sham groups; asterisks indicate P < 0.05
astrocytes has been linked to the activation of P2X7 [21], we did not find any differences in the immunofluorescence of GFAP in WT and P2X7−/− animals (Fig. 5d). Laminin, an extracellular matrix component primarily located in the basement membranes of vascular endothelial cells, is a well-established marker for cerebral blood vessels which is known to be upregulated following brain ischemia [22, 23]. Since P2X7 may influence tissue remodeling and angiogenesis [24, 25], we examined whether genetic deletion of the receptor affects the expression level of laminin. While laminin immunoreactivity in the peri-infarct tissue of WT and P2X7−/− mice and in sham-operated animals was very low, intense immunofluorescence could be measured in determined ROIs within the infarct core (Online Resource 2). However, no difference in laminin-immunofluorescence intensity could be detected between the two groups that underwent tMCAO (Online Resource 3).
others pointed out that the high mortality poses a major challenge [28]. In our study, 1/6 WT and 2/7 P2X7−/− mice had to be euthanized during the first hours after pMCAO. Due to severe neurological deterioration and in the interest of animal welfare, we decided to euthanize the remaining animals 24 h after surgery. We did not detect any relation between the presence of a functional P2X7 receptor and progression of neurological deficits or the resulting need for euthanasia. Significant differences could neither be observed regarding infarct volumes, T2 relaxation times, and ADC of WT and P2X7−/− groups (Table 1, lower panel). However, these findings have to be interpreted keeping the limited group size in mind.
P2X7 does not protect from severe neurological deterioration after pMCAO We had initially chosen to permanently occlude the middle cerebral artery to model human stroke, where recanalization can often not be achieved [26]. While some previous studies reported high survival rates for days after pMCAO (e.g., [27]),
Discussion The aim of the present work was to characterize the role of P2X7 in pathophysiological events following ischemic stroke for a prolonged period of time. We found P2X7−/− mice to rapidly develop a significantly larger brain edema, a condition accompanied by a tendency towards more severe neurological deficits whereby the latter did, however, not reach significance. Microglial activation 72 h after the initial insult was significantly attenuated in animals lacking P2X7.
460
Purinergic Signalling (2016) 12:453–463
Fig. 5 Astroglia activation at the infarct border and surrounding brain tissue 72 h after tMCAO or sham-operation. a Representative confocal image of GFAP immunofluorescence in the infarcted cortex of a WT mouse 3 days after tMCAO. The infarct core region is located towards the left of the image, the peri-infarct tissue towards the right side. The white rimmed rectangle corresponds to a ROI comprising the infarct border and peri-infarct tissue for further analysis. b Representative image of a P2X7−/− mouse. c Quantification of GFAP immunofluorescence. Values
shown are based on measurements from left to right along the complete ROI and were normalized to immunofluorescence intensities measured in contralateral reference region. Grey traces represent the WT, and black traces represent P2X7−/− mice. Continuous traces describe GFAP immunofluorescence after tMCAO, and dotted traces derive from shamoperated animals. d GFAP activation in peri-infarct tissue depicted as ratio of GFAP immunofluorescence between infarcted hemisphere and contralateral reference region. n = 9 for tMCAO groups, n = 4 for sham groups
No differences were observed in neurological scores and infarct size in the initially chosen permanent MCAO model. We did not pursue this approach any further due to serious neurological deficits and lethality. Due to the high mortality after pMCAO, we are unable to draw reliable conclusions regarding the role of P2X7 after permanent cerebral ischemia. The change to transient MCAO was associated with a reduction of infarct size and neurological scores, which displayed a trend towards a better outcome when P2X7 is functionally active. Furthermore, we found that animals deficient for P2X7 were more prone to develop severe brain edema within the first 24 h after onset of transient ischemia. During this early phase of ischemic stroke, cytotoxic brain edema reaches its maximum [29] and is clinically followed by a midline shift that is used as a criterion for the development of brain edema in stroke patients [30]. The mechanisms causing brain edema are only partly understood. As a result of the ischemia, energy depletion in the irrigation area causes the failure of ion pumps and, thus, provokes the intracellular accumulation of excess fluid. Only at later stages of the disease, blood–brain barrier breakdown induces the additional formation of vasogenic brain edema. In contrast, brain pathologies involving direct impairment of the blood–brain barrier primarily entail the formation of vasogenic brain edema. In contrast to our study, genetic deletion of P2X7 has been shown to limit edemic
development in a model of traumatic brain injury. Differences regarding the pathogenesis of edema formation may account for this apparent contradiction [7]. Indeed, in agreement with our data, P2X7 receptor agonists have been reported to improve behavioral dysfunctions in the MCAO model of ischemic stroke 24 h after surgery [12]. The transition from a mainly cytotoxic to a more vasogenic edema between day 1 and day 3 after surgery may also explain why differences between WT and P2X7−/− mice leveled out. Furthermore, a recent study showed that glial cell recruitment takes place early after onset of ischemia and that concentrations of interleukins IL-1β, IL-6, and TNF-α peak 24 h after tMCAO [31]. Since P2X7 is known to affect the expression of all these cytokines, these findings may hint at a greater receptor activity at this time point and thus explain why we detected significant differences only during the early phase of brain ischemia. Both WT and P2X7−/− mice showed increased immunofluorescence signals pointing to microglial and astroglial activation 72 h after tMCAO. The microglial response, however, was significantly attenuated in P2X7-deficient animals. We selected this survival time to evaluate glial responses because previous studies described a peak of glial activation within a few days after MCAO [32]. In future studies, the histological
Purinergic Signalling (2016) 12:453–463
and functional state of either cell type at earlier times after cerebral ischemia may reveal the mechanisms which cause aggravated brain edema in P2X7−/− mice. Although the non-opioid metamizol may be responsible for weak anti-inflammatory effects and attenuation of glial activation, all animals received the same dose of analgesic, and such effects should therefore affect both WT and P2X7−/− mice equally. Possible explanations for our observations have to consider that P2X7 is expressed in microglia as well as in astrocytes and both cell types may exert damaging or favorable effects, depending on their activation state and the actual point of time after ischemia/reperfusion. Microglia are known to be able to differentiate into distinct phenotypes. Although ischemic brain damage initially facilitates transition towards the pro-inflammatory M1 type, microglial cells retain their ability to shift between one and the other. The presence of M2 microglia favors tissue repair and studies involving mice lacking IL-4, which evokes transition towards the M2 phenotype, have shown that the outcome worsens in absence of M2 microglia [33]. In a recent study, it has been shown that IL-4 expression is enhanced in the penumbra of infarcted tissue, where it reaches its peak during the first 24 h [34]. Even though data regarding microglia is not available, it has been proven that M2 macrophages do express P2X7 [35]. It is thus tempting to speculate that the protective M2 phenotype may critically rely on an intact P2X7. Consequently, a lacking functional P2X7 receptor in M2 microglial cells may cause the detrimental effects observed 24 h after MCAO in the absence of a functional P2X7 receptor, as has been proposed before [36]. On the other hand, there is extensive evidence that antagonism of P2X7 exerts neuroprotective effects, presumably by preventing intracellular Ca2+ overloads. Thus, sustained activation of P2X7 has been shown to facilitate glutamate release and cause neuronal and glial cell death [37]. P2X7 activation may also lead to detrimental, prolonged neuroinflammation [37]. The fact that P2X7 antagonism confers beneficial effects by preventing microglial cell death as well as neuroinflammation once more illustrates the dual role played by microglia in pathophysiological events following brain ischemia. Aquaporin 4 (AQP4) is a water permeable channel mainly expressed in astrocytic processes adjacent to cerebral capillaries and plays an important role in bidirectional water transport [38]. The cytotoxic brain edema prevailing in acute cerebral ischemia can be reduced by downregulation of AQP4. Since AQP4 downregulation has been reported to be triggered upon P2X7 activation [39], this mechanism could serve as an explication for the exacerbated brain edema developed in P2X7deficient mice in this study. In cases of vasogenic brain edema, however, AQP4-mediated water movement is crucial for fluid clearance, and AQP4 downregulation
461
by P2X7 would be detrimental [38]. It is also conceivable that P2X7-dependent mechanisms occur downstream of GFAP-related astroglial activation, as has been shown in a recent study comparing WT and P2X7−/− mice subjected to cerebral ischemia. In case of preconditioning to ischemic insults, astrocytic P2X7 was shown to mediate ischemic tolerance and to provide neuroprotection even though no differences between the two genetic phenotypes were found in GFAP immunofluorescence [40]. Similar to our results, infarct sizes analyzed by post mortem 2,3,5, triphenyltetrazolium chloride (TTC) staining were not different between the genotypes in this study. In our study, laminin immunofluorescence was very low in intact brain tissue. This observation is consistent with the results of earlier studies that have also described a considerable increase in signal intensity at sites of cerebral lesions [41]. It has been reported that this phenomenon is provoked by the breakdown of the neurovascular unit, which leaves the corresponding epitopes accessible for the antibodies, a development that is reversed upon the re-establishment of gliovascular junctions [42]. Our results indicate that this process is not strongly influenced by P2X7.
