Work Description

Example data for the chapter "Linking single-cell dynamics to cell fate in differentiating hPSCs" submitted to the voume STEM-CELL-BASED HUMAN EMBRYO MODELS : Methods and Protocols, in SpringerNature's METHODS IN MOLECULAR BIOLOGY Series, by Seth Teague, Zhiyuan Yu, and Idse Heemskerk.

Image data was collected using a Dragonfly/Leica DMI8 spinning disk confocal microscope with a x20 air objective using Andor Fusion software version 2.3.0.31. Image files are in the proprietary Imaris (.ims) file format. MATLAB and Python code for image processing and quantification is provided with the data and at https://github.com/seth414/HeemskerkLabMethods. Processed data originally published in Teague et al., 2024.

The "live" folder contains time-lapse live-cell image data of RUES2 human embryonic stem cells expressing an n-terminal fusion of green fluorescent protein to SMAD4 in the endogenous SMAD4 locus (GFP::SMAD4) and a fluorescent histone nuclear marker (H2B::RFP), produced in Nemashkalo et al., 2017. Cells are in base media of mTeSR without phenol red (STEMCELL Technologies) supplemented with 10 uM ROCK inhibitor. The first channel contains images of the H2B::RFP nuclear marker and the second channel images of GFP::SMAD4. The time lapse consists of 30 time points, with one collected every 10 minutes. Pixels are 0.603 x 0.603 microns. Four z slices spanning 7.5 microns were acquired in each channel at each interval. There are 8 fields of view, constituting two 2x2 montages of overlapping images, with 18% overlap between adjacent images. The first montage is in a well of cells treated with 50 ng/mL BMP4 and 5 uM IWP2, and the second is in a well with the same treatment with the addition of 100 nM LDN193189. Treatment was added to the cells between the 5th and 6th time points.

The "fixed" folder contains immunofluorescence images of the same cells, fixed and bleached immediately after the conclusion of the time lapse. The cells were stained with primary antibodies for phospho-SMAD1/SMAD5/SMAD9 rabbit (Cell Signaling Technologies, #13820S) and OCT3/4 mouse (BD Biosciences, #611202) and the secondary antibodies donkey anti-mouse Alexa Fluor 488 and donkey anti-goat Alexa Fluor 647, as well as for DAPI. The channels are in the following order: DAPI, OCT3/4, pSMAD1/5/9. Eight z slices were collected spanning 13 microns. The same montages/fields of view were used as for the live data.

Because running cellpose is the most computationally intensive step of the image processing and relies on correct installation of cellpose in a conda environment, we provide precomputed cellpose masks that can be used if skipping the cellpose step is desired. Both the "live" and "fixed" folders have a subfolder "sample_cellpose_masks" that contain multipage tiffs of these masks for each position and time point.

The "live" folder also contains .h5 files with precomputed Ilastik pixel classification results in the SMAD4 channel meant to identify cell body vs background. These are intended to improve quantification of cytoplasmic and background intensity for calculation of nuclear to cytoplasmic SMAD4 ratio in the live-cell image data.

References:
Teague, S., Primavera, G., Chen, B. et al. Time-integrated BMP signaling determines fate in a stem cell model for early human development. Nat Commun 15, 1471 (2024).

Nemashkalo, A., Ruzo, A., Heemskerk, I. & Warmflash, A. Morphogen and community effects determine cell fates in response to BMP4 signaling in human embryonic stem cells. Development 144, 3042–3053 (2017).