Abstracts

This is a Late Breaking abstract

Rationale: Outcomes in epilepsy surgery are largely dependent on correctly identifying, mapping and subsequently treating the seizure onset zone (SOZ) and epileptic network. Current non-invasive tools are limited in their ability to identify brain regions involved in the epileptic network. Previous studies have suggested upregulation of glycolysis and subsequent lactate production within epileptic tissue. In this study, we leverage these metabolic changes and present preclinical data using magnetic resonance spectroscopy/imaging of hyperpolarized [1-¹³C] pyruvate (hpMRS/I) to identify elevated lactate in epileptic tissue.

Methods: We used a low Mg²⁺ in vitro model of neuronal hyperactivation to test the ability of hpMRI to identify elevated lactate in culture. Cultures were treated with low Mg²⁺ media for 2 hours for 10 days, resuspended, then treated with 14 mM hyperpolarized [1-¹³C] pyruvate, and scanned using a clinical MR scanner. MR spectroscopy for lactate and pyruvate were measured and lactate-to-pyruvate ratios were identified using the peak integral for the first 60 time-points (180 s) of each metabolite's time-averaged spectrum. To confirm accurate lactate measurement, we collected supernatants from treated cultures and measured lactate concentrations. For in vivo studies, we used a pentylenetetrazol (PTZ) kindling mouse model. Mice were injected with 35mg/kg IP PTZ every other day for 20 days and were compared to vehicle controls. Two days after the last injection, mice were anesthetized, given intravenous (IV) hyperpolarized [1-¹³C] pyruvate, and scanned using a MR scanner. Lactate conversions were measured as above, then compared in control vs. PTZ kindled mice. We also processed mouse brains after kindling and stained for lactate dehydrogenase A (LDHA, enzyme that catalyzed pyruvate to lactate conversion) to confirm expected lactate upregulation.

Results: We were able to use hpMRI to identify significant elevations in lactate in in vitro and in vivo models after repeated neuronal activation and epileptic activity, respectively. Repeated chronic neuronal activation with low Mg²⁺ hpMRI demonstrated a 2-fold elevation in lactate/pyruvate levels compared to control (n=5, p=0.016, paired t-test). This was corroborated with supernatant measurements as lactate concentrations were found to be significantly elevated in low Mg²⁺ treated cells (n=3, p=0079). Furthermore, PTZ kindled mice demonstrated increased lactate conversion 1.5X relative to vehicle treated mice, and LDHA was significantly elevated in PTZ-treated mice (p=0.007). Together, hpMRI was able to detect elevated lactate levels both in vivo and in vitro.

Conclusions: In this study, hpMRI was investigated as a potential tool to detect lactate elevations after chronic neuronal activation and seizure activity. To our knowledge, this study represents the first pre-clinical data suggesting that hpMRI may be useful in identifying epileptic tissue and thus the SOZ and epileptic network. Future experiments will be needed to define the clinical utility of this imaging tool in patients.

Funding: UMB