Conclusions The development of brain edema is a life-threatening complication of cerebral ischemia, the resulting increase in intracranial pressure further compromises adjacent brain structures and exacerbates the disease. Effective therapeutic measures are necessary to improve the functional outcome for the patient. Here, we identified a beneficial role of P2X7 activation in acute ischemic stroke. Our results emphasize the need for a detailed understanding of the receptor’s role in neuroinflammation dependent on the underlying disease, the target cell type, and the time passed since the onset of ischemia in order to decide on a possible therapeutic P2X7 modulation after ischemic stroke. Acknowledgments This work was supported by the Deutsche Forschungsgemeinschaft (DFG) within the framework of Research Group FOR748 and TRR67 (project A6). The authors are grateful to André Rex for technical advice. Expert technical assistance and animal care were provided by Katrin Becker and Anne-Kathrin Krause. Compliance with ethical standard Conflict of interest The authors declare that they have no conflicts of interest. Ethical approval All applicable international, national, and/or institutional guidelines for the care and use of animals were followed. All procedures performed in studies involving animals were in accordance
462
Purinergic Signalling (2016) 12:453–463
with the ethical standards of the institution at which the studies were conducted.
17.
Funding This work was supported by the Deutsche Forschungsgemeinschaft (DFG) within the framework of Research Group FOR748 and TRR67 (project A6).
18.
19.
20.
References 1.
2.
3.
4.
5.
6.
7.
8.
9.
10. 11.
12.
13.
14.
15.
16.
Del Puerto A, Wandosell F, Garrido JJ (2013) Neuronal and glial purinergic receptors functions in neuron development and brain disease. Front Cell Neurosci 7:197 Franke H, Grummich B, Hartig W, Grosche J, Regenthal R, Edwards RH, Illes P, Krügel U (2006) Changes in purinergic signaling after cerebral injury—involvement of glutamatergic mechanisms? Int J Dev Neurosci 24(2–3):123–132 Oliveira JF, Riedel T, Leichsenring A, Heine C, Franke H, Krügel U, Nörenberg W, Illes P (2011) Rodent cortical astroglia express in situ functional P2X7 receptors sensing pathologically high ATP concentrations. Cereb Cortex 21(4):806–820 Domercq M, Perez-Samartin A, Aparicio D, Alberdi E, Pampliega O, Matute C (2007) P2X7 receptors mediate ischemic damage to oligodendrocytes. Glia 58(6):730–740 Nishida K, Nakatani T, Ohishi A, Okuda H, Higashi Y, Matsuo T, Fujimoto S, Nagasawa K (2012) Mitochondrial dysfunction is involved in P2X7 receptor-mediated neuronal cell death. J Neurochem 122(6):1118–1128 Monif M, Burnstock G, Williams DA (2010) Microglia: proliferation and activation driven by the P2X7 receptor. Int J Biochem Cell Biol 42(11):1753–1756 Kimbler DE, Shields J, Yanasak N, Vender JR, Dhandapani KM (2007) Activation of P2X7 promotes cerebral edema and neurological injury after traumatic brain injury in mice. PLoS One 7(7):e41229 Ryu JK, McLarnon JG (2008) Block of purinergic P2X7 receptor is neuroprotective in an animal model of Alzheimer’s disease. Neuroreport 19(17):1715–1719 Lambertsen KL, Biber K, Finsen B (2012) Inflammatory cytokines in experimental and human stroke. J Cereb Blood Flow Metab 32(9):1677–1698 Margaill I, Plotkine M, Lerouet D (2005) Antioxidant strategies in the treatment of stroke. Free Radic Biol Med 39(4):429–443 Alves LA, Bezerra RJ, Faria RX, Ferreira LG, da Silva FV (2013) Physiological roles and potential therapeutic applications of the P2X7 receptor in inflammation and pain. Molecules 18(9):10953–10972 Yanagisawa D, Kitamura Y, Takata K, Hide I, Nakata Y, Taniguchi T (2008) Possible involvement of P2X7 receptor activation in microglial neuroprotection against focal cerebral ischemia in rats. Biol Pharm Bull 31(6):1121–1130 Myer DJ, Gurkoff GG, Lee SM, Hovda DA, Sofroniew MV (2006) Essential protective roles of reactive astrocytes in traumatic brain injury. Brain 129(Pt 10):2761–2772 Suzuki T, Hide I, Ido K, Kohsaka S, Inoue K, Nakata Y (2004) Production and release of neuroprotective tumor necrosis factor by P2X7 receptor-activated microglia. J Neurosci 24(1):1–7 Le Feuvre R, Brough D, Rothwell N (2002) Extracellular ATP and P2X7 receptors in neurodegeneration. Eur J Pharmacol 447(2–3): 261–269 Solle M, Labasi J, Perregaux DG, Stam E, Petrushova N, Koller BH, Griffiths RJ, Gabel CA (2001) Altered cytokine production in mice lacking P2X7 receptors. J Biol Chem 276(1):125–132
21. 22.
23.
24.
25.
26. 27.
28. 29.
30.
31.
32.
33.
34.
35.
