Peripherally restricted viral challenge elevates extracellular glutamate and enhances synaptic transmission in the hippocampus

Abstract

Introduction

Seizures represent a major neuropathological affliction and an important cause of long-term disability. Seizures result from excessive and/or synchronous neuronal activity in the brain. Cerebral inflammation following trauma, ischemia, infections, tumors, etc., has been recognized as an important pathological feature that predisposes and/or elicits seizures (Marchi et al. 2009; Vezzani and Granata 2005; Ravizza et al. 2011). The underlying mechanisms entail the activation of resident innate immune cells, chiefly microglia and astrocytes, as well as the recruitment and activation of peripheral leukocytes leading to the production of a plethora of cytokines, chemokines, prostaglandins and other inflammatory agents. These inflammatory agents may increase excitatory inputs, decrease inhibitory inputs, or both, resulting in hyperexcitability of the neuronal networks, a hallmark of seizures.

Notably, also peripheral inflammation can increase seizure propensity in susceptible individuals (Tellez-Zenteno et al. 2005; Scheid and Teich 2007; Verrotti et al. 2009). The underlying mechanisms involve relaying peripheral innate immunity signals to the brain whereby they induce a “mirror inflammation” (Dantzer and Kelley 2007; Dantzer et al. 2008; Quan and Banks 2007). Several experimental studies dovetail with these clinical data. For example, the simulation of bacterial infection via intraperitoneal (i.p.) injection of a bacterial endotoxin, lipopolysaccharide (LPS) increases seizure susceptibility in mice as seen from a decrease in the threshold of clonic seizures instigated by pentylenetetrazole (PTZ) (Sayyah et al. 2003). In a rat model of inflammatory bowel diseases, intracolonical injection of 2,4,6-trinitrobenzene sulfonic acid (TNBS) increases the susceptibility to PTZ-induced seizures (Riazi et al. 2008). Moreover, experimental arthritis and subcutaneous granuloma decrease the onset and increase the score of PTZ-evoked seizures (Rao et al. 2008).

We have also shown that peripheral viral challenge robustly increases seizure susceptibility (Kirschman et al. 2011; Michalovicz and Konat 2014). In this experimental paradigm, intraperitoneal injection of a viral mimetic, polyinosinic-polycytidylic acid (PIC) results in a several-fold increase in the extent and duration of status epilepticus induced by kainic acid (KA) in mice (Kirschman et al. 2011). This seizure hypersusceptibility is protracted for three days after PIC challenge (Michalovicz and Konat 2014). Of note, PIC is an unstable inflammagen that is rapidly degraded in the bodily fluids (Krasowska-Zoladek et al. 2007), and when injected intraperitoneally does not reach the circulation (Fil et al. 2011). Therefore, PIC challenge represents a bolus stimulation of the innate immune cells within the peritoneal cavity, and these peripherally-generated inflammatory mediators instigate a cerebral response (Konat 2015). In particular, PIC challenge triggers a robust but transient surge of blood cytokines, i.e., interferon β (IFNβ), interleukin 1β (IL-1β), IL-6 and tumor necrosis factor α (TNFα) (Michalovicz and Konat 2014; Cunningham et al. 2007). This cytokine surge instigates a global cerebral response as seen from the upregulation of a myriad of inflammatory genes in all major brain regions (Cunningham et al. 2007; Konat et al. 2009; Fil et al. 2011). In the hippocampus, the ictal site of KA-induced seizures (Ben-Ari and Cossart 2000), PIC challenge dysregulates the expression of over six hundred genes that, in addition to inflammatory and stress proteins, encode several neurotransmission-related proteins and microRNAs (Michalovicz and Konat 2014;Michalovicz et al. 2015). This genomic reprograming undoubtedly underlies the development of seizure hypersusceptibility, albeit specific cellular and molecular pathways have not been defined.

