Part 2 of a two-part series. See also Part 1.
15 February 2010. There’s more to the brain than neurons. Astrocytes and microglia play crucial roles in the development and maintenance of a healthy brain, and both have been studied for their potential protective and deleterious roles in Alzheimer disease. The field grapples with two fundamental questions: What do these cells do, and where do they come from? At Alzheimer’s Disease Beyond Aβ, this year’s Keystone Symposium held January 10-15 at Copper Mountain, Colorado, presenters tried to answer both. The general impression is that researchers are finally getting some traction in understanding the role of these cells and whether or not peripheral phagocytic monocytes can infiltrate the brain and influence pathological cascades (see part 1.[link to part 1 or first story])
One cell type that has been implicated in neurodegeneration is the astrocyte. Less involved in immune responses than microglia, astrocytes lend important trophic support to neurons and protect them from excitotoxicity (see ARF related news story). But they may also be involved in glial activation, and one of the proteins they release, the calcium-binding protein S100B, appears to be upregulated in AD (see Mrak and Griffinbc, 2001). S100B also surfaces around amyloid plaques in APP transgenic mice (see Sheng et al., 2000). To probe the role of the protein in AD, researchers led by Terrence Town, Cedars Sinai Medical Center, Los Angeles, and Takashi Mori at Saitama University, Kawagoe, Japan, crossed Tg2576 APP transgenic mice with animals that overexpress human S100B. At Copper Mountain, Town reported that at 15 and 19 months of age, the double-transgenic animals have deposited more and larger Aβ plaques in the cingulate cortex, hippocampus, and entorhinal cortex compared to Tg2576 controls. Blood vessels in the double- transgenic animals also contain numerous Aβ deposits ,and the mice have higher levels of soluble Aβ and of C-99, and sAPP-β fragments of APP generated by β-secretase (BACE) cleavage. The researchers confirmed elevated BACE activity when they measured it directly.
In addition to more amyloidogenic processing of APP, microgliosis (judged by Iba1 staining) and astrogliosis (judged by GFAP staining) emerged in the double transgenic mice by 19 months of age. But at 9 months, before the emergence of any Aβ pathology, proinflammatory cytokines, including tumor necrosis factor α, interleukin 1b (IL-1b), IL-6 and even mouse S100B were up. This work just appeared in the February Glia (see Mori et al., 2010). The timing of events, with inflammatory signals going off before Aβ begins to accumulate, suggests that brain inflammatory processes are not simply a consequence of plaques but may even drive cerebral β-amyloidosis, suggested Town.
Town also reviewed some of his recent work on TGF-β signaling, which suggests that blocking this pathway can bias peripheral mononuclear phagocytes toward non-inflammatory responses in AD models. The cells’ inflammatory tendencies are can be appeased by shunting TGFβ signaling away from the downstream transcription factors Smad2 and Smad3, and toward Smads 1, 5, and 8 (see ARF related news story). Together, the TGF-β and the S100B data demonstrate a delicate balance when microglia or macrophages come into contact with Aβ, suggested Town. He believes the balance might be struck to enable Aβ clearance without setting off an inflammatory cascade by blocking TGFβ signaling and is currently searching for small-molecule inhibitors that might be suitable for preclinical work. To this end, he has entered into a partnership with Novartis Inc, and with a medicinal chemistry lab at Yale University, Town told ARF.
Town’s data suggest that dialing down TGF-β signaling in peripheral mononuclear phagocytes may open the door for these cells to enter the brain and scavenge Aβ. The role of peripheral macrophages in the brain remains somewhat controversial, however, and its study has been hampered by technical challenges. For one thing, distinguishing brain-resident microglia from infiltrating circulatory macrophages is quite difficult because, when activated, the former express similar markers to the latter. One technique that has been used to explore the role of myeloid cells in the brain is to ablate myeloid-generating bone marrow by irradiation and then transplant new, traceable cells from another animal. In this way, researchers, including Josef Priller, Charité Universitätsmedizin Berlin, Germany, reported that only circulating monocytes expressing the chemokine receptor CCR2 and high levels of the cell surface marker Ly-6C are able to infiltrate the brain (see ARF related news story and Mildner et al., 2009). But questions remain as to whether circulating monocytes enter the brain because the blood brain barrier gets damaged by irradiation, as some researchers suspect, and whether those infiltrating cells have any impact on AD pathology. At Copper Mountain, Priller had some answers.
Priller has tested how bone marrow-derived cells infiltrate the brain using AD mouse models. He irradiated of APP/PS1 and APP23 transgenic mice, then transplanted bone marrow cells. He reported that four months later, bone marrow-derived cells only infiltrated the brain if the transplant was CCR2-positive. But he also found that the even these cells do not enter the brain if it was shielded from the radiation. In this brain-protected setting, plaque morphology was also different—many more microglia were seen to gather around plaques but the plaque load in both cortices and hippocampus was the same. When the whole body was irradiated, microglia appeared to be more distant from plaques and the total amount of Aβ in the brain was reduced, though there was no change in APP processing. The finding suggests that the whole-body radiation itself may leave the brain susceptible to infiltrating cells, perhaps through damage to the vasculature, and it suggests caution in interpreting results obtained with whole-body irradiation. But another interpretation is that the irradiation acts as a wake-up call, sensitizing the brain to microglia. In this regard, Town wondered if the microglia seemed more distant from plaques after total body irradiation because the plaques were being digested by the cells. “I think there could be something profound going on there. Perhaps the injury from irradiation is providing a secondary stimulus to mobilize plaque-clearing mononuclear phagocytes,” he told ARF.
One way to circumvent radiation damage when measuring infiltration of cells into the brain is to use parabiosis, where animals share each other’s circulatory systems (see ARF related news story). Fabio Rossi and colleagues at the University of British Columbia, Vancouver, conjoined a normal mouse to one with green fluorescent protein (GFP)-expressing bone marrow cells. In this scenario, the researchers found no GFP-labeled cells in normal mouse’s brain, suggesting that circulating monocytes rarely cross the blood-brain barrier (see ARF related news story on Ajami et al., 2007). Joseph El Khoury, and colleagues at Massachusetts General Hospital, Charlestown, have used parabiosis with AD transgenic mice. The researchers hooked up female APP/PS1 mice to mice that have their CX3CR1 gene replaced with GFP. At Copper Mountain, El Khoury reported infiltration of green cells into the brain of 6 month-old APP/PS1 animals. El Khoury is not sure why, but when the researchers tried the same approach with regular CX3CR1/GFP knockins, they saw no infiltration into the brain. “We don’t yet know why that is. Nonetheless, the work shows that bone marrow cells can get into the brain without there being any irradiation damage,” he said. Whether APP/PS1 transgenic mice already have some damage to the vasculature that makes them susceptible to infiltration is unclear, but the fact that the researchers did not see any infiltration in regular CX3CR1 KO mice suggests that there is something specific to the APP/PS1 transgenic mice that facilitates infiltration of circulating cells. What imbues that susceptibility remains to be seen.—Tom Fagan.
Part 2 of a two-part series. See also Part 1.