| Adult neurogenesis is a process in which new neurons develop and integrate into the existing neural circuits after embryonic and early postnatal development. Two regions including the olfactory bulb and the hippocampus in the mammalian brain display active adult neurogenesis throughout life. Adult neurogenesis in the hippocampus plays important roles in learning, memory, forgetting, and emotion. Evidence has emerged that adult hippocampal neurogenesis is altered by both physiological factors and pathological conditions.Epilepsy is a common neurological disease, characterized by abnormal neuronal discharge. Temporal lobe epilepsy is the most common type of epilepsy in adult patients; among them,30%-40% of patients cannot get seizure-free by taking currently available anticonvulsant medications. Seizure activity, especially the status epilepticus (SE) transiently enhances hippocampal neurogenesis. Surviving newborn granule cells mature and integrate into the pre-existing neural circuits. According to the locations to which newborn granule cells migrate, granule cells born after SE can be divided into normotopic and ectopic populations. Most ectopically located granule cells are in the hilus, but a small portion of them are present in the molecular layer. Analyzing hilar ectopic granule cells has revealed that not all, but a certain portion of them differ morphologically and electrophysiologically from normotopic granule cells in ways that suggest ectopic granule cells born after SE contribute to the development of epilepsy.Newborn granule cells that migrate to the granule cell layer after SE are far more numerous than ectopic granule cells. Morphological, electrophysiological, and functional studies aimed at exploring the influence of normotopic newborn granule cells in the development of epilepsy have reached somewhat conflicting conclusions. Although morphological studies done in different rodent epilepsy models have found that normotopic granule cells born after SE are morphologically different from naturally born granule cells in several respects, electrophysiological studies have not proven that normotopic newborn granule cells are more excitable than naturally born, age-matched granule cells.By applying retroviral vector-mediated newborn cell labeling, we compared mature granule cells born 5 days after SE to those born naturally with respect to dendritic morphology, intrinsic membrane properties, synaptic connectivity, and magnitude of cellular activity in the resting condition and after being activated by transient seizure activity. The magnitude of cellular activity was indicated by the expression of activity-regulated cytoskeleton-associated protein (Arc), which has been widely used as a marker of neuronal activation.Main methods1. Induction of status epilepticus (SE)Experiments were done in male Sprague-Dawley (SD) rats weighing between 150 and 175 grams. Thirty min after pretreatment with an intraperitoneal injection of scopolamine methyl bromide (2 mg/kg) and terbutaline hemisulfate (2 mg/kg),340 mg/kg of pilocarpine hydrochloride was intraperitoneally injected for induction of SE. Animals were chosen for further studies if the continuous seizure activity was at Racine stage 2 or above and lasted for at least 2 hours. If seizure activity was not initiated until 30 min after the initial pilocarpine dose, an additional dose of 100 mg/kg was given. Controls animals received equal amount of normal saline.2. Construction of CAG-GFP retroviral vectorTwo packing cell lines coupled with plasmids pCAG-GFP and pCMV-VSV-G were used to construct the CAG-GFP retroviral vector. First, Platinum-GP cells were co-transfected by pCAG-GFP and pCMV-VSV-G. Supernatants collected at 48 h and 72 h were centrifuged to yield a virus-containing solution which was used to infect platinum-E cells for generating a stable virus-producing cell line. The virus-producing platinum-E cells were frozen at -80℃ and concentrated virus solution was then obtained from the stable virus-producing cells through two-step ultraspeed centrifugation.3. Intradentate injection of CAG-GFP retroviral vectorFive days after SE induction or saline injection, animals were anesthetized with sodium pentobarbital (25 mg/kg) and ketamine (60 mg/kg) and were individually placed on a stereotaxic frame suitable for rats. For each rat,1μl of retroviral vector was injected into the right dentate gyrus at a rate of 0.1μl/min.4. Transcardial perfusion and tissue preparationAt least 10 weeks after the intradentate injection of retroviral vector, animals were anesthetized with sodium pentobarbital. After complete paralysis, transcardial perfusion was done with 2% paraformaldehyde in phosphate-buffered saline (PBS). The brain was removed from the skull upon completion of transcardial perfusion. The tissue block was post-fixed in the same fixative at 4℃ overnight. Thereafter, the tissue block was sequentially immersed in gradually increased concentration of sucrose in 0.1 M PBS overnight at 4℃. After the tissue block was firmly embedded with a medium, it was coronally cut through the hippocampi into 100 μm or 40μm in thickness with a vibratome5. Dendritic complexity analyses and quantification of dendritic spines in newborn granule cellsThe complexity of dendritic trees in newborn granule cells labeled by GFP was quantitated by Sholl analysis. Sections of 100μm in thickness were prepared and the GFP signals were immunofluorescently amplified to prevent rapid bleaching. Z-series stack of 2μm-thick was taken in GFP-positive cells locating in the suprapyramidal blade with a 20X objective (Zoom= 1). For each rat,3 to 4 cells were scanned. Only