The present study was undertaken to test the hypothesis that hyperexcitability ensuing PIC challenge features dysregulation of glutamate homeostasis. We employed the enzyme-based microelectrode technology for in vivo monitoring of extracellular glutamate levels in the hippocampus to identify neurotransmission-associated events affected by PIC challenge. The characterization of glutamate homeostasis was complemented with an electrophysiological study assessing synaptic transmission and plasticity in acute hippocampal slices.

Materials & Methods

In vivo glutamate measurement

Changes in extracellular glutamate in the hippocampus were monitored using the microelectrode arrays (MEA) technique (Burmeister and Gerhardt 2001) as previously described (Hunsberger et al. 2015a; Hunsberger et al. 2015b). Briefly, the electrodes obtained from Quanteon (Nicholasville, KY) were coated with glutamate oxidase and calibrated, as exemplified in Figure 1. A glass micropipette (Quanteon) was mounted to the arrays for intracranial drug delivery. Mice were anesthetized with isoflurane (1–4% continuous inhalation), placed in a stereotaxic device (David Kopf Instruments, Tujunga, CA) and the MEA/micropipette assemblies were inserted into the hippocampal sub-regions, i.e., dentate gyrus (DG), cornu ammonis 1 (CA1) and cornu ammonis 3 (CA3). The stereotaxic coordinates from the bregma were AP: −2.3 mm, ML: +/−1.5 mm, DV: 2.1 mm for DG, AP: −2.3 mm, ML: +/−2.7 mm, DV: 2.25 mm for CA3 and AP: −2.3 mm, ML: +/−1.7 mm, DV: 1.4 mm for CA1. A reference electrode was implanted under the skin in a remote site. All MEA recordings were performed at 10 Hz using constant-potential amperometry. All measurements and injections were performed after a stable baseline was reached (20–45 min). Both hemispheres were used for drug injection, and sub-regions within a hemisphere, were counterbalanced.

Open in a separate window

Figure 1

In vitro calibration of a self-referencing microelectrode measuring the change in current (pA) on a glutamate oxidase site (GluOx; red) vs. a sentinel site (Sent; blue). Interferents, such as ascorbic acid (AA) and dopamine (DA), did not alter the current at either glutamate oxidase or sentinel sites. Addition of glutamate (Glu) produced a stepwise current increase on the glutamate oxidase site, but no change on the sentinel site. Hydrogen peroxide (H₂O₂) produced an increase in current on both sites. Sensitivity, slope, limit of detection, and R² values were calculated after calibration.

Tonic glutamate levels were calculated in all three sub-regions by averaging extracellular glutamate levels over 10-s periods. Evoked release was induced in a subset of animals by delivering 50–100 nL of 70 mM of potassium chloride (KCl) solution every 2–3 min. The amplitudes of ten reproducible signals were averaged and compared. To measure glutamate uptake, a subset of animals received 1–2 injections at 50 nL increments within a 50–250 nL range of 200 μM glutamate (Sigma-Aldrich, St. Louis, MO) delivered every 2–3 minutes in one hemisphere. Temporal clearance of glutamate was monitored and expressed as the net area under the curve (AUC). Glutamate release in a subset of animals was measured in the opposite hemisphere following inhibition of glutamate uptake with 50–250 nL of 500 μM DL-threo-β-benzyloxyaspartate (TBOA; Tocris, Ellisville, MO). The amperometric data were analyzed using a custom Microsoft Excel software program (MatLab) as previously described (Hunsberger et al. 2015a; Hunsberger et al. 2015b). Data from some hippocampal regions were excluded for reasons including failure of the MEA or clogging of the micropipette. The number of mice per treatment group for glutamate measurements is indicated in Table 1.

Table 1

Number of mice used in glutamate measurement experiments.

