Science: Microglia regulate neurodegeneration in Alzheimer's and Parkinson's diseases

The current view is that amyloid is not necessarily the initiator of neuronal dysfunction and cognitive decline in neurodegenerative diseases
.
Studies of centenarians have found that despite being rich in amyloid in their brains, they still perform
well in cognition.
Epidemiological studies have proposed the concept of cognitive reserve, and individuals with good cognitive reserve can better adapt to pathological changes in the brain, thereby delaying or preventing the onset
.
Therefore, one of the challenges of neurodegenerative disease research is to understand how the brain successfully maintains normal neuronal function
in the presence of amyloid deposits in the brain.
Genetic studies suggest that microglia (the main macrophages in the brain) may be key to
regulating and maintaining neuronal function.
So, in AD and PD, is there impaired microglia-neuronal interaction? What damage causes neurodegeneration?

To answer these questions, Professor Soyon Hong's team from University College London published a review in the "Neurodegeneration" section of Science in October 2020, elaborating on the possible pathways in which microglia cause dysfunction in AD and PD neurons, as "Microglia modulate neurodegeneration in Alzheimer's and Parkinson's" diseases”
。

1Microglia are essential for neuronal function and health

Genomic and proteomic evidence suggests that microglia, similar to macrophages in other tissues, function differently depending on their biological state (i.
e.
, brain region, age, health, and metabolic needs
).
In addition to immune function, microglia are essential for brain development, for example, promoting the formation
of neuronal synapses during brain development.
In the adult brain, microglia perform a variety of functions, including monitoring changes in neuronal activity, regulating learning and memory, and acting as phagocytes and damage sensors
localized to the brain.

Many microglia-neuronal interactions are mediated by cell-cell signaling pathways, including purine signaling, cytokines, neurotransmitters, and neuropeptides
.
These functions usually require high energy expenditure and maintenance of normal mitochondrial metabolism, so microglia exhibit metabolic flexibility
in acute hypoglycemia.
In AD and PD, does chronic mitochondrial dysfunction impair the metabolic adaptability of microglia to monitor and regulate neuronal function? In neuronal mitochondrial damage and neurodegeneration, glial cells accumulate lipid droplets that regulate macrophage function
.
Moreover, the accumulation of lipid droplets in microglia is related to aging, so is abnormal lipid metabolism in aging microglia the basis of susceptibility to AD and PD? Multiple AD and PD risk factors that have been identified, including TREM2, apolipoprotein E (ApoE), GBA1, and stearoyl Coa desaturase (SCD), which have been found to modulate lipid metabolism, lysosomal pathways, and microglial metabolic adaptation, seem to further support this hypothesis
.

2 Microglia mediate AD synaptic loss

In AD, synaptic loss and dysfunction are region-specific and early occurrence, and are closely related
to cognitive impairment.
β-Amyloid (Aβ) oligomers and/or tau have been found to accumulate in synapses and cause synaptic dysfunction and loss
.
More than half of the genetic risk factors for AD are currently expressed by myeloid cells
.

Multiple studies in animal models of AD have shown dysregulation of multiple neuroimmune signaling pathways at synapses, including the classical complement cascade, TREM2, phosphatidylserine (PtdSer), and apolipoprotein E (ApoE) (Figure 1).

So, does the accumulation of local pathological proteins on synapses disrupt neuron-glial cell interactions, which are critical for synaptic health? For example, pathological Aβ or tau accumulated on synapses can upregulate C1q in microglia and promote complement activation and subsequent microglial phagocytosis
in synapses.
Blocking the activation of the classical complement cascade pathway in a mouse model of AD with genes or antibody-based methods has been shown to protect synapses from synaptic loss, dysfunction, and memory loss, suggesting that the microglia-synaptic pathway may be a potential therapeutic target
.
In the beginning, this pathway may be a beneficial process in which microglia engulf abnormal synapses, but slowly, the microglia becomes dysfunctional and damages the neurons
it had tried to save.
This mechanism by which microglia engulf abnormal synapses is known as the microglia-synaptic pruning mechanism
.
Since many microglial functions, including synaptic pruning, appear to be activity-dependent, the hyperactivity of neurons observed in early AD mouse models may indicate the presence
of microglial dysphagocytes.
A deep understanding of the pathways that regulate pruning and specific signals that guide microglia to engulf synapses can help identify potential therapeutic targets for cognitive decline and identify candidate biomarkers to quantify microglial dysfunction associated with synaptic loss
.

