Quantitative Metabolic Signatures of Stem Cell Fate: Multimodal Imaging and Predictive Modeling of iPSC Differentiation

Datum

29.06.2026

Zeit

11:00 - 12:00

Sprecher

Arina Nikitina

Zugehörigkeit

Weizmann Institute of Science, Rehovot, Israel

Sprache

en

Hauptthema

Biologie

Host

Jesse Veenvliet

Beschreibung

Induced pluripotent stem cells offer a powerful route toward patient-specific disease models and regenerative therapies, but their clinical translation depends on the ability to monitor, predict, and control differentiation outcomes. Current quality-control strategies often rely on endpoint measurements or transcriptional and protein markers that emerge only after major cell fate decisions have already occurred. To address this gap we developed experimental and computational approaches to identify early metabolic features associated with pluripotency loss, lineage specification, and differentiation success. First, we developed a multimodal imaging pipeline that co-registers confocal fluorescence microscopy with MALDI mass spectrometry imaging, enabling spatially resolved lipid measurements to be linked with cell-state markers in iPSC colonies. Applying this approach to spontaneous differentiation revealed dynamic phospholipid remodeling during the earliest stages of pluripotency loss. A subset of phosphatidylinositol species emerged as early markers of cell-state transition, preceding changes in canonical pluripotency markers such as Oct4 and revealing metabolic evidence of lineage bifurcation. Perturbation experiments further suggested that phospholipid metabolism is not only a passive marker of differentiation, but may participate in regulating pluripotency maintenance. Finally, I will discuss how real-time oxygen consumption measurements can be used to predict the outcome of directed cardiomyocyte differentiation. By combining live metabolic sensing with machine learning analysis of time-series features, we found that oxygen consumption dynamics during the first several days of differentiation were highly predictive of final cardiomyocyte yield, suggesting a scalable route toward early in-process quality control for cell manufacturing.

Letztmalig verändert: 26.06.2026, 07:39:26