New study links the EHMT2 gene to Kleefstra syndrome

An international research team has linked gene variants EHMT2 with a rare neurodevelopmental disorder that shows significant similarities to Kleefstra syndrome. The work, which was published in Nature Communications on July 1, 2026, it describes seven new genetic variants (de novo), meaning changes not detected in the parents, in individuals with similar clinical and molecular findings.

The authors propose the name «EHMT2-related autosomal dominant Kleefstra syndrome.» To avoid confusion, it is more accurate to speak of Kleefstra syndrome type associated with the EHMT2 gene. It is a newly described, related disorder and not a replacement or revision of the established diagnosis of Kleefstra syndrome type 1, which is associated with EHMT1.

The finding may have diagnostic significance for certain individuals who have a Kleefstra-like phenotype, but no explanation has been found in EHMT1. However, it is not a therapeutic study. No drug, gene, or RNA therapy was tested on humans, no randomized clinical trial was conducted, and no intervention became available to families.

Why is EHMT2 biologically linked to EHMT1

The relationship between the two genes is not coincidental. EHMT1 provides the instructions for the production of the GLP protein, while the EHMT2 encodes the G9a protein. The two proteins work together in a complex involved in epigenetic regulation: it helps control which genes are more or less active, without changing the sequence of the DNA «letters».

The GLP/G9a complex places regulatory chemical marks on histones, the proteins around which DNA is organized. In this way, it influences chromatin structure and access to genes during development. The new study suggests that the disruption of the G9a/EHMT2 side of this common mechanism can lead to a clinical and molecular picture resembling that of the GLP/EHMT1 disorder.

In Kleefstra syndrome type 1, the most common underlying mechanism is deficiency of a functional copy of EHMT1out of the two copies of the gene, one has been deleted or inactivated and the other is insufficient for typical development. Researchers propose a different mechanism for the specific variants EHMT2. The altered G9a proteins appeared to remain structurally stable, but could not properly carry out their methyltransferase activity.

The laboratory results indicate a potential prevailing negative action. Simply put, the altered protein may interfere with the function of the normal complex, even though an unaffected copy of it is also present. EHMT2. This differs from simply producing a smaller amount of protein and does not mean that the two genetic disorders are identical.

Combination of genetic, molecular and preclinical data

The strength of the work lies in the fact that it was not based solely on the discovery of rare variants. The researchers combined clinical genetics with the analysis of characteristic patterns of DNA methylation, histone modifications, and gene expression. These molecular patterns were reported to be similar to those observed in Kleefstra syndrome type 1, providing different yet converging levels of evidence.

The epigenetic signature, meaning a characteristic laboratory DNA methylation pattern, can help evaluate the significance of a genetic variant. It is not in itself a symptom, prognosis, or an indication of treatment response. Similarly, the similarity of gene expression patterns indicates a potential disruption of common biological pathways, but does not prove that the course and all manifestations of the two conditions will be the same.

Researchers also tested in the laboratory whether the mutated G9a proteins retained their normal enzymatic function. They concluded that they were catalytically deficient, likely due to disrupted interactions with histone H3 or with the molecular methyl donor used by methyltransferases. This is a biochemical finding rather than a routine methylation assay or a therapeutic target already validated in patients.

Yet another level of support came from mice carrying a variant EHMT2 derived from a patient. The animals exhibited growth retardation, differences in the skull and facial features, as well as abnormal behavior. The model reinforces the assessment that the variant may affect development in a living organism. However, a mouse does not replicate the entire human condition, and these findings do not predict whether a future treatment would be effective or safe for humans.

What it means for families

For an individual with a confirmed pathogenic variant of EHMT1 or by deletion of the 9q34.3 region which includes the EHMT1, the new publication does not change the diagnosis of Kleefstra syndrome type 1. It does not imply a change in treatments, clinical monitoring, medications, the educational program, or supportive interventions. Care still needs to be personalized by the treating teams.

For families in which there is a Kleefstra-like presentation without a diagnostic finding in EHMT1, the EHMT2 it is now a gene that can be evaluated by the clinical genetics team, particularly when broader genomic testing has identified a rare variant. A result in a genetic report is not automatically pathogenic. Classification by an accredited laboratory, correlation with the clinical picture, and, where indicated, parental testing are required. «Autosomal dominant» does not mean here that a parent was obligate carrier or had symptoms, as the variants were reported as de novo.

The most important immediate contribution of the study is that it broadens the understanding of the GLP/G9a network and offers new tools for research, such as molecular signatures and an animal model. Future, larger patient cohorts will be needed to elucidate the full clinical spectrum, natural history, and potential differences from Kleefstra syndrome type 1.

The human database remains very small. The publication reports seven de novo variants and does not yet allow reliable estimates for the frequency of individual features or for the prognosis. It is also likely that primarily individuals who reached specialized genetic and research centers have been identified, while milder or atypical cases may have gone unrecognized. Furthermore, the available version is characterized by the publisher as a non-finally edited manuscript, therefore its wording and presentation may be updated.

Overall, the work offers strong initial evidence that the loss of normal G9a function through specific variants of EHMT2 it can cause a disorder closely related to Kleefstra. It is significant progress for the diagnosis and biology of the disease, but not proof that intervention in histone methylation will improve symptoms.

Sources

 

Research