The people behind the papers – Yuki Sato

Yuki Sato

Can you give us your scientific biography and the questions your lab is trying to answer?

I worked in Yoshiko Takahashi's group from graduate school until my first postdoctoral fellowship. Since then, I have worked with quail and chicken embryos as model animals. I then joined Scott E. Fraser's lab as a postdoctoral fellow to learn about biological imaging. I participated in the transgenic quail imaging project led by Rusty Lansford and was able to expand my knowledge of avian genetic modification in addition to light microscopy. Since returning to Japan, I have been leading small research groups as an independent investigator. Avian embryos are attractive because various types of experimental manipulations are available, such as tissue transplantation, stage- and tissue-specific gene expression, and live imaging. There is also a sense of speed: a hypothesis made on Monday will be known to be right or wrong by Friday. This allows us to try different things without fear of detours. We just published our first paper on embryonic hematopoiesis. I realized that endothelial-to-hematopoietic transition (EHT) is an interesting phenomenon from both a developmental and a cell biological point of view. We would like to further understand the mechanism of EHT.

What was known about the mechanisms driving the cell morphological changes during EHT before your work?

As seen with trypsin treatment, cultured cells quickly become rounded when cell–cell and cell–substrate adhesions are removed. In contrast, hemogenic endothelial cells undergo cell rounding while maintaining adherens junctions during EHT independently of mitosis. In 1999, Nancy A. Spec and colleagues showed that the transcription factor Runx1 (Cbfa2/AML1) is essential for EHT (the term EHT did not exist at that time and was described as budding from the vessel wall; North et al., 1999) in mouse embryos. Based on electron microscopy, Runx1-positive hemogenic endothelial cells budding from the dorsal aorta had a vacuole-like structure (described as a ‘cystic separation’ in the original paper). However, since this report in 1999, the mechanism of vacuole formation and its role in hemogenic endothelial cells have not been elucidated. We began to focus on these vacuoles about 5 years ago, starting with the localization of aquaporin 1 (AQP1).

What led you to focus on the aquaporin water channels?

Originally, I was interested in how individual cell behavior contributes to the extensive network formation of blood vessels. Vascular endothelial cells change polarity and migrate in response to the mechanical stress caused by blood flow. It seems rational that blood flow stimulation acts as a cue to control endothelial cell migration, because it self-optimizes vessel diameter and branching patterns while maintaining vascular continuity. The elements that create blood flow are the heartbeat and blood. Although the mechanisms of heart formation have been elucidated, we do not know when and where the liquid component (body fluid) of the initial blood is produced. Therefore, we predicted that water channels are involved in fluid accumulation and examined the expression of the aquaporin (AQP) family genes. Among the AQP genes, AQP1 was particularly prominent in the extra-embryonic region (the region where primary hematopoiesis occurs); therefore, we decided to start our analysis with AQP1. We overexpressed AQP1 in the extra-embryonic blood vessels of quail embryos and found that the endothelial cells became progressively rounded and were even released into circulation. This does not normally occur in endothelial cells residing in the extra-embryonic region. Based on this phenomenon, we hypothesized that EHT was caused by cell swelling (we had already forgotten the original issue – fluid accumulation in the initial blood at that moment!). We generated an antibody against avian AQP1, examined its localization, and found that AQP1 localizes to the vacuole membrane of hemogenic endothelial cells. We found this very interesting because vacuoles are known to play multiple functions in plants, but their role in animal cells is not well understood. In summary, we focused on AQP for a completely different purpose, which ultimately led to our interest in the EHT.

Can you summarize your key findings?

EHT is an important phenomenon, in which hematopoietic cells are generated from vascular endothelial cells. During EHT, endothelial cells change from a flattened morphology to a spherical shape, the mechanism of which remains elusive. We found that AQP-mediated water permeation promotes vacuole formation within hemogenic endothelial cells, which in turn causes cell rounding. This suggests that cells involved in the circulatory system have a water-adaptive morphogenetic mechanism.

Did you have any particular result or eureka moment that has stuck with you?

The night I first saw the image of AQP1-overexpressing cells ectopically forming large vacuoles with rounded cell shapes, I could not sleep well because I was so excited that my hypothesis was proven. However, several days after this finding, AQP1 knockout cells showed no effect on EHT, in contrast to AQP1-overexpressing cells. I had seen and heard many cases where single gene knockouts did not show a phenotype, so this was somewhat expected, but the length of the road ahead was beyond my imagination. I did not expect to knock out as many as four AQP genes. Under normal conditions, Runx1-positive hemogenic endothelial cells are hemispherical or completely spherical, so when I first saw the flattened Runx1-positive cell population in AQP multiple knockout embryos, I jumped for joy in a darkened microscope room.

On the flipside, were there any moments of frustration or despair?

In addition to AQP1, AQP5, AQP8 and AQP9, I also performed CRISPR/Cas9 genome-editing experiments on AQP3 and AQP11. For each of these six AQP and Runx1 genes, the entire gene knockout analysis process, including construction of gRNA expression vectors, electroporation, genomic PCR, cloning and sequencing, was performed by myself. I have never cloned so many genes in such a short time. I took care to avoid all possible confusion. Ironically, during the COVID-19 pandemic, conferences, personal meetings, and parties were all canceled, giving me time to lock myself in the lab and clone in silence. I was lucky to have nothing else to do.

Where will this story take the lab and what is next for you after this paper?

We hope to understand the mechanophysiological drivers of EHT. The mechanisms of AQP regulation and cellular responses to water influx are the main topics of current research. In our study, we were unable to elucidate the conditions under which the AQP channel opens. We are trying to clarify whether it is osmotic pressure, hydrostatic pressure, or another unknown mechanism. We are also attempting to uncover the internal changes caused by water influx resulting from artificial AQP overexpression in comparison to normal hemogenic endothelial cells. Furthermore, the relationship between AQP and other EHT-related molecules is a largely unexplored mystery. We are looking for colleagues to solve these issues.

Finally, outside the lab, what do you like to do in your spare time?

I love crafts and currently have a passion for assembling mechanical 3D puzzles. If these had been available in my childhood, my life would have been different.