	DG		CA3		CA1
	PIC	Saline	PIC	Saline	PIC	Saline
Tonic	11	9	11	9	12	9
KCl	5	4	5	4	5	4
Exogenous glutamate	5	6	6	5	4	5
TBOA	6	4	6	4	6	4

Open in a separate window

Results

An increased glutamatergic transmission is a plausible mechanism underscoring PIC-induced hypersusceptibility to KA-induced seizures found in previous studies (Kirschman et al. 2011; Michalovicz and Konat 2014). Here, we assessed glutamate homeostasis in the hippocampus, the ictal site of KA-induced seizures (Ben-Ari and Cossart 2000), using enzyme-based microelectrode technology that allows real-time monitoring of extracellular glutamate in vivo. We used isoflurane to avoid anesthetic-induced changes in resting glutamate levels (Mattinson et al. 2011), and measured glutamate in hippocampal subregions known to be rich in glutamate receptors, i.e., DG, CA1 and CA3 (Nimchinsky et al. 2004;Pettit and Augustine 2000). All measurements were performed 24 h after i.p. injection of PIC or saline. As shown in Fig. 2a, PIC-challenge induced a robust increase in tonic, resting glutamate levels in all three sub-regions. The highest increase of 11-fold over control was observed in DG [F(1,18) = 41.49, p < .0001]. CA1 [F(1,19) = 15.58, p = .0009] and CA3 [F(1,18) = 18.94, p = .0004] featured 9.8-fold and 5.8-fold increase, respectively.

Open in a separate window

Figure 2

Tonic glutamate levels and evoked glutamate release in the hippocampus. Mice were intraperitoneally injected with PIC or saline (control). After 24 h, hippocampal glutamate was analyzed by the enzyme based microelectrode technique in the dentate gyrus (DG), cornu ammonis 1 (CA1) and cornu ammonis 3 (CA3). (a) Extracellular tonic glutamate levels in the hippocampal sub-regions. (b) Baseline-matched representative traces of K⁺ evoked release of glutamate in CA3. (c) The amplitudes of K⁺ evoked release of glutamate in the hippocampal sub-regions. For details see Methods. Bars represent means ± SEMs. Asterisks denote values significantly different from respective controls ***p ≤ .001, ****p ≤ .0001).

Several mechanisms can be considered to account for the increase of tonic glutamate. For example, PIC challenge may alter the capacity or “ceiling” of neuronal terminals to release glutamate (Hinzman et al. 2010). To test this possibility, we used the paradigm of potassium-evoked glutamate release (Day et al. 2006). As shown in Fig. 2b, the injection of KCl induced a transient (approximately 5 s) elevation of extracellular glutamate. No differences were observed between PIC-challenged vs. control mice in any sub-region (DG [F(1,7) = .11, p = .75]; CA1 [F(1,7) = .002, p = .96]; CA3 [F(1,7) = .04, p = .84]; Fig. 2c). These results indicate that PIC challenge does not increase the neurotransmitter content in presynaptic terminals.

A decreased glutamate clearance represents an alternative mechanism for the rise of extracellular glutamate. To test this option, we injected exogenous glutamate, and monitored its clearance by measuring net AUC. We first compared the amplitude of glutamate signals following injection of exogenous glutamate to confirm differences in net AUC between the PIC-challenged and saline-injected mice following application of exogenous glutamate were due to alterations in the uptake and not to differences in the amount of applied glutamate (Hunsberger et al. 2015a; Hunsberger et al. 2015b). Prior to AUC measurement, maximal amplitudes of the glutamate signal were determined to ensure reproducibility of glutamate injection. Fig. 3a shows no significant differences in the maximal amplitudes in any sub-region in control vs. PIC-challenged mice (DG [F(1,9) = .06, p = .81]; CA1 [F(1,7) = .25, p = .63]; CA3 [F(1,9) = .43, p = .53]). Also, no effect of PIC challenge on the diffusion of exogenous glutamate within the tissue expressed as the T_rise values, i.e., the time for the signal to reach maximum amplitude (Sykova et al. 1998), was evident in any sub-region (DG [F(1,9) = .07, p = .80]; CA1 [F(1,7) = .25, p = .63]; CA3 [F(1,9) = .24, p = .63]; Fig. 3b), suggesting any reductions in glutamate uptake were not because of diffusion from the point source (micropipette) to the MEA. Temporal analysis of glutamate levels following its injection revealed a profoundly delayed clearance profile in PIC-challenged mice (Fig. 3c), indicative of an impairment of the neurotransmitter’s uptake. The quantitation of this impairment is shown in Fig. 3d. The greatest increase in net AUC induced by PIC challenge of 8.3-fold over control was found in CA3 [F(1,9) = 11.55, p = .008]. The values for DG [F(1,9) = 13.77, p = .005] and CA1 [F(1,7) = 18.16, p = .004] were 6.7-fold and 3.8-fold, respectively.

