InVivo and InVitro Effects of Pilocarpine: Relevance to Ictogenesis

EEG recording and video monitoring

With the exception of whole brain guinea pig experiments that were performed at the Neurological Institute Besta (Milan), all the experiments were carried out at the Cleveland Clinic, Lerner Research Institute. Stereotactic electrode implantation was performed under pentobarbital anaesthesia (45 mg/kg i.p.), using the Kopf stereotactic frame and a stereotactic atlas of the rat brain. For intrahippocampal recordings, bipolar twisted stainless wire electrodes (0.1 mm diameter, 0.5 mm vertical tip separation, Medwire, New York, New York, U.S.A.) are placed bilaterally in the dorsal hippocampus (LH: left hippocampus, RH: right hippocampus, A: −3.9 mm, L: ± 3.0 mm, D: −2.5 mm from bregma). Stainless steel screws (MX-0090–2, Small Parts Inc., Miami, FI, U.S.A.) were placed bilaterally on the dura mater of the frontal cortex (LC: left frontal cortex, RC: right frontal cortex, A: 1.0 mm, L: ± 2.5 mm from bregma). An additional screw electrode is placed in the frontal sinus and as a referential recording electrode. Rats were left unrestrained for 72 h for recovery from surgery before prolonged EEG recordings are performed. Each rat was kept in separate cage, under 12-h dark–light cycles, with free access to food and water. Digital EEG recordings (five channels per rat) were performed using two Vangard systems (Lamont, Madison, WI, U.S.A.). Each system consists of a recording Hewlett-Packard workstation and a similar review system connected to two 21-inch monitors. EEG data were sampled a rate of 100 Hz.

Animals and tissue sample preparation

Rats were housed in a controlled environment (21 ± 1°C; humidity 60%; lights on 08:00 a.m. – 8:00 p.m.; food and water available ad libitum). Procedures involving animals and their care were conducted in conformity with the institutional guidelines that are in compliance with international laws and policies (EEC Council Directive 86/609, OJ L 358, 1, Dec. 12, 1987; Guide for the Care and Use of Laboratory Animals, U. S. National Research Council, 1996). Rats (male Sprague-Dawley 200–250 g) were injected with methylscopolamine (1 mg/kg, i.p., Sigma-Aldrich, St. Louis, Mo, U.S.A.) 30 min before pilocarpine (350 mg/kg i.p.) to prevent peripheral cholinergic side effects. Animals (n = 6) were anesthetized (isofluorane) during recurrent motor seizures and at onset of SE as assessed by behavioral observation (stages 4 and 5 of Racine’s scale) and EEG recordings. Blood was withdrawn from the right heart ventricles; brain areas (hippocampus and entorhinal cortex) were resected and frozen. Serum samples were obtained after centrifugation at 1500×g for 10 min at 4 °C.

HPLC

Chemicals and reagents: pilocarpine hydrochloride was purchased from Sigma Aldrich. HPLC grade acetonitrile was purchased from Merck (Darmstadt, Germany). Milli-Q water (Millipore, Bedford, MA, U.S.A.) was used throughout. The other chemicals used were of analytical reagent grade. Chromatographic system and conditions: The chromatographic analysis was carried out according to a method previously described by Kuks et al. (1990). The HPLC system used was an Agilent 1100 s HPLC (Agilent, Santa Clara, CA, U.S.A.) consisting of a quaternary pump and a Diode Array Detector (DAD) set at wavelength of 220 nm. The analytical column was a microbondapak C18 3.9^*300 mm from Waters (Milford, MA, U.S.A.). The system was run on isocratic gradient utilizing a mobile phase of 0.6% triethylamine in water (adjusted to pH 2.3 with 85% phosphoric acid) with a run time of 20 min. The chromatographic analysis was performed at room temperature with flow-rate of 1.0 ml/min and injection volume of 5 μl. Data collection and integration were performed by Agilent Chemstation Software.

