The effects of aging on the responsiveness of hippocampal neurons to iontophoretic application of L-glutamate and AMPA were studied in vitro. There were no effects of age on neuronal responses to L-glutamate; however, CA1 pyramidal cells of old rats, but not granule cells in the fascia dentata, showed both a smaller reduction in extracellularly-recorded synaptic responses following application of AMPA (presumably mediated by depolarization), and smaller extracellular "DC" fields (measured by subtracting the DC potentials at the dendrite and soma following AMPA application in the dendrites). To examine the cellular bases of this age-related alteration in AMPA sensitivity, two additional electrophysiological approaches were used: (1) measurement of the amplitude ratios of extracellular EPSP and fiber potential components of the Schaffer collateral-CA1 response; (2) measurement of intracellularly recorded unitary EPSPs and quantal analysis of their fluctuations. The interpretations that would be placed on four hypothetical possible outcomes of such experiments are outlined and assessed in relation to the experimental data. The pattern of results obtained in the present experiments supports the following conclusions: In old rats, individual Schaffer collateral synapses do not appear to have altered AMPA receptor properties, as neither the mean size of the unitary synaptic response nor the apparent quantal size differs between age groups; however, the data do support the conclusion that there are fewer synapses per Schaffer collateral branch in old versus young CA1 pyramidal cells.

1. We recorded extracellular and intracellular CA3-CA1 synaptic responses in hippocampal slices from neonatal rats [postnatal day (P) 15-21 and P29-35]. Presynaptic function was examined by measuring input-output relationships and paired-pulse facilitation and by quantal analysis of minimally evoked responses. 2. Extracellular recording revealed no difference in excitatory postsynaptic potential (EPSP) threshold or the fiber potential response for a given stimulus intensity between the two age groups. However, the slope of the field EPSP was consistently larger in older animals. The increase in EPSP slope was associated with a decrease in paired-pulse facilitation, suggesting an increase in presynaptic function with postnatal development. 3. Extracellular results were confirmed by intracellular recordings that revealed no difference in the minimal stimulation intensity needed to evoke a response, an increase in mean EPSP amplitude with development, and a decrease in paired-pulse facilitation. Quantal parameters were extracted by three separate methods including method of failures, coefficient of variance, and parameter optimization through noise deconvolution. All methods supported presynaptic mediation of facilitation. Comparison of quantal parameters during development indicated an increase in mean quantal content. 4. The results demonstrate that synaptic strength is altered over the course of development because of, at least in part, changes in presynaptic release mechanisms. Developmental differences in presynaptic function provide an explanation of differences in mechanisms for expression of long-term potentiation. The lower initial probability of transmitter release in neonates may permit increased presynaptic change.

Although there is no question that age-related alterations in neural function occur in the central nervous system (CNS) of mammals, these changes tend to be somewhat circumscribed rather than completely global, even in pathological conditions. Examples of this kind of specificity of functional neural change are drawn from experiments on the electrophysiology of the senescent rat hippocampus.

In the Perspective "Too many rodent carcinogens: Mitogenesis increases mutagenesis" by Bruce N. Ames and Lois Swirsky Gold (31 Aug., p. 970), the last paragraph on page 970 (continuing on page 971) was incorrectly printed. It should have read, "One major group of natural chemicals in the human diet are the chemicals that plants produce to defend themselves, the natural pesticides (4). We calculate that 99.9% (by weight) of the pesticides in our diet are natural. Few natural pesticides have been tested in at least one rodent species, and again about half (27/52) are rodent carcinogens. These 27 occur commonly in plant foods (10). The human diet contains thousands of natural pesticides, and we estimate that the average intake is about 1500 mg per person per day (4). This compares to a total of 0.09 mg per person per day of residues of about 100 synthetic pesticides (4). In addition, of the mold toxins tested at the MTD (induding aflatoxin), 11 out of 16 are rodent carcinogens." Also, in paragraph 3 on page 970, "47,000 8-hydroxydeoxyguanosines per cell" should have been "90,000" per cell.