Open in a separate window

Figure 3

The uptake of exogenous glutamate in the hippocampus. Mice were intraperitoneally injected with PIC or saline (control). After 24 h, hippocampal glutamate was analyzed by the enzyme based microelectrode technique in different hippocampal sub-regions as indicated. (a) The amplitude of signals following local injection of 200 μM glutamate. (b) Glutamate diffusion expressed as time to reach maximum amplitude (T_rise). (c) Peak-matched representative traces in the DG. (d) Glutamate uptake expressed as the net area under the curve (AUC). For details see Methods. Bars represent means ± SEM. Asterisks denote significant differences from respective controls (**p ≤ .01).

The augmentation of tonic glutamate may also result from an increased release of glutamate by astrocytes. We inhibited glutamate uptake with TBOA, a competitive non-transportable EAAT blocker (Shimamoto et al. 1998; Montiel et al. 2005; Tovar et al. 2009) to confirm the involvement of these receptors and to unmask the release of glutamate (Jabaudon et al. 1999). Local application of TBOA produced a transient increase in the extracellular glutamate concentration in both PIC-challenged and control mice (Fig. 4a), although the amplitude of this increase differed (Fig. 4b). Thus, DG [F(1,8) = 5.80, p = .04] and CA1 [F(1,8) = 7.06, p = .03] featured an 18- and 17-fold increase in PIC-challenged vs. control mice. In contrast, PIC challenge had no effect on glutamate release in CA3 [F(1,8) = .51, p = .50].

Open in a separate window

Figure 4

Spontaneous release of extracellular glutamate in the hippocampus. Mice were i.p. injected with 12 mg/kg of PIC or saline (control). After 24 h, glutamate uptake was blocked by the application of 500 μM TBOA to unmask glutamate release, and the levels of extracellular glutamate were determined in different hippocampal sub-regions as indicated. (a) A representative trace of transient glutamate release in DG in PIC- and saline-injected mice. (b) The amplitude of extracellular glutamate in the hippocampal sub-regions following TBOA application. For details see Methods. Bars represent means ± SEM. Asterisks denote significant differences from respective controls (*p ≤ .05).

The increased tonic glutamate, the impaired glutamate uptake, and the increased glutamate release implicated that PIC challenge might enhance glutamatergic neurotransmission in the hippocampus. To verify this notion, we examined basal synaptic transmission and synaptic plasticity in hippocampal slices by field recordings. Fig. 5a shows representative traces of EPSPs in hippocampi from PIC challenged vs. control mice. PIC challenge markedly increased the amplitude and slope of the EPSP. As seen from Fig. 5b, PIC challenge profoundly enhanced basal synaptic transmission denoted by input-output responses of the neuronal networks (stimulus response curve) [F(1,41)=30.35, p<0.0001], at each point of measurement [ps>.05]. Throughout the range of stimulus intensities from 50 to 500 μA, the synaptic efficiency increased by over 2.5-fold in slices from PIC-challenged mice as compared to slices from saline-injected mice. A change in basal synaptic transmission may result from alterations in pre- post- and peri-synaptic elements. To further characterize which component across the synapse actually contributed to PIC-induced increased synaptic transmission, we used the PPF protocol that reflects residual calcium levels, a presynaptic mechanism that plays a major role in short-term and long-term plasticity. PIC challenge decreased PPF [F(1,19)=7.391, p=0.014] at the short (50 ms) stimulus interval [p<.05], indicating an increase in presynaptic release probability due to alteration in either presynaptic compartment or astrocyte calcium signaling (Fig. 5c). Albeit, no effect was observed at longer intervals [ps>.0.5]. LTP, a cellular substrate of plasticity that may feature both pre- and post-synaptic expression (Padamsey and Emptage 2014), was significantly increased by PIC challenge [F(1,68)=2.47, p=.0007; Fig. 5d].

