To view this post in german click here/ Für die deutsche Version hier klicken
In October 2015, the Journal of Experimental Medicine published two papers on the effect of peripheral (outside the brain) myeloid cells on Alzheimer’s disease (AD), a disease manifesting itself in the brain. Both publications depleted brain-resident macrophages and allowed the open space (called niche) to be filled with cells from the blood stream. But why would that influence a disease in which neurons are dying? Because evidence suggests that neuroinfammation plays a role in neurodegenerative diseases such as AD, Parkinson’s disease and Huntington’s disease.
The main hallmark of AD pathology is protein aggregates called Aβ plaques. They can be found all over the brain of AD patients and cause neurons to die. Microglia, the macrophages of the brain, are found to associate with plaques but fail to phagocytose (=eat) the toxic aggregates. This lack of response is a huge problem, as removal of plaques by macrophages could slow AD progression. Interestingly, studies have found that peripheral cells such as monocytes and macrophages can infiltrate brain tissue and “eat” Aβ plaques under certain conditions. Could replacing the non-function microglia with fresh peripheral monocytes help to get rid of Aβ plaques?
As it is difficult to distinguish microglia from infiltrating monocytes and macrophages and therefore investigate the exact role each cell type plays, Varvel et al. developed a technique in 2012 which allows the depletion of brain-resident CD11b+ macrophages using the TK mouse model. In these mice, cell expressing the cell surface molecule CD11b also express an enzyme called herpes simplex thymidine kinase, which can transform the antiviral drug ganciclovir into a toxic product which kills the cells. This neat mechanism allows cell-specific depletion of CD11b+ cells after injection of ganciclovir in the brain. Given the local spread of ganciclovir after injection, peripheral CD11b+ cells are not affected.
Varvel et al. and Prokop et al. now crossed TK mice with mice carrying mutated genes (APP and PS1) causing AD, thereby enabling the depletion of microglia in brains suffering from AD pathology. Strikingly, both groups found that peripheral immune cells have no advantage in phagocytosing plaques. Newly recruited macrophages did not even associate with plaques after two weeks. Only after months of being within the brain environment did the cells start to engage the plagues- but still, they did not phagocytose them. This was the same in mice already suffering from deposits at the time of depletion and in mice depleted before the start of pathology.
Trying to identify why peripheral cells did not engage plaques right away Varvel et al. analysed Trem2 expression in the microglia using microscopy. While plaques-associated microglia in non-depleted animals showed Trem2 expression, newly recruited non-associated macrophages did not. Long-term recruited macrophages that started to engage plaques also up-regulated Trem2, suggesting that the expression of the molecule is regulated by contact with plaques. Interestingly, increases in Trem2 expression have been linked to up-regulated phagocytosis in macrophages – however in this study this link was not seen.
Nonetheless it is striking that long-term engrafted peripheral cells nearly completely acquire the functions and morphology of resident-brain microglia. Given their different origin – microglia stem from progenitors in the embryo and are self-renewing throughout life, while peripheral blood monocytes are short-lived with bone marrow origin – this study suggests that at least in their repopulation model the environment trumps origin cues for final differentiation. For AD treatment, however, this means that replacement of microglia with peripheral cells may not be a good therapeutic target to reduce Aβ load.
But what if Aβ plaques simply do not contain enough stimuli to activate macrophages? To answer this, Prokop et al. linked cell replacement with vaccination against Aβ to try and encourage macrophages to engage plaques. Interestingly, they found that injecting anti-Aβ antibodies led to a mild decreased plaque load. However so did the injection of the control antibody, which was not directed against Aβ – suggesting that unspecific stimulation of peripheral cells before their recruitment to the brain may be beneficially. More work will be needed on what could prime macrophages to attack Aβ plaques before cell replacement-based therapies can be taken forward.