By continuing to browse this site you agree to us using cookies as described in About Cookies
Notice: Due to essential maintenance the subscribe/renew pages will be unavailable on Wednesday 26 October between 02:00- 08:00 BST/ 09:00 – 15:00 SGT/ 21:00- 03:00 EDT. Apologies for the inconvenience.
Numerous studies of ischemia-induced neuronal damage were primarily focused on general mechanisms of cell death or survival whereas much less attention was paid to the modifications of neuronal connectivity and synaptic morphology that could have an important impact on cell function and survival. It is known that hippocampal CA1 pyramidal neurons are selectively vulnerable to ischemia while adjacent CA3 neurons are relatively resistant. The CA3→CA1 hippocampal projection is significant both for its synaptic plasticity and for the extent of convergence and divergence, which is greater than elsewhere in the trisynaptic circuit (Shepard and Haris,1998). Hippocampal CA1 stratum radiatum excitatory spine synapses were studied here because the vast majority is CA3→CA1 type of synapses arising from ipsilateral Schaffer collaterals and contralateral commissural axons (Johnston and Amaral,1997).
A hallmark event in the early post-ischemic period is enhanced permeability of mitochondrial membranes, but the precise mechanisms by which mitochondrial function is disrupted are still unclear. Mitochondria play a critical role in the pathogenesis of cerebral ischemia. Brain mitochondria undergo ultrastructural changes after transient and during permanent cerebral ischemia. After transient focal ischemia, cortical neuronal mitochondria become injured, which is manifested by condensation, increased matrix density, and deposits of electron-dense material, finally resulting in disintegration. In contrast, permanent ischemia causes increasing loss of matrix density, associated with mitochondrial swelling, which disappears after 24 hr of focal ischemia (Solenski et al.,2002). A further response to cerebral ischemia is the permeabilization of at least the outer mitochondrial membrane. These results suit in the liberation of proapoptotic proteins such as cytochrome c, caspase 9, and second mitochondria-derived activator of caspases (Namura et al.,2001; Sugawara et al., 2002). Elevation of cytosolic calcium concentration during ischemia/reperfusion might be a signal for permeabilization of the mitochondrial membrane.
The mechanisms of ischemic neuronal injury are not yet precisely understood. Recent studies have strongly suggested that mitochondria are the major targets of ischemia-mediated adverse reactions. On the basis of several reports, oxidative phosphorylation activities of brain mitochondria markedly deteriorate due to ischemia and reperfusion, leading to irreversible cellular damage. Moreover, evidence has been presented that a mitochondrial permeability transition may be a key step in ischemic cell damage (Kobayashi et al.,2003). This involves the activation of a permeability transition pore in the inner mitochondrial membrane, which short-circuits the membrane to H+ and allows calcium to leave the mitochondria. The formation of such a megapore is typically caused by oxidative stress and mitochondrial calcium accumulation. Recent studies have shown that brain mitochondria undergo swelling and release proteins located in the matrix or the intermembrane space when exposed to various stimuli (Kobayashi et al.,2003).
The aim of our study was ultrastructural comparative analysis of the morphological organization of mitochondria in the CA1 and CA3 hippocampal pyramidal cells following transient global cerebral ischemia. Evaluation of the temporal profiles of morphological changes in presynaptic CA1 and CA3 terminals was also done with transmission electron microscopy (TEM) up to 7 days following 5-min ischemia in Mongolian gerbils.
MATERIAL AND METHODS
Adult male Mongolian gerbils (Meriones unguiculatus, 60–75 g) were submitted to 5 min duration of cerebral ischemia. Groups of four gerbils per cage (Erath, FRG), were housed in an air-conditioned room, at a temperature of 23°C ± 2°C, with 55% ± 10% humidity, and with lights on 12 hr/day (7:00–19:00). The gerbils were given commercial food and tap water ad libitum. All experimental manipulations were performed during the light phase, between 9:00 and 15:00 under identical conditions. Animals used for procedures were treated in strict accordance with the NIH Guide for Care and Use of Laboratory Animals (1985) and European Communities Council Directive (86/609/EEC), as well as with approval of the local Ethical Committee. All efforts were made to minimize animal suffering and to reduce the number of animals used.
