The scientists microinjected into rat hippocampus these constructs: the full-length mutant human APPswe gene, various control constructs and, most importantly, an Aβ gene construct originally developed by Todd Golde called BRI-Aβ42 fusion. It is modified in such a way that Aβ42 gets expressed in the secretory pathway without having to be cleaved first from APP. This allowed a comparison of the effects of APP overexpression, which includes the generation of all its attendant cleavage products, with just Aβ42 alone.

A month later, the rats expressed high (but not yet quantified) levels of Aβ42 in the subiculum, CA1, CA3 and granule cell layers of the dentate gyrus of the hippocampus. The peptide also appeared in the entorhinal cortex, probably because some projecting neurons picked it up in the hippocampus and transported it back. The pathology of these rats is not yet fully analyzed.

The key finding of the study was a robust decline in the performance of Aβ42-expressing rats in the Morris water maze, where they showed deficits at both three and six months after the injection. Rats expressing the full mutant APPswe behaved like the controls in this test. Intriguingly, the poster noted that a trend toward poorer performance showed up three days after injection, but that data was not shown. Moreover, the Aβ42-expressing rats also showed a clear deficit in a water maze test of working memory.

Curiously, both the APPswe- and Aβ42-expressing rats were much more physically active than were controls, but it is not clear whether they ran around more because of anxiety or because there really is some mysterious effect of these transgenes on locomotor activity, Bland said.—Gabrielle Strobel.