Memory impairment in rats by hippocampal administration of the serine protease subtilisin.

Since the serine protease subtilisin has been reported to generate a novel form of long-term depression (LTD) in rat hippocampal slices, the present work was designed to determine whether it has any effect on learning and memory processes. Rats were used to examine the effects of subtilisin, injected directly into the dorsal hippocampus, on task performance in a step-through inhibitory avoidance of a mild footshock. The administration of 100 ng of subtilisin into each hippocampus, immediately after training, was sufficient to induce a detectable learning deficit with a footshock stimulus of 0.5 mA. Higher doses produced dose-related impairments in memory consolidation. These effects were not the result of irreversible toxicity, since rats trained with a higher amplitude footshock (0.75 mA) were able to perform as control animals; therefore, the amnesic effect was not further evident. Furthermore, the administration of subtilisin before avoidance training did not produce any detectable effect on performance during the training or test sessions, indicating that neither acquisition nor consolidation was affected. It is concluded that the post-training administration of a serine protease inhibitor is able to produce robust deficits of memory consolidation consistent with its ability to generate LTD, raising the possibility that related molecules could play physiological or pathological roles in the modulation of learning and memory.

Discovery and characterization of cnidarian peptide toxins that affect neuronal potassium ion channels.

Peptides have been isolated from several species of sea anemones and shown to block currents through various potassium ion channels, particularly in excitable cells. The toxins can be grouped into four structural classes: type 1 with 35-37 amino acid residues and three disulphide bridges; type 2 with 58-59 residues and three disulphide bridges; type 3 with 41-42 residues and three disulphide bridges; and type 4 with 28 residues and two disulphide bridges. Examples from the first class are BgK from Bunodosoma granulifera, ShK from Stichodactyla helianthus and AsKS (or kaliseptine) from Anemonia sulcata (now A. viridis). These interfere with binding of radiolabelled dendrotoxin to synaptosomal membranes and block currents through channels with various Kv1 subunits and also intermediate conductance K(Ca) channels. Toxins in the second class are homologous to Kunitz-type inhibitors of serine proteases; these toxins include kalicludines (AsKC 1-3) from A. sulcata and SHTXIII from S. haddoni; they block Kv1.2 channels. The third structural group includes BDS-I, BDS-II (from A. sulcata) and APETx 1 (from Anthropleura elegantissima). Their pharmacological specificity differs: BDS-I and -II block currents involving Kv3 subunits, while APETx1 blocks ERG channels. The fourth group comprises the more recently discovered SHTX I and II from S. haddoni. Their channel blocking specificity is not yet known but they displace dendrotoxin binding from synaptosomal membranes. Sea anemones can be predicted to be a continued source of new toxins that will serve as molecular probes of various K(+) channels.