📄 developmental biology

Distinct neurogenic progenitor cell populations balance cell type production in the embryonic mouse retina

This study reveals that robust production of diverse retinal cell types in the embryonic mouse retina is achieved through Notch-mediated lateral inhibition, which balances two distinct, fate-biased neurogenic progenitor populations (Galanin+ and Olig2+) that arise from asymmetric divisions and generate specific neuronal subsets.

Original authors: Bushnell, H., Cepko, C.

Published 2026-08-04

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Bushnell, H., Cepko, C.

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ⚕️ This is an AI-generated explanation of a preprint that has not been peer-reviewed. It is not medical advice. Do not make health decisions based on this content. Read full disclaimer

Imagine the human body as a bustling construction site, but instead of bricks and mortar, it's built from tiny, living cells. In the early days of life, before you even have eyes to see with, a special team of "master builders" called progenitor cells gets to work. These cells are like versatile apprentices; they haven't decided what kind of worker they want to be yet. They could become a skin cell, a muscle cell, or, in the case of the eye, a neuron that helps you see. The big mystery scientists have been trying to solve is: how does this chaotic construction site end up with the perfect number of every single type of worker? If the site produces too many electricians and not enough plumbers, the building won't work. In the eye, specifically the retina (the part at the back that acts like the film in a camera), there are over 100 different types of neurons that need to be built in the exact right proportions to let you see the world. For decades, scientists wondered if these master builders just rolled dice to decide their fate, or if there was a strict, hidden blueprint guiding them.

This paper dives into that mystery by looking at the construction site of a mouse eye while it's still being built in the womb. The researchers discovered that the master builders aren't just rolling dice; they are actually splitting into two distinct teams with very different job descriptions. One team, identified by a protein called Galanin, is biased to build the "messengers" of the eye (retinal ganglion cells) and the "local coordinators" (amacrine cells). The other team, marked by a protein called Olig2, is biased to build the "light sensors" (cones) and the "horizon detectors" (horizontal cells).

Here is the clever part: these two teams don't just appear out of nowhere. They are created through a process called asymmetric division. Imagine a master builder splitting in two. Instead of making two identical apprentices, one half stays a master builder (ready to keep the team growing), while the other half immediately picks a side, becoming either a Galanin-team member or an Olig2-team member. This ensures that the construction site always has a steady supply of builders while simultaneously producing the specific workers needed for the next stage.

But how do the builders know which team to join? The paper suggests a game of "hot potato" played with a chemical signal called Notch. Think of the Notch signal as a loud buzzer. When a builder on the Olig2 team (the light sensor builders) is active, it sends out a strong signal to its neighbor, effectively saying, "Hey, don't be like me! Be the opposite!" This signal pushes the neighbor to become a Galanin-team member (the messenger builder). In return, the Galanin team member, which has a low level of this signal, allows its neighbor to stay or become an Olig2 member. This back-and-forth, known as lateral inhibition, acts like a local feedback loop. It ensures that the two types of builders are balanced right next to each other, preventing the construction site from accidentally making only one type of worker.

The researchers tested this idea by turning the "buzzer" up and down in lab experiments. When they silenced the signal, the builders all rushed to join the Olig2 team. When they cranked the signal up, they all joined the Galanin team. This proves that the Notch signal is the switch that decides which team a builder joins. While the paper doesn't claim to have solved the entire mystery of how the eye is built, it strongly suggests that this local, neighbor-to-neighbor conversation is the key to keeping the eye's construction site perfectly balanced, ensuring that every cell type is produced in the right amount to give us clear vision.

Problem
A central question in developmental biology is how cellular diversity is reliably generated in correct proportions during mammalian development, particularly in tissues like the retina where over 100 neural cell types arise from a common pool of multipotent retinal progenitor cells (RPCs). While transcription factors required for specific cell fates are known, the mechanisms by which individual progenitor cells select among multiple fate options and coordinate these decisions locally to ensure robust tissue-level composition remain unresolved. Specifically, while fate-biased progenitor populations producing cones and horizontal cells (marked by Olig2) have been identified, the progenitor population responsible for generating retinal ganglion cells (RGCs) and amacrine cells during the same early embryonic window (E12–E16) in mice had not been defined. Furthermore, the mechanisms balancing these distinct neurogenic outputs, particularly the role of Notch signaling in coordinating neighboring progenitor behaviors, required further elucidation.