Engel O, Kolodziej S, Dirnagl U, Prinz V (2011) Modeling stroke in mice - middle cerebral artery occlusion with the filament model. J Vis Exp 47:2423 Menzies SA, Hoff JT, Betz AL (1992) Middle cerebral artery occlusion in rats: a neurological and pathological evaluation of a reproducible model. Neurosurgery 31(1):100–107 Dijkhuizen RM, Nicolay K (2003) Magnetic resonance imaging in experimental models of brain disorders. J Cereb Blood Flow Metab 23(12):1383–1402 Ito D, Tanaka K, Suzuki S, Dembo T, Fukuuchi Y (2001) Enhanced expression of Iba1, ionized calcium-binding adapter molecule 1, after transient focal cerebral ischemia in rat brain. Stroke 32(5): 1208–1215 Gever JR, Cockayne DA, Dillon MP, Burnstock G, Ford AP (2006) Pharmacology of P2X channels. Pflügers Arch 452(5):513–537 Eriksdotter-Nilsson M, Bjorklund H, Olson L (1986) Laminin immunohistochemistry: a simple method to visualize and quantitate vascular structures in the mammalian brain. J Neurosci Methods 17(4):275–286 Ji K, Tsirka SE (2012) Inflammation modulates expression of laminin in the central nervous system following ischemic injury. J Neuroinflammation 9:159 Gu BJ, Wiley JS (2006) Rapid ATP-induced release of matrix metalloproteinase 9 is mediated by the P2X7 receptor. Blood 107(12): 4946–4953 Hill LM, Gavala ML, Lenertz LY, Bertics PJ (2010) Extracellular ATP may contribute to tissue repair by rapidly stimulating purinergic receptor X7-dependent vascular endothelial growth factor release from primary human monocytes. J Immunol 185(5): 3028–3034 Rha JH, Saver JL (2007) The impact of recanalization on ischemic stroke outcome: a meta-analysis. Stroke 38(3):967–973 Tanaka Y, Fukumitsu H, Soumiya H, Yoshimura S, Iwama T, Furukawa S (2012) 2-decenoic acid ethyl ester, a compound that elicits neurotrophin-like intracellular signals, facilitating functional recovery from cerebral infarction in mice. Int J Mol Sci 13(4): 4968–4981 Davis MF, Lay C, Frostig RD (2013) Permanent cerebral vessel occlusion via double ligature and transection. J Vis Exp (77) Yeung PK, Lo AC, Leung JW, Chung SS, Chung SK (2009) Targeted overexpression of endothelin-1 in astrocytes leads to more severe cytotoxic brain edema and higher mortality. J Cereb Blood Flow Metab 29(12):1891–1902 Walberer M, Blaes F, Stolz E, Muller C, Schoenburg M, Tschernatsch M, Bachmann G, Gerriets T (2007) Midline-shift corresponds to the amount of brain edema early after hemispheric stroke – an MRI study in rats. J Neurosurg Anesthesiol 19(2):105–110 Lenglet S, Montecucco F, Denes A, Coutts G, Pinteaux E, Mach F, Schaller K, Gasche Y, Copin JC (2014) Recombinant tissue plasminogen activator enhances microglial cell recruitment after stroke in mice. J Cereb Blood Flow Metab 34(5):802–812 Franke H, Günther A, Grosche J, Schmidt R, Rossner S, Reinhardt R, Faber-Zuschratter H, Schneider D, Illes P (2004) P2X7 receptor expression after ischemia in the cerebral cortex of rats. J Neuropathol Exp Neurol 63(7):686–699 Cherry JD, Olschowka JA, O’Banion MK (2014) Neuroinflammation and M2 microglia: the good, the bad, and the inflamed. J Neuroinflammation 11:98 Zhao X, Wang H, Sun G, Zhang J, Edwards NJ, Aronowski J (2015) Neuronal interleukin-4 as a modulator of microglial pathways and ischemic brain damage. J Neurosci 35(32):11281–11291 Zanin RF, Braganhol E, Bergamin LS, Campesato LF, Filho AZ, Moreira JC, Morrone FB, Sevigny J, Schetinger MR, de Souza Wyse AT, Battastini AM (2012) Differential macrophage activation
Purinergic Signalling (2016) 12:453–463 alters the expression profile of NTPDase and ecto-5'-nucleotidase. PLoS One 7(2):e31205 36. Apolloni S, Amadio S, Montilli C, Volonte C, D’Ambrosi N (2013) Ablation of P2X7 receptor exacerbates gliosis and motoneuron death in the SOD1-G93A mouse model of amyotrophic lateral sclerosis. Hum Mol Genet 22(20):4102– 4116 37. Pedata F, Dettori I, Coppi E, Melani A, Fusco I, Corradetti R, Pugliese AM (2015) Purinergic signalling in brain ischemia. Neuropharmacology. In Press. 38. Zador Z, Stiver S, Wang V, Manley GT (2009) Role of aquaporin-4 in cerebral edema and stroke. Handb Exp Pharmacol 190:159–170 39. Lee M, Lee SJ, Choi HJ, Jung YW, Frokiaer J, Nielsen S, Kwon TH (2008) Regulation of AQP4 protein expression in
463 rat brain astrocytes: role of P2X7 receptor activation. Brain Res 1195:1–11 40. Hirayama Y, Ikeda-Matsuo Y, Notomi S, Enaida H, Kinouchi H, Koizumi S (2015) Astrocyte-mediated ischemic tolerance. J Neurosci 35(9):3794–3805 41. Franciosi S, De Gasperi R, Dickstein DL, English DF, Rocher AB, Janssen WG, Christoffel D, Sosa MA, Hof PR, Buxbaum JD, Elder GA (2007) Pepsin pretreatment allows collagen IV immunostaining of blood vessels in adult mouse brain. J Neurosci Methods 163(1): 76–82 42. Szabo A, Kalman M (2004) Disappearance of the post-lesional laminin immunopositivity of brain vessels is parallel with the formation of gliovascular junctions and common basal lamina. A double-labelling immunohistochemical study. Neuropathol Appl Neurobiol 30(2):169–177
ORIGINAL ARTICLE
Lack of functional P2X7 receptor aggravates brain edema development after middle cerebral artery occlusion Melanie Kaiser 1 & Anja Penk 2 & Heike Franke 1 & Ute Krügel 1 & Wolfgang Nörenberg 1 & Daniel Huster 2 & Michael Schaefer 1
Received: 26 November 2015 / Accepted: 23 March 2016 / Published online: 5 April 2016 # Springer Science+Business Media Dordrecht 2016
Abstract Effective therapeutic measures against the development of brain edema, a life-threatening complication of cerebral ischemia, are necessary to improve the functional outcome for the patient. Here, we identified a beneficial role of purinergic receptor P2X7 activation in acute ischemic stroke. Involvement of P2X7 in the development of neurological deficits, infarct size, brain edema, and glial responses after ischemic cerebral infarction has been analyzed. Neurologic evaluation, magnetic resonance imaging, and immunofluorescence assays were used to characterize the receptor’s effect on the disease progress during 72 h after transient middle cerebral artery occlusion (tMCAO). Sham-operated animals were included in all experiments for control purposes. We found P2X7-deficient mice to develop a more prominent brain edema with a trend towards more severe neurological deficits 24 h after tMCAO. Infarct sizes, T2 times, and apparent diffusion coefficients did not differ significantly between wild-type and P2X7−/− animals. Our results show a characteristic spatial distribution of
Electronic supplementary material The online version of this article (doi:10.1007/s11302-016-9511-x) contains supplementary material, which is available to authorized users.
reactive glia cells with strongly attenuated microglia activation in P2X7−/− mice 72 h after tMCAO. Our data indicate that P2X7 exerts a role in limiting the early edema formation, possibly by modulating glial responses, and supports later microglia activation. Keywords Brain ischemia . MCAO . Microglia . P2X7 . Purinergic receptor
Abbreviations ADC Apparent diffusion coefficient AQP4 Aquaporin 4 CCA Common carotid artery DWI Diffusion weighted image ECA External carotid artery ICA Internal carotid artery MCA Middle cerebral artery MCAO Middle cerebral artery occlusion P2X7−/− P2X7 receptor knockout NMR Nuclear magnetic resonance pMCAO Permanent middle cerebral artery occlusion RARE Rapid acquisition with relaxation enhancement ROI Region of interest tMCAO Transient middle cerebral artery occlusion VOI Volume of interest WT Wild-type
* Michael Schaefer [email protected]
Introduction 1
2
Rudolf Boehm Institute of Pharmacology and Toxicology, Medical Faculty, Universität Leipzig, Härtelstr. 16-18, 04107 Leipzig, Germany Institute of Medical Physics and Biophysics, Medical Faculty, Universität Leipzig, Leipzig, Germany
Ischemic stroke is one of the most important vascular diseases, which causes loss of nervous tissue and may lead to severe disability or even death. A major complication is the development of brain edema, which
454
exerts pressure on neighboring tissues and further aggravates the disease. Despite the clinical importance of the disease, knowledge gaps concerning the pathophysiology of cerebral ischemia continue to complicate the search for an effective therapy. The purinergic receptor P2X7, an ATP-gated cation channel belonging to the purinergic P2 receptor family, has been shown to affect pathophysiological events following brain ischemia. P2X7 is a low-affinity ATP receptor mainly expressed on immune and glial cells. Regarding the central nervous system, P2X7 is known to be present on microglia, astrocytes, oligodendrocytes, and some neuronal populations [1]. Inflammation and tissue trauma may entail high extracellular ATP concentrations [2] and trigger the activation of P2X7 [3]. Subsequent Ca2+ entry may cause neuronal and oligodendrocyte cell death [4, 5], but activation of P2X7 has also been shown to induce microglia proliferation [6]. Consistently, the receptor is involved in pathophysiological events following ischemic brain damage, but studies to clarify its exact role yielded inconsistent results. On the one hand, a detrimental pro-inflammatory role is attributed to P2X7 in experimental brain injury, which was reversible by antagonists or diminished by gene knockout [7, 8]. These results are in agreement with the fact that the maturation and release of IL-1β as well as the generation of reactive oxygen species, both pathologic stimuli in experimental and human stroke [9, 10], are triggered upon the activation of the receptor [11]. On the other hand, results obtained by pharmacological modulation of P2X7 point towards a beneficial function of the receptor, presumably conveyed by activating neuroprotective responses in glial cells [12]. In fact, P2X7-bearing microglia and astrocytes may contribute to the reduction of neuronal cell death [13, 14]. A primary in vivo study did not argue for an influence of P2X7 on the infarct volume 24 h after ischemia [15]; however, no further infarct parameters or time points were investigated. In this context, we evaluated the contribution of P2X7 to the outcome after middle cerebral artery occlusion (MCAO), an established model of ischemic stroke, in mice after surgery for the subsequent 72 h. We applied clinical, nuclear magnetic resonance (NMR) imaging, and immunofluorescence methods to obtain data regarding neurological deficits, infarct size, brain edema, and glial cell activation. Our results are compatible with a protective role of P2X7 during initial phases after the onset of ischemia. At later time points, microglia activation was supported by P2X7, but no significant differences were observed with respect to the clinical outcome.
