Human glioma cells induce hyperexcitability in cortical networks

Abstract

Purpose: Patients with gliomas frequently present with seizures, but the factors associated with seizure development are still poorly understood. In this study, we assessed peritumoral synaptic network activity in a glioma animal model and tested the contribution of aberrant glutamate release from gliomas on glioma-associated epileptic network activity.

Methods: In vitro brain slices were made from glioma-implanted mice. Using extracellular field recordings, we analyzed peritumoral epileptiform activity induced by Mg(2+)-free medium in slices from tumor-bearing animals and sham-operated controls. We assessed the effect of sulfasalazine (SAS), a blocker of system and glutamate release, on spontaneous and evoked activity in tumor-associated slices.

Key findings: Tumor-associated cortical networks were hyperexcitable. The onset latency of Mg(2+)-free-induced epileptiform activity was significantly shorter in tumor-bearing slices, and the incidence of Mg(2+)-free-induced ictal-like events was higher. Block of glutamate release from system decreased the response area of evoked activity and completely blocked Mg(2+)-free-induced ictal-like, but not interictal-like events.

Significance: Control of seizures in patients with gliomas is an essential component of clinical management; therefore, understanding the origin of seizures is vital. This work provides evidence that peritumoral synaptic network activity is disrupted by tumor masses resulting in network excitability. We show that blocking glutamate release via system with SAS, a drug already approved by the U.S. Food and Drug Administration (FDA), can inhibit Mg(2+)-free-induced ictal-like epileptiform events similar to other chemicals used to decrease seizure activity. We, therefore, suggest that further studies should consider SAS a promising agent to aid in the treatment of seizures associated with gliomas.

Figure 1

Tumor-bearing slices show spontaneous epileptiform activity. Representative bioluminescence imaging of tumors in vivo. (A) Example of luciferase expressing U251ffluc glioma cells (mouse 3–5) or vehicle alone (mouse 1 and 2) implanted intracranially and visualized with an in vivo imaging using a Xenongen IVIS Bioluminescent Imaging System. Cresyl violet staining of sham-operated (B) and tumor-bearing cortex (C). Tumor-bearing slice show a pronounced tumor mass (in black box) traversing the cortical layers. Extracellular field recording electrodes positioned adjacent to tumor mass in layer II/III and stimulating electrode was placed in deeper layer IV. Electrodes were placed in a comparable region in sham-operated slices. (D) Representative samples of spontaneous extracellular field recordings from a sham (left) and tumor-bearing slice (right), scale bar: 50 sec, 0.5 mV. Inset shows individual interictal-like event in tumor-bearing slice on a faster timescale. Scale bar: 1 sec, 0.5 mV.

Figure 2

Extracellular field recording in sham-operated and tumor-bearing slices. Representative traces of field recordings from sham-operated (A) and tumor-bearing (B) slices in response to increasing stimulation intensities. Upper trace in (A) shows individual trace from sham and from tumor slices in (B) in response to half maximal stimulation intensity. Graphical display of input/output curves of fPSPs recorded in superficial cortical layer II/III of the excitatory response amplitude (C) and excitatory response area (D) as a function of stimulation intensity in sham-operated (squares) compared to tumor-bearing slices (circles). Tumor-bearing slices showed significantly larger amplitude (p < 0.01) and area (p < 0.01) of the “early” excitatory component. Input/output curve of the “late” inhibitory response area (E) amplitude (F).

Figure 3

Mg²⁺-free solution induced epileptiform activity in control and tumor-bearing cortical slices. Representative traces of extracellular field recordings of spontaneous epileptiform events induced by Mg²⁺-free solution in a sham-operated (A) and tumor-bearing slice (B). Insets show traces on a magnified timescale to clearly demonstrate individual epileptiform activity in each group. (C) Mean onset latency of epileptiform event in Mg²⁺-free solution for each group revealed a significantly shorter latency in tumor-bearing slices compared to sham-operated. (D) Summary of the frequency of ictal-like and interictal-like events in sham-operated and tumor-bearing slices showing a significant increase in ictal-like events (p < 0.05) and a decrease in interictal-like events in tumor-bearing slices compared to controls. (E) Mean duration of epileptiform activity is similar in both groups.

Figure 4

Effect of SAS on synchronous activity in the peritumoral network. (A) Representative traces of field recordings in ACSF, after 30 min in Mg²⁺-free, followed by co-application of SAS (250 μM) and Mg²⁺-free and following the addition of APV (20 μM) in tumor-bearing slices (B). Summary bar graphs of the effect of SAS on the response duration of Mg²⁺-free-induced epileptiform activity. (C) Summary data of the experiment in (A) showing that SAS (250 μM) blocked the induction of ictal-like but not interictal-like events.

Figure 5

SAS blocks the induction of Mg²⁺-free ictal-like activity. Sample traces of recordings in ACSF(A), after application of SAS (250 μM) for 30 min and following co-application of SAS (250 μM) and Mg²⁺-free. Insets show events on a faster timescale. (B) Summary bar graphs of the experiment in (A) showing that pretreatment with SAS (250 μM) blocked Mg²⁺-free-induced ictal-like but not interictal-like events.

Figure 6

SAS decreases evoked Mg²⁺-free-induced epileptiform activity. Representative traces of evoked fPSP recordings elicited by half-maximal stimulation intensity obtained from tumor-bearing slices (A) in ACSF, in the presence of Mg²⁺-free solution, and following co-application of SAS (250 μM). In ACSF tumor-bearing evoked responses were large with long durations. Application of Mg²⁺-free solution induced long-lasting polysynaptic activity and SAS (250 μM) decreased the (B) response area (p < 0.01) towards control levels but there was no change in the response amplitude (p > 0.05).

Figure 7

SAS enhances inhibitory later “inhibitory” response in Mg²⁺-free solution. (A) Evoked responses to increasing stimulation intensity from tumor-bearing slices in Mg²⁺-free and co-application of Mg²⁺-free and SAS (250 μM). Summary I/O curves of the effect of SAS on the “early” excitatory response area (B) and amplitude (C) of Mg²⁺-free-induced activity in tumor-bearing slices. Input/output curves of the “late” inhibitory response area (D) and amplitude (E).