Efficient differentiation of human primordial germ cells through geometric control reveals a key role for Nodal signaling
- Department of Cell and Developmental Biology, University of Michigan Medical School, United States
- Department of Biomedical Engineering, University of Michigan, United States
- Center for Organogenesis, University of Michigan Medical School, United States
- Department of Internal Medicine, Gastroenterology, University of Michigan Medical School, United States
- Department of Physics, University of Michigan, United States
Figures
Figure 1 with 3 supplements
Primordial germ cell-like cells (PGCLCs) form at the interface between extraembryonic and primitive streak-like cells.
(A, B) Immunofluorescence for different marker genes (maximal intensity projection along z). Yellow arrowhead in (B) points to higher NANOG expression in PGCLCs than pluripotent cells in the colony …
Figure 1—figure supplement 1
Quantitative relationships between marker genes with immunofluorescence (IF) and scRNA-seq.
(A–F) Additional stainings of micropatterned colonies and spatial distribution of positive cells. (G) Breakdown of cell populations corresponding to Figure 1G and (E, F). (H, I) Normalized radial …
Figure 1—figure supplement 2
Additional scRNA-seq analysis, including correlation with CS7 human gastrula.
(A, B) Comparison of immunofluorescence (IF) for ISL1, TBXT, SOX2 with their expression on the PHATE visualization of the scRNA-seq data. (C) Primordial germ cells (PGCs) budding off as a lineage …
Figure 1—figure supplement 3
Primordial germ cell-like cell (PGCLC) differentiation in different male and female cell lines.
Each row shows immunofluorescence (IF) and quantification for PGC markers in different cell lines. (A–C) ESI017, also shown in Figure 1B, shown here for comparison. (E–F) PGP1, male induced …
Figure 2 with 2 supplements
Primordial germ cells (PGCs) are specified by 42 hr but continue to mature while endoderm arises between 42 and 72 hr.
(A, B) Immunofluorescence over time showing a stable PGC population and later emergence of endoderm. (C) Quantification of marker expression at different times showing the emergence of endoderm …
Figure 2—figure supplement 1
Additional images and quantification for time series up to 96 hr.
(A) DAPI staining for colonies shown in Figure 2A and B. (B) Radial profile of marker expression at different times. (C) Overlay of colonies from Figure 2A and B, including DAPI, showing location of …
Figure 2—figure supplement 2
Additional images for pluripotency markers over time.
(A, B) Separate channels for Figure 2G. (C) Scatterplots of corresponding quantification.
Figure 3 with 1 supplement
Primordial germ cell-like cells (PGCLCs) require sustained BMP, Nodal, and FGF but only brief Wnt signaling.
Each row shows staining and quantification of PGC markers after perturbation of different pathways. Error bands in spatial distributions are omitted for clarity but are similar in magnitude to Figure…
Figure 3—figure supplement 1
Additional images and quantification for signaling perturbations in Figure 3.
(A) Overlay and separate channels for Figure 3A. (B) scatterplot of PRDM1 vs. TBXT for BMP4-treated colonies with or without BMP-receptor inhibition after 24 hr, colored for the condition. (C) …
Figure 4 with 2 supplements
Exogenous Activin rescues primordial germ cells (PGCs) in the absence of endogenous Wnt or Nodal in a dose- and time-dependent manner.
(A, B) Wnt inhibition (WNTi, IWP2, 5 µM) with different doses of Activin, for example, A3 = 3 ng/ml Activin. (C–E) Activin rescue of WNTi at 0 hr vs. 12 hr with LEF1 staining. (F–H) Like (C–E) but …
Figure 4—figure supplement 1
Additional images of primordial germ cell (PGC) rescue by Activin after WNT inhibition.
(A, B) Individual channels for Figure 4A and C. (C) IWR-1 does not block PGC specification at 10 µM. (D) Individual channels for Figure 4F (IWR-1 inhibition at 50 µM and rescue of PGCs by Activin). …
Figure 4—figure supplement 2
Additional images of the effect of Activin timing and dose on primordial germ cell (PGC) specification.
(A–C) Images of individual channels for Figure 4K, M, and R. (A) corresponds to Figure 4K, (B) to Figure 4M, (C) to Figure 4R. (D, E) Effect of exogenous Activin timing on cell fate in NodalKO …
Figure 5 with 1 supplement
Control of colony size dramatically impacts the fraction of primordial germ cell-like cells (PGCLCs).
(A–D) Different diameter colonies stained for TFAP2C, SOX17, EOMES at 42 hr and quantification, (B) SOX17 vs. TFAP2C scatterplot colored for colony size. (C) Same plot colored for EOMES expression. …
Figure 5—figure supplement 1
Additional images of primordial germ cell (PGC) specification on small micropatterns.
(A, B) Staining for PGC markers and quantification after treatment with Activin for 24 hr followed by inhibition of TGFb signaling by SB431542 (compare Figure 4R). (C) Scatterplots corresponding to …
Figure 6 with 1 supplement
A network of cross-repressive cell fates qualitatively explains Nodal perturbations.
(A) Input signaling profile in space and time. (B) Diagram of the model. (C) Definition of the model. (D) Phase diagram showing predicted expression of cell fate markers at steady state for …
Figure 6—figure supplement 1
Simpler mathematical models.
Each row shows a model and its predicted patterns for the perturbations we wanted to explain. (A) Stable primordial germ cell-like cells (PGCLCs) through autoregulation of SOX17 in combination with …
Author response image 1
Author response image 2
Micropatterned PGCLCs.
