Research
Research interests
I want to understand how gene expression is controlled in early development: at chromatin, by transcription factors, protein complexes and epigenetic regulators, and in the cytoplasm, by mRNA–protein condensates that localise maternal mRNAs and time their translation. My training spans chromatin and metabolism in fly embryos, single-molecule RNA imaging, protein phase separation, organelle biology, enzyme kinetics and behaviour, with quantitative imaging and code throughout.
01Core interest
Chromatin-level control of gene expression in early development
Zygotic genome activation is set up at the level of chromatin. Pioneer factors such as Zelda and GAF open nucleosome-bound enhancers and recruit co-activators such as CBP; Polycomb and Trithorax complexes, histone acetylation and methylation, and marks inherited through the germ line bias which genes respond; and enhancer–promoter loops and domains begin to form before transcription starts. Current work asks which of these are instructive for the timing of activation, and how metabolic state reaches the enzymes that write these marks.
Open questions
- Which factors and complexes make an enhancer competent before the genome is activated, and which act only once transcription starts?
- How do Polycomb, Trithorax and acetylation states inherited from the egg combine with marks written in the embryo to set the timing of activation?
- Do changes in metabolite supply alter histone marks at specific loci during the maternal-to-zygotic transition, or only globally?
My work: Stage 5 and the MZT · Current project: metabolism and histone marks
Landmark work
- Sun et al. 2015Zelda overcomes the nucleosome barrier at enhancers during ZGA
- Hug et al. 2017Chromatin architecture emerges during ZGA independent of transcription
- Espinola et al. 2021Cis-regulatory loops arise before TADs and gene activation
- Ghavi-Helm et al. 2014Enhancer loops appear stable during development
- Davidson et al. 2019DNA loop extrusion by human cohesin
- Galouzis et al. 2025Chip (Ldb1) bridges the pioneer factor Zelda to the co-activator CBP during zygotic genome activation
- Maziak et al. 2026Three-dimensional genome reorganization foreshadows zygotic genome activation in Drosophila
- Freund et al. 2026Pioneer-factor activity requires stable chromatin occupancy through sequence-specific binding and disordered domains
02Core interest
Spatiotemporal gene regulation by mRNA–protein condensates
Many maternal mRNAs are regulated by where they are as much as by how much there is. In the oocyte and early embryo, mRNAs such as oskar and nanos assemble with RNA-binding proteins into RNP condensates that are transported, anchored at specific sites and kept translationally silent until needed. Their assembly depends on RNA–RNA as well as RNA–protein contacts, and their material state, from liquid to solid-like, is regulated and essential for development. Current work links the composition and physical state of condensates to when, and where, the mRNAs inside them are translated.
Open questions
- How do the composition and material state of an RNP condensate set when the mRNAs inside it are translated?
- How are condensates delivered to and held at specific sites, such as the oocyte posterior, and what releases their mRNAs?
- Which metabolic, ionic or environmental factors tune condensate formation in embryos?
My work: oskar RNA condensates and sequential smFISH · Prion protein phase separation
Landmark work
- Brangwynne et al. 2009Germline P granules are liquid droplets
- Bose et al. 2022Liquid-to-solid transition of oskar RNP granules is essential for development
- Trcek et al. 2015Drosophila germ granules contain homotypic mRNA clusters
- Little et al. 2015Independent and coordinate trafficking of single germ plasm mRNAs
- Banani et al. 2017Biomolecular condensates: organizers of cellular biochemistry
Connected interests
03
Metabolism, chromatin and developmental timing
Chromatin-modifying enzymes use metabolites as donors and cofactors, so glycolytic and mitochondrial flux can tune acetylation, methylation and lactylation. In embryos, metabolic state also behaves as a signal that can set developmental tempo.
Open questions
- Does lactate supply histone lactylation in the early embryo, and does it change which genes switch on and when?
- How do metabolic gradients and fluxes encode timing during development?
