Amyloid β Induces Lipid Droplet-Mediated Microglial Dysfunction in Alzheimer’s Disease

  1. Priya Prakash
  2. Palak Manchanda
  3. Evi Paouri
  4. Kanchan Bisht
  5. Kaushik Sharma
  6. Prageeth R. Wijewardhane
  7. Caitlin E. Randolph
  8. Matthew G. Clark
  9. Jonathan Fine
  10. Elizabeth A. Thayer
  11. Alexis Crockett
  12. Nadia Gasmi
  13. Sarah Stanko
  14. Richard A. Prayson
  15. Chi Zhang
  16. Dimitrios Davalos
  17. Gaurav Chopra

Reviewed by

Read the full article See related articles

Listed in

Log in to save this article

Abstract

Several microglia-expressed genes have emerged as top risk variants for Alzheimer’s disease (AD). Impaired microglial phagocytosis is one of the main proposed outcomes by which these AD-risk genes may contribute to neurodegeneration, but the mechanisms translating genetic association to cellular dysfunction remain unknown. Here we show that microglia form lipid droplets (LDs) upon exposure to amyloid-beta (Aβ), and that their LD load increases with proximity to amyloid plaques in brains from human patients and the AD mouse model 5xFAD. LD formation is dependent upon age and disease progression and is more prominent in the hippocampus in mice and humans. Despite variability in LD load between microglia from male versus female animals and between cells from different brain regions, LD-laden microglia exhibited a deficit in Aβ phagocytosis. Unbiased lipidomic analysis identified a substantial decrease in free fatty acids (FFAs) and a parallel increase in triacylglycerols (TAGs) as the key metabolic transition underlying LD formation. We demonstrate that DGAT2, a key enzyme for the conversion of FFAs to TAGs, promotes microglial LD formation, is increased in microglia from 5xFAD and human AD brains, and that inhibiting DGAT2 improved microglial uptake of Aβ. These findings identify a new lipid-mediated mechanism underlying microglial dysfunction that could become a novel therapeutic target for AD.

Article activity feed

  1. PREreview

    This Zenodo record is a permanently preserved version of a PREreview. You can view the complete PREreview at https://prereview.org/reviews/8208261.

    This review reflects comments and contributions by Anna Oliveras, Mayank Chugh, Bhargy Sharma and Femi Arogundade. Review synthesized by Anna Oliveras.

    This preprint studies how microglia cells react to AB exposure to regulate lipid droplet (LD) load and its consequences for their phagocytic capacity. The authors use 5xFAD AD mice model, primary microglia cultures and complemented with human brain tissue samples. The research offers significant understanding of the reasons behind microglial dysfunction in Alzheimer's disease (AD). It emphasizes the role of lipid droplets (LDs) and their connections to Amyloid-beta (Aꞵ) in hindering the ability of microglia to perform phagocytosis effectively. The discovery of DGAT2 as a crucial enzyme that facilitates the formation of lipid droplets (LDs) and leads to impaired microglial function implies that modulating this pathway could hold promise as a potential therapeutic approach for Alzheimer's disease (AD).

    Major comments:

    • Strong data: microglia in contact with AD plaques preferentially accumulates LD in the AD mice model.

    • Weak data: mechanistic link between chronic AB exposure, LD build-up and phagocytic dysfunction

    • Relevant data: inhibition DGAT1 enzyme, involved in LD formation, recovers the phagocytic capacity of 5xFAD mice.

    • The manuscript will benefit from clear postulating of hypothesis and experimental testing framework. Currently, the findings presented in different sections are disorganised and disrupt the message and flow. For example, the first section does not really describe region-dependence while it is in the section title and figure 1. Another example of flow disruption is constant switching between mouse, human, and in vitro sample. How are these three systems relevant in this study can be introduced in the introduction. 

    • The figures in the main text containing fluorescence microscopy images and SRS images are very small and not with sufficient resolution for readers. The panels in the figures are too small, making it difficult to read them. The manuscript would benefit if readers can identify LDs, plaques, cells visually in addition to quantitative measurements. Many figures require fixing labels and legends to be readable.

    • First section in the results does not show much data in region-dependence , hence it is recommended that the section title be modified or more information included within this section.

    • An upward trend for plaque burden is seen from 3 months old mice, but a significant neuronal loss in these 5xFAD mice is reported at 9 months of age. It is unclear why this 5-7 months aging point was chosen for this study.

    • Provide proper citation for using BODIPY dye. What is the threshold for BODIPYhi and BODIPYlo? Include this information in the text and cite appropriate references.

    • It is unclear how the TAG lipid species were quantified to measure relative amounts.

    Minor comments:

    • Figure 1e graph looks identical to 1c and I believe it has been a mistake (based on the explanation of the data).

    • Certain conclusions are not well supported with experimental evidence. The authors should emphasize the significance of their experimental findings in this study, instead of focusing on potential extrapolation to inconclusive inferences.

    • Full-form of DGAT should be mentioned in Abstract.

    • In the last subsection of results, using phagocytic activity seems indirect measurement as phagocytosis is not really shown. Uptake indicative of increased phagocytic activity would be a better measure.

    • Figure 1h : Can they enlarge SRS lipid channels?  Can they show the SRS protein channel as a control? This can be added to the supplementary.

    • Figure 2a : Plaque distances are not labelled.

    • Refering Figure 2e, f : Can authors please describe in detail either here or in methods how exactly the plaque proximal and distant distant is calculated? In my understanding, 3D reconstruction is done before but where is the reference for the distance? Can they provide a pictorial representation in the supplementary perhaps?

