Establishment of a human pluripotent stem cell-derived sensory neuron model for diabetic peripheral neuropathy and pharmacological evaluation

For millions of people living with diabetes, a quiet and painful complication can take hold in the nerves of their hands and feet. This condition, known as diabetic peripheral neuropathy, begins with a loss of sensation, often described as a numbness that creeps up the limbs like a stocking or a glove. Over time, the damage deepens, causing the long, thin fibers that carry signals from the skin to the brain to wither and die from the outside in. While doctors can manage the pain, there is currently no treatment that can stop or reverse this nerve degeneration. The search for such a cure has been stalled by a fundamental problem: scientists have lacked a reliable way to study human nerve cells in the lab. The cells inside a mouse or a rat behave differently than those in a person, and taking nerve tissue directly from human patients is often impossible. Without a human model that accurately mimics the disease, testing new drugs is like trying to fix a complex engine without ever seeing the engine itself.

A team of researchers at Beijing Shunyi District Hospital has now built that missing engine. They created a laboratory model using human stem cells that grow into mature sensory neurons, the specific type of nerve cell damaged in diabetic neuropathy. By exposing these human cells to the high sugar levels and oxidative stress found in diabetes, the scientists were able to watch the nerve damage happen in real time. Their work confirms that the damage is caused by the toxic effects of sugar on the cell's energy systems, not just by the physical pressure of high fluid levels. Most importantly, they showed that this model can predict whether a drug will protect the nerves. When they tested three existing medications, the cells survived, and the long, delicate branches of the nerves remained intact, offering a new path forward for finding treatments that could actually repair the damage.

To understand how this new model works, one must first look at the source of the cells. The researchers started with human pluripotent stem cells, which are a special type of cell capable of turning into any tissue in the body. Using a carefully timed sequence of chemical signals, they guided these stem cells to become sensory neurons. This process took thirty-five days. The resulting cells were not a mixed bag of different types; they were a pure population of sensory neurons, verified by the presence of specific markers that identify them as such and the absence of markers for motor neurons. This purity was essential, as it ensured that any damage observed later was happening specifically to the sensory nerves involved in the disease, rather than being confused by other cell types.

With a reliable supply of human sensory neurons, the team set out to recreate the conditions of diabetic neuropathy in a dish. They knew that patients with the condition suffer from chronic high blood sugar, which eventually leads to a buildup of harmful molecules called reactive oxygen species that damage the cell's power plants, the mitochondria. To mimic this, they first exposed the cells to a high concentration of glucose for twenty-four hours. They then added a burst of hydrogen peroxide to trigger an acute oxidative stress, simulating the sudden spikes in damage that occur in the body. A critical step in their design was to rule out a common confusion in previous studies: they wanted to know if the damage was caused by the sugar itself or simply by the fact that high sugar makes the liquid in the dish thicker and more pressurized. To test this, they added a substance called mannitol to a control group to create the same physical pressure without the sugar. The cells in the sugar group suffered significant damage, while those in the pressure-only group remained healthy. This proved that the injury was driven by the metabolic toxicity of the glucose, not by physical pressure.

Once the injury model was established, the researchers tested whether it could identify drugs that might protect the nerves. They selected three compounds known for their potential to reduce stress or improve cell health: N-acetylcysteine, minocycline, and metformin. Before exposing the cells to the damaging stress, they gave the cells a two-hour head start with one of these drugs. The results were clear. The untreated cells, overwhelmed by the stress, showed a massive surge in damaging molecules and a collapse of their mitochondrial function. In contrast, the cells treated with the drugs maintained their internal balance. The damaging molecules were kept in check, and the energy-producing mitochondria remained stable. All three drugs successfully prevented the immediate death of the cells, with N-acetylcysteine showing the strongest effect in restoring the cell's internal environment.

However, the true test of a model for diabetic neuropathy is not just whether it can keep a cell alive for a few hours, but whether it can prevent the slow, progressive death of the nerve's long branches. In patients, the disease is characterized by a "dying-back" pattern where the far ends of the nerves break down first. To see if their model could capture this, the researchers let the experiment run for twenty-one days. In the group that received no drug treatment, the network of nerve fibers began to fragment and disappear, losing nearly seventy percent of its structure over the three weeks. This mirrored the wasting away seen in human patients. But in the groups where the cells had received the early drug treatment, the nerve networks remained robust and connected. The drugs had not just kept the cells alive; they had preserved the intricate architecture of the nerves, preventing the long-term degeneration that defines the disease.

This study provides a significant step forward in the fight against diabetic neuropathy by offering a tool that speaks the language of human biology. The researchers have demonstrated that they can grow pure human sensory neurons, subject them to the specific metabolic stresses of diabetes, and observe the resulting damage with high precision. By proving that the damage is metabolic rather than physical, and by showing that the model can detect both immediate cellular rescue and long-term structural protection, they have created a platform that is ready for use. This system allows scientists to screen potential therapies with a level of confidence that was previously impossible, bridging the gap between laboratory experiments and the complex reality of human disease. The work suggests that with the right tools, the search for treatments that can halt or reverse nerve damage is no longer just a hope, but a tangible possibility grounded in human biology.