🔬 physics

Cell size heterogeneity controls crystallization of the developing fruit fly wing

This study reveals that cell size heterogeneity, rather than tissue shear flow, is the primary control parameter governing the transition from disordered to crystalline cellular packing in the developing fruit fly wing, with shear flow serving only to accelerate the ordering process.

Original authors: Kartik Chhajed, Franz S. Gruber, Raphaël Etournay, Natalie A. Dye, Frank Jülicher, Marko Popović

Published 2026-05-26

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Kartik Chhajed, Franz S. Gruber, Raphaël Etournay, Natalie A. Dye, Frank Jülicher, Marko Popović

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine a fruit fly's wing as a bustling construction site made of thousands of tiny, soft, squishy bricks (the cells). When the wing first starts forming, these bricks are a chaotic mess: some are big, some are small, and they are jumbled together in a disordered pile. But as the wing develops, something magical happens: the bricks suddenly line up into a perfect, honeycomb-like pattern, like soldiers standing in perfect rows.

For a long time, scientists thought this perfect ordering was caused by the wind blowing through the construction site—specifically, the physical "shear flows" (stretches and pushes) that happen as the wing grows. They thought the wind forced the bricks to line up.

This paper says: Not quite. The wind helps, but it's not the main boss. The real secret ingredient is uniformity in size.

Here is the story of how they figured it out, using simple analogies:

1. The "Mismatched Puzzle" Problem

Imagine you are trying to build a perfect hexagonal mosaic (a honeycomb pattern) using puzzle pieces.

Scenario A: You have a box of pieces that are all exactly the same size. It's easy to snap them together into a perfect, ordered pattern.
Scenario B: You have a box where some pieces are huge and some are tiny. No matter how hard you try, they won't fit together neatly. They will stay jumbled and disordered.

The researchers found that the fruit fly wing starts out like Scenario B. The cells are all different sizes (this is called "size heterogeneity" or "polydispersity"). As the wing develops, the cells somehow become more uniform in size. Once they reach a certain level of uniformity, they naturally snap into that perfect honeycomb order.

2. The "Wind" vs. The "Bricks" Experiment

To prove that the "wind" (tissue shear flow) wasn't the main cause, the scientists did some clever experiments:

The "No-Wind" Test: They surgically cut the wing to stop the physical stretching and flowing. They expected the bricks to stay messy. Instead, the bricks still lined up into perfect local groups (crystallites). The wind wasn't necessary to make the bricks fit together; the bricks just needed to be the same size.
The "Wind" Effect: However, when they did have the wind (in normal wings), the local groups of bricks didn't just line up; they all turned to face the same direction. The wind acted like a conductor, making sure all the little groups marched in the same direction, creating a large-scale, organized army. Without the wind, the groups formed but pointed in random directions.

3. The "Too-Big" Failure

They also tested a scenario where the bricks never became uniform. They used a specific genetic mutation (a "proximal ablation") that kept the cells varying wildly in size. In this case, nothing happened. The bricks remained a chaotic mess, and no honeycomb pattern ever formed. This confirmed that if the sizes don't get uniform, the order never happens, no matter what else is going on.

4. The "Magic of Growing Up"

So, how do the cells get uniform? The paper suggests that as the cells divide and grow, they naturally tend to even out their sizes. It's like a group of people who start with different heights but, through a process of growing and dividing, end up all being roughly the same height. Once they hit that "Goldilocks" zone of uniformity, the crystallization kicks in automatically.

The Big Picture

The Main Driver: The transition from chaos to order is controlled by cell size uniformity. When cells become similar in size, they naturally organize into a crystal-like structure.
The Helper: Tissue shear flow (the "wind") doesn't create the order, but it acts as a magnet that aligns all the small ordered groups so they point in the same direction, creating a large-scale pattern.
The Blocker: If the cells remain different sizes (high "polydispersity"), the order is impossible, and the tissue stays disordered.

In short: To get a perfectly organized fruit fly wing, you don't need a strong wind to push the cells into place. You just need the cells to stop being so different from one another. Once they are all the same size, they naturally fall into line.

Technical Summary: Cell Size Heterogeneity Controls Crystallization of the Developing Fruit Fly Wing

Problem Statement
A fundamental challenge in developmental biology is understanding how complex tissue patterns and shapes emerge from the collective behavior of large numbers of cells. While previous studies have established that insect ommatidia, vertebrate somites, and fly wing hairs exhibit regular organization, the mechanisms governing the transition from disordered to ordered cellular packings in soft, deformable tissues remain unclear. Specifically, in the developing Drosophila melanogaster wing, cellular packing transitions from a disordered state to a hexagonal, crystalline arrangement during pupal development. Previous hypotheses suggested that large-scale tissue shear flows were the primary driver of this ordering. However, the role of cell size heterogeneity (polydispersity) in controlling this phase transition had not been systematically quantified or integrated into theoretical models of epithelial mechanics.