Standard samples and calibration curves

Stock solutions (10 mM) of pilocarpine hydrochloride were freshly prepared in water and working solutions (0.1 nM–10 mM) were prepared by diluting stock solutions with water.

FACS and cytokine ELISA

Blood samples were withdrawn before development of SE (n = 6). For fluorescent-activated cell sorting (FACS) analysis, 150 μL of whole blood (or brain homogenized) were incubated with a combination of specific antibodies recognizing different population of WBC. In particular mouse anti-rat CD8 a-FITC, CD3-R-PE and CD4-PE-Cy5 (BD Pharmingen, San Diego, CA, U.S.A.) were added (2 μL, 5 μL, and 5 μL, respectively) to blood samples. Mouse anti-rat CD11b-FITC, CD3-R-PE, and CD45-PE-Cy5 were added (2μL, 5μL, and 5μL, respectively) for the determination of neutrophils/monocytes. Blood samples were analyzed by the FACS-Core facility at LRI. IL-1β, TNF-α, and IL-6 ELISA–tests (rats n = 6 control, n = 6 Stage IV, n = 6 SE) were purchased from Pierce Biotechnology Inc (Rockford, IL, U.S.A.) and performed as described by the vendor.

Slice electrophysiology

We used a total of 40 viable slices obtained from 10 rats. Slices are judged morphologically by looking at swollen cells and recovery of the same over time, and by detecting extracellularly evoked field potential recordings in response to stimulation of white matter. In most of the experiments performed for the purposes described here, CA1 activation was used (Janigro et al., 1997; D’Ambrosio et al., 1999). Rats were deeply anesthetized with isoflurane and then decapitated. The brain was quickly removed from the skull and immediately placed into ice-cold oxygenated (95% O₂; 5% CO₂) cutting solution, composed in mM of 2.5 KCl, 28 NaHCO₃, 1.25 NaHPO₄, 7 Dextrose, 7 MgCl₂, 0.5 CaCl₂, 235 sucrose, 1 ascorbic acid, 3 Na-Pyruvate. The brainstem and cerebellum were dissected out. The central section of the cerebral hemisphere containing somatosensory cortical regions and dorsal or dorsoventral hippocampi was subsequently cut in the coronal plane into slices of 450 μm thickness using a Vibratome 3000 (The Vibratome Co., St. Louis, MO, U.S.A.). Prior to recording, slices were kept at room temperature and pH 7.4 in a preincubation chamber containing oxygenated artificial cerebrospinal fluid (aCSF) of the composition (in mM) 120 NaCl, 3.1 KCl, 3 MgCl₂, 26 NaHCO3, 1.25 KH₂PO4, 10 Dextrose, 1 CaCl2, 1 ascorbic acid, 3 Na-Pyruvate. After 1 h in recovery, the slices were transferred to a chamber containing recording aCSF [(in mM) 120 NaCl, 3.1 KCl, 1 MgCl₂, 26 NaHCO₃, 1.25 KH₂PO₄, 10 Dextrose, 2 CaCl₂ and allowed to equilibrate to the temperature achieved in the recording chamber. In the experiment at 6 mM K⁺ the aCSF had the following composition: [(in mM) 120 NaCl, 4.75 KCl, 1 MgCl₂, 26 NaHCO₃, 1.25 KH₂PO₄, 10 Dextrose, 2 CaCl₂. For electrophysiological recordings, a slice was placed in a submerged recording chamber and constantly superfused with oxygenated recording aCSF at a flow rate of 2 ml/min at 32°C. Concentrations ranging from 10–500 μM of pilocarpine were dissolved into the recording aCSF and were, thus, applied to the slices. Extracellular field potentials (FP) were recorded using borosilicate glass micropipettes (0.5–2 MΩ) filled with aCSF.

