A 19-day human macrophage from iPSCs clears amyloid faster than standard microglia and signals harder through TREM2

Human induced pluripotent stem cells (iPSCs) can now be driven in 19 days into cultures that are more than 90% CD45⁺CD11b⁺ macrophage-like cells, which the authors call iMacs. Existing iPSC-derived microglia (iMGL) protocols rely on multi-step cytokine cocktails and longer maturation, or on transducing multiple transcription factors, both of which introduce variability and complicate genetic manipulation. Across three iPSC donor lines (WTC11, KOLF2.1J, ID1), iMac purity, viability and reproducibility were comparable to iMGLs while differentiation took roughly half as long.

iMacs sit closer to brain macrophages transcriptionally and are enriched for clearance programs tied to disease-associated microglia (DAM) and lipid-associated macrophages (LAM). Relative to iMGLs generated in parallel in the same lab, iMacs upregulate the scavenger receptors F13A1, MRC1, CD36 and CD163, and downregulate homeostatic microglial genes CX3CR1, P2RY12 and CST3 along with MHC class II genes. Bulk RNA sequencing of iMacs from three WTC11 clones placed them closest to CNS macrophages among published human tissue-resident macrophage datasets from brain, liver, lung and atherosclerotic plaque; among mature models they were most similar to iPSC-derived and monocyte-derived macrophages and most divergent from immortalized THP-1 and HMC3 lines and transcription-factor-driven iMGLs.

iMacs take up more of several substrates and carry more lysosomal mass than iMGLs, and respond more strongly to TREM2 engagement. Inducing DAM-like states in iMGL systems previously required substrate exposure, transcription factor overexpression or engineered knockout lines. After 24 hours, 39.2% of iMacs versus 24.6% of iMGLs were positive for carboxylated latex beads by flow cytometry; live imaging gave 24-hour area-under-curve fold changes of 7.9 for amyloid-beta 1-42, 4.9 for amyloid-beta 1-40 and 1.6 for zymosan; LysoTracker measured 3.9-fold more lysosomal mass; and antibody crosslinking of TREM2 produced 8.15-fold greater induction of phosphorylated PLCγ2 than in iMGLs.

The protocol is aimed at researchers who need a human cell system for studying phagocytic clearance and AD risk genes, and it is suited to in vitro monoculture with bulk readouts. iMacs mature without brain parenchymal signals such as TGF-β and IL-34, so their state is shaped by culture conditions rather than by a specific tissue environment; they share key features with BAMs and DAM but the resemblance is partial. The authors propose re-introducing specific environmental factors to test whether iMacs can be directed toward distinct tissue-resident macrophage subsets, and integrating them into co-cultures, organoids or in vivo xenotransplantation models.

The authors note that all transcriptomic profiling used bulk RNA sequencing, so heterogeneity within iMac cultures remains uncharacterized and single-cell approaches would be needed to tell whether the BAM- and DAM-associated signatures reflect a uniform population or distinct subpopulations. The efferocytic phenotype was characterized with a limited set of substrates, leaving materials such as myelin and apoptotic cells to be tested. Genetic tractability has so far been shown for a single target gene, MEF2C, and suitability for larger-scale perturbation such as a pooled CRISPR screen remains to be established. iMacs also do not fully match any single BAM subpopulation, and LYVE1 expression was low across samples with no difference between cell types.