Breakthrough moments: Yoshiki Sasai’s discoveries in the 3^rd dimension

Abstract

Yoshiki Sasai pioneered the organoid field with his idea of mimicking embryonic development in 3D. We shine a spotlight on his seminal work describing how the innate ability of embryonic stem cells to self-organize into layers and grow in a polarized fashion fosters their appropriate differentiation and response to morphogens.

As Hans Clevers highlighted in 2016, the use of the term “organoid” witnessed a revival in recent years and acquired a new meaning. In the past four decades, “organoids” were mainly used in classic developmental biology experiments in the effort to study organogenesis by cell dissociation-reaggregation experiments. Now the term refers to three-dimensional (3D) aggregates obtained from embryonic stem cells (ESCs), induced pluripotent stem cells (iPSCs), and tissue-specific adult stem cells (aSCs), implying an organized system where undifferentiated cells develop in a coordinated fashion. Yoshiki Sasai pioneered this new conceptualization.

Sasai’s group was the first to propose the idea that pluripotent cells can differentiate and generate highly complex and organized structures mimicking normal tissue and organ development when grown in 3D. He envisioned that, under 3D culture conditions, cell differentiation could be optimized and more physiological because the cells, free from physical constaints imposed by the plastic surface of a culture dish, have the ability to rearrange themselves and self-organize in layered and polarized stuctures as dictated by their own surface properties and internal genetic programs. More than 10 years ago, Sasai’s laboratory designed conditions with tighly controlled medium and cell number to re-create the right environment for cells to spontaneously organize in the 3D space and prefiguring a complex organ, such as the brain. They pioneered the generation of 3D forebrain structures from mouse and human ESCs, quickly followed by optic cup, neocortex, cerebellum, hippocampus, adenohypophysis, ventral telencephalon and pituitary gland (Kadoshima et al., 2013; Muguruma et al., 2015; Ozone et al., 2016; Sakaguchi et al., 2015; Sasai, 2013). This was a stunning demonstration of Sasai’s idea that ESCs realize a higher level of spatiotemporal organization when growing in a “multicellular society”, where their interactions actuated their own internal genetic programs, possibly through local signaling systems (Sasai, 2013). Among all Sasai’s protocols, common parameters include an initial aggregation of stem cells, a general neural induction, and in many cases, brief periods of exposure to specific morphogens for tissue-specific patterning and growth factor stimulation, followed by an extended period in growth factor-free medium to favor cell cycle exit and differentiation. Under these conditions, the community of ESC-derived neural progenitors goes though a complex series of stereotypied events resulting in tissue-like stratifications, compartment formation, and coordinated cell movements (like in the formation of an optic cup) in the absence of further exogenous signaling. Even though Sasai’s team did not call their 3D structures “organoids”, their basic procedural layout is followed by organoid protocols today. Building on the long-born concept that in the absence of epidermal and mesodermal inducing signals (mimicked in vitro by Nodal, BMP and Wnts inhibitors), ESCs will intrinsically acquire rostral neural fate by a “default model of neural induction”, Sasai’s team achieved the differentiation of dorsal telencephalic precursors from aggregates of mouse ESCs, called embryoid bodies (EBs) under conditions that minimize the addition of external pattern-inducing molecules (Watanabe et al., 2005)). A few years later, they developed a 3D-aggregation culture system, which they called SFEBq (serum-free floating culture of EB-like aggregates with quick reaggregation) in which aggregates of mouse and human ESCs spontaneously self-organized to form apico-basally polarized cortical tissue (Sasai, 2013). These aggregates could be further patterned to rostral or caudal cortical fate by the addition of molecules that created intra-aggregate morphogen gradients, such as treatment with FGF8-promoting rostral pallial fate. In his seminal paper on SFEBq culture, Sasai’s group used mouse and human ESCs. Inspired by this work, we applied his protocol to human iPSCs. Our findings (Mariani et al., 2012) suggested that iPSCs could be used with Sasai’s procol to generate bona fide human cortical neurons. This was later confirmed by Sasai’s group (Kadoshima et al., 2013) and opened the way to other scientific teams, including ours, for generating stem cell-derived dorsal forebrain organoids in a more reproducible and higher thoroughtput fashion (Mariani et al., 2015; Qian et al., 2016).

A decade after Sasai established the foundations for the field, organoid culture methods have diversified. A variety of organoid models are used today: unpatterned organoids containing a variety of brain regions and sometimes including sensory cells such as retina (Lancaster et al., 2013); patterned organoids, where stem cells are directed to differentiate into a specific regional fate (spinal cord, cortex, basal ganglia, cerebellum) by the addition of extrinsic morphogens; and organoids derived from adult tissue-specific stem cells, such as intestinal stem cells. Each system has its own advantages and its use is dependant on the experimental questions being asked. Yet, we believe it was Sasai’s seminal concept of self-organization and his emphasis on the power of intercellular signaling via endogenously generated morphogens that laid the foundations for the “organoid” field, including his realization that it could be harnessed for a variety of different cell and organ types. Brain organoids have become very powerful and invaluable research tools to understand/mimic normal and abnormal brain development beginning from its very early stages in a dynamic way. Moreover, the ability to generate “personal” organoids from patient-specific iPSCs has opened new frontiers for disease modeling, drug discovery, and gene therapy.

The field of brain organoids is still in its infancy. Being able to mimic later stages of brain development and maturation is, and is going to be, one of the main efforts of many scientists in the next few years. Also, the development of robust and, at the same time, scalable in vitro methods for the generation of brain organoids is on the top list of many laboratories. Understanding the stengths and limitations of each in vitro modelling system will help the field move forward. At the same time, a key question that we should constantly keep asking is to what extent organoids mimic their in vivo counterpart. Ultimately, as Sasai undoubtedly realized, organoids should not be taken as a system that reproduces comprehensivelly and faithfully the developing brain, but instead as a model for key aspects of brain development. Organoids may help us unravel, in an experimentally tractable fashion, developmental principles that would otherwise be impossible to understand in a human system as well as help us disentangle the collective cellular, molecular and genetic events undelying morphogenesis.