Experimental Procedure—Occlusion of Common Carotid Arteries
Since mature gerbils lack posterior communicating arteries, that normally connects the posterior circulation of the brain from the vertebral arteries with the anterior circulation from the carotid arteries within the circle of Willis, occlusion of both common carotid arteries results, reproducibly, in global forebrain ischemia. The Mongolian gerbils were anaesthetized by ketamine (75 mg/kg body weight) and xylazine (2 mg/kg body weight) and placed in the dorsal position. The neck area was shaved then both common carotid arteries were exposed carefully by blunt dissection and clamped for 5 min with microaneurysm clips. After removing the clips, reperfusion was confirmed visually, and the skin was sutured by 3–4 loose silk stitches. For sham-operated animals, both common carotid arteries were exposed but not occluded. Post ischemic temperature was carefully monitored due to the fact that the gerbil model typically shows an intrinsic hyperthermic response during the initial hours of recirculation (Kuroiwa et al.,1990). The rectal temperature of animals was carefully monitored and maintained at 38°C by an external heating lamp. Since the changes in body temperatures are known to have an impact on the consequences of global ischemia, it was maintained at 37°C ± 0.3°C throughout the surgical procedure by a feedback-controlled heating pad (TR-100, PS-100, Fine Science Tool, North Vancouver, Canada). Gerbils were allowed to recover in their home cages for 2 hr under a Xenon heating lamp and then returned to animal quarters.
The gerbils were divided into experimental groups (N = 6 per group). Control groups were intact and sham-operated whereas the treatment group was submitted to 5-min ischemia. Intact gerbils were not submitted to any type of surgical procedure and served as a control for operation stress whereas sham-operated gerbils were exposed to the same surgical intervention as ischemic gerbils, but without occlusion of both common carotid arteries. The animals were allowed to survive for 4 and 7 days after occlusion.
Light and Electron Microscopy
Animals were fixed by transcardiac perfusion with 4% paraformaldehyde and 2.5% glutaraldehyde in 0.1 M phosphate buffer (pH 7.4) under deep anesthesia. Brains were removed from the skull. The hippocampus was isolated and cut into 500-μm thick transverse slices which were post-fixed in the same fixative for 1.5 hr and in 1% OsO4 for 1 hr. Tissue slices were then dehydrated in ascending series of ethanol followed by absolute alcohol and embedded in EPON resin. Toluidine blue-stained semi-thin (1 μ) sections of perfusion-fixed brains were used to localize hippocampal CA1 and CA3 areas (Fig. 1A). For electron microscopy, ultra-thin sections (70 nm) from the middle portion of CA1 stratum radiatum and CA3 were stained with uranyl acetate and lead citrate and examined using either JEM-100CX (Jeol, Japan), or Tecnai G212 transmission electron microscope (TEM—FEI Company, Eindhoven, Netherlands) equipped with a digital camera Mega View III Soft Imaging Systems (Munster, Germany). The photos taken in JEM-100CX electron microscope at a magnification of ×10,000 were scanned with a flatbed scanner at 600 dpi and 256 gray levels (ScanMarker i900, MICROTEK). The images in Tecnai G212 electron microscope were taken at a magnification of ×24,000.
Morphometric Analysis of Mitochondria
We performed morphometric analysis of mitochondria in total number of electron microscopic (EM) images per exponential group ranging from 71 to 170 (Table 1, Data, n EM images). From each obtained EM image, we analyzed following morphometric parameters: (1) Density of presynaptic terminals—Estimation of synapse density was carried out on each EM image by counting all asymmetric spine synapses (determined by the presence of a spine head with prominent postsynaptic density or elements of spine apparatus and docked synaptic vesicles in the active zone of a presynaptic terminal; Fig. 1D) on the micrograph surface limited by the counting frame (36 μm2 test area) followed by calculation of the number of synapses per unit area. For each experimental group, the total area of 1.080–1.440 μm2 was analyzed. All other parameters were calculated in a similar manner. (2) Mitochondrial frequency in CA1/CA3 terminals was calculated from following parameters: Number of “empty” terminals—terminals without mitochondria profile, number of terminals with one mitochondrion, and number of terminals with two or more mitochondria (Figs. 2, 3). (3) Estimation of a volume fraction or volume of mitochondria per volume of presynaptic terminal: We performed a morphometric evaluation of mitochondria volume in the each terminal using COREL Photo Paint on digitized micrographs, using the grid and the point counting technique (Fig. 4A). Point counting was used to estimate areas, which were themselves estimators of volume. (4) Frequency of damaged mitochondria was evaluated by the ratio between intact, “normal” (healthy looking mitochondria) and damaged mitochondria (Figs. 5, 6). For the present study, we have used following criteria to classify mitochondrial damage: mitochondria were either normal or damaged. Damaged mitochondria included any type of ultrastructural changes (broken cristae, membrane disruptions, electron-dense mitochondria, intra-mitochondrial edema, hypertrophic “giant” mitochondria, partially or completely damaged mitochondria) indicating a spectrum of degeneration from normal to damaged mitochondria (previously described by Aliev et al.,2009).
