henhaobio
Sign in

NatureDevelopmental BiologyPeer-reviewed journal article

A comprehensive series of temporal transcription factors in the fly visual system

Nikolaos KonstantinidesiD, Isabel HolgueraiD, Anthony M. RossiiD, Aristides Escobar, Liébaut DudragneiD, Yen‐Chung CheniD, Thinh Ngoc TraniD, Azalia M. Martínez JaimesiD, Mehmet Neset ÖzeliD, Félix SimoniD, Zhiping ShaoiD, Nadejda M. TsankovaiD, John F. FullardiD, Uwe WalldorfiD, Panos RoussosiD, Claude DesplaniD

PDFDOIPreprintCodeData
In-depth full-text digest available · Jump to digest ↓

Abstract

The brain consists of thousands of different neuronal types that are generated through multiple divisions of neuronal stem cells. These stem cells have the capacity to generate different neuronal types at different stages of their development. In Drosophila, this temporal patterning is driven by the successive expression of temporal transcription factors (tTFs). While a number of tTFs are known in different animals and across various parts of the nervous system, these have been mostly identified by informed guesses and antibody availability. We used single-cell mRNA sequencing to identify the complete series of tTFs that specify most Drosophila medulla neurons in the optic lobe. We tested the genetic interactions among these tTFs. While we verify the general principle that tTFs regulate the progression of the series by activating the next tTFs in the series and repressing the previous ones, we also identify more complex regulations. Two of the tTFs, Eyeless and Dichaete, act as hubs integrating the input of several upstream tTFs before allowing the series to progress and in turn regulating the expression of several downstream tTFs. Moreover, we show that tTFs not only specify neuronal identity by controlling the expression of cell type-specific genes. Finally, we describe the very first steps of neuronal differentiation and find that terminal differentiation genes, such as neurotransmitter-related genes, are present as transcripts, but not as proteins, in immature larval neurons days before they are being used in functioning neurons; we show that these mechanisms are conserved in humans. Our results offer a comprehensive description of a temporal series of tTFs in a neuronal system, offering mechanistic insights into the regulation of the progression of the series and the regulation of neuronal diversity. This represents a proof-of-principle for the use of single-cell mRNA sequencing for the comparison of temporal patterning across phyla that can lead to an understanding of how the human brain develops and how it has evolved.

In-depth summary

Claude Opus 5.5 · xhigh effort · 8 Nov 2026 · from the full text of version 2 (7 Nov 2021) · CC BY 4.0

Drosophila larval optic lobe single-cell RNA-seq and pseudotime analysis define a near-complete medulla neuroblast temporal transcription factor series, with Eyeless and Dichaete acting as regulatory hubs.

Temporal transcription factors (tTFs) had been found piecemeal through antibody availability or candidate genes, so full temporal series were unknown and cross-species comparison was fragmentary. This work screens every TF along a neuroblast trajectory, validates new tTFs and their epistasis in vivo, and links them to neuronal identity and a shared differentiation program.

Key results

  1. ShownAn unbiased pseudotime screen of all transcription factors in medulla neuroblasts recovered every known tTF plus new candidates whose proteins mark restricted windows. Together they give 12 conservatively defined temporal windows.

    Among 629 FlyBase-annotated TFs, 39 had temporally restricted pseudotime expression: 14 at high levels (the 6 known tTFs plus eight new) and 25 at low, fluctuating levels; four of these 25 tested in vivo were not temporal. Antibodies confirmed windows for Opa (two waves), Erm, Hbn, Scro and BarH1; Esg was salt-and-pepper and Oaz-Gal4 labeled all neuroblasts. · A comprehensive temporal series of transcription…; Trajectory analysis: identification of candidate…; Discussion; Figure legends

  2. ShownMutant clones and RNAi epistasis revealed a branched network rather than a linear relay. Erm and Hbn activate the next tTF and repress the previous one, Opa times Ey onset, Scro links Ey to D, and Ey and Dichaete (D) act as integrating hubs.

    erm clones lost Ey and expanded Hth; opa clones delayed Ey (and later tTFs) without affecting Hth or Erm; ey clones lost Scro and Hbn and extended Erm and Opa; scro RNAi kept Ey to the last division and lost D; hbn clones extended Ey and lost Slp and the second Opa wave; D clones lost BarH1; tll clones left BarH1 unchanged (at least four brains per experiment). · Early unit; Middle unit; Late unit; Immunohistochemistry

  3. ShowntTFs also specify neuronal identity: most persist as proteins in neurons born in their window, Hbn and Opa are needed for specific downstream neuronal TF populations, and tTF expression assigns medulla cell types to their temporal windows of origin.

    hbn mutant clones lacked Traffic-jam and Otd neurons; opa clones had fewer TfAP-2 neurons than adjacent wild-type tissue. Opa and BarH1 proteins persisted in newborn neurons although their mRNAs were not detected there. Hey-based Notch calls: 78 neuronal types Notch-OFF, 52 Notch-ON; all Dm types traced to Ey/Hbn windows. · Temporal transcription factors often remain…; Tables

  4. ShownOptic lobe neurons follow a shared differentiation program that does not depend on their identity. They transcribe neurotransmitter and other functional genes within roughly the first 20 hours of neuronal life, yet ChAT and VGlut proteins were not detected in the L3 optic lobe.

    Mi1 trajectory from L3 to P70: modules enriched for translation (FDR=10^-141) and axon guidance (FDR=10^-20), with neurotransmitter (FDR=10^-4) and membrane-potential (FDR=10^-8) genes rising from L3; these GO signatures appeared in all neuronal branches incl. lamina and lobula plate; ChAT and VGlut proteins absent in L3 optic lobe but present in larval VNC neurons. · The implementation of neuronal identity occurs…; Merging of larval and pupal Mi1 and DE analysis…; GO enrichment analysis

  5. SuggestedCross-species comparison suggests that human cortical neurons share the order of differentiation steps. The fly medulla tTF series, however, does not appear as strict temporal windows in mouse cortical radial glia.