Methodology
The authors employed a combined approach of quantitative live imaging, mechanical and genetic perturbations, and computational vertex modeling.

Experimental Quantification: Using large-scale time-lapse microscopy of fluorescently labeled E-cadherin in the Drosophila wing blade, the authors tracked 5,000–10,000 cells over 16 hours of pupal development. They developed a robust metric for local hexagonal order (cell hexatic, $\psi_6$ ) that is unbiased by tissue elongation. This involved applying a pure shear transformation to cell center positions within local patches to remove average cell elongation before calculating the hexatic order parameter. They also measured tissue-scale order via the tissue hexatic ( $\langle \psi_6 \rangle$ ) and quantified cell size polydispersity ( $\Delta$ ) as the coefficient of variation of cell areas.
Perturbation Experiments: To test the necessity of shear flows and cell divisions, the authors analyzed:
- Distal Ablation (DA): Laser ablation of the distal wing margin to reduce tissue shear flow.
- Proximal Ablation (PA): Laser ablation of the proximal wing to disrupt mechanical cues.
- Genetic Mutants: cdc2 mutants (inhibiting cell division), dumpy mutants (reducing shear flow), and Planar Cell Polarity (PCP) mutants.
Vertex Modeling: The authors implemented a vertex model of epithelial tissues incorporating cell size polydispersity and active bond tension fluctuations (modeled as an Ornstein-Uhlenbeck process). They simulated both quasistatic relaxation (to map phase diagrams) and overdamped dynamics (to study ordering kinetics). The model allowed for the control of polydispersity ( $\Delta$ ) and the imposition of shear strain.

Key Contributions and Results

Decoupling Local and Global Order: The study reveals a distinction between local crystallite formation and tissue-scale alignment. In wild-type wings, both local hexatic order ( $\langle |\psi_6| \rangle$ ) and tissue hexatic order ( $|\langle \psi_6 \rangle|$ ) increase over time. However, in distal ablation experiments where shear flow is largely inhibited, local hexatic order still increases significantly (crystallites form), but tissue-scale order remains low. This indicates that shear is not required for crystallization but is essential for the alignment of crystallites.
Cell Size Polydispersity as a Control Parameter: The authors demonstrate that cell size polydispersity decreases during normal wing development. Crucially, they identify a critical threshold for polydispersity ( $\Delta \approx 0.2$ ). When $\Delta$ falls below this threshold, local hexagonal order emerges; when it remains high, the tissue remains disordered.
Phase Transition in Vertex Models: Simulations confirm that cell size polydispersity drives a melting transition. Below a critical $\Delta$ , the system settles into a crystalline state; above it, the system is disordered (liquid). The transition exhibits characteristics of the KTHNY theory of two-dimensional melting, with sequential unbinding of dislocation and disclination defects.
Kinetics and Shear: Dynamic simulations show that while reducing polydispersity drives the system toward the crystalline phase, the kinetics of ordering are slow. Tissue shear flows significantly accelerate the alignment of locally ordered crystallites, explaining the rapid emergence of tissue-scale order in wild-type wings compared to shear-inhibited conditions.
The Role of Cell Division: Inhibiting cell division (via cdc2 mutants) does not prevent crystallization, though it alters the timing of polydispersity reduction. Conversely, proximal ablation prevents the reduction of polydispersity, and consequently, crystallization fails to occur. This suggests that the reduction of polydispersity, rather than division itself, is the primary driver of ordering.

Significance and Claims
The paper claims to identify cell size heterogeneity (quantified by polydispersity) as the fundamental control parameter governing the transition between disordered and ordered cellular packings in the developing fruit fly wing.

The authors argue that while tissue shear flows are not the cause of crystallization (as crystallites form even in their absence), they play a critical secondary role in facilitating the alignment of these crystallites to achieve tissue-scale order. The reduction of cell size polydispersity over time acts as a "quench" that drives the tissue across a phase transition from a liquid-like disordered state to a solid-like crystalline state.

This work challenges the prevailing view that tissue shear is the primary mechanism for ordering in the fly wing, proposing instead that the homogenization of cell sizes is the prerequisite for crystallization. The findings suggest that the physical principles of particle crystallization, specifically the role of polydispersity, are directly applicable to understanding the morphogenesis of biological tissues. The study concludes that the positional regularity of adult wing hairs likely results from this crystallization process driven by the reduction of cell size heterogeneity during the pupal phase.

1. The "Mismatched Puzzle" Problem

2. The "Wind" vs. The "Bricks" Experiment

3. The "Too-Big" Failure

4. The "Magic of Growing Up"

The Big Picture

More like this