My work: Current project and its background
Landmark work
- Wellen et al. 2009ATP-citrate lyase links cellular metabolism to histone acetylation
- Zhang et al. 2019Metabolic regulation of gene expression by histone lactylation
- Li et al. 2024Lactate regulates major ZGA by H3K18 lactylation in mammals
- Miyazawa et al. 2025Glycolytic metabolites control the timing of embryo segmentation
- Hayashi et al. 2025Histone lactylation during Drosophila gametogenesis
04
Post-transcriptional control and the timing of expression
In the early embryo much of the regulation happens after transcription: maternal mRNAs are stored, localised, translated or cleared on a schedule, and the balance between their decay and the onset of zygotic transcription sets the timing of the maternal-to-zygotic transition.
Open questions
- What sets the half-lives of maternal mRNAs, and how are clearance pathways coordinated with genome activation?
- How do RNA-binding proteins and miRNAs encode timing in 3′UTRs, and how variable is that timing between embryos?
My work: oskar RNA condensates
Landmark work
- Tadros et al. 2007SMAUG is a major regulator of maternal mRNA destabilization in Drosophila
- Bushati et al. 2008Temporal reciprocity of miRNAs and their targets during the MZT
- Lécuyer et al. 2007Global analysis of mRNA localization
- Bothma et al. 2014Transcriptional bursts in living Drosophila embryos
05
Gut microbiota and host chromatin
Short-chain fatty acids and lactate made by gut microbes change host histone modification states, and in flies the microbiota alters oogenesis and the timing of the maternal-to-zygotic transition in the offspring.
Open questions
- Which microbial metabolites reach the germ line, and in what chemical form?
- Can microbe-dependent chromatin states persist from oocyte to embryo?
Landmark work
- Krautkramer et al. 2016Diet–microbiota interactions mediate global epigenetic programming
- Fellows et al. 2018Microbiota-derived SCFAs promote histone crotonylation in the colon
- Elgart et al. 2016Impact of gut microbiota on the fly’s germ line
- Iatsenko et al. 2018Microbiota-derived lactate activates intestinal Nox and shortens lifespan
06
Metabolic state and behaviour
Diet, commensal bacteria and metabolism change how much flies move and sleep, what they eat and how they reproduce. Activity and sleep can be measured in many individual flies over days, which makes behaviour a sensitive read-out of metabolic state, and a bridge to the same metabolic inputs that act on chromatin.
Open questions
- How do metabolic signals reach the neurons that set activity, sleep and food choice?
- Do metabolic states that alter chromatin in the germ line also leave behavioural traces?
Landmark work
- Keene et al. 2010Clock and cycle limit starvation-induced sleep loss in Drosophila
- Leitão-Gonçalves et al. 2017Commensal bacteria and essential amino acids control food choice and reproduction
- Hudry et al. 2019Sex differences in intestinal carbohydrate metabolism promote food intake and sperm maturation
07
Computational biology: quantitative imaging and code
These questions are answered by measurement: segmenting embryos and nuclei, counting single molecules, quantifying marks and following transcription live. I write analysis code in Python, R and the ImageJ macro language, and build tools for reproducible quantification.
Open questions
- How can image-based measurements of histone marks and RNA be made comparable across embryos, sessions and labs?
- Where do general segmentation models fail on embryos, and what does it take to fix them?
My work: Image lab · Nuclei counter · Statistics helper
Landmark work
- Garcia et al. 2013Quantitative imaging of transcription in living Drosophila embryos
- Schindelin et al. 2012Fiji: an open-source platform for biological-image analysis
- Berg et al. 2019ilastik: interactive machine learning for bioimage analysis
- Stringer et al. 2021Cellpose: a generalist algorithm for cellular segmentation
08
Genetic tools, biosensors and biotechnology
Precise tools turn questions into experiments: tissue-specific expression and CRISPR editing in living animals, and engineered proteins that light up when they meet a molecule, so a reaction can be followed as it happens.
Open questions
- How can enzyme activity and metabolites be measured in real time, in the test tube and in living tissue?
- How can genetic perturbations in single tissues separate cause from correlation in whole animals?
My work: Phosphate biosensor and FtsZ
Landmark work
- Brand & Perrimon 1993Targeted gene expression (GAL4/UAS)
- Port et al. 2014Optimized CRISPR/Cas tools for Drosophila
- Brune et al. 1994Real-time measurement of phosphate release with a fluorescent protein probe