In vitro guinea pig brain preparation

Brains were isolated from young adult Hartley guinea pigs (150–200 g; Charles River Laboratories, Calco, Italy) according to the standard technique described in detail elsewhere (de Curtis et al., 1998b; de Curtis et al., 1991). After induction of barbiturate anesthesia, the brain was carefully isolated and transferred to the incubation chamber. The basilary artery was cannulated with a polyethylene cannula to ensure arterial perfusion with a complex saline solution (composition: NaCl 126 mM, KCl 3 mM, KH₂PO₄ 1.2 mM, MgSO₄ 1.3 mM, CaCl₂ 2.4 mM, NaHCO₃ 26 mM, glucose 15 mM, 3% dextran MW 70,000) oxygenated with a 95% O₂; 5% CO₂ gas mixture, pH 7.3. The arterial perfusion rate was 5.5 ml/min. Brain isolation was performed at low temperature (15 °C), and brain temperature was slowly increased to and maintained at 32 °C during the experiments. Adequate measures were taken to minimize animal pain and discomfort, in accordance with the International guidelines on animal research. The experimental protocol was approved by the Ethical Committee of the Istituto Nazionale Neurologico Besta in Milano. Extracellular recordings were simultaneously performed in the piriform cortex (PC), in the lateral and medial entorhinal cortex (EC) and in the CA1 region of the hippocampus with glass micropipettes filled with a 0.9% NaCl solution (2–5 MOhm input resistance). The functional viability of these cortical regions and the preservation of the BBB in the isolated brain preparation has been extensively demonstrated (de Curtis et al., 1998), (Librizzi et al., 2001, 2006). Bipolar silver stimulating electrodes positioned on the lateral olfactory tract were utilized to evoke potentials. Data were digitized and stored on a personal computer. The software for data acquisition and analysis (ELPHO) was developed by Dr. Vadym Gnatkovsky.

FITC-albumin solution composition

A recently developed microangiographic technique based on the injection of fluorescent probes enables confocal visualization of the entire brain micro- and macrovasculature (Cavaglia et al., 2001). Simultaneous observation of functional changes in BBB permeability is also possible if appropriate microangiographic molecules are employed. To this end, fluorescent albumin was used, since albumin does not extravasate abluminally if the BBB is intact (Cavaglia et al., 2001). The fluorescent albumin solution was prepared by reconstituting 500 mg of bovine desiccate albumin-fluorescein isothiocyanate (FITC-albumin, MW 69 KD; 12 mol/mol albumin; Sigma, St. Louis, MO, U.S.A.) in 50 ml of phosphate buffered saline (0.1M PBS) lacking magnesium and calcium ions. The FITC solution was stirred at room temperature in the dark for 5 min prior to injection. FITC albumin was directly infused in the left ventricle (Cavaglia et al., 2001). 10 ml of solution were injected at rate of 1 ml/min under general anesthesia. Presence of vascular leakage in pilocarpine-treated and control animals was evaluated by visual inspection and by confocal-microscopy (Carvey et al., 2005; Serduc et al., 2006; Yuan et al., 2006).

BBB permeability to radiotracers

A concentrated bolus of [¹⁴C] Diazepam and [³H] Sucrose (2 μC each) also containing 1 μM pilocarpine was injected into the left ventricle of an anesthetized animal that had been prepared as for FITC-albumin injection. The diffusion into the brain was monitored 15 min after the injection. Blood and brain samples containing samples of [¹⁴C] sucrose or [³ H] Diazepam were introduced into vials with 4 ml of Ready Protein Beckman scintillation cocktail (Packard Ultima Gold, ECN, Costa Mesa, CA, U.S.A.). Radioactivity was counted with an LS 6500 scintillation counter (Beckman).