Open in a separate window

Figure 5

Synaptic transmission in hippocampal slices. Mice were intraperitoneally injected with PIC or saline (control), and after 24 h, hippocampal slices were prepared. The Schaffer collateral pathways of CA3 were stimulated, and fEPSPs evoked in the striatum radiatum of CA1 were recorded. (a) Representative traces of fEPSPs evoked at a stimulus intensity of 200μA. (b) Basic synaptic transmission as the fEPSP slope measured at increasing stimulus intensity (c) Paired-pulse facilitation expressed as the change of ratios of the second stimulus fEPSP slopes to the first stimulus fEPSP slopes plotted as a function of interstimulus intervals. (d) Long term potentiation calculated as the fEPSP ratio over time. Inset: Representative traces shown include data collected from saline-injected and PIC-challenged animals during baseline recordings (black) overlayed on traces during 55–60 minute LTP (brown). For details see Methods. Symbols represent means ± SEMs from 3 mice (12 slices) per group. Asterisks denote significant differences from respective controls (Student’s t-test; * p ≤ .05, ***p ≤ .001).

Discussion

The major finding of our study is that peripheral PIC challenge disrupts cerebral glutamate homeostasis, resulting in a robust increase in the basal extracellular glutamate concentration (Fig. 2a). This increase is likely to underlie hippocampal hypersusceptibility to KA-induced seizures in PIC challenged mice (Kirschman et al. 2011; Michalovicz and Konat 2014). In support of this notion, increased tonic glutamate levels have been shown to correlate positively with the severity of focal motor seizures induced by intrahippocampal injection of 4-aminopyridine (4-AP) (Stephens et al. 2014). Also, astrocytic release of glutamate has been shown to facilitate the initiation of seizures (Kang et al. 2005), while the suppression of glial glutamate release leads to decreased seizure susceptibility (De Bundel D. et al. 2011). In addition, human epileptogenic hippocampi exhibit augmented basal glutamate levels during interictal periods that may contribute to seizure generation (Cavus et al. 2005). Moreover, the overflow of extracellular glutamate has been recognized as a key factor in the development of neuronal hyperexcitability (Featherstone and Shippy 2008). For example, dysregulation of extracellular glutamate homeostasis has been directly linked to hyperexcitability of cortical and spinal cord neurons at diverse pathological conditions (Campbell et al. 2012; Campbell et al. 2014; Campbell and Hablitz 2004; Campbell and Hablitz 2008; Putatunda et al. 2014). Glutamate-induced hyperexcitability is chiefly mediated by ionotropic glutamate receptors, in particular, NMDA receptors, but the involvement of metabotropic glutamate receptors has also been implicated (Featherstone and Shippy 2008). The mechanism entails a direct ligation of the receptors, although other, more circuitous pathways may also be involved. Consequently, we posit that the amplified response of the CA1 pyramidal cells induced by the upsurge of extracellular glutamate in PIC challenged mice contributed to the enhanced synaptic transmission (Fig. 5b), PPF (Fig. 5c), and LTP (Fig. 5d). However, a possibility that PIC challenge might also reduce inhibitory signaling can be considered. For example, a reduction in the number of inhibitory synapses was observed in the cortex of mice following repeated LPS injections (Chen et al. 2014).