Table 1. Quantitative parameters of the comparative ultrastructural analysis of the mitochondria in the CA1 and CA3 hippocampal pyramidal cells following global ischemia in Mongolian gerbils
DAR - days after reperfusion.
n EM images
Σ empty terminals/percentage
Ratio mitochondria per terminal ± SEM
0.67 ± 0.10
0.50 ± 0.03
0.54 ± 0.08
0.63 ± 0.07
1.04 ± 0.08
1.54 ± 0.10
0 mit per terminal ± SEM
3.37 ± 0.20
5.14 ± 0.21
5.32 ± 0.28
3.19 ± 0.24
3.19 ± 0.23
1.93 ± 0.18
1 mit per terminal ± SEM
1.18 ± 0.10
2.65 ± 0.12
2.68 ± 0.23
1.11 ± 0.13
2.54 ± 0.15
1.28 ± 0.14
> 1 mit per terminal ± SEM
0.34 ± 0.05
0.47 ± 0.07
0.49 ± 0.09
0.72 ± 0.11
1.56 ± 0.11
1.66 ± 0.12
Mean number of terminals ± SEM
4.89 ± 0.24
4.25 ± 0.23
3.49 ± 0.42
5.02 ± 0.29
4.86 ± 0.28
4.29 ± 0.20
Σ “good;” mitochondria/percentage
Σ “damaged” mitochondria/percentage
Statistical analysis was performed using Statistic software (version 5, StatSoft, Tulsa, OK). Values are shown as mean ± standard error of the mean (SEM). The two-tailed Kolmogorov–Smirnov test was used to assess the differences between samples (P < 0.05 was considered to indicate statistical significance).
Cerebral ischemia leads to diverse structural changes in neural cells. Our early EM studies showed transient mitochondrial swelling, desegregation of polyribosomes, decreases in rough endoplasmic reticulum and Golgi apparatus in post-ischemic hippocampal neurons (Nikonenko et al.,2009).
For this study, we employed transmission electron microscopic technique (TEM) to examine ischemia-induced effects on the mitochondrial morphological organization within the presynaptic terminals. We analyzed mitochondria in control (sham-operated) and post-ischemic asymmetric synapses in the gerbil's hippocampal CA1 stratum radiatum and CA3 areas (Fig. 1). At the ultrastructural level, we distinguished excitatory or asymmetric synapses, from inhibitory or symmetric ones by their prominent PSD, widened synaptic cleft and round synaptic vesicles of larger size in the presynaptic terminals (Peters et al.,1991; Kovalenko et al.,2006; Fig. 1D). In the middle portion of the CA1 stratum radiatum analyzed in the study, the great majority of synaptic inputs is excitatory whereas the proportion of inhibitory inputs is low (˜3%), and they preferentially terminate on the dendritic shaft (Megias et al.,2001; Kovalenko et al.,2006; Fig. 1B,C). Therefore, we could focus our analyses on the pool of excitatory CA1 spine synapses as well as CA3 terminals. Our quantitative parameters are presented in Table 1.
Ultrastructural analysis of mitochondrial frequency in CA1 hippocampal terminals had shown significantly increased number of empty terminals (terminals without mitochondria, Fig. 2A) and terminals with one mitochondrion (Fig. 2B) compared to control, both 4 and 7 days after reperfusion (Fig. 2D, indicated by * symbol). But, comparative analysis between results obtained 4 and 7 days after occlusion revealed no statistically significant differences in CA1 area (2D, gray and black bars). Ultrastructural analysis performed in CA3 area had reveled significantly increased number of terminals with one and more than one mitochondrion (Fig. 2C) 4 days after occlusion compared to control (Fig. 2E, gray bars). On the other hand, 7 days after occlusion it was found decreased number of empty terminals and increased number of terminals with more than one mitochondrion compared to control (Fig. 2E, black bars). Comparative ultrastructural analysis between results obtained 4 and 7 days after occlusion revealed difference in the number of empty terminals and terminals with one mitochondrion (Fig. 2E, indicated by ▴ symbol). Finally, comparative analysis of mitochondrial frequency between terminals in CA1 and CA3 area had shown statistically significant differences in all measured parameters (Fig. 2D,E, indicated by # symbol).