Purinergic Signalling (2016) 12:453–463
Material and methods Animals Mouse breeding and experiments were performed in the animal facility of the Faculty of Medicine, University of Leipzig according to the ARRIVE guidelines as well as the European (Council Directive 2010/63/EU) and German guidelines for the welfare of experimental animals and were approved by the local authorities (Landesdirektion Leipzig; ID: TVV 64/ 13). All possible efforts were made to replace, reduce, and refine experiments including laboratory animals. Male mice (25–35 g) with P2X7−/− or wild-type (WT) genotype of C57BL/6 J background originally purchased from the Jackson Laboratory (Bar Harbor, ME, USA) were used [16]. Mice were housed under standard conditions with free access to food and water and a 12-h light–dark cycle with lights on from 7:00 a.m. From first NMR session on, animals were housed individually. Surgical procedures Anesthesia was induced with 2.5 % of isoflurane (Baxter, Unterschleissheim, Germany) in oxygen and maintained with 1.5 % of the anesthetic. Analgesia was assured by subcutaneous injection of 200 mg/kg metamizol (WDT, Garbsen, Germany), a non-opioid analgesic with only weak antiinflammatory effects. Metamizol was administered before surgery and every 6 hours afterwards. MCAO was essentially achieved as described before with minor modifications [17]. Briefly, after a midline neck incision, the left common carotid artery (CCA) and its bifurcation into the external (ECA) and internal carotid artery (ICA) were exposed and carefully separated from the adjacent vagus nerve. The CCA as well as the ECA were ligated (Suprama Silk 7–0, Feuerstein, Berlin, Germany) while blood flow through the ICA was impeded by a microvascular clip. A polysiloxane-coated filament (Suprama Nylon 8–0, Feuerstein) coated with Xantopren M mucosa (Heraeus Kulzer GmbH, Hanau, Germany) was then inserted through a small opening cut into the CCA distal to the ligature. Upon removal of the microvascular clip, the filament was advanced approximately 11 mm along the ICA into the circle of Willis, where it blocked the origin of the middle cerebral artery (MCA). The filament was immediately removed in sham-operated animals or after 60 min in mice undergoing transient MCAO (tMCAO). It remained in place in animals belonging to the permanent MCAO (pMCAO) group. The ICA was then ligated (Suprama Silk 5–0, Feuerstein) and the neck incision closed. Neurological score Neurological deficits were evaluated daily, and a neurological score based on a scale described before [18] was assigned to
Purinergic Signalling (2016) 12:453–463
each animal. Initially, mice were suspended by their tail above the surface of a table. Animals able to extend both forelimbs towards the surface were scored 0. If the right forelimb remained flexed and subsequent tests did not reveal further neurological deficits, the score was set to 1. Mice given a score of 2 showed decreased loading of the right forelimb when moving without the support of the hind limbs as by elevating them by the tail. If fixation of the animal by the tail caused them to move in circles, they were scored 3. Circling behavior when allowed to move freely resulted in a score of 4. Severely reduced general condition and inability to move resulted in a score of 5 and immediate euthanasia. Magnetic resonance imaging protocol and image analysis One and 3 days after MCAO, mice were anesthetized with 75 mg/kg ketamine (WDT) and 6 mg/kg xylazine (Bayer Health Care, Leverkusen, Germany). Subsequently, each mouse underwent MRI measurements, which were performed on a Bruker DRX 300 (7.05 T) wide bore NMR spectrometer equipped with a microimaging unit including a mouse bed (MicWB40 Probe using a single channel birdcage resonator with 30 mm inner diameter and a Micro2.5 Gradient System, Bruker BioSpin GmbH, Rheinstetten, Germany) providing a maximal gradient strength of 730 mT/m. The cerebral brain region was mapped using nine coronal slices with a slice thickness of 1 mm (field of view 20 mm × 20 mm) and no gap in between (Online Resource 1A). Three different scanning protocols were used. High resolution, intermediate weighted images were acquired using a fast spin-echo sequence (rapid acquisition with relaxation enhancement, RARE) with a RARE factor of 8 (repetition time 1200 ms; echo time 9.2 ms; effective echo time 32.4 ms; matrix 256 × 256; averages 2; total scanning time 76 s). A multi slice multi echo (MSME) sequence was used to measure the T2 time (repetition time 2500 ms; echo time 10.2 ms; number of echo images 12; matrix 256 × 128; total scanning time 320 s), and five diffusion weighted images (DWI) for each slice were acquired using a pulsed gradient spin echo to measure the apparent diffusion coefficient (ADC) along the rostral-caudal direction (repetition time 1200 ms; echo time 23.7 ms; time between gradient pulses (Δ) 12.2 ms; gradient duration (δ) 6 ms; matrix 128 × 64; b values of 52.2, 157.1, 355.8, 638.6, and 1110.3 s/mm2 (diffusion gradient 35–200 mT/m); total scanning time 384 s). Image analysis was performed in a randomized, blinded fashion using custom MATLAB scripts. First, RARE images were used to create a mask for each of the nine slices including only the brain to speed up analysis. These masks were then downsampled for further analysis of T2 and DWI image trains to the desired resolution (256 × 128 for T2 and 128 × 64 for DWI), including only pixels in the low resolution mask if at least half of the corresponding pixels formed part of the high-
455
resolution mask from the RARE. T2 and ADC maps were calculated using a mono-exponential fitting function. However, as the images show a significant noise floor due to limited scanning time, the fitting functions for T2 and ADC were corrected for an offset term (determined as mean value of an area including pure noise). T2 maps were then adjusted in brightness and contrast to define a region of interest (ROI) with altered T2 values. All ROIs were then summed up to a volume of interest (VOI) representing the size of the infarct. A corresponding VOI on the contralateral site was created using mirroring (slightly adjusted to not include physiologically hyperintense neuroanatomic regions) (Online Resource 1C). For these two VOIs (ipsi- and contralateral), mean values for T2 times and ADC were calculated. Mice with no observable altered region in the whole brain were analyzed in the same manner by drawing VOIs similar to those in animals with cerebral infarction. Second, the position of the midline was determined using neuroanatomic landmarks such as the fissura longitudinalis, lateral, and third ventricles in three consecutive slices (Online Resource 1B). The percentage of the infarcted hemisphere as a proportion of the total brain volume in the middle slice was used to describe oedematous brain swelling. Eventually, swelling of the infarcted hemisphere was put in relation to the size of the ischemic lesion.
