An injectable gel carrying the bacterium L. johnsonii improved fertility measures in rats after endometrial injury. Animals given the gel showed implantation and produced newborns, while untreated injured controls had neither. The treatment also brought the damaged uterine lining closer to the thickness of sham-operated animals.
The material, called LAT@L, was designed to combine prolonged uterine retention with changes to the local tissue environment. In the rat experiment, animals were randomly assigned to five groups: blank sham-operated rats, untreated injury controls, HA hydrogel, LAT hydrogel and LAT@L hydrogel. All treatments were delivered by direct intrauterine injection.
A gel built to stay in place
That design depended on the gel staying where it was placed. LAT@L adhered firmly to uterine surfaces and achieved intrauterine retention exceeding 14 days. The supplied analysis does not give a full distribution of retention times or adhesion variability, so it does not show how consistently that performance was reproduced across animals.
On the tissue measure, LAT@L moved the injured lining close to the blank-group level. Endometrial thickness rose from 285 µm in controls to a maximum of 576 µm after treatment, approximately 93.5% of blank-group thickness. The fibrotic area was reduced as well.
Signals inside cells and tissue
In a separate cell assay, fibroblasts were exposed to hydrogen peroxide to create oxidative stress. LAT@L produced a substantial reduction in green fluorescence from a probe used to track reactive oxygen species, or ROS, inside cells. The result is consistent with intracellular ROS scavenging, but it remains a cell-based surrogate rather than a direct measure of fertility.
Initial compatibility tests found that the LAT hydrogel did not significantly reduce fibroblast numbers. L. johnsonii released from LAT@L also showed no significant viability difference from free bacteria in vitro. These tests did not establish clinical effectiveness or safety.
Several tissue markers also shifted toward the blank-group pattern. ER and PR expression in LAT@L-treated rats was more similar to the blank group, supporting a receptivity-related recovery signal. The same group showed lower CD86 and IL-6 and higher CD163 and IL-10, a pattern consistent with M2-associated and anti-inflammatory activity.
The uterus's microbial profile shifted too. LAT@L-treated uteri were closer to the blank group, with higher Bacillota relative abundance and lower Escherichia coli and Enterococcus faecalis relative abundance. These are compositional microbiome readouts, so they show a change in the mix of detected microbes, not that the change caused the fertility outcome.
Fertility came with caveats
The fertility test put the largest number on the treatment's reported effect. Untreated controls had no implantation and no newborns, whereas LAT@L improved implantation and raised the average number of neonates to approximately 11. The supplied analysis does not report exact implantation rates, group denominators or variability for this outcome, limiting how precisely the size of the effect can be judged.
Safety checks in the rats found no significant weight loss, major-organ or uterine histologic adverse effects, or changes in serum markers for cardiac, liver or renal function. That is a limited safety finding within the tested measures, but it does not establish the absence of rare or long-term harms. The reported follow-up also does not establish long-term reproductive, offspring or microbiological safety.
For the morphology analysis, the reported sample size was six, with results presented as mean plus or minus standard deviation and tested with one-way ANOVA followed by Tukey's post-test. The stated significance threshold was a p-value below 0.001. The supplied analysis does not provide the matching standard deviation, confidence interval or full effect-size calculation for the thickness change.
What the study cannot answer
The work remains preclinical. It combines a rat injury model with cell and bacterial experiments, so the reported results do not establish efficacy or safety in women. It also does not establish whether the microbiome changes, immune-cell pattern or receptivity-marker shifts were necessary for fertility recovery.
The authors acknowledged support from the Guangzhou Municipal Science and Technology Project and a joint fund of the Third Affiliated Hospital of Sun Yat-sen University and Chaozhou Central Hospital. Study data are available from the corresponding author upon reasonable request.
Paper data and sources
Original title: Lactobacillus-Loaded Bioadhesive Hydrogel with Prolonged Retention and Microenvironmental Remodeling for Endometrial Repair and Fertility Restoration.
Authors: Lifa Chen, Hui Zhou, Yifei Li et al.
Journal/Repository: Advanced science (Weinheim, Baden-Wurttemberg, Germany)
Status: Peer-reviewed
First online: 2026-08-21
DOI: 10.1002/advs.77229
Original paper