Index | Round
Tables | Selected
Abstracts | News
Cellular/Animal Models I
by Brian J. Cummings (UC Irvine)
July 18, 1998 Soto (abstract 295) presented data documenting the ability of their 11 and 5 amino-acid inhibitors to inhibit initial Abeta peptide aggregation and its ability to dissociate already formed Abeta fibrils in vitro. In addition to the assays mentioned in the abstract, this inhibition of fibril formation is also evident by EM analysis and protection via live cell/dead cell staining has been extended to 6 days in culture where 50 micromolar concentrations of inhibitor alone were not toxic to cells. He also presented additional data on the effects of Abeta injections into rat amygdala. Injections alone (still an N=9) resulted in deposition of Congo positive deposits of Abeta as well as detection of fibrils at the EM level. Cresyl violet staining shows hypertrophied cells (and frank cell death) and staining for microglia (IL1beta) and astrocytes (GFAP) revealed both local and more distant increases resulting from the injection. When inhibitor was co-injected with Abeta, the amyloid deposits were 50% smaller and Congo red and Thioflavine negative. (It was not clear whether the glial response was also reduced).
While it is already counter intuitive that any peptide could actually reverse already formed fibrils (as suggested by their in vitro data), this effect was also demonstrated in vivo! Additional animals were injected with Abeta alone and 8 days later, the iAbeta5 inhibitor was injected into the original site. Even under these conditions, there was an extensive reduction in the area occupied by Abeta. Soto concluded with the tantalizing suggestion that this approach might also by applied to other protein conformational folding disorders (e.g. prions) and hinted that work in progress is promising. During the Q&A, he clarified that to be effective in vitro, the inhibitors were used in a 1 to 1 molar ratio with Abeta while in vivo they used a 20 molar excess. Dr. Soto also indicated that these inhibitors can cross the blood brain barrier and that there are some modifications to the peptide that can be made (have they been made?) to increase its influx. So my question is: what's the name of your start-up company and when is the IPO?
Xu (abstract 296) reported that 17beta estrodiol was effective in reducing both Abeta40 and Abeta42 levels as measured via mass spectrometry, and that the effects of estrogen were time-dependant (over the course of many days). He also reported that pretreatment over many days resulted in increases in APP/Abeta within vesicles in the TGN and hypothesized that this "sequesterization" of APP/Abeta may reduce the amount of available substrate for beta and gamma secretases, thus explaining estrogen's protective effect. Dennis Selkoe asked about a dose response effect (these data were not shown) and the level of effect, because his group has experienced high variability in Abeta levels between different neuroblastoma cell lines. From the data Dr. Xu described, this issue appears to be unresolved. Another person asked if 17alpha estradiol has a similar effect. The answer with regards to APPs is yes, but the investigators have not measured Abeta yet.
Suzuki (abstract 297) presented evidence that phosphorylation of APP plays a role in neurite development. I missed the first part of his talk concerning the role of APP770/Arg672 in regulating the metabolism of APP, but returned to hear that the T668E mutation's effects on neurogenesis were demonstrated by measuring neurite number and neurite length via immunocytochemistry for MAP5 (both of which were reduced compared to controls). Someone asked if the Arg672 site is close to the motif for trafficking via the clathrin pathway. The short answer was that in non-neuronal cells, internalization of APP is normal.
Baumeister (abstract 302) presented data on the effects of mutations in sel-12 on cellular function, morphology and behavior. Sel-12 is the C. elegans homologue to human presenilin. Since there is a caspase cleavage site on PS, it is a potential target site for therapeutic intervention. Sel-12 mutant animals have a lethal egg-laying deficit which results in the eggs hatching inside the mother. Expression of human PS1 or PS2 in sel-12 mutant C. elegans can block this lethal effect, but if mutant (FAD) human PS was used, there was no rescue effect. This also indicates that C. elegans is capable of processing human PS (and PS mutations) despite sequence differences. They also made mutants to block the caspase cleavage site on PS1 and PS2 and found no effect on the function of PS in C. elegans. Immunocytochemically, sel-12 was shown to be expressed in a variety of cell types, including the CNS (if you can call 302 neurons a CNS). There were no substantial differences within the CNS of sel-12 mutants, but subtle differences were observed in the neurite branching pattern of specific neurons. Since theses neurons are known to play a role in "temperature memory", they tested mutants and found they were deficient on this task. They were also impaired in the reversal response to noxious stimuli. They have not looked for synaptic changes yet.
Following Baumeister's talk, there was a spirited discussion between Christian Haass and Rudy Tanzi on whether caspase cleavage plays a role in the generation of Abeta. Haass argued that caspase cleavage plays no role while Tanzi remained unconvinced. Perhaps at basal levels of expression, cells might not need caspases to generate Abeta, but the story might be different in stressed cells [whose opinion?]. They finally agreed to jointly state that "caspase cleavage is not required for basal production of Abeta in healthy cells" and concurred that "the caspase derived c-terminal fragment of PS1 is definitely BAD for cells". Stay tuned to the two labs for details [further developments?]