Immunofluorescence staining Animals were deeply anesthetized and transcardially perfused (2 % paraformaldehyde in sodium acetate buffer containing 10 IU/ml heparin, pH 6.5, followed by 2 % paraformaldehyde in sodium borate buffer containing 0.15 % glutaraldehyde, pH 8.5). Brains were immediately removed and stored in the latter solution. In preparation for immunofluorescence staining, coronal sections (50 μm) were cut using a vibratome (Leica, Typ VT 1200S, Nussloch, Germany) and collected as free-floating slices. Unspecific protein binding was prevented by blocking for 30 min with 5 % fetal calf serum (Sigma-Aldrich, St. Louis, MO, USA) dissolved in 0.05 M Tris-buffered saline (TBS;pH 7.6; Sigma-Aldrich) and 10 % Triton X-100 (Sigma-Aldrich). Slices were then incubated with primary antibodies rabbit anti-ionized calcium binding adaptor molecule 1 (Iba1; 1:1000; Wako Chemicals GmbH, Neuss, Germany), mouse anti-glial fibrillary acidic protein (GFAP; 1:1000; Dako Cytomation, Glostrup, Denmark), and rabbit anti-laminin (1:50; Dako) for 48 h at 4 °C. Following subsequent washing steps with TBS, slices were exposed to secondary antibodies Cy3-conjugated donkey anti-rabbit IgG (1:800; Jackson ImmunoResearch, West Grove, PA, USA) or Cy5-conjugated donkey anti-mouse IgG (1:800; Jackson ImmunoResearch) for 2 h at room temperature. Slices were then mounted on slide glasses, dehydrated in ethanol, cleared
456
in n-butylacetate, embedded in Entellan (Merck, Darmstadt, Germany), and covered. Confocal microscopy and quantification Immunofluorescence was examined using a confocal laser scanning microscope (LSM 510 Meta, Zeiss, Oberkochen, Germany) equipped with two helium/neon lasers emitting at 543 and 633 nm, respectively. Cy5-labeled signals were color coded in green. Images taken with a Plan-Neofluar 10x/0.3 objective allowed visualization of infarct core, penumbra, and surrounding tissues and served as a template to define regions of interest (ROIs). While the infarcted tissue was characterized by loss of cellular integrity and only permitted measurements of unspecific immunofluorescence, the infarct border itself was strongly demarcated by activated microglia, and intense GFAP immunofluorescence could be seen somewhat further from the infarct core. In order to establish spatial profiles of glial responses, we thus positioned ROIs so that they encompassed the infarct border and the surrounding tissue. Immunofluorescence intensity along three such ROIs per animal was measured, corrected for the respective background signals, and normalized to unaffected brain regions to compensate for uneven staining in different slices. Similarly, three ROIs were defined to count microglial cells. C-Apochromat 40x/1.2 and C-Apochromat 63x/1.2 water immersion objectives were used to obtain detailed images of glial cell morphology. Data evaluation and statistical analysis All data are expressed as mean ± SEM, obtained in n animals. One and two-way repeated measures ANOVA, and subsequent pairwise multiple comparison procedures were applied to compare variables in more than two groups. The two-tailed Student’s t test was used to test for statistically significant differences between two experimental groups. Values for P < 0.05 were considered significant.
Results P2X7 reduces rapid development of brain edema after ischemic infarction We evaluated the progression of neurological deficits, infarct size and brain swelling during 72 h after tMCAO. Mice lacking P2X7 tended to initially (6 h) experience more severe neurological problems than those of the WT (P = 0.169), but these findings only constitute a non-significant trend that leveled out until the end of the observation (Fig. 1a). Furthermore, the postischemic interval was characterized by a significantly greater midline shift in the P2X7-deficient
Purinergic Signalling (2016) 12:453–463
group at 24 h after reperfusion (P = 0.026), (Fig. 1c). Whereas the WT group expressed a significant enlargement of the midline shift over time (P = 0.046), the midline was not further shifted in the P2X7-deficient animals until 72 h after surgery. Of note, the values represented in Fig. 1c correspond to the overall swelling of the infarcted hemisphere. In order to calculate these values, we determined the line separating both hemispheres, and for this task, we used high resolution RARE images and T2 maps (also see Online Resource 1). Due to resolution, these two types of images were superior to DWI images regarding the recognizability of anatomical structures marking the midline. Since only the position of these anatomical landmarks (and the total brain volume as restricted by the intact mouse skull) affected the results depicted in Fig. 1c, it is not possible to deduct the respective share of cytotoxic and vasogenic brain edema from these measurements. Because the infarct size appeared to be slightly larger in P2X7-deficient mice (Fig. 1b), one could speculate the corresponding pronounced midline shift to be a mere result of an increased volume of infarcted tissue. We therefore established the ratio between swelling of the infarcted hemisphere and the respective lesion volume and obtained strongly increased values for P2X7-deficient animals at 24 h that confirm our aforementioned results (P = 0.011, Fig. 1d). Consistent with the neurological scores after 72 h, differences regarding brain edema were no longer detectable. At this time point in addition to the midline shift, the relative brain swelling of WT mice had significantly increased (P = 0.032). Although T2 relaxation times and ADC did not differ at any time between WT and P2X7−/− mice, both displayed significantly prolonged T2 relaxation times and decreased ADC when compared with sham-operated animals (Fig. 2; Table 1, upper panel), indicating the development of a cytotoxic edema, i.e., the shift of extracellular water to intracellular compartments [19]. Furthermore, T2 relaxation times significantly grew between 24 and 72 h after MCAO (WT: P = 0.009; P2X7−/−: P = 0.005), indicating interstitial water accumulation in all animals that underwent tMCAO (Table 1, upper panel). Sham-operated animals did not show neurological deficits. T2 relaxation times as well as ADC values of the ipsi- or contralateral brain hemispheres in sham-operated mice did not differ from those obtained in the contralateral hemisphere of animals that underwent tMCAO. Hence, contralateral values served as a reference for unaffected tissue in Table 1. Microglia activation is attenuated in animals lacking P2X7 The ionized calcium binding adaptor protein 1 (Iba1), a marker expressed in microglia and macrophages that is upregulated after tissue damage, was used to evaluate microglial activation 72 h after tMCAO [20]. Activated astrocytes were monitored by expression of glial fibrillary acidic protein (GFAP). Both Iba1 and GFAP expression were highly increased after
Purinergic Signalling (2016) 12:453–463
457
Fig. 1 Evaluation of neurological deficits, infarct size, and brain swelling at different time points after tMCAO. a Neurological scores indicating the severity of neurological deficits at various time points after tMCAO in WT (grey bars) and P2X7−/− animals (black bars). b Infarct sizes were determined in vivo by NMR imaging. Consecutive image slices covering the whole brain were evaluated to obtain the respective infarct volumes. c Brain swelling, resulting in a midline shift and compression of the contralateral hemisphere, was assessed by calculating the volume occupied by the infarcted hemisphere as percentage of the total brain volume.d Swelling of the infarcted hemisphere normalized to respective infarct size. Of note, relative brain swelling augmented significantly between 24 and 72 h in WT mice. n = 8–12; asterisks indicate P < 0.05