Endothelial cells

The isolation and culture of rat brain microvascular endothelial cells has been described (Stanness et al., 1997). Brain tissue was obtained from 2 week old male Sprague-Dawley rats (20 g; six rats/experiment) anesthetized with 1% isoflurane (Abbott Labs, Chicago, IL, U.S.A.). Brain cortical tissue was homogenized, filtered through a 147-μm filter (Sefar America, Kansas City, MO, U.S.A.) and enzymaticaly processed with protease (50 μg/ml for 30 min) (Roche Diagnostics Corp., Indianapolis, IN, U.S.A.). Dextran (15%) (Sigma U.S.A.) was used to remove fat from the tissue prior to additional 2 h digestion with collagenase (100 μg/ml). Brain microvascular endothelial cells, separated by a Percoll gradient (1,580 g) were cultured in 75-cm² tissue culture flasks precoated with fibronectin (3 μg/cm²) (Biomedical Technologies, Stoughton, MA, U.S.A.). Brain microvascular endothelial cells were characterized by the uptake of acetylated low-density lipoprotein labeled with DiI Ac-L-DL (Biomedical Technologies) and by von Willebrand factor staining, visualized by the Vector Stain Elite kit (Vector Labs, Burlingame, CA, U.S.A.). On day-7, when flasks were confluent, more than 96% of the cells were positive for DiI Ac-L-DL and von Willebrand factor. WBC were isolated as described (Krizanac-Bengez et al., 2006a, 2006b). Approximately 10⁶ cells were added to each well.

Statistical analysis was performed with aid of Origin 7.0 (MicroCal, Northampton MA, U.S.A.); data were considered to be significantly different when p < 0.05 (by ANOVA).

Penetration of pilocarpine across the blood–brain barrier

We measured pilocarpine levels in rat serum and brain tissue following a systemic injection of 350 mg/kg. This was done either acutely (e. g., 2–3 min after the injection and in absence of behavioral modifications) or in correspondence of behavioral and EEG changes typical of Stage IV. The details of the HPLC procedure and calibration methods are described in the methods. In the acutely exposed group, rats were sacrificed 3 to 5 min after pilocarpine administration and labeled as “naive” in Fig. 2A. We evaluated the permeability of the BBB to the hydrophilic molecule ¹⁴C Sucrose (m. w. 324), the lipophilic tracer ³ H-diazepam (m. w. 284), and pilocarpine (m. w. 208). Brain pilocarpine levels were significantly lower than predicted by the lipophilicity of the drug (Fig. 2A). Unexpectedly, pilocarpine’s brain levels were on average similar to sucrose, and significantly lower than those for diazepam (n = 6) in spite of the significant difference in oil:water partition coefficients (Log P: −5.19, and 1.02 for sucrose and pilocarpine, respectively); Diazepam (Log P = 2.98) was, as expected, significantly more permeant than sucrose. Note that at approximately stage IV, brain levels of pilocarpine were significantly lower than those measured in serum (Fig. 2B), suggesting low BBB permeability even in presence of obvious behavioral and EEG changes. In conclusion, these results show that pilocarpine’s penetration across the BBB was significantly lower than expected based on its structural properties (lipophilicity and low molecular weight).

In vitro effects of pilocarpine

The concentrations of pilocarpine found in rat brain at around stage IV were used in rat hippocampal slice experiments to assess their effects on CA1 pyramidal cell excitability (Fig. 3A). After establishment of a baseline field potential recording (usually 20–30 min), slices were exposed either individually or sequentially to pilocarpine concentrations ranging from 10 to 500 μM (n = 30 slices from seven rats). Specifically, 10 slices were exposed to increasing concentrations of pilocarpine (Fig. 3A). In an additional set of experiments, each slice was exposed to only one concentration of pilocarpine (Fig. 3B). A total of 20 slices (four slices each concentration) was used. Overall, in the 87% of slices tested, pilocarpine failed to exert any significant effect on spontaneous FP (e. g., traces in Figs. 3A and B). Seizure-like activity characterized by direct current (DC) shifts or repetitive bursting, interictal-like bursts, or depolarizing shifts were never observed regardless of the concentration of pilocarpine used. Modest increases in firing rate were observed in the remaining 13% slices. Individual slices were exposed to a single concentration of pilocarpine to rule out desensitization or habituation. Initial responses to pilocarpine were again minimal, but burst activity became more pronounced at steady state (usually 30 min of exposure, Fig. 3B). Again, no seizure-like events were detected in these experiments.

