Nerve Regeneration Surgery Leaves Distinct Molecular Fingerprints Across the Nervous System in Rats

When surgeons amputate a limb, one of the most stubborn problems they face is not the missing hand or foot itself, but the nerve stump left behind. Severed peripheral nerves, deprived of their targets, often sprout tangled masses of regenerating axons called neuromas, which can produce devastating chronic pain. In recent years, a surgical technique known as the regenerative peripheral nerve interface, or RPNI, has emerged as a promising way to give those homeless axons somewhere productive to go. Now, a team of researchers in China has taken one of the most detailed looks yet at what this procedure does to the nervous system as a whole, and their findings paint a picture that is both encouraging and carefully hedged.

The study, published in BMC Neuroscience by Yujie Chen, Xinyi Gu, Sirui Tang, Huiran Zang and Guangxue Li of Peking University People’s Hospital and collaborators, set out to answer a deceptively simple question: when an RPNI is built on the end of a transected nerve, how does the rest of the nervous system respond? The technique works by wrapping the cut end of a nerve in a small, free graft of muscle. The denervated muscle fibers within the graft become reinnervated by the regenerating axons, effectively creating a living biological target that absorbs the sprouting nerves and, in principle, prevents the chaotic growth that drives neuroma pain.

What makes the new work unusual is its ambition to look beyond the surgical site. Most studies of nerve interfaces focus on the local environment of the graft, examining how well axons regenerate and how the muscle responds. The Chinese team instead performed RNA sequencing on four tissues from each animal: the reconstructed nerve itself, the dorsal root ganglion where the sensory neuron cell bodies live, the thalamus, and the hippocampus. The thalamus serves as a critical relay station for sensory signals on their way to the cortex, while the hippocampus, though best known for memory, is increasingly recognized as a site where chronic pain leaves measurable traces. By sampling all four tissues simultaneously, the researchers hoped to catch echoes of the peripheral reconstruction propagating into the central nervous system.

The experimental design was rigorous for an exploratory study of this size. Thirteen male rats were randomized into three groups. Three animals received a sham procedure, five underwent sciatic nerve transection with no reconstruction, leaving an untreated nerve stump, and five received sciatic nerve transection followed by RPNI reconstruction. Eight weeks after surgery, the team harvested the four tissues from every animal, generating fifty-two RNA sequencing libraries in total. The researchers used DESeq2, a standard statistical framework for differential expression analysis, across all fifty-two libraries, and then repeated key analyses on an overlapping thirty-six-library subset as a sensitivity check to make sure their results were not artifacts of one particular statistical choice.

The molecular results were strikingly tissue-specific. Comparing RPNI animals directly with unreconstructed stump animals, the team found 189 differentially expressed genes in the nerve itself, sixty-seven in the dorsal root ganglion, only two in the thalamus, and a mere six in the hippocampus, all using a false discovery rate threshold of five percent and a minimum absolute log2 fold change of one. In other words, the further upstream one travels from the site of reconstruction, the quieter the transcriptional signal becomes. The nerve and its associated sensory neurons show a robust molecular response to the presence of the regenerative interface, while the brain regions sampled show only faint traces of change at the gene expression level.

Yet the histology told a more nuanced story. When the researchers examined thalamic tissue under the microscope, they found that RPNI animals had significantly fewer Iba1-positive cell bodies, a marker of activated microglia, the resident immune cells of the brain that proliferate and change shape when neural injury or inflammation is present. They also counted fewer TUNEL-positive nuclei, an indicator of cell death, in the thalamus of reconstructed animals. Both differences reached statistical significance in Tukey-adjusted comparisons, with P values of 0.01602 and 0.01283 respectively. Standard hematoxylin and eosin and Nissl staining did not establish differences in overall cell counts. The microglial and cell-death findings suggest that, at the cellular level, reconstruction may be associated with a calmer central environment, even if that calm is not loudly reflected in the transcriptome of the thalamus.

Behaviorally, the picture was more ambiguous. The team scored autotomy, a behavior in which animals self-mutilate a denervated limb and a common proxy for neuropathic pain in rodent models, at four and eight weeks after surgery. Median autotomy scores were one in the RPNI group versus seven in the stump group at week four, and two versus nine at week eight. Those raw differences look dramatic, but with only five animals per group and Holm-adjusted P values of 0.1786 and 0.09524 across six comparisons, the differences did not reach conventional statistical significance. The authors are explicit that these results do not establish an analgesic effect of the procedure, even though the direction of the trend is consistent with what clinicians hope the technique can achieve.

Perhaps the most sobering findings came from the study’s cross-checks. When the researchers compared their RNA sequencing point estimates with quantitative reverse transcription PCR measurements for twenty-four candidate genes, the two methods gave estimates with opposite signs for twelve of them, half of the candidates tested. And when they searched for genes whose expression might be reversed by reconstruction across multiple tissues, they found no single gene shared across all tissues, and external comparisons against public rat nerve and mouse dorsal root ganglion datasets all yielded Benjamini-Hochberg adjusted q values of 0.30, far from significance. These results underscore how fragile exploratory signals can be when sample sizes are small and multiple tissues are interrogated at once.

The authors are admirably candid about the limits of their work. In their conclusions, they state that RPNI was associated with tissue-specific expression changes and selected thalamic histological differences, but that these exploratory findings do not establish global normalization, cross-tissue communication, or analgesic efficacy. That restraint matters in a field where enthusiasm for nerve interfaces sometimes outpaces the evidence. The study was funded by the Beijing Municipal Science and Technology Commission and the Peking University People’s Hospital Research and Development Funds, and it was approved by the institution’s animal ethics committee. The work was published open access on 28 September 2026 under a Creative Commons license.

For the growing community of surgeons and neuroscientists working on regenerative nerve interfaces, the study offers both a template and a caution. The template is the multi-tissue design itself: sampling peripheral nerve, sensory ganglia, and brain regions in the same animals provides a systems-level view that single-tissue studies cannot match. The caution is statistical: with five animals per group, only the largest effects survive correction, and the divergence between RNA sequencing and PCR measurements for half of the tested candidates shows how easily exploratory signals can mislead. Larger, adequately powered studies will be needed to determine whether the quieter thalamic microglia and the lower autotomy trends seen here mature into reliable evidence that RPNI reconstruction genuinely tames the central consequences of nerve injury. For now, the molecular echoes of reconstruction remain exactly what the authors call them: exploratory.

Subject of Research: Multi-tissue molecular and histological responses to regenerative peripheral nerve interface reconstruction after sciatic nerve transection in rats

Article Title: Exploratory multi-tissue profiling after regenerative peripheral nerve interface reconstruction in rats

Article References: Chen, Y., Gu, X., Tang, S., Zang, H., & Li, G. (2026). Exploratory multi-tissue profiling after regenerative peripheral nerve interface reconstruction in rats. BMC Neuroscience. https://doi.org/10.1186/s12868-026-01055-w

Image Credits: AI Generated

DOI: 10.1186/s12868-026-01055-w

Keywords: regenerative peripheral nerve interface, peripheral nerve injury, sciatic nerve transection, RNA sequencing, dorsal root ganglion, thalamus, microglia, neuropathic pain, autotomy, differential gene expression, neuroma prevention, rat model