cerebral ischemia in animals that underwent tMCAO when compared with sham-operated mice and represent the characteristic spatial pattern of glia cells with respect to the infarct site (Fig. 3, left panel). In order to obtain information regarding the distribution of activated glial cells, we defined a cortical ROI that encompasses the penumbra and adjacent brain tissue (Fig. 4a, b, Fig. 5a, b) and measured immunofluorescence intensity along this ROI. Since cellular integrity was almost completely lost in the infarct cores, immunofluorescence signals in these regions were inconsistent and
Fig. 2 Calculated T2 (upper panel) and ADC maps (lower panel) overlaid with RARE images from NMR. Representative examples for tMCAO and sham groups of WT and P2X7-deficient mice 24 h after surgery are shown. Identified ischemic regions are shown grey rimmed. Scale bar 2 mm. The linear color coding bar represents values of 0–100 ms for T2 and 0– 250 × 10−5 mm2/s for the ADC
considered non-specific. Consistent with microglial activation, immunofluorescence of Iba1 was highly enhanced at the infarct border in both WT and P2X7−/− mice, where microglia literally formed a wall between the infarct core and the peri-infarct area (Fig. 4c). This microglial response was significantly attenuated in the absence of P2X7 (P = 0.016, Fig. 4d). Notably, this effect was not caused by a decreased microglial cell number but reflected a lower per-cell Iba1 immunoreactivity in P2X7−/− mice compared with their WT (302.4 ± 15.7 vs. 300.0 ± 7.8 microglial cells per ROI in WT
458 Table 1
Purinergic Signalling (2016) 12:453–463 Clinical and NMR parameters obtained in WT and P2X7−/− mice after tMCAO and pMCAO
T2 relaxation time, ms Apparent diffusion coefficient, mm2 s−1 10−5
Time after tMCAO (h) 24 72 24 72
WT infarct 46.51 ± 0.94 54.66 ± 2.08* 49.07 ± 2.60 47.55 ± 2.75
Time after pMCAO (h) WT infarcte Neurological score Infarct size, mm3 T2 relaxation time, ms Apparent diffusion coefficient, mm2 s−1 10−5
P2X7−/− infarct 45.53 ± 0.60 54.61 ± 3.76* 51.36 ± 1.82 49.14 ± 3.51
WT contralat. P2X7−/− contralat. n 32.00 ± 0.28 31.74 ± 0.18 64.36 ± 0.93 64.69 ± 3.19
31.89 ± 0.25 31.23 ± 0.30 64.03 ± 2.36 66.42 ± 1.14
8 5–7
P2X7−/− infarct WT contralat. P2X7−/− contralat. n
6 24 24
4.0 ± 0.32 4.4 ± 0.24 3.6 ± 0.51 4.2 ± 0.37 114.43 ± 17.33 91.97 ± 9.85
– – –
– – –
4
24 24
54.75 ± 0.98 39.34 ± 0.79
32.13 ± 0.56 69.21 ± 7.88
31.30 ± 0.19 76.63 ± 3.46
4 3
54.91 ± 0.84 45.80 ± 5.66
5
*P < 0.05 different from mean values at 24 h
and P2X7−/− mice, respectively; 220.1 ± 4.3 and 204.3 ± 11.5 microglial cells per ROI in sham-operated WT and P2X7−/− mice, respectively). This finding indicates no strong effect on microglia recruitment or local proliferation, but a limitation of microglial activation when the function of P2X7 is missing. Under resting conditions, microglia possess a ramified morphology with small somata and extensive processes that facilitate surveillance. While this phenotype was visible in sham-
operated animals, transient cerebral ischemia caused morphologic changes, and microglia adopted a reactive, in part even amoeboid-like phenotype (Fig. 3, magnified images of Iba1). Of note, there were no discernible morphological differences between reactive microglia and astrocytes of WT and P2X7−/− mice (Fig. 3, magnified images of Iba1 and GFAP). Reactive astrocytes were seen in a broader region outside the infarct core (Fig. 5c). Although both activation of microglia and
Fig. 3 Confocal images of double immunofluorescence to characterize microglia and astroglia distribution in peri-infarct tissue and corresponding brain sections of sham-operated mice 72 h after surgery. The upper rows, corresponding to WT and P2X7−/− mice after tMCAO, illustrate how Iba1-labeled, reactive microglia (arrows) delimit the ischemic core
(asterisks) while increased astroglial GFAP-expression (arrowheads) prevails in the surrounding tissue over a wider brain area. Sham-operated mice, as represented in the lower rows, depict a homogeneous distribution of glial cells in their resting state. Merged images are included for better spatial visualisation. Scale bars 100 μm
Purinergic Signalling (2016) 12:453–463
459
Fig. 4 P2X7-dependent microglia activation in ischemic lesion and surrounding brain tissue 72 h after tMCAO or sham-operation. a Representative confocal image of Iba1 immunofluorescence in the infarcted cortex of a WT mouse 3 days after tMCAO. The core region is located towards the left of the image, the peri-infarct tissue towards the right side. The white rimmed rectangle corresponds to a ROI defined for further analysis. b Representative image of a P2X7 −/− mouse. c Quantification of Iba1 immunofluorescence. Values shown are based on
measurements along the complete ROI and were normalized to immunofluorescence intensities measured in a distant and apparently unaffected region b. Grey traces represent the WT mice and black traces represent P2X7−/− mice. Continuous traces describe Iba1 immunofluorescence after tMCAO, and dotted traces derive from sham-operated animals. d Microglia activation at infarct border depicted as ratio of Iba1 immunofluorescence between infarct-delimiting region Ba^ and region Bb^. n = 8– 10 for tMCAO groups, n = 4 for sham groups; asterisks indicate P < 0.05
astrocytes has been linked to the activation of P2X7 [21], we did not find any differences in the immunofluorescence of GFAP in WT and P2X7−/− animals (Fig. 5d). Laminin, an extracellular matrix component primarily located in the basement membranes of vascular endothelial cells, is a well-established marker for cerebral blood vessels which is known to be upregulated following brain ischemia [22, 23]. Since P2X7 may influence tissue remodeling and angiogenesis [24, 25], we examined whether genetic deletion of the receptor affects the expression level of laminin. While laminin immunoreactivity in the peri-infarct tissue of WT and P2X7−/− mice and in sham-operated animals was very low, intense immunofluorescence could be measured in determined ROIs within the infarct core (Online Resource 2). However, no difference in laminin-immunofluorescence intensity could be detected between the two groups that underwent tMCAO (Online Resource 3).
others pointed out that the high mortality poses a major challenge [28]. In our study, 1/6 WT and 2/7 P2X7−/− mice had to be euthanized during the first hours after pMCAO. Due to severe neurological deterioration and in the interest of animal welfare, we decided to euthanize the remaining animals 24 h after surgery. We did not detect any relation between the presence of a functional P2X7 receptor and progression of neurological deficits or the resulting need for euthanasia. Significant differences could neither be observed regarding infarct volumes, T2 relaxation times, and ADC of WT and P2X7−/− groups (Table 1, lower panel). However, these findings have to be interpreted keeping the limited group size in mind.
P2X7 does not protect from severe neurological deterioration after pMCAO We had initially chosen to permanently occlude the middle cerebral artery to model human stroke, where recanalization can often not be achieved [26]. While some previous studies reported high survival rates for days after pMCAO (e.g., [27]),
Discussion The aim of the present work was to characterize the role of P2X7 in pathophysiological events following ischemic stroke for a prolonged period of time. We found P2X7−/− mice to rapidly develop a significantly larger brain edema, a condition accompanied by a tendency towards more severe neurological deficits whereby the latter did, however, not reach significance. Microglial activation 72 h after the initial insult was significantly attenuated in animals lacking P2X7.