Author response image 3
Author response image 4
Author response image 5
Author response image 6
Tables
Table 1
Cell signaling reagents.
| Reagent | Nickname | Vendor, catalog # | Dose | Function |
|---|---|---|---|---|
| rhBMP4 | BMP4 | R&D Systems, #314BP/CF | See figures | Activate BMP pathway |
| rhActivin | A | R&D Systems, #AFL338 | See figures | Activate TGFb pathway |
| CHIR-99021 | C | Tocris, #4423 | See figures | Canonical Wnt agonist |
| IWP 2 | WNTi | Tocris, #3533 | 5 µM unless stated otherwise | Block Wnt secretion |
| IWR-1 | cWNTi | Thermo Fisher, 50-101-4191 | 50 µM unless stated otherwise | Block canonical Wnt signaling |
| LDN-193189 | BMPRi | MedChemExpress, # HY-12071 | 250 nM | Block BMP signaling |
| SB-431542 | TGFBRi | Apexbio, #A8249 | 10 µM | Block TGFb signaling |
| PD-0325901 | MEKi | ESIBIO, #ST10009 | 5 µM | Block MEK signaling |
| PD-173074 | FGFRi | MedChemExpress, #HY-10321 | 1 µM | Block FGF signaling |
Table 2
Primary antibodies used for immunostaining.
| Protein | Species | Dilution | Catalog # | Vendor |
|---|---|---|---|---|
| ISL1 | Mouse | 1:200 | 39.4D5 | DSHB |
| SOX2 | Rabbit | 1:200 | 3579S | Cell Signaling Technology |
| TBXT (BRA) | Goat | 1:300 | AF2085 | R&D Systems |
| PRDM1 (BLIMP1) | Rat | 1:50 | SC-47732 | Santa Cruz Biotechnology |
| SOX17 | Goat | 1:200 | AF1924 | R&D Systems |
| TFAP2C | Mouse | 1:150 | SC-12762 | Santa Cruz Biotechnology |
| NANOG | Goat | 1:100 | AF1997 | R&D Systems |
| EOMES (TBR2) | Rabbit | 1:500 | AB23345 | Abcam |
| POU5F1 | Mouse | 1:400 | 611,202 | BD Biosciences |
| LEF1 | Rabbit | 1:200 | C12A5 | Cell Signaling Technology |
Table 3
Secondary antibodies.
| Protein | Species | Dilution | Catalog # | Vendor |
|---|---|---|---|---|
| Alexa Fluor 647 anti-goat | Donkey IgG | 1:500 | A21447 | Thermo Fisher Scientific |
| Alexa Fluor 555 anti-goat | Donkey IgG | 1:500 | A21432 | Thermo Fisher Scientific |
| Alexa Fluor 488 anti-mouse | Donkey IgG | 1:500 | A21202 | Thermo Fisher Scientific |
| Alexa Fluor 647 anti-rat | Whole IgG | 1:500 | 112-605-167 | Jackson ImmunoResearch |
| Alexa Fluor 647 anti-rabbit | Donkey IgG | 1:500 | A31573 | Thermo Fisher Scientific |
| Alexa Fluor 555 anti-rabbit | Donkey IgG | 1:500 | A31572 | Thermo Fisher Scientific |
Table 4
qPCR primers.
| GAPDH | ACAACTTTGGTATCGTGGAAGG | GCCATCACGCCACAGTTTC |
| SOX17 | GTGGACCGCACGGAATTTG | GGAGATTCACACCGGAGTCA |
| NANOS3 | CTTTGACCTGTGGACAGATTACC | GCCTGGTTTCAGGACCCTC |
| DPPA3 | TTAATCCAACCTACATCCCAGGG | AGGGGAAACAGATTCGCTACTA |
| DDX4 | TTGTTGCTGTTGGACAAGTGGGTG | GCAACAAGAACTGGGCACTTTCCA |
| EOMES | CGCCACCAAACTGAGATGAT | CACATTGTAGTGGGCAGTGG |
| PRDM1 | CTACCCTTATCCCGGAGAGC | GGACATTCTTTGGGCAGAGT |
Appendix 1—table 1
Model parameters.
| Parameter | Value | Meaning |
|---|---|---|
| n | 2 | Hill function coefficient |
| 0.1733 | Protein dilution + degradation rate | |
| 0.173 | Production rate for and | |
| 1 | Autoactivation steady state for and | |
| (0.313, 0.338, 0.371, 1.2) | Signaling thresholds | |
| Inhibition thresholds | ||
| (0.269, 0.135) | Maximum signaling-driven protein level | |
| Autoactivation thresholds | ||
| 0.5 | activation threshold for on and | |
| 100 | dimerization constant for |
Additional files
-
Supplementary file 1
List of gastrulation genes used for visualization and clustering.
- https://cdn.elifesciences.org/articles/72811/elife-72811-supp1-v2.xlsx
-
Supplementary file 2
Most differentially expressed genes in clusters found in scRNA-seq data.
- https://cdn.elifesciences.org/articles/72811/elife-72811-supp2-v2.xlsx
-
Supplementary file 3
Most differentially expressed genes from Supplementary file 1 in clusters found in scRNA-seq data.
- https://cdn.elifesciences.org/articles/72811/elife-72811-supp3-v2.xlsx
-
Transparent reporting form
- https://cdn.elifesciences.org/articles/72811/elife-72811-transrepform1-v2.docx
