Alexander Glushakov

Glutamate, the major excitatory neurotransmitter, is critical for normal brain development and function. Both extremes of glutamate receptor activity are detrimental for the brain. Glutamate's role in excitotoxicity has driven the search for receptor antagonists as neuroprotective agents, most of which have failed to achieve clinical, i.e. efficacious and safe, neuroprotection. High selectivity and potency provide potential explanations for this failure. For example, targeting individual glutamate receptor subtypes leaves other pathways of glutamatergic excitotoxicity intact. Furthermore, potent depression of glutamate receptor activity causes clinical side effects, such as the symptoms of schizophrenia produced by NMDA receptor antagonists. To produce efficacious neuroprotection devoid of significant side effects, it may be necessary to normalize the function of all components of the glutamatergic system, instead of blocking a single type of glutamate receptors. Halogenated der...

An increasing body of evidence supports the hypothesis that diminished function of N-methyl-D-aspartate (NMDA) receptors and the associated increase in glutamate release and overstimulation of alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA)/kainate receptors are critical elements of the pathophysiology of schizophrenia. Here, we describe a halogenated derivative of the aromatic amino acid L-phenylalanine that 1) activates NMDA receptors, 2) depresses presynaptic glutamate release, and 3) blocks AMPA/kainate receptors. The experiments were conducted in rat cerebrocortical cultured neurons by using the patch-clamp technique. 3,5-Dibromo-L-phenylalanine (3,5-DBr-L-Phe) augmented NMDA miniature excitatory postsynaptic currents (mEPSCs) and activated the steady-state current, effects that were eliminated by NMDA receptor antagonists DL-2-amino-5-phosphonopentanoic acid and MK-801 (dizocilpine maleate; 5H-dibenzo[a,d]cyclohepten-5,10-imine). 3,5-DBr-L-Phe was a partial agonist at the glutamate-binding site of NMDA receptors with an EC50 of 331.6 +/- 78.6 microM and with an efficacy of 30.5 +/- 4.7% compared with NMDA. 3,5-DBr-L-Phe depressed both amplitude and frequency of AMPA/kainate mEPSCs. The IC50 of 3,5-DBr-L-Phe to inhibit AMPA/kainate mEPSC frequency was 29.4 +/- 4.3 microM. 3,5-DBr-L-Phe significantly decreased paired pulse depression of AMPA/kainate EPSCs and attenuated current activated by AMPA with higher efficacy at lower concentration of AMPA. 3,5-DBr-L-Phe neither affected GABA miniature inhibitory postsynaptic currents nor elicited action potentials. By enhancing NMDA receptor function, reducing glutamate release and blocking AMPA/kainate receptors 3,5-DBr-L-Phe represents a new type of polyvalent modulator of glutamatergic synaptic transmission with potential therapeutic applications.

This paper reviews recent results of our investigation of the mechanisms whereby hyperphenylalaninemia may cause brain dysfunction in classical phenylketonuria (PKU). Acute applications of L-Phe in rat and mouse hippocampal and cerebrocortical cultured neurons, at a range of concentrations found in PKU brain, significantly and reversibly depressed glutamatergic synaptic transmission by a combination of pre- and postsynaptic actions: (1) competition for the glycine-binding site of the N-methyl-D-aspartate (NMDA) receptors; (2) attenuation of neurotransmitter release; (3) competition for the glutamate-binding site of (RS)-amino-3-hydroxy-5-methyl-4-isoxazolepropioinic acid and kainate (AMPA/kainate) receptors. Unlike L-Phe, its non-tyrosine metabolites, phenylacetic acid, phenylpyruvic acid, and phenyllactic acid, did not produce antiglutamatergic effects. L-Phe did not affect inhibitory gamma-aminobutyric (GABA)-ergic transmission. Consistent with this specific pattern of effects caused by L-Phe in neuronal cultures, the expression of NMDA receptor NR2A and AMPA receptor Glu1 and Glu2/3 subunits in brain of hyperphenylalaninemic PKU mice (Pah(enu2) strain) was significantly increased, whereas expression of the NMDA receptor NR2B subunit was decreased. There was no change in GABA alpha1 subunit expression. Considering the important role of glutamatergic synaptic transmission in normal brain development and function, these L-Phe-induced changes in glutamatergic synaptic transmission in PKU brain may be a critical element of the neurological symptoms of PKU.