A possible explanation for pilocarpine’s lack of effect on CA1 firing may have depended on the deafferented nature of the slice preparation itself (see, for example, Avoli et al., 2000). To address this, we performed field potential recordings in the isolated guinea pig brain preparation perfused with high concentrations of pilocarpine (100–1000 μM). The effects of kainic acid, applied at 4 μM, were used as a positive control (Fig. 3C1). Note that while kainic acid produced the expected epileptiform effects on the brain, high dosages of pilocarpine produce low-voltage oscillations in the gamma range that resembled those seen in limbic structures after local application of carbachol (Fig. 3C2; (Dickson et al., 2003)). Seizure-like events were never observer under these conditions. Evoking seizures required millimolar concentrations of pilocarpine or manipulation of BBB permeability (data not shown; see (Uva et al., 2006)).

Peripheral actions of pilocarpine

Expression of muscarinic receptors in WBC (Costa et al., 1990; Rinner and Schauenstein, 1993; Cook and Persinger, 1999) may also have constituted a target for pilocarpine’s action after systemic administration. This was investigated by fluorescence-based analysis of immune cells collected from blood samples immediately after pilocarpine injection, before or during SE (Fig. 4). Note that, a significant decrease of CD4-expressing cells was observed shortly (within 30 min) after pilocarpine injection and prior to SE. At the same time point, we also measured the serum levels of the proinflammatory cytokine Il-1β (Fig. 5A). These were greatly elevated at early time points after pilocarpine injection and remained elevated throughout the experiment. TNF-α levels remained unaffected (Fig. 5B).

In vitro effects of methyl-scopolamine

Since the experiments described so far were performed in animals pretreated with methyl-scopolamine, we tested its efficacy in preventing peripheral activation of muscarinic receptors in two separate experimental paradigms. For this purpose, we used the levels of pilocarpine measured in serum of animals (see Fig. 2B) and a concentration of scopolamine that was comparable to the one used in vivo (i.e., a 1/350 ratio against pilocarpine). We first exposed rat brain microvascular endothelial cells (Krizanac-Bengez et al., 2006a, 2006b) to pilocarpine in the presence or absence of methyl-scopolamine (Fig. 5C). These experiments were complemented by a set of trials where WBC were also added to the culture system (Fig. 5C). Regardless of the presence or absence of methyl-scopolamine, pilocarpine caused a significant release of Il-1β comparable in time course to what observed in vivo (Fig. 5A). We also measured the effects of methyl-scopolamine on cell morphology. To this end, we challenged endothelial cell cultures with pilocarpine exposures and compared the results to naive, time-matched cultures. Note that methyl-scopolamine failed to prevent the profound changes in cell morphology and membrane integrity induced by pilocarpine (Fig. 5D).

Pilocarpine’s effect on the blood-brain barrier

Since the data so far presented suggested that the CNS levels of pilocarpine that can be achieved after peripheral administration are not sufficient to cause seizures, we tested the hypothesis that other systemic changes may trigger neuronal hyperexcitability. We first tested the intactness of the endothelial cell BBB, a crucial determinant of homeostatic control of the neuronal microenvironment. At pilocarpine stage IV, that is, at the time serum and brain levels of pilocarpine were measured, and before the development of SE, BBB integrity was measured by comparing ¹⁴C-Sucrose permeability in naive versus pilocarpine treated rats. On average, ¹⁴C-Sucrose penetrated pilocarpine-treated rat brains in a fashion comparable to naïve brain (Fig. 6A). The permeability to sucrose was only slightly increased, suggesting little damage to the BBB at this stage. However, FITC extravasation analysis demonstrated a reproducible pattern of patchy leaks in pilocarpine-treated hippocampi before SE (Fig. 6B–D and Figs. 7 & 8). BBB leakage was observed in the frontal neocortex, hippocampus, EC, and dentate gyrus. Interestingly, neurons in pyramidal and granule cell layers were filled with FITC-albumin, suggesting avid uptake by these neurons (e. g., Fig. 7B1–2 and 8C).