    Human GW19 cortical plate snRNA-seq (3,363 cells, six modules): DNA replication (FDR=10^-44), axon guidance (FDR=10^-7), synapse organization (FDR=10^-11), but no translation enrichment. In mouse E12-E15 radial glia (GSE118953), only the Ey ortholog Pax6 showed age bias (adjusted p = 1.969240e-08); Foxg1 decline was not significant. · Common differentiation trajectory in Drosophila…; Drosophila tTF expression in mouse cortical…; Analysis of human data; Figure legends

Methods, data and reproducibility

10x Chromium scRNA-seq of L3 optic lobes was integrated with Seurat v3 and merged with annotated pupal data. Monocle pseudotime on medulla neuroblasts screened FlyBase TFs; candidates were validated by immunostaining, MARCM mutant clones and MZVUM-Gal4 RNAi. Monocle3 gene modules and GO analysis were run on fly Mi1 and human fetal cortex trajectories.

  • 10x Chromium scRNA-seq of microdissected wandering L3 optic lobes (Canton-S), ten libraries · Drosophila melanogaster · n = 49,893 cells from 40 L3 optic lobes · GSE167266
  • Annotated early pupal (P15) optic lobe scRNA-seq from the authors' prior work, merged to label larval neuronal trajectories · Drosophila melanogaster
  • Mi1 neuron clusters from pupal P15, P30, P40, P50 and P70 datasets merged with L3 Mi1 cells for a differentiation trajectory · Drosophila melanogaster
  • Adult optic lobe scRNA-seq (authors' prior work) used to confirm neurotransmitter identity of cell types · Drosophila melanogaster
  • MARCM mutant clones and Vsx-domain RNAi knockdowns in wandering L3 optic lobes, immunostained for tTFs and neuronal TFs · Drosophila melanogaster · n = at least four brains per experiment
  • Hashed snRNA-seq of fetal human cortical plate (anterior frontal lobe, prenatal autopsy, gestational week 19), via PsychENCODE Knowledge Portal · human · n = 3,363 cells after QC (40,000 hashed nuclei loaded)
  • Published mouse cortex scRNA-seq (Telley et al.) of radial glia at embryonic days 12 to 15 · mouse · GSE118953
Methods & tools
10x Genomics Chromium Single Cell 3' v2 (fly) and v3 (human), Cell Ranger 3.0.1 (BDGP6.88) and 3.1.0 (GRCh38), Seurat v3 integration and clustering, Monocle with DDRTree pseudotime, Monocle3 find_gene_modules, CITE-seq-Count 1.4.2, TotalSeq-A nuclear hashing with FANS, Gene Ontology Resource enrichment (Fisher's exact test, FDR), FlyBase transcription factor annotation, MARCM mosaic mutant clones, MZVUM-Gal4 (Vsx-Gal4) driven UAS-RNAi, Immunohistochemistry with Leica SP8 confocal and Fiji
Reagents & environment
Cell Ranger 3.0.1 (BDGP6.88) for fly and 3.1.0 (GRCh38) for human; Seurat v3; Monocle and Monocle3; CITE-seq-Count 1.4.2; Chromium 3' v2 (fly) and v3 (human) chemistry.
Reproducibility gaps
Scripts and R objects are shared via a single Google Drive link rather than a versioned repository; human snRNA-seq data are released through PsychENCODE subject to its data-access and attribution requirements.
Limitations (6)
  • Oaz could not be confirmed as a tTF: no working antibody existed, and the Oaz-Gal4 reporter labeled all medulla neuroblasts, possibly due to Gal4 perdurance. Esg appeared salt-and-pepper and is likely not a bona fide tTF. · stated by the authors
  • The role of the second Opa temporal window remains unresolved; the authors could not determine whether it regulates Hbn or Slp. · stated by the authors
  • Neurotransmitter transcripts at L3 were not matched by detectable ChAT or VGlut protein, and why functional genes are transcribed days before use is unknown; commitment to identity is offered only as a hypothesis. · stated by the authors
  • The 12 temporal windows are a conservative assignment, and the authors draw the per-window neuroblast divisions in their tTF schematic as hypothetical rather than observed. · stated by the authors
  • All 25 low-level candidates were discarded after only four were checked in vivo. Clone phenotypes (e.g., the reported TfAP-2 reduction) are shown as images without reported cell counts or statistical tests. · noted in this summary
  • The human conservation claim rests on a single gestational stage (week 19, 3,363 cells) and qualitative overlap of GO trends, with no formal similarity test. The text calls mouse Pax6 older-enriched, but the Extended Data legend calls it young-enriched. · noted in this summary

Read this if you study neural stem cell temporal patterning, want a validated tTF regulatory network for the fly medulla, or plan to use scRNA-seq trajectories to find temporal factors in other species.

Citations
147
FWCI
9.30
Field percentile
top 1%

Reproducibility ledger

Nobody has reported trying to reproduce this paper yet. Did you replicate an experiment or run the code? Your report saves the next lab a week.

Sign in to file a reproducibility report.

Related papers

Similarity Map
SelectedSame fieldCross-field
THISTreese '26Zhu '22Filippopoul '26Ray '22Özel '22Marques '21
78% semantic matchNeuroscience
McKenzie Treese, Yen‐Chung Chen, Abigail Tyree+9 more·Proceedings of the National Academy of Sciences·Aug 19, 2026· 5 citations