Figure 1.
Complement-mediated synaptic loss in AD microglia

Another biologically and therapeutically important question is: Do microglia target and clear specific synapses? Proteomic studies of synaptosomes in human and mouse AD brains have found mitochondrial dysfunction
of synapses.
However, it is unclear
whether complement factors, including C1q and C3b, target specific synapses (i.
e.
, dysfunction and/or impairment).
Lipid signaling in neuronal-glial cell interactions can be a key determinant
.
For example, in the mouse hippocampus, TREM2, a key damage sensor for microglia, has been shown to be involved in the synaptic breeding process
.
One of the ligands of TREM2 is Ptdser
, which is exposed outside the neuronal membrane.
Thus, microglia TREM2 may sense damaged synapses
in AD via PtdSer signaling.
Gangliosides, which are abundant in the brain, have also been associated
with Aβ-induced synaptic dysfunction in mouse brains.
Aβ-bound GM1 gangliosides are enriched on neuronal membranes of the early AD brain
.
In addition, anti-GM1 ganglioside antibodies can immobilize complement to neuronal membranes, and the use of targeted C1q antibodies in AD models can improve anti-ganglioside antibody-mediated neuronal damage
in mice.
These findings raise a new question about whether gangliosides are associated
with synaptic loss of AD and complement-mediated phagocytosis of synapses by microglia.

In addition, in the neuroimmune and lipidation pathways, ApoE
needs to be focused.
Previous studies have suggested a possible link between astrocyte ApoE and microglial synaptic pruning: the rate at which astrocyte phagocytosis synapses are regulated by the ApoE allele and are associated with
normal synaptic plasticity.
This rate gradually decreases during aging, making synapses susceptible to complement-mediated microglia
phagocytosis.
Notably, ApoEe4 can increase pathological Aβ
in synapses in the human AD brain.
Recent studies have found that ApoE binds to C1q and modulates the activation
of the classical complement cascade.
The above results show that ApoE plays an important role in astrocyte-neuron-microglial synaptic interaction, especially in AD, the specific dysregulation of ApoE is closely related
to lipid metabolism pathways such as cholesterol.

TREM2 and ApoE are two major risk factors for late-onset AD, and studies of these two factors have shown that abnormal lipid metabolism of microglia is a manifestation
of the sensitivity of the brain immune system to chronic accumulation of amyloid.
For example, TREM2-deficient microglia develop chronic demyelination and cannot metabolize lipids
normally.
In addition, TREM2 appears to be a key regulator of
the main lipid transporter, ApoE.
The study found that ApoE transports excess lipids from overactive neurons into lipid droplets of astrocytes, where it is metabolized, suggesting that ApoE can improve lipid toxicity
caused by neuronal overactivity.
What needs to be clear is the relationship between amyloid-related neuronal hyperactivity and lipid metabolism in astrocytes and microglia, and why this relationship weakens in older brains or brains with mutations
in AD risk genes such as TREM2.

3PD glial cells: regulators of αSyn toxicity

The pathological changes in PD are often accompanied by a pronounced accumulation of neuronal proteins α synuclein (αSyn) in astrocytes and microglia, which has recently been considered a prominent feature
in mouse models of Parkinson's disease.
In addition, improving the microglia-astrocyte interaction can mitigate PD-like pathology
in the αSyn aggregation model.
These studies suggest that glial cells play a direct role
in the uptake and mediation of αSyn's neurotoxicity.
Notably, neurophagocytosis – microglia phagocytosis of neurons – is demonstrated
in PD by the accumulation of neuromelanin within microglia.
Synapses in PD tissue are rich in pathological αSyn aggregates, which may indicate the presence of synaptic phagocytosis mechanisms
similar to those observed in AD in PD.
However, whether complement and microglia mediate synaptic loss of PD is currently unclear
.

From a genetic point of view, the link between PD and microglia is not as obvious
as the link between AD and microglia.
Familial synucleinopathies may be related to the expression level of total neuronal αSyn, however, in sporadic Parkinson's disease, neurodegeneration is closely related to certain biologically active forms of αSyn and not
to the total level of αSyn.
In addition, three synucleinopathies—PD, dementia with Lewy bodies (DLB), and multiple system atrophy (MSA)—are all characterized by amyloid alphaSyn loading, but they clearly exhibit different patterns
of brain region-specific protein accumulation and neuronal dysfunction.
This significant region-specific αSyn diffusion pattern is thought to be induced by prion-like diffusion of specific extracellular αSyn aggregates or strains, similar to prion disease
.
Genetic studies of PD have shown that there are many genetic risk factors
for diseases in sphingolipid metabolism.
The risk genes GBA1, SMPD1, GALC, ASAH1, CTSD, SPTLC1 and SLC17A5 of PD are involved in the degradation of lysosomes, suggesting that the dysfunction of degradation of αSyn aggregates may be an important mechanism
for the occurrence and development of diseases.

