Meta-analysis of 20 rodent RNA-seq studies links enriched environment to synaptic and Fos/Jun gene activity

Researchers from the Institute of Neurosciences in Spain and the University of Warsaw in Poland carried out a systematic review and meta-analysis of transcriptomic studies on environmental enrichment in rodents, as reported in Molecular Neurobiology. springer.comMolecular Neurobiologyspringer.comOpen source ↗ The PubMed search covered publications up to February 2025 and returned 323 articles. After manual screening for relevance and data availability, 20 RNA-seq reports remained.

Enrichment paradigms give animals more sensory, cognitive, social and motor stimulation. Earlier rodent experiments showed that this improves learning, memory and neuroplasticity. Many transcriptomic studies had examined the molecular changes, but a unified view of the genes and pathways that are consistently modulated was missing.

Method: 20 datasets, one pipeline

Inclusion criteria were an enrichment paradigm, rodents, genome-wide transcriptomic methods, and examination of brain regions or neuronal populations. The authors reprocessed the 20 datasets with a single analysis pipeline and applied three complementary statistical methods.

Results: synapses, plasticity and Fos/Jun

Despite considerable heterogeneity between studies, the analysis identified consistent gene expression signatures linked to synaptic function, plasticity and their transcriptional regulation. springer.comMolecular Neurobiologyspringer.comOpen source ↗ The activity-dependent transcriptional program was upregulated, including Fos and Jun family members.

According to the publication's retelling, these genes form components of the AP-1 protein complex, which regulates genes involved in signaling between neurons and in remodeling of their connections. The same group had earlier reported in mice that AP-1 takes part in mechanisms by which early-life experience can long influence learning and memory.

Limitations and open data

Substantial methodological heterogeneity across the source studies increased variability in the outcomes. Stressors or disease models, particularly in rat studies, introduced a major confounding factor and limited reliable comparison between species. Overall, the studies showed a low to moderate risk of bias.

The authors stress that the work was done on rodents, so the findings cannot yet be transferred directly to humans. The patterns found also do not prove which molecular mechanisms cause the improvement in learning and memory. To support further research, the team built an open web application, mEEtaBrain, which lets users explore changes in individual genes and in different brain regions. springer.comMolecular Neurobiologyspringer.comOpen source ↗