460
Purinergic Signalling (2016) 12:453–463
Fig. 5 Astroglia activation at the infarct border and surrounding brain tissue 72 h after tMCAO or sham-operation. a Representative confocal image of GFAP immunofluorescence in the infarcted cortex of a WT mouse 3 days after tMCAO. The infarct core region is located towards the left of the image, the peri-infarct tissue towards the right side. The white rimmed rectangle corresponds to a ROI comprising the infarct border and peri-infarct tissue for further analysis. b Representative image of a P2X7−/− mouse. c Quantification of GFAP immunofluorescence. Values
shown are based on measurements from left to right along the complete ROI and were normalized to immunofluorescence intensities measured in contralateral reference region. Grey traces represent the WT, and black traces represent P2X7−/− mice. Continuous traces describe GFAP immunofluorescence after tMCAO, and dotted traces derive from shamoperated animals. d GFAP activation in peri-infarct tissue depicted as ratio of GFAP immunofluorescence between infarcted hemisphere and contralateral reference region. n = 9 for tMCAO groups, n = 4 for sham groups
No differences were observed in neurological scores and infarct size in the initially chosen permanent MCAO model. We did not pursue this approach any further due to serious neurological deficits and lethality. Due to the high mortality after pMCAO, we are unable to draw reliable conclusions regarding the role of P2X7 after permanent cerebral ischemia. The change to transient MCAO was associated with a reduction of infarct size and neurological scores, which displayed a trend towards a better outcome when P2X7 is functionally active. Furthermore, we found that animals deficient for P2X7 were more prone to develop severe brain edema within the first 24 h after onset of transient ischemia. During this early phase of ischemic stroke, cytotoxic brain edema reaches its maximum [29] and is clinically followed by a midline shift that is used as a criterion for the development of brain edema in stroke patients [30]. The mechanisms causing brain edema are only partly understood. As a result of the ischemia, energy depletion in the irrigation area causes the failure of ion pumps and, thus, provokes the intracellular accumulation of excess fluid. Only at later stages of the disease, blood–brain barrier breakdown induces the additional formation of vasogenic brain edema. In contrast, brain pathologies involving direct impairment of the blood–brain barrier primarily entail the formation of vasogenic brain edema. In contrast to our study, genetic deletion of P2X7 has been shown to limit edemic
development in a model of traumatic brain injury. Differences regarding the pathogenesis of edema formation may account for this apparent contradiction [7]. Indeed, in agreement with our data, P2X7 receptor agonists have been reported to improve behavioral dysfunctions in the MCAO model of ischemic stroke 24 h after surgery [12]. The transition from a mainly cytotoxic to a more vasogenic edema between day 1 and day 3 after surgery may also explain why differences between WT and P2X7−/− mice leveled out. Furthermore, a recent study showed that glial cell recruitment takes place early after onset of ischemia and that concentrations of interleukins IL-1β, IL-6, and TNF-α peak 24 h after tMCAO [31]. Since P2X7 is known to affect the expression of all these cytokines, these findings may hint at a greater receptor activity at this time point and thus explain why we detected significant differences only during the early phase of brain ischemia. Both WT and P2X7−/− mice showed increased immunofluorescence signals pointing to microglial and astroglial activation 72 h after tMCAO. The microglial response, however, was significantly attenuated in P2X7-deficient animals. We selected this survival time to evaluate glial responses because previous studies described a peak of glial activation within a few days after MCAO [32]. In future studies, the histological
Purinergic Signalling (2016) 12:453–463
and functional state of either cell type at earlier times after cerebral ischemia may reveal the mechanisms which cause aggravated brain edema in P2X7−/− mice. Although the non-opioid metamizol may be responsible for weak anti-inflammatory effects and attenuation of glial activation, all animals received the same dose of analgesic, and such effects should therefore affect both WT and P2X7−/− mice equally. Possible explanations for our observations have to consider that P2X7 is expressed in microglia as well as in astrocytes and both cell types may exert damaging or favorable effects, depending on their activation state and the actual point of time after ischemia/reperfusion. Microglia are known to be able to differentiate into distinct phenotypes. Although ischemic brain damage initially facilitates transition towards the pro-inflammatory M1 type, microglial cells retain their ability to shift between one and the other. The presence of M2 microglia favors tissue repair and studies involving mice lacking IL-4, which evokes transition towards the M2 phenotype, have shown that the outcome worsens in absence of M2 microglia [33]. In a recent study, it has been shown that IL-4 expression is enhanced in the penumbra of infarcted tissue, where it reaches its peak during the first 24 h [34]. Even though data regarding microglia is not available, it has been proven that M2 macrophages do express P2X7 [35]. It is thus tempting to speculate that the protective M2 phenotype may critically rely on an intact P2X7. Consequently, a lacking functional P2X7 receptor in M2 microglial cells may cause the detrimental effects observed 24 h after MCAO in the absence of a functional P2X7 receptor, as has been proposed before [36]. On the other hand, there is extensive evidence that antagonism of P2X7 exerts neuroprotective effects, presumably by preventing intracellular Ca2+ overloads. Thus, sustained activation of P2X7 has been shown to facilitate glutamate release and cause neuronal and glial cell death [37]. P2X7 activation may also lead to detrimental, prolonged neuroinflammation [37]. The fact that P2X7 antagonism confers beneficial effects by preventing microglial cell death as well as neuroinflammation once more illustrates the dual role played by microglia in pathophysiological events following brain ischemia. Aquaporin 4 (AQP4) is a water permeable channel mainly expressed in astrocytic processes adjacent to cerebral capillaries and plays an important role in bidirectional water transport [38]. The cytotoxic brain edema prevailing in acute cerebral ischemia can be reduced by downregulation of AQP4. Since AQP4 downregulation has been reported to be triggered upon P2X7 activation [39], this mechanism could serve as an explication for the exacerbated brain edema developed in P2X7deficient mice in this study. In cases of vasogenic brain edema, however, AQP4-mediated water movement is crucial for fluid clearance, and AQP4 downregulation
461
by P2X7 would be detrimental [38]. It is also conceivable that P2X7-dependent mechanisms occur downstream of GFAP-related astroglial activation, as has been shown in a recent study comparing WT and P2X7−/− mice subjected to cerebral ischemia. In case of preconditioning to ischemic insults, astrocytic P2X7 was shown to mediate ischemic tolerance and to provide neuroprotection even though no differences between the two genetic phenotypes were found in GFAP immunofluorescence [40]. Similar to our results, infarct sizes analyzed by post mortem 2,3,5, triphenyltetrazolium chloride (TTC) staining were not different between the genotypes in this study. In our study, laminin immunofluorescence was very low in intact brain tissue. This observation is consistent with the results of earlier studies that have also described a considerable increase in signal intensity at sites of cerebral lesions [41]. It has been reported that this phenomenon is provoked by the breakdown of the neurovascular unit, which leaves the corresponding epitopes accessible for the antibodies, a development that is reversed upon the re-establishment of gliovascular junctions [42]. Our results indicate that this process is not strongly influenced by P2X7.
Conclusions The development of brain edema is a life-threatening complication of cerebral ischemia, the resulting increase in intracranial pressure further compromises adjacent brain structures and exacerbates the disease. Effective therapeutic measures are necessary to improve the functional outcome for the patient. Here, we identified a beneficial role of P2X7 activation in acute ischemic stroke. Our results emphasize the need for a detailed understanding of the receptor’s role in neuroinflammation dependent on the underlying disease, the target cell type, and the time passed since the onset of ischemia in order to decide on a possible therapeutic P2X7 modulation after ischemic stroke. Acknowledgments This work was supported by the Deutsche Forschungsgemeinschaft (DFG) within the framework of Research Group FOR748 and TRR67 (project A6). The authors are grateful to André Rex for technical advice. Expert technical assistance and animal care were provided by Katrin Becker and Anne-Kathrin Krause. Compliance with ethical standard Conflict of interest The authors declare that they have no conflicts of interest. Ethical approval All applicable international, national, and/or institutional guidelines for the care and use of animals were followed. All procedures performed in studies involving animals were in accordance
462
Purinergic Signalling (2016) 12:453–463
with the ethical standards of the institution at which the studies were conducted.
17.
Funding This work was supported by the Deutsche Forschungsgemeinschaft (DFG) within the framework of Research Group FOR748 and TRR67 (project A6).
18.
19.
20.
References 1.
2.
3.
4.
5.
6.
7.
8.
9.
10. 11.
12.
13.
14.
15.
16.