From the obtained parameters we calculated mean number of mitochondria per terminal (Fig. 3). In CA1 area, post-ischemic decrease in the number of mitochondria per terminal was found (Fig. 3A). On the other hand, significant increase in the number of mitochondria per terminal was found in CA3 area (Fig. 3B,C). This finally resulted in a significant difference between the measured parameter in CA1 and CA3 area in post-ischemic, as well as control conditions (Fig. 3D, indicated by # symbol).
Morphometric evaluation of mitochondria volume fraction in the single terminal (Fig. 4A) in CA1 hippocampal area had showed a significant increase compared to control both 4 and 7 days after reperfusion (Fig. 4, indicated by * symbol). Again, our analysis revealed a significant difference in the mitochondrial volume fraction in presynaptic CA1 and CA3 terminals in post-ischemic, as well as control conditions (Fig. 4B, indicated by # symbol). Mitochondrial swelling is one of the initial post-ischemic changes. In mild cases swollen mitochondria regain their normal shape soon. In severe cases, these organelles demonstrate condensation, increased matrix density, and deposits of an electron-dense material followed by the disintegration of mitochondria (Solenski et al.,2002; Kovalenko et al.,2006; Fig. 5).
We also analyzed mitochondrial post-ischemic damage (Fig. 5) 4 and 7 days after reperfusion in presynaptic CA1 and CA3 terminals (Fig. 6). Post-ischemic conditions compromised the function and structural integrity of mitochondria, thereby contributing to cerebral metabolic dysfunction and cell death. Mitochondria were assessed with respect to the ultrastructure: rupture of mitochondrial outer membrane, swelling of mitochondria, distention of mitochondria, the inner matrix of the slightly swollen mitochondria showed increased electron density, etc. (Aliev et al.,2009; Fig. 5). We found post-ischemic vacuolization of mitochondria, endoplasmic reticulum and Golgi apparatus, as well as fragmentation and disintegration of neurofilaments which followed later. Our ultrastructural study revealed a difference in control conditions—we found more damaged mitochondria in CA3 terminals than in CA1 terminals. On the other hand, in post-ischemic conditions it was more damaged mitochondria in CA1 terminals compared to CA3 terminals (Fig. 6).
This study characterized for the first time the quantitative and qualitative pattern of mitochondria organization in CA1/CA3 hippocampal neurons affected by ischemia. Our results demonstrated that ultrastructural morphometric analysis provides criteria for documenting post-ischemic mitochondrial damage.
An important information about the timing and early events of the ischemia-induced cell damage can be provided by morphological examination of the impaired tissue. The post-ischemic changes included ultrastructural modifications within a presynaptic terminal, which could contribute to alterations of synaptic transmission were reported in ischemia (Jourdain et al.,2002; Nikonenko et al.,2003; Briones et al.,2005; Kovalenko et al.,2006). As synaptic modifications appeared shortly after the ischemic insult, well before neuronal degeneration, they could be involved in the mechanisms leading to cell damage and death. Delayed ischemic brain damage is associated with mitochondrial dysfunction, but the underlying mechanisms are not known in detail. Recent data suggest that the process is associated with multidirectional changes in the activities of various proteins located in mitochondria (Broughton et al.,2009).
Synaptic remodeling takes place in hippocampal tissue early after ischemia. A short ischemic episode induces a progressive decrease in the numbers of synaptic terminals in the rat CA1 area exceeding 30%—one day, and 65%—seven days after the occlusion. Nearly half of the remaining synapses display degenerative features (Kovalenko et al.,2006). In addition, there is a rapid ischemia-related increase in the number of perforated synapses and multiple synapse boutons (Kovalenko et al.,2006). Perforations are thought to correlate with reactive synaptogenesis, supporting a suggestion that few remaining synapses increase their efficacy in response to ischemia (Jourdain et al.,2002).
Our comparative analysis of excitatory spine synapses in CA1 and CA3 hippocampal areas revealed a significant difference in the ultrastructural organization of mitochondria in control conditions as well as following global cerebral ischemia. Mitochondrial frequency and the mean number of mitochondria per terminal in these hippocampal areas are different which is closely connected with the function of pyramidal neurons in control conditions and with the sensitivity to ischemic insult. Distinct populations of hippocampal neurons demonstrate a differential vulnerability to ischemia (Schmidt-Kastner and Freund,1991). These data suggest a link between vulnerability and synaptic plasticity. The metabolic correlate of this synaptic plasticity is increased number of mitochondria at synaptic terminals.