The elevation of tonic glutamate could be due to increased glutamate release, decreased glutamate uptake, or both. Local application of potassium evoked the same amounts of glutamate in PIC and control hippocampi (Fig. 2c), indicating no alteration in the capacity or ceiling of presynaptic release of this neurotransmitter. However, the clearance of injected glutamate was profoundly hampered (Fig. 3d), indicating an impairment of glutamate uptake by PIC challenge. Inflammatory cytokines upregulated in the hippocampus in response to PIC challenge might mediate this impairment. For example, IL-1β and TNFα inhibit astrocytic glutamate uptake (Hu et al. 2000; Ye and Sontheimer 1996), and the Tnfa and Il1b gene expression is upregulated in the hippocampi of PIC-challenged mice as compared to controls (Michalovicz and Konat 2014).

The excitatory amino acid transporters 1 and 2 (EAAT1/2) expressed almost exclusively in astrocytes play the major role in the uptake of glutamate (Niciu et al. 2012). The application of TBOA, a specific competitive inhibitor of EAAT1/2 (Shimamoto et al. 1998), induced a transient increase of extracellular glutamate (Fig. 4), indicating the involvement of these transporters. However, the contribution of neuronal transporters cannot be ruled out. Moreover, in DG and CA1 of PIC challenged animals, local TBOA application elicited much greater glutamate spikes than in control tissues. Because blocking of EAAT unmasks glutamate release (Jabaudon et al. 1999), our results suggest that PIC challenge not only impairs glutamate uptake but also augments glutamate release from astrocytes. This result is consistent with previous studies showing the enhancement of astrocytic glutamate release by inflammatory mediators, i.e., IL-1β (Casamenti et al. 1999), TNFα and prostaglandins (Bezzi et al. 2001). Altogether, our results strongly implicate astrocytes as cellular targets for inflammatory mediators generated in response to PIC challenge.

In contrast to DG and CA1, no difference between PIC challenged and control mice in the post-TBOA glutamate amplitude was detectable in CA3 (Fig. 4), suggesting that the release of glutamate in this subregion is not affected by the inflammatory milieu instigated by PIC challenge. Because presynaptic glutamate release was not altered in CA3 (Fig. 2c), the astrocytic release mechanisms are likely candidates to account for this region-specificity. Astrocytes release glutamate through different mechanisms, e.g., Ca²⁺-dependent exocytosis, glutamate exchange via the cystine–glutamate antiporter (Xc-) and reversal of uptake by glutamate transporters (Malarkey and Parpura 2008). It is tempting to speculate that unlike DG and CA1 astrocytes, CA3 astrocytes use mechanisms that are not susceptible to the inflammatory milieu induced by PIC challenge. Such a differential response is buttressed by a previous observation that in the presence of TBOA, tetraethylammonium chloride elicits a several-fold greater glutamate release in CA3 than in CA1 or DG (Chiba et al, 2010). However, specific mechanisms may vary between these two paradigms.

Field recordings in hippocampal slices prepared from PIC-challenged vs. control mice are congruent with the dysregulated glutamate homeostasis observed in vivo. Thus, the robust increase in the basal glutamatergic synaptic transmission (Fig. 5b) likely resulted from the elevation of extracellular glutamate (Fig. 2a). The underlying mechanisms might involve the activation of extrasynaptic glutamate receptors (Petralia 2012). PPF, an index of short-term plasticity, reflects synaptic efficacy determined by the probability of presynaptic neurotransmitter release (Zucker and Regehr 2002). PIC challenge significantly reduced PPF (Fig. 5c), indicating that increased probability of glutamate release at the terminals of the Schaffer collaterals might contribute to the increased synaptic transmission. However, this increased presynaptic activity is in divergence with unchanged potassium-evoked glutamate amplitudes (Fig. 2c), another facet of the presynaptic neurotransmitter release. A plausible explanation is that the glutamate amplitudes measure the maximum capacity for release by depleting presynaptic glutamate pool with large doses of potassium, and these measurements may not be compatible with the physiological/functional release measured by PPF. Furthermore, LTP that can be expressed at postsynaptic as well as presynaptic loci (Padamsey and Emptage 2014) was increased by PIC challenge (Fig. 5d), indicating an enhancement of synaptic strength. Altogether, these results show that PIC challenge increases both basal synaptic transmission and synaptic plasticity.