A possible explanation of the differences found between P_sucrose (little change) and extravasation of plasma protein may be due to the fact that while P_sucrose was measured in whole brain homogenates, FITC leakage was assessed histologically. In other words, it is possible that hemodynamic changes may have accounted for a deficient perfusion of radioactive tracer (¹⁴C-Sucrose) thus leading to an underestimate of actual permeability. This was tested by reexamining FITC perfusion patterns in naïve versus stage IV animals (Fig. 6). Note that, the hippocampal formation and adjacent areas were dramatically hypoperfused, as demonstrated by lack of visible, FITC-albumin filled vessels. Taken together, these results suggest that pilocarpine caused two different cerebrovascular changes, one consisting of increased BBB permeability and the other related to hemodynamic changes (Saad et al., 2003).

Synergism between [K]_out and pilocarpine

Since the results so far presented suggested little direct effects of pilocarpine on hippocampal neurons but increased BBB permeability, we tested the hypothesis that slightly elevated [K]_out, as one may expect in the proximity of focal cerebrovascular leaks, may unveil proepileptogenic actions of muscarinic agonists (Fig. 9). As expected based on the work by Avoli et al. (2000) and Rutecki and Yang (1998), we found that even slight elevation of [K]_out (to 6 mM) unmasked a powerful proepileptogenic effect of low concentrations of pilocarpine (50 μM). This effect was remarkable compared to what seen in low (4.35 mM) [K]_out under identical experimental conditions (Fig. 9A versus B, 10 slices obtained from three rats).

Pilocarpine’s permeability across the blood–brain barrier

Our experiments show that pilocarpine does not efficiently permeate the BBB. Direct comparison with an impermeant and permeant compound (sucrose and diazepam) revealed that pilocarpine belongs to the low-permeation lipophilic agents, which are commonly extruded by multiple drug resistance proteins (Marroni et al., 2003). Comparison between pilocarpine and sucrose/diazepam was performed in control animals, leaving open the possibility that BBB permeability changes occurring after exposure of pilocarpine may have caused increased drug transfer into the brain. This was clearly not the case since pilocarpine did not significantly enter the brain at EEG Stage IV. It therefore appears that pilocarpine is a poor permeant across the BBB as also suggested by previous radiotracer studies (Omori et al., 2004).

Patchy leaky spots could be seen in the brain by FITC extravasation. It is thus possible that local concentrations of pilocarpine were slightly higher in those regions. However, while the brain levels of pilocarpine achieved appear to be low compared to dose initially injected intraperitoneally, our results cannot rule out the hypothesis of focally increased pilocarpine (Egorov et al., 2003). Under this scenario, initiation of seizures depends on BBB loss in regions highly sensitized to cholinergic agonists. Concomitantly and perhaps synergistically, failure of BBB protection could determine an alteration of the brain homeostasis thus facilitating the entry of blood-borne molecules/ions (e. g., K⁺) (Janigro, 1999; Grant and Janigro, 2004; Oby and Janigro, 2006). We have recently demonstrated that focal motor seizures can be triggered by BBB openings induced by intracarotid injection of a hyperosmolar bolus of mannitol (Marchi et al., 2007). In this study, BBB opening was achieved immediately after the first brain passage of the 1.4 M bolus of mannitol injected directly in the cerebral circulation. The probability of seizures occurrence was directly correlated with extent of BBB opening and seizures occurred minutes after a successful BBB opening. Recent evidences points to the same etiological role of BBB dysfunction in contributing to seizures occurrence (Ivens et al., 2007; van Vliet et al., 2007). However these results cannot be directly compared. For example, van Vliet and colleagues investigated the role of BBB integrity in the progression of temporal lobe epilepsy using a different experimental paradigm consisting of a systemic (and not intracarotid) administration of mannitol. Mannitol was injected at concentrations that are commonly used to prevent post-traumatic events including epileptogenesis. Ivens and colleagues identified a selective glial uptake of FITC-albumin after a chronic opening of the BBB. In our acute experiments, as well as in other animal models including the aforementioned van Vliet paper, albumin accumulated in neurons and glia. In spite of these minor discrepancies, our data suggest that early focal BBB damage observed in the pilocarpine model may contribute to the development of seizures allowing entrance into the brain of serum-derived factors.

