A series of 3-substituted acetic and propanoic acid derivatives of 5,7-dichlorokynurenic acid were prepared and evaluated as antagonists of the glycine site on the NMDA receptor complex. They were found to possess reduced affinity for the glycine site relative to 5,7-dichlorokynurenic acid and the analogous β-carboxyindole-acetic acid series of glycine antagonists. The involvement of N-methyl-D-aspartate (NMDA) receptor activation in a variety of neurodegenerative and cognitive processes has been well established. Recent studies have lent good pharmacological evidence for the therapeutic potential of modulators of the NMDA receptor which act via the glycine site. For example, activation of the glycine receptor is required for induction of LTP in the neonatal hippocampus. D-Cycloserine, a glycine partial agonist, has been shown to enhance task learning in rats. Another partial agonist, 1-aminocyclopropane-1-carboxylic acid (ACPC) effectively blocked NMDA-induced convulsions without eliciting any overt behavioral effects of its own. Furthermore, glycine antagonists have proven active in a variety of seizure models. Other reports have shown glycine antagonists protect against NMDA induced neurotoxicity in vitro and neurodegeneration in vivo. 5,7-Dichlorokynurenic acid (1b) and 2-(2-carboxy-6-chloroindol-3-yl)acetic acid (2c) are two potent and selective antagonists of the glycine site. Recently, McDonald and coworkers have reported on the increased potency of the 4,6-dichloro substituted 3-(2-carboxyindol-3-yl)propanoic acid 2e relative to 2c. Additionally, they have described a kynurenic acid analogue 1c which possesses a heteroatom-linked acetic acid group at C-4. In an effort to further explore the structural commonality between these series, we have synthesized and evaluated kynurenic acid derivatives with an acetic or propanoic acid moiety appended to the 3-position as represented by 3a-c. The 2-(2-carboxy-4-hydroxyquinolin-3-yl)acetic acids 3a and 3b were prepared as shown in Scheme I. The appropriate aniline was condensed with diethyl ethoxalylsuccinate (4a) in the presence of 10 mol% glacial acetic acid to give the methyl 1-anilino-2-carboxy-1,2-dihydrosuccinates 5a and 5b. The water which formed was azeotroped off with refluxing benzene using a Dean and Stark trap. The intermediates 5a and 5b were then purified by flash chromatography before cyclizing in refluxing diphenyl ether to afford 6a and 6b, respectively. Saponification gave the corresponding diacids 3a and 3b. The 3-propanoic acid derivative 3c was prepared analogously except that methyl 2-oxalylglutarate (4b) was used in the initial condensation step. Compound 8, which lacks a 4-hydroxy substituent, was obtained by refluxing 6a in POCl3 to yield 7 followed by hydrogenolysis and saponification. The 2-carboxyindole-3-acetic acid derivatives 2b and 2d were synthesized using standard methodology as summarized in Scheme II. In the case of 2d, the hydrazone intermediate was isolated prior to cyclization with p-toluenesulfonic acid in refluxing benzene. All compounds exhibited physical chemical data and elemental analyses consistent with their assigned structures. The compounds were evaluated for their ability to displace [3H]-glycine at a screening dose of 10 μM. Ki values were determined for those compounds which showed greater than 50% displacement at 10 μM (Table 1). Compounds possessing affinity for the glycine site were shown to be antagonists by virtue of their ability to inhibit [3H]-MK801 binding. As previously reported, we found 2-(2-carboxyindol-3-yl)acetic acid (2b) possessed affinity for the glycine site approximately equal to that of kynurenic acid (1a) while quinoline 8, a ring expanded version of 2b, lacked appreciable affinity for the glycine site. Addition of the 3-acetic acid side chain to kynurenic acid (1a) to give 3a led to a modest loss of affinity for the glycine site, while substitution of 5,7-dichlorokynurenic acid (1b) with the 3-acetic acid side chain to give 3b resulted in a marked decrease in affinity for the glycine site. Further homologation of the 3-acetic side chain to give the 3-propanoic acid derivative 3c led to a further loss of affinity for the receptor. This is in contrast to the 4,6-dichloro substituted indole series where a modest increase in affinity is associated with the introduction of the 3-acetic acid moiety 2d or the 3-propanoic acid moiety 2e. Gray et al. showed a similar increase in potency upon introduction of the 3-acetic acid moiety to the 6-chloroindole series. These results suggest that the 3-acetic acid moiety appended to 5,7-dichlorokynurenic acid is unable to reach the same distal hydrogen bonding site that has been postulated for the favorable interaction of the indole 3-acetic acid group.