In addition, the slice experiments provide compelling evidence for the intrinsic nature of the hippocampal alterations induced in PIC-challenged mice. For instance, peripheral inflammation might have increased the permeability of KA and/or glutamate through the blood-brain barrier (BBB), leading to hyperexcitability of hippocampal networks that would manifest as seizure hypersusceptibility. However, the robustly augmented excitatory synaptic transmission in the perfused slices from PIC-challenged as compared to control animals shows that the hyperexcitability indeed originates in the hippocampal parenchyma. This finding corroborates our previous study showing greatly increased spontaneous ictal activity elicited with 4-aminopurine in hippocampal slices from PIC-challenged vs. control mice (Konat et al. 2012). As discussed above the augmentation of synaptic transmission results from the elevation of extracellular glutamate. Ergo, the slice studies also indirectly verify the intrinsic nature of glutamate dysregulation observed in vivo.

We have recently shown that PIC challenge profoundly upregulates expression of the complement in the hippocampus, and that this upregulation is commensurate with the period of seizure hypersusceptibility (Michalovicz et al. 2015). The complement is a major mediator of synaptic modifications (Stevens et al. 2007; Schafer et al. 2012; Stephan et al. 2013), and complement proteins have proconvulsive activity when injected into the hippocampus (Xiong et al. 2003). Therefore, it’s tempting to speculate that the alteration of glutamate homeostasis and hyperexcitability might be induced by the complement proteins. The mechanisms of such alterations might entail anaphylatoxins generated through the complement activation. Anaphylatoxins can activate their cognate receptors on microglia, astrocytes and neurons resulting in the generation of inflammatory factors that affect function of the postsynaptic terminals. For example, Il-1β (Viviani et al. 2003; Yang et al. 2005), IL-6 (Xiaoqin et al. 2005; Samland et al. 2003), TNFα (Beattie et al. 2002; Stellwagen et al. 2005), IFNβ (Hadjilambreva et al. 2005), CXCL10 (Ragozzino et al. 1998), CXCL1/2 (Giovannelli et al. 1998; Ragozzino et al. 1998) and the prostaglandin PGE2 (Chen and Bazan 2005) can enhance glutamatergic synaptic transmission. Thus, in addition to the disruption of glutamate homeostasis discussed formerly, inflammatory factors may induce hyperexcitability of the hippocampal neurons. Alternatively, complement proteins or their derivatives/complexes might bind to synaptic structures resulting in functional impairment of surface receptors that control glutamate homeostasis and/or synaptic transmission.

Recently, seizure hypersusceptibility of rats subjected to colonic inflammation (Riazi et al. 2008) has been linked to an increased synaptic transmission in the hippocampus (Riazi et al. 2015), albeit the extent of this increase was much less than the increase observed here. In contrast to our study, the colonic inflammation reduced LTP in hippocampal slices. Therefore, it seems that hippocampal hyperexcitability may be a common mechanism by which peripheral inflammation increases seizure susceptibility, but the effects on synaptic plasticity vary depending on the inflammatory paradigm.

In conclusion, our results indicate that inflammation instigated by peripheral PIC challenge enhances excitatory synaptic transmission and plasticity in the hippocampus by elevating extracellular glutamate concentration and increasing presynaptic activity. These putative pathway are likely responsible for the development of seizure hypersusceptibility. Our results warrant a comprehensive investigation of the underlying mechanisms at both the cellular and molecular level to provide a foundation for the development of therapeutic strategies for the management of inflammation-related seizures.

Acknowledgments

Abbreviations

Footnotes

Reference List