Lack of efficacy with methyl-scopolamine pretreatment

Previous results by others (Dehaye et al., 1984; Wang et al., 1999) question the usefulness of methyl-scopolamine as blocker of peripheral actions by pilocarpine. In fact, the levels of pilocarpine used to induced SE, as well as those we measured in blood, are well in excess of those capable of fully displacing antagonist’s binding. This was confirmed by our in vitro studies that demonstrated that pilocarpine’s effects on WBC or endothelial cells are not significantly antagonized by methyl-scopolamine. This was observed in spite of a 1–2 h interval between administration of the antagonist and that of pilocarpine. This insufficient blockade by methyl-scopolamine may suggest that this pretreatment, while useful in decreasing mortality and in diminishing a number of behavioral events (e. g., body tremors, salivation, chromodacryorrhea, and diarrhea), may not be capable of fully counteracting the systemic effects of pilocarpine.

In vitro action of pilocarpine

Our electrophysiological evidence demonstrates that concentrations of pilocarpine close to the range of activation of muscarinic receptors or to those measured in vivo do not cause any considerable epileptiform activity at physiologic [K]_out. In vitro seizure-like activity was obtained only in condition of elevated extracellular K⁺ (6 mM). Previous findings by others have indirectly implied the same, since seizure-like activity in vitro was obtained only when cofactors, such as high K⁺ were applied (Rutecki and Yang, 1998). In similar experiments but using the EC-hippocampal slice, low Mg⁺⁺ was used to trigger seizure-like activity with pilocarpine (Massimo Avoli, personal communication and Avoli et al., 2000). The induction of epileptiform discharges by intraparenchymal, intraventricular, or intrahippocampal application of pilocarpine also required much greater amounts of the drug in view of the high affinity for muscarinic receptors (Dehaye et al., 1984; Leite et al., 2002; Cavalheiro et al., 2006).

Work by Alonso et al. demonstrated that the cholinergic agonist carbachol induces seizure-like in slices (Dickson and Alonso, 1997). For several reasons, these findings are not comparable to ours because a different agent has been used. For example, carbachol appeared to specifically target interneurons versus principal cells. Paulsen and colleagues (Fisahn et al., 1998) failed to reproduce these findings, and at the same concentration carbachol did not cause seizure-like activity in vitro. Interestingly, and consistent with our hypothesis, the only major difference between the two approaches is the fact that Alonso and colleagues used a slightly elevated concentration of potassium (5 mM) while when no epileptic activity was seen, the concentration of K⁺ was 3 mM. The latter concentration is comparable to the potassium present in CSF while the former is expected in the brain after BBB opening. Egorov et al., (Egorov et al., 2003) observed localized depolarizing effects of carbachol, but focal injections required 10 mM of the drug, clearly a much higher concentration than what was used by the authors in the other papers. Finally, Magistretti and colleagues did not observe significant field potential changes suggestive of epileptiform activity after exposure of brain slices to muscarinic agents in spite of elevated potassium concentrations (Magistretti et al., 2004). Thus, previously published data and our presented results strongly suggest that cofactors are required to produce cholinergic seizures.

A limitation of the slice approach for pharmacological studies is the use of largely nonphysiological conditions (Schwartzkroin and Prince, 1976). We thus expanded our studies to the in vitro whole brain preparation. This preparation allows the investigation of drug or circuitry effects under controlled conditions and functional preservation of the BBB that closely mimic what the in vivo/in situ condition (de Curtis et al., 1991; Muhlethaler et al., 1993; Librizzi et al., 1999, 2001, 2006). Application of extreme concentrations of pilocarpine (up to 1 mM) did not cause epileptiform activity, but rather induced fast oscillations in the gamma range that were comparable to those induced by activation of cholinergic synapses by carbachol (Fisahn et al., 1998; van Der et al., 1999; Dickson et al., 2003). The effect of pilocarpine on FP recorded in CA1 and entorhinal cortices following stimulation of the lateral olfactory tract in the isolated guinea pig brain were evaluated to overcome the limitations of the hippocampal slice preparation. Pilocarpine reduced the conditioned response in a pairing test and reduced the propagation of lateral olfactory tract (LOT)-evoke activity into CA1. This suggests that the muscarinic effect of pilocarpine is inhibitory on field activity. In addition, pilocarpine’s effect in the brain was limited to gamma rhythm, while seizure-like events were never observed at concentrations similar to those seen prior to SE in vivo (Uva et al., 2006).

