Student Theses and Dissertations


Iain Martyn

Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)


In amniotes, gastrulation is marked by the creation of the primitive streak (PS) and is largely controlled by WNT, BMP, and ACTIVIN/NODAL signalling. Despite detailed characterization in model organisms, the human PS and the role these pathways play in its formation and patterning remains a mystery. In this work I focused on understanding the role and control of the WNT pathway in human PS development. Due to the ethical limitations of working with human embryos, I used an in vitro human embryonic stem cell (hESC) micropatterned “gastruloid” system. I first showed that in the human PS there is a conserved BMP → WNT→ NODAL signalling initiation hierarchy, and that WNT is necessary and sufficient for PS formation. Next, I found that structured subpopulations of endoderm and mesoderm emerge and self-organize depending on different BMP, WNT, and ACTIVIN/NODAL levels, and that by comparison to the mouse embryo I could arrange these subpopulations along an anterior-posterior axis. With the development of a new cell tracking technique, I was also able to identify and characterize robust cell migrations from the PS region of each gastruloid that depended on which fates the cells would ultimately adopt. Putting these pieces together, I was able to derive a rudimentary first fate map of the human PS, as well as a rough picture of the BMP, WNT, and ACTIVIN/NODAL signalling gradients that determine it. One interesting and unforeseen result from this fate map was the hint of a human “organizer” cell fate that emerged under joint WNT and ACTIVIN/NODAL stimulation. To characterize and functionally prove this organizer’s existence, I devised an ex ovo cross-species transplantation strategy grafting treated gastruloids into chick embryos. The assay demonstrated that the human cells induce and contribute autonomously to a secondary axis while inducing neural fate in the host, thus fulfilling the most stringent criteria for an organizer. This work adds an important milestone to the research program begun in 1924 with the first famous organizer experiment of Hilde Mangold and Hans Spemann, and the methods I developed have opened a door to new functional explorations and tests of early human development. Having learned more about the role of WNT in determining cell fates in the gastruloid model, I next endeavoured to understand how the spatial extent and duration of the WNT signal itself was controlled. With the use of various CRISPR/Cas9 knockout lines, I discovered that DKK1 and E-CADHERIN were the two dominant factors, with E-CADHERIN transducing boundary forces to focus WNT signalling to colony border at early times, and DKK1 controlling the late WNT pattern via cell non-autonomous negative feedback. With the help of time-lapse imaging of a fluorescent reporter line and mathematical modelling, I showed that these two factors mediate a wave of WNT signalling that spreads across the tissue to be patterned, and that this wave is a generic property of a bistable system and thus likely generalizable to other instances in development. While limited by the use of hESCs, taken together my findings provide a first glimpse into the role and control of WNT signalling early on in our own, human development.


A Thesis Presented to the Faculty of The Rockefeller University in Partial Fulfillment of the Requirements for the degree of Doctor of Philosophy

Available for download on Thursday, December 12, 2019

Included in

Life Sciences Commons