neurons with intact dendritic arborization were analyzed. After GFP-positive soma and apical dendrites were traced under ImageJ with the NeuronJ plugin, Sholl analysis was carried out by using ImageJ with the "Sholl analysis" plugin. The interval between concentric circles was 25μm with the center point at the soma.The immunofluorescently amplified sections were also used for quantitative analysis of dendritic spines. For each rat,3 newborn granule cells located in the suprapyramidal blade were chosen for this analysis, and total of 6 rats were used each group. Z-series stacks were made in 9 dendritic segments (approximately 50μm each) each neuron with a 63x objective; three were in the middle molecular layer and another three were in the outer molecular layer. Other parameters set for z-stacks were as follows:Zoom 4,0.12μm of z-series thickness. A 3-dimension image was reconstructed from individual z-stack files. Spines in each dendritic segment were marked and the numbers of mushroom-like spines and the rest were counted by using ImageJ. The densities of total and mushroom-like spines were calculated by dividing the numbers of spines with the length of dendritic segment. A mushroom spine was defined if the diameter of the head was greater than the width of the neck.6. Patch clamp analysis of GFP-labeled newborn granule cells6.1 Acute hippocampal slice preparationHippocampal slices were prepared at least 10 weeks after intradentate injection of rctroviral vector. Animals were individually anesthetized with ether and then decapitated. After removal, the brain was immersed in artificial cerebrospinal fluid (ACSF) equilibrated with 95% O2/5% CO2 at 6℃. The brain was then cut through the midline. The brain block containing the right hippocampus was cut into 410μm-thick coronal slices with avibratome. The sections were then incubated in a 95%O2/5%CO2-saturated ACSF as mentioned above at 34.5℃ for 30 min. Thereafter, sections were stored in the standard ACSF equilibrated with 95% O2/5% CO2 at room temperature.6.2 Slice perfusion, cell visualization and patch clamp recordingEach slice was transferred to a RC-26 recording chamber and continuously perfused with 95% O2/5% CO2-equilibrated standard ACSF (2.5 ml/min) at room temperature. Slices were visualized with an Olympus BX51WI fluorescence microscope, equipped with the far infrared-differential interference contrast (DIC) optics, a charge-coupled device camera, and a 40X water-immersion objective. After a GFP-positive granule cell was fluorescently identified, the cell was switched to DIC visualization and patch-recorded. Parameters recorded were spontaneous and evoked firing and miniature post-synaptic currents.6.3 Spontaneous and evoked firingTo record spontaneous and evoked firing, we used a potassium gluconate-based internal solution. After a high resistance seal (>2 GΩ) was formed, spontaneous firing was recorded for 5 min in cell-attached mode by holding the cell at-70 mV. Whole cell recording was then achieved through membrane rupture and the resting membrane potential was recorded. Membrane time constant, input resistance, and membrane capacitance were obtained from the current deflection in response to a 10 mV step hyperpolarization from the resting membrane potential for 500 ms. At least 5 min after membrane rupture, cellular firing pattern was determined by stepped current injection. The threshold potential, amplitude, width, amplitude of the after depolarization (ADP) and the ADP area of action potentials were calculated by analyzing the first action potential evoked by current injection. A burst firing was defined as the spikes appeared in trains containing not one but two or more spikes. A liquid junction potential of 10 mV was corrected.6.4 Recording miniature inhibitory and excitatory post-synaptic currentsTo record miniature synaptic currents, GFP-positive cells were patched and currents were recorded at-70 mV in the whole-cell mode at room temperature. The recording electrode was filled with a cesium chloride-based internal solution.Miniature inhibitory postsynaptic current (mIPSC) were recorded in the presence of 1μM tetrodotoxin (TTX),10μM 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX), and 50 μM D-2-amino-5-phosphonopentanoate (D-AP5). Miniature excitatory postynaptic currents (mEPSCs) were recorded in the presence of 1μM tetrodotoxin (TTX) and 30 μM bicuculline methiodide. D-AP5, CNQX and TTX were purchased from Tocris Bioscience. In each animal, only one cell was recorded. Recordings were made for 2.5 min and miniature post-synaptic current events were manually analyzed by using MiniAnalysis, as described previously. Between the control and SE rats, the amplitude, frequency,10-90% rise time, decay time, and area of current events were statistically compared.7. Activity of newborn granule cells following pilocarpine-induced status epilepticusBecause enhanced neuronal activity is associated with an increased Arc expression at both mRNA and protein levels, Arc immunoreactivity was used to indicate the activation of newborn granule cells after pilocarpine-induced status epilepticus in the resting condition and in response to a transient seizure episode. At the 5th day after pilocarpine (SE) or saline injection (control),1μl of CAG-GFP retroviral vector was injected into the dentate gyrus to label SE-induced or naturally born granule cells with a protocol as mentioned above. At least 70 days after viral vector injection, pentylenetetrazol (PTZ) at a dose of 20 mg/kg was intraperitoneally injected to SE (PTZ-treated SEs) or control (PTZ-treated controls) rats. PTZ-untreated SE and PTZ-untreated control