A complementary mechanism of action of pilocarpine?

The fact that pilocarpine-induced activity can be triggered in condition of high extracellular K⁺ has been already demonstrated by others (Rutecki and Yang, 1998; Avoli et al., 2000). Increased extracellular K⁺ mimics the consequences of in vivo opening of the BBB and subsequent loss of brain homeostasis. What remains to be elucidated is the extent of K⁺ increase in the brain parenchyma after opening of the BBB. These changes are difficult to predict or model considering the buffering capacity of glial cells. We previously demonstrated that small, but significant changes in [K]_out are sufficient to account for post-traumatic excitability changes (D’Ambrosio et al., 1998, 1999). Thus, it is possible that a similar change in [K]_out occurred in pilocarpine brain since both traumatic brain injury and pilocarpine have a profound effect on BBB integrity (data presented herein and Schmidt and Grady, 1993).

Under nonphysiological conditions, the delicate balance between large depolarizing and hyperpolarizing potentials occurring in the hippocampus is lost, and synchronous, repetitive neuronal depolarizations may occur. While most studies have concentrated on epileptogenesis following failure of inhibitory potentials or exaggerated excitation, nonsynaptic contributions to epileptogenesis have been relatively neglected despite the fact that several features of abnormal neuronal hyperexcitability and synchronization can be understood by models which focus on inadequate K⁺ clearance, excessive accumulation mechanisms (Lux et al., 1986; Traynelis and Dingledine, 1988; Gabriel et al., 1998) and/or alterations in the size of the extracellular space. Elevated potassium, as seen after loss of brain homeostasis, is a potent convulsant (Traynelis and Dingledine, 1988) and also acts as a vasoactive agent capable of both dilation (at low [K]_out) or constriction (Nguyen et al., 1998). There is a significant difference between brain and plasma [K]_out, favoring potassium entry into the CNS when the BBB is breached (Keep et al., 1993; Grant and Janigro, 2004). Thus, if a secondary effect of pilocarpine is an inflammatory change in BBB permeability, even focal changes in [K]out may be responsible in part for the initiation of seizures. Of course other mechanisms can be involved, since many epileptogenic agents are more concentrated in serum than brain (e. g., glutamate (O’Kane et al., 1999; Grant et al., 2003)).

A less well-understood possibility is a direct action of inflammatory mediators, such as Il-1β, on neurons. Admittedly, attempts to tie interleukins to ion channels is in its infancy, but several reports have suggested that elevation of Il-1β may cause sudden rapid changes in excitability of both inhibitory and excitatory neurons (Plata-Salaman and Ffrench-Mullen, 1992; Sawada et al., 1992; Yang et al., 2005). Thus, focal entry of Il-1β via the para-cellular way left open by the initial proinflammatory insult by pilocarpine may constitute a possible contributor for pilocarpine seizures after systemic injection. If a peripheral inflammatory process contributes to pilocarpine-induced seizures then pretreatment with antiinflammatory drugs should protect from the onset of SE. The effect of non-steroidal antiinflammatory drugs on seizures elicited by pilocarpine was investigated (Ikonomidou-Turski et al., 1988). Interestingly, only mefenamic acid was able to reduce the severity of seizures while sodium salicylate and phenylbutazone exacerbated the effects of pilocarpine. These controversial results demonstrate the complexity of pilocarpine’s action but point to an inflammatory component.