1. Cell-permeant Ca2+ chelators such as 1,2-bis-(2-amino-phenoxy)ethane- N,N,N',N'-tetraacetic acid acetoxymethyl ester (BAPTA-AM) protect neurons against excitotoxic and ischemic neuronal injury in vitro and in vivo. Here we provide the first steps toward characterizing the mechanisms by which these agents produce their neuroprotective effects. 2. Cultured mouse spinal neurons were simultaneously loaded with the Ca2+ indicator fura-2 and with one of three permeant chelators derived from the fast Ca2+ buffer BAPTA, or with ethylene glycol-bis(beta-aminoethyl ether)-N,N,N',N'-tetraacetic acid acetoxymethyl ester (EGTA-AM). Adding these chelators did not interfere with the fluorescence spectrum of fura-2 and had no effect on baseline [Ca2+]i. 3. The neurons were challenged with 250 microM L-glutamate for 50 min, producing a marked transient [Ca2+]i increase followed by a decay of [Ca2+]i to a lower "plateau." About 80% of control neurons succumbed to this excitotoxic insult. Neurons that survived adjusted their plateau [Ca2+]i to lower levels than those that succumbed. 4. Neurons that were pretreated with permeant Ca2+ chelators became more resistant to these neurotoxic challenges. 5. We examined whether this reduction in glutamate neurotoxicity could be related to the given buffer's known Ca2+ affinity (Kd), its Ca2+ binding kinetics, and its ability to attenuate glutamate-induced [Ca2+]i increases. 6. Pretreatment of neurons with BAPTA analogues having Kds ranging from 100 to 3,600 microM 1) attenuated the amplitude and 2) lengthened the time constant describing the rise and decay of the glutamate-evoked [Ca2+]i transient. The magnitude of these effects paralleled the affinity of the chelator for Ca2+. 7. BAPTA-AM and its analogues dramatically attenuated the early neurotoxicity of glutamate, reducing cell deaths by up to 80%. However, in contrast with the graded effects of chelators having different Ca2+ affinities on Ca2+ transients, all BAPTA analogues were equally protective. These protective effects did not relate to the chelators' Ca2+ affinity within a Kd range of 100 nM (for BAPTA) to 3,600 nM (for 5,5'-dibromo BAPTA). 8. BAPTA-AM protected neurons in a concentration-dependent manner with 50% protection obtained with 10 microM, a concentration having no effect on the [Ca2+]i transient amplitude. 9. EGTA, a slow Ca2+ buffer with a similar Ca2+ affinity to BAPTA produced the same effects as BAPTA on [Ca2+]i transient kinetics. However, it was far less protective than BAPTA. 10. The time course of early glutamate neurotoxicity was altered by the BAPTA analogues, but not EGTA. BAPTA analogues caused a small increase in cell deaths in the first minutes of each experiment, followed by relative sparing from further neurodegeneration. 11. The ability of low Ca2+ affinity chelators such as 5,5'-dibromo BAPTA to protect neurons without markedly attenuating measured [Ca2+]i increases conflicts with the hypothesis that global elevations in [Ca2+]i are responsible for triggering neurotoxicity.(ABSTRACT TRUNCATED AT 400 WORDS)