APOE4 Shapes Neutrophil–Microglia Crosstalk in Alzheimer’s Disease: My First-Author Work in Nature Medicine

I am excited to share a deeper look into my shared first-author publication in Nature Medicine, a project that grew out of my Master’s thesis and went on to become one of the cornerstone studies of my early research career. In this work, we investigated how the Alzheimer’s disease risk gene APOE4 drives dysregulated communication between neutrophils and microglia, ultimately contributing to cognitive decline.

My contributions included all computational data analysis, spanning bulk and single-cell RNA-seq, machine-learning modeling, biomarker discovery, and transcriptomic integration. I also wrote large portions of the manuscript, and designed, organized, and produced the figures that present the central concepts of the study.

Background: APOE4 as an Immune Risk Gene

APOE4 is the strongest genetic risk factor for late-onset Alzheimer’s disease. While traditionally viewed through the lens of lipid metabolism and amyloid biology, increasing evidence suggests that APOE4 reshapes immune function, both in the central nervous system and the periphery.

Our study reveals that APOE4 disrupts immune homeostasis by altering the behaviour of neutrophils, which in turn affects microglial activation states in the brain. According to the model and schematics in the paper (see illustrated immune cascade on pg. 2 of the PDF), APOE4 drives a heightened innate immune tone that primes microglia toward dysfunctional activation during AD progression.

A Machine-Learning Discovery: IL18R1+ Neutrophils as a Biomarker

(derived from transcriptomics and ML classification — see heatmap and feature-selection analysis on pg. 7)

A central part of my master’s thesis was the application of machine learning to peripheral immune data to discover cell populations and molecular signatures associated with APOE4.

Using:

feature selection methods
classification algorithms
transcriptomic integration across datasets
dimensionality reduction and clustering
pathway enrichment

I identified a subset of IL18R1-positive neutrophils that were pronounced in APOE4 carriers, especially in women, mirroring the sex- and genotype-specific patterns seen in microglial dysfunction.

These IL18R1+ neutrophils displayed:

pro-inflammatory signatures
altered chemokine receptor landscapes
associations with cognitive impairment scores
transcriptional alignment with disease-associated microglia (DAM/MGnD signatures)

This finding is highlighted as a potential peripheral biomarker, offering a window into CNS immune dysfunction through accessible blood measurements.

Microglia Reflect Neutrophil Dysregulation

(see microglial single-cell UMAP and pathway overlays on pg. 4)

Our analysis revealed that APOE4 disrupts not only peripheral immunity but also central microglial states. The transcriptomic maps show:

increased IFN-response microglia (IRM/IEM)
dysregulated lipid metabolism
impaired transition into beneficial phagocytic states
elevated inflammatory tone in women carrying APOE4

This immune axis — from blood neutrophils to brain microglia — underscores a multi-system nature of AD risk.

What Makes IL18R1+ Neutrophils a Promising Biomarker?

From the combined machine-learning and RNA-seq analyses:

IL18R1 expression robustly discriminated APOE4 carriers
It correlated with microglial activation states in brain tissue
It was detectable in peripheral blood, making clinical translation feasible
It captured sex-dependent immune variation, which may refine risk stratification

This discovery offers a novel bridge between peripheral diagnostics and central immune pathology in Alzheimer’s disease.

My Contributions to the Study

This project was deeply meaningful to me, both scientifically and personally. My key contributions included:

Shared first authorship
Full computational analysis, including:
- bulk RNA-seq
- single-cell RNA-seq
- clustering, DEG analysis, pathway analysis
- machine learning for biomarker discovery
- integration of neutrophil and microglial datasets
Conceptual development and model building
Manuscript writing, including drafting several key sections
Designing and producing the figures, including the multi-layered immune interaction diagrams
Data interpretation, helping shape the narrative that connects APOE4, neutrophils, and microglia

This publication played a formative role in shaping the direction of my PhD work on neuroimmune regulation and transposable element biology.

A Step Toward Immune-Based Precision Medicine in Alzheimer’s

By identifying a sex- and genotype-specific immune signature and mapping its influence from the periphery to the brain, this study opens the door toward future:

blood-based biomarkers
stratified clinical trials
targeted immunomodulatory therapies

Our findings highlight that Alzheimer’s disease is not only a disorder of neurons and protein aggregates but also of immune communication—and that intervening in these pathways may offer entirely new treatment strategies.

If you would like to read the full article or learn more about the data analyses behind the study, feel free to reach out. A more detailed technical breakdown of the machine-learning pipeline will be posted soon.