So, are glial cells involved in preventing or facilitating the transmission of extracellular αSyn aggregates in different brain regions, leading to region-specific synucleinopathy? The answer is yes
.
The only known uptake receptor for extracellular αSyn aggregates, LAG3, is mainly expressed
by microglia.
In addition, in recent synucleinopathic models, disruption of the clearance of extracellular αSyn by microglia leads to dysfunction
of dopaminergic neurons.
These studies suggest that glial uptake and processing is key
to regulating alphaSyn activity.
Thus, while glial cells can serve as sites for neurons to clear protein misfolds in the physiological environment, glial cells can be a double-edged sword in disease: the uptake and processing of non-toxic αSyn by glial cells may actually be a process of producing specific toxic αSyn through autophagy and defective lysosomal degradation (Figure 2).

Pathological modification of extracellular αSyn by microglia mediated by imbalance of sphingolipid metabolism may be a key determinant of
chronic αSyn dysfunction in PD, DLB, or MSA.

Figure 2.
Glial cells act as regulators of αSyn toxicity in PD

4 extrabrain observation of Parkinson's disease: macrophage-neuronal signaling in the gut

The latest preclinical and genetic data suggest that in PD pathology, the enteric nervous system (ENS) – the so-called enteric brain – may be involved in
PD pathology.
Researchers have observed αSyn aggregation in ENS and believe that αSyn spreads from the gut to the brain
in a cell-to-cell transsynaptic fashion.
Cutting the vagus nerve trunk in mice was found to prevent pathological αSyn from entering the brain, suggesting that the vagus nerve is a potential conduit
for delivering αSyn.
It is worth noting that intestinal injection of Syn not only induces αSyn phosphorylation in intestinal neurons, but also stimulates the production of CX3CL1 and CSF1, ligands
that bind to CX3CR1 and CSF1R on intestinal macrophages.
A recent study identified a specific type of tissue-resident macrophage in the ENS, which is similar to microglia in the brain and has a long lifespan that is important for neuronal survival and intestinal function (Figure 3).

These intestinal macrophages residing on the ENS express high levels of transcripts, including Gba1 and Lrrk2, which are involved in vesicle trafficking and the endolysosomal pathway
.
Mutations in the LRRK2 gene are a common cause of autosomal dominant PD; However, exactly how LRRK2 causes Syn pathology and PD-like symptoms is currently unclear
.
Notably, macrophages lacking Lrkk2 showed higher proteolytic activity and expressed higher levels of lysozyme, suggesting that LRRK2 regulates lysosomal function and phagosome maturation
.
In addition, LRRK2 interacts with actin remodeling factor WAVE2 to regulate phagocytic function
of macrophages.
Therefore, assessing how intestinal macrophages are affected in PD and regulating the clearance of pathological αSyn along the gut-brain axis is of great
significance for the diagnosis and treatment of PD.

Figure 3.
Macrophage-neuronal interaction along the gut-brain axis in patients with Parkinson's disease

5 Outlook

In AD and PD, macrophages in the brain and intestine do not perceive dysfunctional neurons, which can lead to pathological disruption
of neuronal homeostasis and normal function.
Research from patient tissues and animal models clearly points out that glial cells are not just phagocytes of amyloid, they also manage and regulate neuronal health
.
Therefore, the development of methods to monitor glial cell-neuron interactions in the living brain and the evaluation of pathways associated with the disease is particularly important
for the early diagnosis and treatment of neurodegenerative diseases.

Whether neuroinflammation is beneficial or harmful requires problem-specific analysis
.
For example, the classical complement cascade helps reduce amyloid, but also mediates synaptic loss
.
Enhancing TREM2 activity may be beneficial in brain tissue with amyloid accumulation, but not in brain tissue with nerve fiber tangles
.
Therefore, for chronic multifactorial diseases such as AD and PD, we need to carefully consider the specific body state
while evaluating function and impact.
Using different strategies at different stages of the disease can target specific biological processes for effective therapeutic purposes
.