To explore the hypothesis that L-phenylalanine (L-Phe) depresses glutamatergic synaptic transmission and thus contributes to brain dysfunction in phenylketonuria (PKU), the effects of L-Phe on spontaneous and miniature excitatory postsynaptic currents (s/mEPSCs) in rat and mouse hippocampal and cerebrocortical cultured neurons were studied using the patch-clamp technique. L-Phe depressed the amplitude and frequency of both N-methyl-D-aspartate (NMDA) and non-NMDA components of glutamate receptor (GluR) s/mEPSCs. The IC(50) of L-Phe to inhibit non-NMDAR mEPSC frequency was 0.98 +/- 0.13 mM, a brain concentration seen in classical PKU. In contrast, D-Phe had a significantly smaller effect, whereas L-leucine, an amino acid that competes with L-Phe for brain transporter, had no effect on mEPSCs. Unlike GluR s/mEPSCs, GABA receptor mIPSCs were not attenuated by L-Phe. A high extracellular concentration of glycine prevented the attenuation by L-Phe of NMDAR current, activated by exogenous agonist, and of NMDAR s/mEPSC amplitude, but not of NMDAR s/mEPSC frequency. On the other hand, L-Phe significantly depressed non-NMDAR current activated by low but not high concentrations of exogenous agonists. Glycine-independent attenuation of NMDAR s/mEPSC frequency suggests decreased presynaptic glutamate release caused by L-Phe, whereas decreased amplitudes of NMDAR and non-NMDAR s/mEPSCs are consistent with competition of L-Phe for the glycine- and glutamate-binding sites of NMDARs and non-NMDARs, respectively. The finding that GluR activity is significantly depressed at conditions characteristic of classical PKU indicates a potentially important contribution of impaired GluR function to PKU-related mental retardation and provides important insights into the potential physiological consequences of impaired GluR function.

The effects of the halogenated aromatic amino acid 3,5-dibromo-D: -tyrosine (3,5-DBr-D: -Tyr) were studied in rat models of stroke and epileptic seizures caused by middle cerebral artery occlusion (MCAo) through respective intracerebral injection of endothelin-1 (ET-1) and intraperitoneal (i.p.) injection of pentylenetetrazole (PTZ). 3,5-DBr-D: -Tyr was administered as three bolus injections (30 or 90 mg/kg, i.p.) starting at 30, 90, and 180 min after ET-1 administration or as a single bolus (30 mg/kg, i.p.) 15 min prior to PTZ administration. Neurological deficits and infarct volume were estimated 3 days after ET-1 administration and seizure score was assessed during the first 20 min after PTZ administration. The safety of 3,5-DBr-D: -Tyr was evaluated in control animals using telemetry to measure cardiovascular parameters and immunostaining to assess the level of activated caspase-3. 3,5-DBr-D: -Tyr significantly improved neurological function and reduced infarct volume in the brain even when the treatment was initiated 3 h after the onset of MCAo. 3,5-DBr-D: -Tyr significantly depressed PTZ-induced seizures. 3,5-DBr-D: -Tyr did not cause significant changes in arterial blood pressure, heart rate and spontaneous locomotor activity, nor did it increase the number of activated caspase-3 positive cells in the brain. We conclude that 3,5-DBr-D: -Tyr, by alleviating the deleterious effects of MCAo and PTZ in rats with no obvious intrinsic effects on cardiovascular parameters and neurodegeneration, exhibits promising potential as a novel therapeutic direction for stroke and seizures.