How a Fly’s Ear Wires Itself to Its Exoskeleton During Pupal Life

Scientists have mapped, hour by hour, how the fruit fly’s tiny hearing organ physically anchors itself to the exoskeleton during pupal development, revealing a sequence of cellular events that had remained largely hidden until now. The study, published in Discover Developmental Biology, tracks the formation of the Johnston’s organ and its attachment to the a2-a3 joint of the antenna, the structure that turns sound-induced vibrations into the neural signals females use to recognize male courtship songs.

In Drosophila melanogaster, hearing does not rely on eardrums. Instead, flies detect sound through their antennae. Males produce species-specific courtship songs by vibrating their wings, and females use their antennae to recognize these conspecific songs, which facilitates copulation. This form of acoustic communication is thought to play an important role in speciation. Sound vibrates the feather-like arista attached to the third antennal segment, causing rotation of that segment. The rotation is sensed by the Johnston’s organ, an array of mechanoreceptors housed in the second antennal segment and attached to the cuticle of the a2-a3 joint.

The Johnston’s organ is built from functional subunits called scolopidia. Each scolopidium contains mechanosensory neurons, a ligament cell, a scolopale cell, a cap cell, and a dendritic cap. The dendritic cap is an extracellular matrix secreted by the scolopale cell that covers the tips of the neurons and attaches to the cuticle. Because no muscles are inserted into the a2-a3 joint, its movement is driven directly by the sound-induced rotation of the distal segment, and active oscillations of the Johnston’s organ neurons contribute to mechanical amplification of sound reception. Mutations that disrupt the elaborate joint structure are associated with hearing defects, making the development of this connection a central question in fly auditory biology.

While previous studies had revealed gene expression patterns in larval antennae and the final morphological defects in mutant adults, the morphogenetic processes during the pupal stage, which lasts about five days, had remained largely unknown. The research team, Shan Jiang and Reiko Tajiri, tracked antenna development throughout pupation using fluorescent markers, confocal microscopy, and stains for chitin and actin, examining three antennae at each time point.

The researchers first documented how the joint itself takes shape. The antennal primordium appears as a flat disc in the late third instar larva and gradually deforms. At approximately 12 hours after puparium formation, the antennae sit dorsal to the eyes with dome-like shapes, and the a2-a3 joint appears as a simple epithelial constriction. The antennae then move toward the anterior tips of the head, accompanied by elongation of the arista, whose tip remains anchored to the pupal cuticle. Movement ceases and the arista reaches its maximum length at about 22 hours after puparium formation.

After that, the joint gradually decreases in diameter and invaginates dorsally into the second segment. The invagination proceeds asymmetrically, with the lateral region sinking deeper than the medial region, an asymmetry already visible at 23 to 25 hours after puparium formation. By 36 to 38 hours, the joint is oriented parallel to the dorsoventral axis, and in three dimensions it forms a ring-like structure. At 54 hours, the inner diameter of the joint is approximately 10 micrometers. Cuticle formation follows a distinct schedule: at 45 hours no chitin is detected in the antenna, at 47 hours patchy chitin appears on bristles of the external surface but not in the joint, and by 54 hours cuticle has formed throughout the apical surface of the joint epidermis.

The team then charted how the sensory organ reaches toward this forming joint. At 10 hours after puparium formation, each scolopale cell extends a thin protrusion containing a rod-like actin structure, likely a scolopale rod. By 14 hours, the NompA protein, a marker of the dendritic cap, is detected in a subset of scolopale cells and extends beyond the cell membrane, indicating extracellular deposition, while cap rods extend beyond the NompA structures. At 21 hours, the dendritic caps of all scolopidia lining the epidermal invagination have elongated to reach the cap rod tips at the level of the epidermal apical surface. From that stage onward, the tips of the scolopidia remain precisely positioned at the deepest point of the epidermal invagination.

Quantitative measurements revealed a striking burst of growth: the dendritic caps elongate rapidly between about 38 and 44 hours after puparium formation. The authors suggest this rapid elongation may partly reflect transport of NompA from scolopale cell bodies to their protrusions, since the protein’s signal in cell bodies diminishes substantially by 44 hours. They also propose that mechanical tension, potentially generated as the Johnston’s organ pulls on the epidermis, may drive cap elongation. The study raises the possibility that the sensory organ induces epidermal invagination by mechanically pulling the epidermis or by sending a biochemical signal, an idea supported by earlier findings that atonal mutants, which fail to form the Johnston’s organ, show incomplete epidermal invagination at the joint.

The relationship appears to run in both directions. Loss of the transcription factor cut causes cuticular defects in the joint and deafness, while scolopidia that form normally in these mutants subsequently degenerate. Similarly, flies mutant for spalt-major and spalt-related show joint cuticle defects and later degeneration of scolopidia. These observations suggest that normal development of the joint cuticle is crucial for maintaining the sensory organ, and that scolopidia require anchorage to stay in position even before cuticle forms, since displaced scolopidia were observed in cut mosaic antennae at around 40 hours after puparium formation, prior to the onset of cuticle deposition.

As a step toward identifying the genetic program governing the organ-exoskeleton connection, the researchers screened Notch signaling target genes for expression in the joint epidermis. An enhancer trap of big brain, a Notch target gene, was expressed in a circular band of epidermal cells immediately distal to the a2-a3 joint early in pupation, and from 17 to 19 hours onward in both distal and proximal sides of the joint invagination, while remaining excluded from the sensory organ itself. Knockdown of big brain with RNAi produced no apparent morphological abnormalities in the joint or the organ under light microscopy, though the authors note that subtle defects could require ultrastructural analysis and hearing tests to detect. The findings provide a developmental framework for understanding how the fly’s auditory machinery is physically wired to the exoskeleton that transmits sound.

Subject of Research: Pupal morphogenesis of the Drosophila antennal auditory organ and its attachment to the exoskeleton

Article Title: Morphogenesis of the Drosophila antennal auditory structure in the pupa, with a focus on the connection between the sensory organ and the exoskeleton

Article References: Jiang, S., & Tajiri, R. (2026). Morphogenesis of the Drosophila antennal auditory structure in the pupa, with a focus on the connection between the sensory organ and the exoskeleton. Discover Developmental Biology, 236(1), Article 6. https://doi.org/10.1007/s00427-026-00734-2

Image Credits: AI Generated

DOI: 10.1007/s00427-026-00734-2

Keywords: Drosophila, Johnston's organ, hearing, pupal development, morphogenesis, a2-a3 joint, scolopidia, dendritic cap, NompA, Notch signaling, cuticle, antenna