Methodology
The study employed a multi-modal approach combining single-cell RNA sequencing (scRNA-seq), genetic lineage tracing, and functional perturbation:

scRNA-seq Integration: The authors integrated published scRNA-seq datasets from E13.5 and E14.5 mouse retinas (7,016 cells) to identify transcriptional heterogeneity within neurogenic progenitor cells (NPCs). They used Louvain clustering to resolve subpopulations based on cell cycle markers and fate determinants.
Genetic Lineage Tracing: To validate and characterize the identified populations, the authors utilized Gal-Cre mice crossed with Ai140 (tdTomato) and Confetti reporter strains. This allowed for the permanent labeling of Galanin (Gal)-expressing cells and their progeny, enabling clonal analysis of cell type output. Sparse labeling of embryonic RPCs was achieved using Mcm2-CreER crossed with Ai140 to analyze the clonal relationships between Gal+ and Olig2+ NPCs.
Spatial Validation: Hybridization Chain Reaction RNA fluorescence in situ hybridization (HCR RNA-FISH) and immunostaining were used to validate marker expression (Gal, Olig2, Dll1, Dll4, cleaved Notch1 Intracellular Domain [N1ICD]) and cell cycle status (EdU) in tissue sections.
Functional Perturbation: Embryonic retinal explants (E13.5) were electroporated with plasmids to manipulate Notch signaling. Notch activity was inhibited using the $\gamma$ -secretase inhibitor LY411575 or activated via overexpression of the Notch Intracellular Domain (NICD).
Transcriptomic Profiling: Following perturbation, electroporated cells were isolated via FACS and subjected to bulk RNA-seq to assess transcriptional shifts. Quantitative analysis of NPC populations in explants was performed using HCR RNA-FISH.

Key Contributions and Results

Identification of Distinct NPC Populations: The study identified a previously undefined population of NPCs expressing Galanin (Gal). These Gal+ NPCs are transcriptionally distinct from the previously characterized Olig2+ NPCs.
- Gal+ NPCs: Express RGC fate determinants (Dlx1, Dlx2, Isl1) and are biased to produce RGCs and amacrine cells. Clonal analysis revealed that Gal+ NPCs undergo terminal divisions, predominantly generating RGC-RGC, RGC-amacrine, or amacrine-amacrine pairs. These clones also include intrinsically photosensitive RGCs (ipRGCs) and displaced cells, suggesting Gal+ NPCs are multipotent within the RGC/amacrine lineage but do not produce photoreceptors.
- Olig2+ NPCs: Confirming previous work, these cells are biased to produce cones and horizontal cells.
Mechanism of NPC Genesis: Clonal analysis of sparsely labeled RPCs revealed that Gal+ and Olig2+ NPCs rarely arise as sibling pairs from the same division. Instead, they predominantly arise from asymmetric divisions where an RPC produces one fate-biased NPC (Gal+ or Olig2+) and one multipotent RPC sibling. This suggests a model where multipotent RPCs sequentially generate different fate-biased NPC states.
Notch Signaling as a Regulator of NPC States: The study established that Gal+ and Olig2+ NPCs correspond to distinct Notch signaling states:
- Olig2+ NPCs represent a NotchLow state and are the primary source of Delta-family ligands (specifically Dll4) in the embryonic retina.
- Gal+ NPCs represent a NotchHigh state, expressing high levels of canonical Notch targets (Hes1, Hes5, etc.) and low levels of Delta ligands.
- Lateral Inhibition: The spatial arrangement of these cells supports a lateral inhibition mechanism. Dll4-high Olig2+ cells neighbor N1ICD-high Gal+ cells, suggesting Olig2+ cells activate Notch in neighbors to drive them into the Gal+ state, while Gal+ cells (NotchHigh) inhibit neighbors from adopting the Gal+ state, allowing them to become Olig2+.
Causal Role of Notch: Perturbation experiments demonstrated that Notch signaling causally specifies NPC populations.
- Inhibition of Notch (LY411575) shifted the balance toward Olig2+ NPCs and photoreceptor differentiation markers.
- Activation of Notch (NICD overexpression) increased the proportion of Gal+ NPCs and markers of RPC proliferation and nascent amacrine/horizontal cell identity. Neither perturbation induced RGC markers; however, NICD expression significantly reduced levels of RGC markers (Rbpms, Isl1, and Pou4f2), suggesting that while Notch activity promotes the Gal+ NPC state, the subsequent transition to the RGC fate or expression of RGC differentiation markers may require downregulation of Notch activity.

Significance
The paper proposes a model where the robust production of diverse retinal cell types is achieved through local feedback mechanisms that balance discrete, fate-biased NPC populations. Rather than relying solely on stochastic cell-intrinsic programs, the retina utilizes Notch-mediated lateral inhibition between neighboring progenitors to stabilize complementary states (Gal+ NotchHigh vs. Olig2+ NotchLow). This mechanism ensures that the production of RGCs/amacrine cells and cones/horizontal cells is coordinated locally, providing a solution to how variable clonal outputs can result in reproducible tissue-level cell type proportions. The findings reconcile previous contradictory reports on Notch's role in retinal fate by suggesting that Notch acts at multiple sequential decision points, specifically directing the choice between distinct neurogenic NPC states.

More like this