Del Puerto A, Wandosell F, Garrido JJ (2013) Neuronal and glial purinergic receptors functions in neuron development and brain disease. Front Cell Neurosci 7:197 Franke H, Grummich B, Hartig W, Grosche J, Regenthal R, Edwards RH, Illes P, Krügel U (2006) Changes in purinergic signaling after cerebral injury—involvement of glutamatergic mechanisms? Int J Dev Neurosci 24(2–3):123–132 Oliveira JF, Riedel T, Leichsenring A, Heine C, Franke H, Krügel U, Nörenberg W, Illes P (2011) Rodent cortical astroglia express in situ functional P2X7 receptors sensing pathologically high ATP concentrations. Cereb Cortex 21(4):806–820 Domercq M, Perez-Samartin A, Aparicio D, Alberdi E, Pampliega O, Matute C (2007) P2X7 receptors mediate ischemic damage to oligodendrocytes. Glia 58(6):730–740 Nishida K, Nakatani T, Ohishi A, Okuda H, Higashi Y, Matsuo T, Fujimoto S, Nagasawa K (2012) Mitochondrial dysfunction is involved in P2X7 receptor-mediated neuronal cell death. J Neurochem 122(6):1118–1128 Monif M, Burnstock G, Williams DA (2010) Microglia: proliferation and activation driven by the P2X7 receptor. Int J Biochem Cell Biol 42(11):1753–1756 Kimbler DE, Shields J, Yanasak N, Vender JR, Dhandapani KM (2007) Activation of P2X7 promotes cerebral edema and neurological injury after traumatic brain injury in mice. PLoS One 7(7):e41229 Ryu JK, McLarnon JG (2008) Block of purinergic P2X7 receptor is neuroprotective in an animal model of Alzheimer’s disease. Neuroreport 19(17):1715–1719 Lambertsen KL, Biber K, Finsen B (2012) Inflammatory cytokines in experimental and human stroke. J Cereb Blood Flow Metab 32(9):1677–1698 Margaill I, Plotkine M, Lerouet D (2005) Antioxidant strategies in the treatment of stroke. Free Radic Biol Med 39(4):429–443 Alves LA, Bezerra RJ, Faria RX, Ferreira LG, da Silva FV (2013) Physiological roles and potential therapeutic applications of the P2X7 receptor in inflammation and pain. Molecules 18(9):10953–10972 Yanagisawa D, Kitamura Y, Takata K, Hide I, Nakata Y, Taniguchi T (2008) Possible involvement of P2X7 receptor activation in microglial neuroprotection against focal cerebral ischemia in rats. Biol Pharm Bull 31(6):1121–1130 Myer DJ, Gurkoff GG, Lee SM, Hovda DA, Sofroniew MV (2006) Essential protective roles of reactive astrocytes in traumatic brain injury. Brain 129(Pt 10):2761–2772 Suzuki T, Hide I, Ido K, Kohsaka S, Inoue K, Nakata Y (2004) Production and release of neuroprotective tumor necrosis factor by P2X7 receptor-activated microglia. J Neurosci 24(1):1–7 Le Feuvre R, Brough D, Rothwell N (2002) Extracellular ATP and P2X7 receptors in neurodegeneration. Eur J Pharmacol 447(2–3): 261–269 Solle M, Labasi J, Perregaux DG, Stam E, Petrushova N, Koller BH, Griffiths RJ, Gabel CA (2001) Altered cytokine production in mice lacking P2X7 receptors. J Biol Chem 276(1):125–132
21. 22.
23.
24.
25.
26. 27.
28. 29.
30.
31.
32.
33.
34.
35.
Engel O, Kolodziej S, Dirnagl U, Prinz V (2011) Modeling stroke in mice - middle cerebral artery occlusion with the filament model. J Vis Exp 47:2423 Menzies SA, Hoff JT, Betz AL (1992) Middle cerebral artery occlusion in rats: a neurological and pathological evaluation of a reproducible model. Neurosurgery 31(1):100–107 Dijkhuizen RM, Nicolay K (2003) Magnetic resonance imaging in experimental models of brain disorders. J Cereb Blood Flow Metab 23(12):1383–1402 Ito D, Tanaka K, Suzuki S, Dembo T, Fukuuchi Y (2001) Enhanced expression of Iba1, ionized calcium-binding adapter molecule 1, after transient focal cerebral ischemia in rat brain. Stroke 32(5): 1208–1215 Gever JR, Cockayne DA, Dillon MP, Burnstock G, Ford AP (2006) Pharmacology of P2X channels. Pflügers Arch 452(5):513–537 Eriksdotter-Nilsson M, Bjorklund H, Olson L (1986) Laminin immunohistochemistry: a simple method to visualize and quantitate vascular structures in the mammalian brain. J Neurosci Methods 17(4):275–286 Ji K, Tsirka SE (2012) Inflammation modulates expression of laminin in the central nervous system following ischemic injury. J Neuroinflammation 9:159 Gu BJ, Wiley JS (2006) Rapid ATP-induced release of matrix metalloproteinase 9 is mediated by the P2X7 receptor. Blood 107(12): 4946–4953 Hill LM, Gavala ML, Lenertz LY, Bertics PJ (2010) Extracellular ATP may contribute to tissue repair by rapidly stimulating purinergic receptor X7-dependent vascular endothelial growth factor release from primary human monocytes. J Immunol 185(5): 3028–3034 Rha JH, Saver JL (2007) The impact of recanalization on ischemic stroke outcome: a meta-analysis. Stroke 38(3):967–973 Tanaka Y, Fukumitsu H, Soumiya H, Yoshimura S, Iwama T, Furukawa S (2012) 2-decenoic acid ethyl ester, a compound that elicits neurotrophin-like intracellular signals, facilitating functional recovery from cerebral infarction in mice. Int J Mol Sci 13(4): 4968–4981 Davis MF, Lay C, Frostig RD (2013) Permanent cerebral vessel occlusion via double ligature and transection. J Vis Exp (77) Yeung PK, Lo AC, Leung JW, Chung SS, Chung SK (2009) Targeted overexpression of endothelin-1 in astrocytes leads to more severe cytotoxic brain edema and higher mortality. J Cereb Blood Flow Metab 29(12):1891–1902 Walberer M, Blaes F, Stolz E, Muller C, Schoenburg M, Tschernatsch M, Bachmann G, Gerriets T (2007) Midline-shift corresponds to the amount of brain edema early after hemispheric stroke – an MRI study in rats. J Neurosurg Anesthesiol 19(2):105–110 Lenglet S, Montecucco F, Denes A, Coutts G, Pinteaux E, Mach F, Schaller K, Gasche Y, Copin JC (2014) Recombinant tissue plasminogen activator enhances microglial cell recruitment after stroke in mice. J Cereb Blood Flow Metab 34(5):802–812 Franke H, Günther A, Grosche J, Schmidt R, Rossner S, Reinhardt R, Faber-Zuschratter H, Schneider D, Illes P (2004) P2X7 receptor expression after ischemia in the cerebral cortex of rats. J Neuropathol Exp Neurol 63(7):686–699 Cherry JD, Olschowka JA, O’Banion MK (2014) Neuroinflammation and M2 microglia: the good, the bad, and the inflamed. J Neuroinflammation 11:98 Zhao X, Wang H, Sun G, Zhang J, Edwards NJ, Aronowski J (2015) Neuronal interleukin-4 as a modulator of microglial pathways and ischemic brain damage. J Neurosci 35(32):11281–11291 Zanin RF, Braganhol E, Bergamin LS, Campesato LF, Filho AZ, Moreira JC, Morrone FB, Sevigny J, Schetinger MR, de Souza Wyse AT, Battastini AM (2012) Differential macrophage activation
Purinergic Signalling (2016) 12:453–463 alters the expression profile of NTPDase and ecto-5'-nucleotidase. PLoS One 7(2):e31205 36. Apolloni S, Amadio S, Montilli C, Volonte C, D’Ambrosi N (2013) Ablation of P2X7 receptor exacerbates gliosis and motoneuron death in the SOD1-G93A mouse model of amyotrophic lateral sclerosis. Hum Mol Genet 22(20):4102– 4116 37. Pedata F, Dettori I, Coppi E, Melani A, Fusco I, Corradetti R, Pugliese AM (2015) Purinergic signalling in brain ischemia. Neuropharmacology. In Press. 38. Zador Z, Stiver S, Wang V, Manley GT (2009) Role of aquaporin-4 in cerebral edema and stroke. Handb Exp Pharmacol 190:159–170 39. Lee M, Lee SJ, Choi HJ, Jung YW, Frokiaer J, Nielsen S, Kwon TH (2008) Regulation of AQP4 protein expression in
463 rat brain astrocytes: role of P2X7 receptor activation. Brain Res 1195:1–11 40. Hirayama Y, Ikeda-Matsuo Y, Notomi S, Enaida H, Kinouchi H, Koizumi S (2015) Astrocyte-mediated ischemic tolerance. J Neurosci 35(9):3794–3805 41. Franciosi S, De Gasperi R, Dickstein DL, English DF, Rocher AB, Janssen WG, Christoffel D, Sosa MA, Hof PR, Buxbaum JD, Elder GA (2007) Pepsin pretreatment allows collagen IV immunostaining of blood vessels in adult mouse brain. J Neurosci Methods 163(1): 76–82 42. Szabo A, Kalman M (2004) Disappearance of the post-lesional laminin immunopositivity of brain vessels is parallel with the formation of gliovascular junctions and common basal lamina. A double-labelling immunohistochemical study. Neuropathol Appl Neurobiol 30(2):169–177