    • First sub-section of results: "These results illustrate that following prolonged Aβ exposure, microglia exhibit: i) a unique lipid signature that likely facilitates the formation and accumulation of LDs within them in a sex-distinct manner, and ii) dysregulation in several key metabolic pathways, that could also impact their functions in the context of AD." : It is unclear what the authors mean by a "unique" signature as there is a huge variability between lipidomes. Is the unique lipid signature is related to the female samples which are almost identical in the PCA? The point made in this sentence is unclear. Additionally, the authors should be more specific here when referencing to the metabolic pathways. Relevance of their results to any specific function or any particular metabolic pathway should be mentioned, instead of unlcear broad-stroke outlooks.

    • Figure 1 k, l: How was n = 6 decided? Please provide the details of the power analysis calculation.

    • In the first line of discussion, the authors comment "a plethora of studies have revealed important microglial contributions….", however a single review article is referenced in support. We suggest to include some original research articles to support the statement.

    • Sub-section 4 of results: Regarding the statement "This suggests that Aβ plays a direct role in upregulating... " - Explain the link between those metabolites/metabolic pathways and LD formation. Is the lipidome from 5xFAD microglia and AB-treated ex vivo WT microglia not overlapping? Why?

    • Fig S13a : The metabolomics data as well as the lipidomics data shows a biphasic trend, instead of a continuous transition. Discuss how to interpret it.

    • Regarding differential regulation of lipids in microglia (presumably Table ST7) : Why is there an initial decrease (12h) of TAG and CE and then and upregulation after 24h? Needs further discussion.

    • Referencing of Figures 1i-m in text does not match to labeling in the figure. e.g. there is no 1m figure which is referenced in the text.

    • Referencing results of Figure 2i, j : " This could indicate that since 5xFAD mice progressively develop Aβ plaques starting at 2 months of age, chronic microglial exposure to Aβ possibly alters their functional abilities and overall state by the age of 5-7 months, when amyloid deposition is extensive throughout the brain." : The authors need to be more specific with the conclusions. Which type of functional abilities? One would assume that the lipid storage capacities are already overloaded in 5xFAD mice.

    • Referencing results of Figure 2l ,k and S9b-c , "In conclusion, these experiments demonstrate that LD-laden microglia that are chronically exposed to Aβ exhibit a dysfunctional phagocytic phenotype" : To directly draw this conclusion and show a mechanistic link, the experimental workflow needs to be modified. First label LD, select for LD-rich cells in WT and 5xFAD mice and then assess phagocytic function. Using the current workflow, the results can be overinterpreted: the population of WT LD+ microglia contains already the population of with higher phagocytic capacity.

    • How many cells per mice were analysed for results in Fig 2d?

    • Figure 2a: What is the significance of showing plaque in shades of blue. If relevant, what is the intensity scale? Additionally, correlation between LD and plaque in cortex and CA1 does not look positive as per the figure 2A. It may be better to explain this after referencing 2b and S7a.

    • Figure 2b: Plaque- distance is not labelled.

    • What is the difference between Fig 2f and Fig S7d? The data seems the same but the supplemental has more data points. Clarify that in figure legend (also for Fig 2g and Fig S7e)

    • Referencing results in Fig 2 : Can authors please describe in detail either here or in methods how exactly the plaque proximal and distant distance is calculated? In my understanding, 3D reconstruction is done before but where is the reference for the distance? Can they provide a pictorial representation in the supplementary perhaps?

    • Referencing results of Fig 2c-d : "Importantly, this also translated in human microglia in the hippocampus of AD patients (both male and female), where plaque-proximal microglia had a significantly higher number of LDs than plaque-distant microglia, which decreased with distance from the nearest plaque" : Discuss why phenotype seems more prominent in subiculum than cortex or hippocampus. Authors should also discuss findings from Cortex and CA1 and how that relates to what is observed in subiculum

    • Figures are improperly referenced in the text "Even though aged WT brains had a high number of LDs in microglia (as also previously reported, 20), and in other cells (Fig. 1i; S4a), 5xFAD brains had significantly higher overall LD density and total LD area in cortex and parts of the hippocampus (CA1 and subiculum) compared to WT (Fig. S4b-c)." : Figure reference is misleading- Is number of LD in young and older mice compared? I couldn't find this data. Or it is only cited from previous literature? Figure 1i and S4a should be referenced at the end of this sentence instead of Fig. S4b-c.

    • Regarding "chronic 5xFAD" : What is the time-frame categorised as chronic 5xFAD? what does chronic mean in this context? which age?

    • SRS full form should be mentioned earlier in the text. It is already used in Fig1 caption.

    Comments on reporting:

    • Analysis codes for Lipidomics and Metabolomics data are not easy to access, an explanation document to reference in the folders in Github would be helpful. 

    • Lipidomics data link is not updated on their webpage  http://microgliaomics-chopralab.appspot.com .

    • Supplementary Tables were not found with this manuscript. The present review doesn't include data presented in them.

    • What is the composition of control vehicle used in the study? It  should at least be mentioned in methods section.

    Suggestions for future studies:

    • In reference to Fig. 4c and S14 a,b : The contradictory findings - Downregulation of Dgat2 mRNA and abundance of DGAT2+ microglia - need explanation. Possibly a supplementary western blot result would be useful to avoid doubts of staining artifacts. 

    • The rationale in the results could be explained better - e.g.  asking 'if' exposure to Ab induced LD formation should precede testing 'how' it does. 

    • Future research should explore the impact of inhibiting DGAT2 on cognitive function and the progression of AD pathology in animal models of Alzheimer's disease. Additionally, it is crucial to investigate any potential side effects or unintended consequences that may arise from the use of DGAT2 inhibitors.

    Competing interests

    The author declares that they have no competing interests.

  2. Version published to 10.1101/2023.06.04.543525v1 on bioRxiv