11 Nov 1998

Since the development of transgenic models of β-amyloid plaque formation, considerable debate has developed over the apparent paucity of neuronal cell loss found in the majority of these models. One hypothesis for why there is an overall lack of neurotoxicity in these models is that mice overexpressing βAPP are unable to mount an efficient inflammatory response to the overproduction of the Aβ peptide. Previous work has shown that the β-amyloid peptide activates the complement cascade by binding to the A chain of C1q, a collagen-like region of this molecule that typically mediates non-antibody-mediated complement cascade activation. In a study presented by David Cribbs, the ability of mouse C1q to bind human Aβ peptide and activate a complement response was examined (Abstract 592.8). Sequence analysis revealed that the mouse C1q A chain is missing two of the three arginine residues that are essential for complement activation by the human C1q A chain. Further, biochemical data demonstrated that approximately twice as much Aβ peptide is required by mouse C1q as is required by human C1q for equivalent complement activation. This observation does not reflect a general deficiency of mouse complement, as the antibody-dependent activation of mouse C1 is actually more efficient than that in the human.

These data may suggest that additional modifications of the mouse genome may be necessary to mimic the activation of inflammatory mechanisms by the Aβ peptide observed in the human Alzheimer's brain. An additional presentation on this topic by K. L. Wright (Abstract 592.9) supports this contention. In this work, the complement sufficiency of PS1/APP double transgenic mice was enhanced by backcrossing onto a BUB/BnJ background. The BUB/BnJ mouse strain has been shown to exhibit a much greater capacity for complement activation than other mouse strains. First-generation backcross mice exhibited a dramatic increase in microglial response to congophilic Aβ deposits and an overall increase in microglial response in cortex as observed by the distribution of immunoreactivity. No change in either Aβ deposition or GFAP staining was noted. Further studies may yield additional insights into the importance of inflammatory mechanisms in AD pathogenesis.—Aileen Anderson