Abstracts

Rationale: The reversal potential of GABA receptors (E_GABA) is dependent upon the chloride concentrations on both sides of the neuronal membrane. The chloride on the extracellular aspect of the membrane is canonically considered to equal the chloride in the bulk cerebrospinal fluid.  However, neurons are surrounded by an extracellular matrix comprised of variably sulfated glycosaminoglycans that can alter local chloride concentration and can be hydrolyzed by active matrix metalloproteinases released by tissue injury.

Methods: We synthesized a single-wavelength, pH-insensitive chloride-sensitive fluorophore and constrained it to the extracellular space by conjugation with 10 kilodalton dextran. We then used 2-photon Fluorescence Lifetime IMaging (FLIM) to measure extracellular chloride in acute and organotypic cultures of hippocampal slices, with subsequent confirmation in vivo through both murine and porcine cortical windows. We used slice injury as well as 2-photon photolysis of single neurons within organotypic slices to model acute brain injuries and again confirmed in vivo.

Results: We first tested whether the extracellular sulfate moieties change the baseline local chloride concentration.  We found that the extracellular chloride ([Cl^-]_o) between neurons in the depths of acute hippocampal slices and at all depths of organotypic hippocampal slice cultures was only half of the bulk CSF chloride. We confirmed this finding in vivo. We next measured [Cl^-]_o after the sulfate moieties in the matrix were freed by either exogenous chondroitinase ABC or endogenous matrix metalloproteinases (MMPs) after brain injury. We found a strong dependence of [Cl^-]_o vs distance from injury, with Cl concentration increasing to the ACSF levels near the injured surface of acute slices or proximity to photolysed neurons in organotypic slices, respectively. These changes in [Cl^-]_o should also alter the neuronal intracellular chloride via the activity of the high-velocity equilibrative membrane chloride transporters.  We have previously reported such changes in slices and confirmed them in both injury models here, in addition to an in vivo piglet model for which preliminary data is presented.  If the injury-induced increases in extra- and intracellular chloride were due to release of extracellular sulfates and replacement by chloride, these sulfates should be released to the perfusate, and we confirmed this using colorimetric assays of chondroitin sulfate.  Finally, the release of sulfates should be inhibited by MMP antagonists, and we confirmed that broad-spectrum inhibition using the zinc chelator ZX-1 or the more specific MMP-2/9 inhibitor SB3CT also reduced the extracellular and intracellular chloride concentration and neuronal volume at the surface of cut slices and in proximity to photolysed neurons.

Conclusions: [Cl^-]_o is partially displaced by sulfates in the extracellular matrix. Damage to the extracellular matrix following brain injury alters the distribution of chloride in both the extra- and intracellular spaces. These findings have immediate implications for the treatment of cytotoxic edema and seizures after acute brain injury.

Funding: R35116852: Neuronal ion and volume shifts after acute brain injury