21 Nov 2003

The role of inflammation in Alzheimer’s disease—beyond serious dispute, yet still enigmatic—gained much attention at the 33rd Annual Meeting of the Society for Neuroscience, held last week in New Orleans. While immunotherapy labs are retooling their approaches (see ARF related news story), other studies approached the functions of a key inflammatory operative—the microglial cell—by means of gene expression analysis. For example, Anna Parachikova, working with Carl Cotman at University of California, Irvine, presented a poster (629.26) suggesting that MHC class 2 antigens are upregulated in hippocampus early on in AD. She extracted mRNA from postmortem frontal cortex and hippocampus of seven people with mild dementia/MCI and 14 controls, and compared expression of 12,000 gene sequences. For further analysis, Parachikova used the genes in which expression changes came up as significant in two separate statistical analyses. Overall, those genes in which expression changed in the cortex tended to be upregulated (many of those were inflammatory markers), whereas gene changes in the hippocampus tended to be downregulations, Parachikova explained. A noticeable exception was MHC class 2 genes, in which expression in the hippocampus increased, as did their protein levels in Western blots. It is unclear whether this increased expression of genes involved in antigen presentation in the hippocampus occurs only on microglia or perhaps also on neurons, Parachikova said (see ARF related news story; see also Comparative Genomics section in ARF related news story). She suggests her study has picked up a region-specific increase in MHC 2 expression at the threshold to AD, which indicates that early in the disease the hippocampus contains microglia capable of devouring cells that present antigens together with MHC 2, and reinforces the view that inflammation is not a late event in the AD cascade.

That phagocytosing microglia would be prowling the AD hippocampus is neither new nor surprising, given that they are the resident macrophages in the brain. After all, most brain pathologies stimulate these cells in some way. Still, their role in neuroprotection versus destruction are ill-defined. Many scientists think that microglial activation is initially “adaptive” but gets out of hand in chronic neurodegeneration, in AD possibly egged on by the persistent presence of amyloid deposits. When does a protective activation tip over into a damaging one? Equally unclear are the functions of a microglial cell compared to those of infiltrating macrophages. To get a molecular handle on such questions, Monica Carson of the Scripps Research Institute in La Jolla, California, compared gene expression in resting versus activated microglia to identify genes that microglia express in response to various inflammatory stimuli (527.5). She then used in-situ hybridization to characterize expression of these genes in normal mice and models of multiple sclerosis and AD.

Among the genes upregulated in response to AD-related cytokines, she found the microglial receptor TREM-2 and its adaptor protein DAP-12. The function of these genes is poorly understood, but Carson said they rev up the microglia’s ability to present antigens and activate T cells. TREM-2 and DAP-12 expression in normal brain is highest in entorhinal cortex and hippocampus, and the APP23 transgenic mice from Novartis Pharmaceuticals showed robust expression of these genes in microglia near amyloid plaques. Overall, the profiles of gene expression among microglial populations were highly heterogeneous, varying by brain region and surrounding pathology. In addition to T cell reactions (Monsonego et al., 2003), individual variations in these microglial responses might help explain why a subset of patients developed inflammation in Elan’s halted vaccine trial, Carson suggested.

Among dozens of presentations on microglial biology and activation, several corroborated this notion of microglial heterogeneity. To name but one, Michael Dailey and colleagues at the University of Iowa in Iowa City imaged slices of mouse and rat hippocampus with two-channel fluorescence and time-lapse confocal microscopy (743.13). These authors noted marked differences in the expression of early activation markers such as NFkB or leukocyte function antigen (LFA-1), as well as in the microglia’s motility in response to injury. Together, these studies reinforce the complexity of the task at hand, namely, of turning to therapeutic advantage, or at least disarming, these still-enigmatic cells.

Does brain pathology lure outside macrophages? Another lingering question about microglia appears close to an answer, however. It is this: Are the activated microglia found near amyloid deposits prior residents of the brain, or have at least some of them recently immigrated from across the blood-brain barrier? Microglia derive from blood-derived monocytes made throughout life in the bone marrow. Tarja Lappetelainen, working with Jari Koistinaho and others at the University of Kuopio, Finland, brought classic immunology to bear on this issue (945.4). Using irradiation, the scientists first wiped out the immune system of 21-month-old APP/PS1 transgenic mice and non-transgenic controls, and then transplanted back the bone marrow from mice overexpressing green fluorescent protein. Four weeks later, they measured by flow cytometry whether the graft had taken hold and was producing blood cells; it was. Fourteen weeks later, they checked for green fluorescent cells in various brain areas. Both mice strains had equal numbers of GFP cells, indicating that bone marrow-derived cells continuously filter into the brain. While some of these cells appeared to cluster around amyloid plaques, on the whole this study detected no difference between the wild-type and APP/PS1-trangenic mice in the amount of infiltrated GFP-positive cells. Does that mean there is no specific response by the peripheral immune system to CNS amyloid pathology? Probably not. “The mice used in this study were very old, with fully established Aβ plaque pathology and related gliosis already at the start of the experiment. Therefore, it is not surprising that the extent of infiltration is not greater in the transgenic mice. Clearly, more studies are needed to investigate the effects of blood-derived microglia on the development and progression of Aβ pathology,” comments coauthor Milla Koistinaho. You can view abstracts mentioned in this story at the SfN/ScholarOne website.—Gabrielle Strobel.