rats received an equal volume of saline. Diazepam at a dose of 5mg/kg was intraperitoneally injected into rats 15 min after PTZ treatment or saline injection. Two hours after the administration of diazepam, rats were transcardially perfused with 2% paraformaldehyde in 0.1 M PBS. Arc was immunofluorescently stained in coronally cut sections (40μm in thickness) in which at least one GFP-expressing soma was included. Sections were scanned by using a Zeiss LSM 780 confocal microscope equipped with a 40×objective at a Zoom 1. The relative intensities of Arc immunoreactivity were measured in the GFP-positive soma and surrounding 10 GFP-negative granule cell-like somata by using ImageJ. The intensity of Arc immunoreactivity in the GFP-positive soma was normalized to the averaged Arc immunoreactivity in 10 surrounding GFP-negative granule cells. Main results1. Effectiveness of cell labeling with CAG-GFP retroviral vector in the rat hippocampal dentate gyrus More than 10 weeks after viral vector injection, GFP-expressing somata with processes could be seen in both the suprapyramidal and infrapyramidal blades of control or SE rats. These cells were scattered along the granule cell layer-hilus border. Occasionally, cells with a single basal dendrite that extended into the hilus were noticed; more often they were seen in SE rats.2. Dendritic complexity and spine density of mature granule cells born after status epilepticus The apical dendrite of granule cells born in a control rat had 4-5 branch orders and their distal branches always reached the outer molecular layer. The apical dendrites of granule cells born after SE had very similar dendritic branch orders and arborizations also extended into the outer molecular layer. Statistical analysis revealed that the dendritic branching in the mature granule cells born after SE (n= 8 rats) was not significantly different from those granule cells born naturally (n= 8 rats); however, the segment close to the soma appeared to be less branched in the SE rats.For the dendritic segments located in the middle molecular layer, both total and mushroom-like spine densities were not statistically different between the controls (n = 6) and SEs (n= 6). However, mushroom-like spine density in the dendritic segments located in the outer molecular layer in SE rats (n= 6) was significantly denser than that in the control rats (n= 6).3. Intrinsic membrane properties and firing characteristics of mature granule cells born after status epilepticusIn cell-attached mode at near resting membrane potential, none of the newborn granule cells was found to have persistent spontaneous firing; the percentages of newborn granule cells with occasional spontaneous firing are 11.8%(2/17) in the control group and 17.6%(3/17) in the SE group (n= 17) (P> 0.1, by Chi-square test). After membrane rupture, resting membrane potentials were -76.0±1.1 mV in the controls (n= 6), and -75.8 mV±0.9 mV in the SE rats (n= 6), respectively. The threshold potentials for firing were not significantly different between the control and SE groups. Action potential parameters, including the amplitude, width, ADP, and area of ADP were not statistically different between the control (n= 6) and SE (n= 6) groups.4. Miniature inhibitory and excitatory postsynaptic currentsBy holding cells at-70 mV, both mIPSCs and mEPSCs were recorded at room temperature. It appeared that mIPSCs recorded from newborn granule cells in the SE group were similar to those in the control group; statistical analyses revealed that neither frequency, amplitude,10-90% rise time, decay time, area, nor interval time were significantly different between the control (n= 6) and SE (n= 6) groups. For mEPSCs, the amplitude in the SE group (n= 6) was significantly smaller than that in the control group (n= 6). In addition, there was a trend toward lower event frequency in the SE group5. Cellular activity of mature granule cells born after status epileticus measured by Arc expression in the resting condition or during the period of pentylenetetrazol (PTZ)-induced seizureIn PTZ-untreated controls, dentate granule cells constitutively expressed Arc at low level. Arc immunoreactivity in granule cells of PTZ-untreated SE animals appeared more intense than that of granule cells in PTZ-untreated controls; however, the intensity of Arc immunoreactivity in the GFP-labeled newborn cells did not exceed that of nearby GFP-negative granule cells. Treatment of controls with PTZ dramatically enhanced Arc immunoreactivity in the soma and apical dendrites of almost every granule cell, but again, a difference between GFP-labeled granule cells and granule cells surrounding them could not be noticed. Treatment of SE rats with PTZ also increased Arc immunoreactivity in both the soma and apical dendrites of almost all granule cells and it appeared that the intensity of Arc immunoreactivity in GFP-labeled newborn granule cells was similar to that in adjacent GFP-negative granule cells. Quantitative analyses that compared Arc immunoreactivity in GFP-labeled cells to the averaged Arc immunoreactivity of surrounding granule cells clearly shows that the intensities of Arc immunoreactivity in GFP-positive somata and dendrites were not different from surrounding granule cells in the resting condition or during a transient seizure episode.ConclusionIn conclusion, upon maturation, granule cells born after pilocarpine-induced status epilepticus do not appear hyperactive in comparison with those born in physiological conditions. Without continuous abnormal neurogenesis, the transient enhancement of neurogenesis would not sustain hyperactivity. |
PDF Full Text Request