11 January 2009. Microglia, the phagocytes of the brain, are drawn to plaques where they can gobble up amyloid-β (Aβ) and clear it. However, the immune responders also spew out harmful proinflammatory cytokines and chemokines, fueling inflammation that can damage neurons. Two studies probing the interactions of microglia and Aβ suggest that, for better or worse, the cells are quick to react to amyloid.
The first study indicates that microglia are attracted to amyloid plaques by the angiogenic vascular endothelial growth factor (VEGF), acting through the Flt-1 receptor on microglia. VEGF elevation has been reported in AD (Tarkowski et al., 2002) and VEGF is found associated with amyloid plaques (Yang et al., 2004). The new work, from James McLarnon and colleagues at University of British Columbia in Vancouver, suggests that one function of VEGF could be to recruit proinflammatory microglia, which then damage neurons. Blocking Flt-1 partially inhibited microglia recruitment and neuron death in a rat model of Aβ-induced neuroinflammation, consistent with the fatal attraction model of microglial action. That study appears in the January 7 issue of the Journal of Neuroscience.
A second study, also in the Journal of Neuroscience, presents in vivo imaging data to show that in mice, microglia mount a rapid response to passive immunization with Aβ antibodies. That paper, from the lab of David Holtzman and colleagues at Washington University in St. Louis, Missouri, appeared in the December 24 Journal of Neuroscience. The speedy activation may be behind the observed Aβ clearance after passive immunization (Bard et al., 2000), but could also give rise to unwanted effects, such as microhemorrhage (Racke et al., 2005).
Calling All Microglia
The elevation of VEGF in Alzheimer disease was first recognized over a decade ago, but its purpose remains unclear. VEGF signals through several receptors on endothelial cells and immune cells, and the new study focuses on Flt-1, which is found on inflammatory immune cells including monocytes/macrophages, but has not thus far been looked at in neurodegeneration. First author Jae Ryu found that, like VEGF, Flt-1 expression is increased in brain tissue from AD patients, and specifically in microglia. Microglia isolated from normal human brain expressed little VEGF or Flt-1, but when the researchers treated the cells with Aβ, both mRNAs were induced. Immmunohistochemical staining colocalized Flt-1 and VEGF with the microglia marker HLA-DR, and with Aβ deposits.
To probe the role of VEGF/Flt-1 in microglial response to Aβ, Ryu and colleagues used a model of Aβ-driven inflammation, namely injection of Aβ42 into rat hippocampus. They showed that Flt-1 expression increased one and three days after peptide injection, and appeared to localize mostly to microglia.
Flt-1 mediates immune cell chemotaxis, and studies of microglial migration both in vitro and in vivo suggest that Flt-1 supports microglia chemotaxis in response to Aβ. Coinjecting a Flt-1 antibody into hippocampus along with Aβ reduced the number of recruited microglia by about 20 percent. The antibody also appeared to reduce the neurotoxicity of Aβ as measured by staining for the neuronal marker NeuN, consistent with the microglia causing a neurotoxic neuroinflammation.
In a different in vivo model, the investigators injected Aβ into rat hippocampus, and three days later injected green fluorescent protein-labeled microglia at a separate site. They then tracked the movement of microglia toward the Aβ. When microglia were first incubated with anti-Flt-1, their migration was retarded compared to untreated cells. The role of Flt-1 was also detected in vitro, where migration of microglia in culture was stimulated by media from Aβ-treated cells, and that effect was reduced about 40 percent by anti-Flt-1 antibody.
“This work is the first to point out the role of VEGF and its receptor Flt-1 in mobilizing microglia giving them the chemokine stimulus to move them into sites of plaque,” McLarnon told ARF.
The work adds another candidate signaling pathway for microglia recruitment and activation in AD, in addition to the previously identified chemokine receptor Ccr2 (see ARF related news story). Understanding these pathways may help selectively modulate microglia function to reduce harmful neuroinflammation, without affecting beneficial functions of microglia, or other physiological functions of VEGF, which may include neuroprotection (see ARF related news story).
The findings do not rule out additional roles of VEGF in modulating blood vessel function in AD brain, including effects on angiogenesis and inflammation. McLarnon says his lab is actively investigating that question.
Passive Immunization, Aggressive Response
As phagocytes, microglia have been shown to take up and degrade Aβ, and their activation leads to clearance of plaque-laden brain tissue in response to delivery of Aβ antibodies directly into brain. However, active or passive Aβ immunization involves peripheral antibodies, and their effect on microglia activation was not clear. To get a close-up look, first author Jessica Koenigsknecht-Talboo of the Holtzman lab collaborated with Brian Bacskai and Bradley Hyman at Massachusetts General Hospital, Charlestown, to perform in vivo multiphoton microscopic imaging of microglia in live mice passively immunized with an antibody to aggregated Aβ. To aid visualization, the investigators crossed PDAPP mice with animals that express GFP in microglia. After injecting dextran to label blood vessels and a dye to stain plaques, they viewed brain tissue through a cranial window.
Their results paint a picture of rapid activation of microglia after peripheral antibody treatment. In young mice without any plaques, the microglia appear dispersed. Later on, plaques form with microglia clustered around them. The morphology of plaque-associated microglia revealed fewer processes and larger cell bodies than microglia not associated with amyloid. To look at microglia dynamics, the investigators took photos every hour, and saw that plaque- associated microglia showed less movement of their processes than microglia not exposed to aggregated amyloid.
Within three days of peripheral administration of the Aβ antibody m3D6 (a humanized version of this antibody is now in Phase 3 trials as Bapineuzumab [see ARF related news story]), there were 50 to 80 percent more microglia around plaques and associated with amyloid on blood vessels. The total number of microglia in brain and the number of processes per cell increased. The microglial recruitment required aggregated Aβ, as no such increases were observed in young mice that lacked plaques, or with an antibody that only recognizes soluble Aβ. The effects also required intact antibody and were not seen with an m3D6 version lacking the Fc domain necessary for interacting with microglia.
“It is possible that the rapid changes in microglia after antibody treatment reported in this study may explain some of the positive responses (such as plaque clearance) as well as negative responses (such as vasogenic edema and CAA-associated hemorrhage) that have been observed in other anti-Aβ antibody studies,” the authors write. However, it remains to be determined how much, if any, of therapeutic effects of Aβ vaccination, including cognitive improvement, depend on microglial responses. In addition, neither study looked closely at the phenotype of the activated microglia, which can produce either neurotrophic cytokines or inflammatory mediators. Further work is needed to determine if antibodies can be optimized to enhance favorable microglial responses while avoiding harmful activation.—Pat McCaffrey.
Ryu JK, Cho T, Choi HB, Wang YT, McLarnon JG. Microglia VEGF receptor response is an integral chemotactic component in Alzheimer’s disease pathology. J. Neurosci. 2009 Jan 7;29(1):3-13. Abstract
Koenigsknecht-Talboo J, Meyer-Luehmann M, Parsadanian M, Garcia-Alloza M, Finn MB, Hyman BT, Bacskai BJ, Holtzman DM. Rapid microglial response around amyloid pathology after systemic anti-Abeta antibody administration in PDAPP mice. J Neurosci. 2008 Dec 24;28(52):14156-64. Abstract