br Introduction The amino acid glycine functions as a

Introduction The amino 3-bromo-5-phenyl Salicylic Acid glycine functions as a neurotransmitter in the vertebrate central nervous system: as an inhibitory neurotransmitter and as a co-agonist of the NMDA-subtype of excitatory glutamatergic receptors (Eulenburg et al., 2005). Under a current hypothesis that schizophrenia reflects dysfunction in glutamatergic synapses of the central nervous system, it has been proposed that increasing the tonic level of glycine at glutamatergic synapses may be therapeutic for schizophrenia (Marino and Conn, 2002, Sur and Kinney, 2007, Marek et al., 2010). Glycine is removed from synapses by two Na+,Cl--coupled transporters, GlyT1 and GlyT2. Both are expressed in glycinergic brain regions, GlyT1 primarily by glial cells and GlyT2 by glycinergic neurons (Eulenburg et al., 2005, Marek et al., 2010). GlyT1 is also expressed in brain regions rich in glutamatergic synapses. Confocal microscopy co-localizes GlyT1 with vGLUT1, a vesicular glutamate transporter and marker of glutamatergic neurons. Furthermore, anti-GlyT1 serum co-immunoprecipitates NMDA receptor whereas anti-GlyT2 and pre-immune sera do not (Cubelos et al., 2005, Raiteri and Raiteri, 2010). GlyT1 inhibition does indeed enhance NMDA receptor currents in rat brain slices, even when the slices are perfused with saturating concentrations of glycine (Bergeron et al., 1998). Under the glutamatergic hypothesis, these data suggest that GlyT1 inhibitors are potential therapeutics for schizophrenia. Small clinical trials have been reported for both natural agonists of the Gly site on NMDA receptor, and for the GlyT1 inhibitor sarcosine (Javitt, 2009, Marek et al., 2010). Because glycine plays multiple roles in the central nervous system, the potential exists for adverse side effects. Perry et al. (2008) reported that rats dosed with the GlyT1 inhibitors ALX-5407 or LY2365109 exhibited locomotor hyperactivity (impaired gait, compulsive toe walking) and impaired respiration (labored breathing). On the other hand, Depoortère et al. (2005) reported no such effects in rats dosed with the GlyT1 inhibitor SSR504734. This complexity highlights the need to attend to factors that may affect the balance between therapeutic efficacy and side-effect liability of GlyT1 inhibitors. Alternative promoter usage and splicing of GlyT1 mRNA generates three different NH2-terminal isoforms in both rodents and humans (Kim et al., 1994, Adams et al., 1995). Two of these variants, GlyT1(652) and GlyT1(633), have differential tissue-specific patterns in rat white matter and grey matter (Borowsky et al., 1993; isoforms there designated Gly-T1 and Gly-T2). Recombinant expression in some polarized cell types reveals that different isoforms are targeted to different cell surfaces (Poyatos et al., 2000). While these data suggest that different GlyT1 isoforms may differentially impact different systems, their physiological roles are not well understood. Investigation of the roles of GlyT1 isoforms is hampered by the lack of pharmacological tools, namely isoform-specific GlyT1 inhibitors. In the literature, inhibitor potencies are usually reported against a single GlyT1 isoform. Atkinson et al. (2001) reported that ALX-5407 inhibited all three GlyT1 isoforms with equal potency, but two of the three isoforms used in that work were the incorrect isoforms discussed under Experimental Procedure. To enable the discovery of isoform-specific GlyT1 inhibitors, we have developed cell lines stably expressing each of the NH2-terminal isoforms of GlyT1. We report here the validation of GlyT1 uptake assays in these cell lines. We further report that reference GlyT1 inhibitors, and antipsychotic drugs that are weak GlyT1 inhibitors, provide only minimal differentiation among GlyT1 isoforms.
Results The three isoforms of GlyT1 are designated GlyT1(633), GlyT1(652), and GlyT1(706). The numbers in parentheses denote the number of amino acid residues in each isoform. CHO-K1 cell lines expressing each isoform were selected, and their GlyT1 activity was measured by uptake of [3H]Gly. For each isoform of GlyT1, two cell lines were chosen for detailed evaluation, one with higher GlyT1 activity than the other. Km and Vmax values were determined for all six cell lines (Table 1). The Km values for the substrate Gly varied only modestly, and subsequent assays were conducted at [Gly]total=60μM, close to the Km value for all six cell lines. We verified that the time course of Gly uptake was linear through 20min at [Gly]total = 60μM for all six cell lines (Fig. 1). Subsequent assays were conducted as 20min endpoint assays.