A stress-response gene shows up in odontogenic differentiation

Duan, Su, Yang, and colleagues at the West China Hospital of Stomatology, Sichuan University, went looking for genes that control reparative dentin formation, the barrier tissue that seals an exposed or injured pulp. They ran RNA-seq on dental pulp stem cells (DPSCs) through 1, 3, 7, and 14 days of odontogenic differentiation induction. FKBP5, the gene encoding FK506-binding protein 51, sat in the top three differentially expressed genes at every one of the four timepoints, and it was the only protein in that set that stayed elevated across the whole period (N = 6 per group) (Duan et al. 2026, PMID 42522268). An independent dataset in GEO (GSE244057) showed the same rise at day 9 of induction. Notably, lipopolysaccharide stimulation alone did not move FKBP5, so this is a differentiation signal rather than a generic inflammatory one.

FKBP5 tracks pulpitis and injury in tissue

The expression pattern fits the repair setting. In human pulp tissue, FKBP5 was low in healthy and carious pulp but significantly elevated in pulpitis samples (P < .01, N = 4), with signal in the odontoblast layer. In a rat pulp injury model, FKBP5 rose within 3 hours of injury, stayed high from 12 hours to 3 days, and returned to baseline by 7 days (Duan et al. 2026). A transient, injury-induced pulse in the exact cell population expected to build the dentin bridge is the right shape for a repair regulator.

Gain and loss move the dentin program in step

The causal tests used lentiviral knockdown and overexpression in DPSCs. Overexpressing FKBP5 increased mineralized nodule formation, alkaline phosphatase activity, and the dentin matrix proteins DSPP and DMP-1 after induction; knocking it down suppressed all of them (Duan et al. 2026). Transplanted with treated dentin matrix into the subcutis of nude mice, overexpressing cells built a distinct odontoblast-like cell layer with DSPP, connexin 43, and ZO-1 signal that knockdown and control cells did not. On inflammation, knockdown raised IL-1β and IL-8 after LPS exposure while overexpression lowered their protein levels, though the authors state plainly that the mechanism of this anti-inflammatory leaning is unresolved.

The promoter is the point of control

Mechanistically, the paper lands on histone acetylation rather than DNA methylation. ChIP-qPCR showed that odontogenic induction increased H3K9ac enrichment at the FKBP5 promoter, with no comparable change for H3K27ac, and reduced-representation bisulfite sequencing found no methylation difference. Targeted dCas9-p300 recruited acetyltransferase activity to the promoter and raised FKBP5 expression, tying the chromatin mark directly to transcription (Duan et al. 2026).

A pharmacological test in the injury model

The in vivo arm used the rat pulp injury model with drugs rather than gene transfer. Vorinostat (SAHA), a broad HDAC inhibitor, raised H3K9ac and H3K27ac at the FKBP5 promoter and increased FKBP5 expression in pulp tissue at 3 days; the p300/CBP inhibitor C646 had no significant effect. At 28 days, micro-CT showed high-density mineralized tissue resembling reparative dentin in the pulp cavity near the injury site, with bone volume fraction, trabecular number, trabecular thickness, and bone mineral density all elevated in the SAHA group (N = 7 to 8) (Duan et al. 2026). Histology showed a reparative dentin layer with a defined boundary and collagen by Masson staining, plus increased DSPP and DMP-1 and reduced IL-1β at 7 days.

Boundary

Everything causal here is preclinical. The human data are expression patterns across a handful of pulp samples (N = 4 per group), which show association, not mechanism in patients. The gain and loss experiments run in cultured DPSCs and an ectopic nude mouse transplantation site, not in an intact human pulp. And SAHA is a generic HDAC inhibitor, so it demonstrates that histone acetylation at this promoter is a workable lever for reparative dentin in rats; it is not a FKBP5-specific therapy, and FKBP5 itself is a pleiotropic protein with a long record in stress and glucocorticoid signaling that cuts both ways for drug targeting. For the pulp-dentin repair program, the contribution is a specific, testable chain: injury raises H3K9ac at the FKBP5 promoter, FKBP5 pushes odontoblast differentiation and dentin matrix deposition, and pharmacologically raising acetylation at that promoter thickens the reparative barrier in a small-animal model. Whether an epigenetic probe can be aimed at the pulp without off-target effects is the question the paper opens, not one it answers.

Provenance: primary source read in full from the open-access record at PubMed Central, PMC13415464 (PMID 42522268, DOI 10.1093/stcltm/szag047); every quantitative claim above was checked against that full text. Analysis written for the daily pass. See /method/ for the site’s evidence standards.