In this study, we have shown pronounced post-ischemic damage of mitochondrial membranes, which is more evident in the synaptic terminals of hippocampal CA1 area compared to the terminals in CA3 area. Disorganization of mitochondrial membranes is a first step of neuronal cells apoptosis. Mitochondrial released cytochrome c combines with apoptosis protease activation factor-1, procaspase-9, and ATP in the cytosol, producing active caspase-9, the activation of this initiator caspase then leads to the proteolytic activation of caspase-3, the primary effectors caspase of the cell. This pathway often involves the release of additional mitochondrial proteins such as a second mitochondrial-derived activator of caspase/direct IAP-associated binding protein with low PI (Smac/DIABLO) and HtrA2/Omi, which antagonize inhibitor of apoptosis (IAP) proteins, and apoptosis inducing factor and Endonuclease G, which contribute to DNA fragmentation. A second extrinsic pathway of caspase activation exists, where the binding of a death ligand such as Fas to it death receptor triggers auto-processing of caspase-8 to its active form, which then directly activates caspase-3. The proapoptotic Bcl-2 family protein Bid bridges the two pathways by translocation to mitochondria and releasing cytochrome c after truncation by caspase-8 (Polster et al.,2004; Broughton et al.,2009).
Our ultrastructural study has shown that transient global cerebral ischemia induces a dramatic swelling of the mitochondria evidenced by the changes in the volume fraction parameter. These changes are more obvious in the synaptic terminals of hippocampal CA1 area compared to the terminals in CA3 area. Mitochondria undergo swelling following exposure to high calcium and oxidative stress. The swelling results from the opening of the mitochondrial permeability transition pore in the inner membrane. Solute entry precipitates water influx, leading to expansion of the matrix space within the highly convoluted inner membrane and dissipation of the electrochemical gradient. The comparatively rigid outer mitochondrial membrane eventually ruptures, releasing the contents of the intermembrane space including cytochrome c (Polster et al.,2004).
Mitochondrial damage and neuronal death are close connected with ion homeostasis. The loss of cellular high-energy compounds during ischemia causing the loss of the Na+/K+ gradient, virtually eliminates three of the four mechanisms of cellular calcium homeostasis. This, in turn, causes a massive and rapid influx of calcium into the cell. Mitochondrial sequestration, the remaining mechanism, causes overloading of the mitochondria with calcium and diminished capacity for oxidative phosphorylation. Elevated intracellular calcium activates membrane phospholipases and protein kinases. A consequence of phospholipase activation is the production of free fatty acids, including the potent prostaglandin inducer, arachidonic acid. The degradation of the membrane by phospholipases damages membrane integrity, further reducing the efficiency of calcium pumping and leads to a further calcium overload and a failure to regulate intracellular calcium levels following the ischemic episode (Niizuma et al.,2009).
Our results also revealed temporal differences in the ultrastructural mitochondrial organization in CA1 and CA3 hippocampal areas 4 and 7 days after reperfusion. This could be explained by the fact that re-oxygenation also restores ATP levels, and this may in turn allow active uptake of calcium by the mitochondria, resulting in massive calcium overload and destruction of the mitochondria (Niiyama et al.,2005).
Taken together, the present study indicates novelty that the ultrastructural morphological alteration of mitochondria in CA1/CA3 synapses is one of the post-ischemic changes could be an indicator of hippocampal metabolic dysfunction and synaptic plasticity. We found different structural organization of mitochondria—the number of empty terminals and terminals with one mitochondrion increased more in CA1 area in the period of reperfusion. On the other hand, the number of terminals with two and more mitochondria increased in the same period more evident in CA3 area. Also, the mean number of mitochondria per terminal was significantly higher in CA3 area comparing with CA1. Our suggestion is that this quantitative difference might be considered as sign of neuroprotection and that increased number of mitochondria in CA3 area might defend the terminals from post-ischemic damage. Moreover, the volume fraction of mitochondria in CA1/CA3 terminals was different suggesting that these structures are more functionally active in post-ischemic period. In these two different synaptic terminals, the activity of the mitochondria, structures that provides energy necessary for various cellular activities, may play different role in post-ischemic phenomenon.
Several of morphological changes have been reported in association with synaptic plasticity and might relate to mechanisms activated under both physiological and pathological conditions. The parameters of synaptic contact, such as its dimensions, curvature and shape, size of presynaptic terminal, synaptic vesicles density and distribution, are all related to the efficacy of the synaptic transmission (Kovalenko et al.,2006).
The investigation of structural aspects of ischemic injury in the hippocampus may provide insight into the pathogenetic mechanisms and help to develop new therapeutic strategies.