NSD2 deficiency disrupts the epigenetic landscape of cochlear hair cells to enhance ECM expression and to perturb cochlear hair cell function

Hearing is a feat of microscopic engineering. Inside the snail-shaped cochlea of the inner ear, thousands of tiny sensory cells called hair cells stand in neat rows, waiting to catch sound waves. These cells do not just sit there; they are anchored and supported by a complex scaffolding of proteins and fibers known as the extracellular matrix. This scaffolding acts like the mortar between bricks, holding the delicate structures in place and ensuring they move in perfect unison when sound arrives. If this scaffolding becomes too stiff, too loose, or accumulates too much material, the hair cells cannot vibrate correctly, and the signal for sound is lost. For decades, scientists have known that genetic mutations can break the machinery of these cells, but a new study suggests that the instructions for building the scaffolding itself might be just as critical as the bricks.

Researchers have long suspected that chemical tags on our DNA, which act like volume knobs for genes, play a role in how these ear cells develop. One such tag, a chemical modification called H3K36me2, is added to DNA by an enzyme named NSD2. While NSD2 is known to be important for general development, its specific job inside the hearing cells of the ear has remained a mystery. A team of scientists at Shanghai Jiao Tong University and other institutions set out to solve this puzzle. They wanted to know what happens when the ear cells lose the ability to produce NSD2. By creating mice that lacked this enzyme specifically in their hair cells, they discovered a chain reaction that starts with a missing chemical tag and ends with a buildup of structural material that physically disrupts hearing.

The researchers began by observing when NSD2 is naturally present in the developing ears of mice. They found that the enzyme is most active during the earliest stages of hair cell maturation, right when the cells are organizing themselves into their precise rows. This timing suggested that NSD2 is essential for setting up the ear's architecture. To test this, the team generated a line of mice where the gene for NSD2 was switched off only in the hair cells. The results were immediate and severe. These mice suffered profound hearing loss, unable to detect sounds across a wide range of frequencies. When the scientists examined the ears of these mice under a microscope, they saw that the hair cells were still alive, but their arrangement was chaotic. The neat, parallel rows of cells had spread apart, creating wide, empty gaps between them. The tiny hair-like bundles on top of the cells, which usually form a perfect W-shape to catch sound, were flattened and disorganized.

To understand why the cells had fallen out of place, the team looked at the molecular instructions inside the cells. They found that without NSD2, the chemical tag H3K36me2 disappeared from the DNA. This loss triggered a massive change in the cell's genetic activity. Genes that should have been turned down were turned up, and vice versa. Most notably, the cells began producing an excessive amount of extracellular matrix proteins, particularly collagen. In a healthy ear, collagen provides necessary support, but in these mutant mice, the cells were overproducing it. This excess collagen accumulated in the spaces between the hair cells, effectively pushing them apart and distorting the delicate structure of the organ of Corti. The researchers described this as a failure of the cell's internal regulation, where the absence of a single enzyme led to a runaway production of structural material that physically jammed the hearing mechanism.

The team confirmed this mechanism using a laboratory cell line that mimics hair cells. When they removed the NSD2 gene from these cells, the cells also began to produce too much collagen and lost their normal shape. They found that the cells behaved differently depending on how stiff the surface they were growing on was, suggesting that the excess collagen was altering how the cells sensed their environment. To prove that this buildup was the direct cause of the hearing loss, the researchers performed a rescue experiment. They used a harmless virus to deliver a genetic "silencer" into the ears of the deaf mice, targeting the specific genes responsible for the excess collagen. This intervention reduced the amount of collagen, allowed the hair cells to return to a more normal arrangement, and significantly restored the mice's hearing.

This study reveals a new pathway for hearing impairment, showing that the problem is not always a broken cell, but sometimes a broken instruction that leads to too much support material. The findings suggest that the enzyme NSD2 acts as a gatekeeper, preventing the overproduction of the extracellular matrix during ear development. When this gatekeeper is missing, the scaffolding grows out of control, disrupting the precise geometry required for hearing. The research offers a potential new direction for treating hearing loss, suggesting that therapies aimed at reducing excess collagen or restoring the function of NSD2 could help repair the ear's structure. While the study was conducted in mice, the discovery highlights a fundamental biological principle: the health of our senses depends not just on the cells themselves, but on the precise chemical balance that keeps their environment in order.