Abstract
Background:
Olanzapine is a commonly used drug in the treatment of schizophrenia, but the mechanism of abnormal bone metabolism caused by olanzapine is still unclear. The change of microflora may be an important factor leading to the change of bone metabolism. Therefore, the purpose of this study was to explore a plausible hypothesis that olanzapine may aggravate abnormal bone metabolism and cause bacterial imbalance in patients with schizophrenia.
Methods:
This study intervened in mice by gavage with olanzapine to detect changes in alveolar bone tissue and oral microbiota. The effect of related bacteria on osteogenesis was further examined.
Results:
The results showed that Enterococcus increased, the bone mass and type I collagen of alveolar bone decreased. Enterococcus lipoteichoic acid (LTA) inhibited osteogenic differentiation and up-regulated SAA1 gene expression. SAA1 gene can down-regulate the expression of COL1A1 gene, and the proteins encoded by the two may interact.
Conclusions:
Olanzapine may increase the relative abundance of oral Enterococcus, whose components are plausibly linked to increased expression of SAA1 gene and inhibition of bone formation through extracellular matrix-related pathways. These exploratory findings support further exploration of microbiota-based strategies to alleviate skeletal complications and promote oral health. The clinical research presented in this paper has been registered on ClinicalTrials.gov, a platform of the U.S. National Institutes of Health (Registration Number: NCT06123897; URL: https://clinicaltrials.gov/ct2/show/NCT06123897), with the registration date of November 9, 2023.
1 Introduction
With increasing social pressure, the incidence of psychiatric diseases such as schizophrenia has risen significantly (Charlson etĀ al., 2018; Jauhar etĀ al., 2022). As one of the top ten causes of disability worldwide, schizophrenia has placed a substantial economic burden on society. Atypical antipsychotic drugs are currently the most common treatment for schizophrenia, and olanzapine is one of the most widely used among them. It has minimal extrapyramidal reactions and anticholinergic effects. As a dopamine-serotonin antagonist, olanzapine binds with high affinity to the 5-hydroxytryptamine receptor 2A (5-HT2A) and effectively alleviates psychotic symptoms in patients. However, the use of olanzapine can further exacerbate abnormal bone metabolism in schizophrenia patients (Maagensen etĀ al., 2021), with up to two-thirds of patients experiencing impaired bone mineral density after taking the drug (De Hert etĀ al., 2016; Sabe etĀ al., 2022). Therefore, the impact of olanzapine on bone metabolism cannot be overlooked.
Atypical antipsychotics can worsen bone metabolism through various pathways. The prevailing view is that antipsychotics primarily cause abnormal bone metabolism by up-regulating prolactin levels (Naidoo etĀ al., 2003). However, some studies have found that patients with schizophrenia do not always exhibit significant changes in serum prolactin levels after taking olanzapine, suggesting that olanzapine may induce abnormal bone metabolism through non-hyperprolactinemic pathways. Further research is required to explore the mechanisms by which olanzapine affects bone formation in schizophrenia.
Microbiota has been shown to be closely linked to bone formation. More than half of the bacteria in the human body are colonized in the gastrointestinal tract (29%) and oral cavity (26%) (Peterson etĀ al., 2009). Both the intestinal and oral microbiota have been proven to influence bone metabolism (Lu etĀ al., 2021; Zhao etĀ al., 2022; Duffuler etĀ al., 2024; Xu etĀ al., 2024), and changes in intestinal microbiota may also affect the oral microbiota (Park etĀ al., 2021). Additionally, it has been reported that atypical antipsychotic drugs possess certain antibacterial properties and can cause disturbances in the intestinal microbiota (Nehme etĀ al., 2018; Chait etĀ al., 2020; Chen etĀ al., 2023), although there are few studies focusing on the oral microbiota. The changes in oral microbiota following olanzapine administration and their effects on bone metabolism require further investigation.
Therefore, this study aims to investigate the clinical phenomenon of olanzapine exacerbating abnormal bone metabolism and causing microbiota imbalances in schizophrenia patients. This will help explore the role and mechanisms of oral microbiota dysbiosis, providing new therapeutic targets and research directions for addressing abnormal bone metabolism in schizophrenia patients after medication use.
2 Results
2.1 Olanzapine treatment results in a significant decrease in bone mass
In the clinical study, male patients taking olanzapine had significantly lower Z-scores for the lumbar spine (ā0.36 vs. 0.91), femoral neck (ā0.49 vs. 1.34), and hip (ā0.09 vs. 1.06) than controls (P<0.05). In females, Z-scores for the lumbar spine (ā0.51 vs. 0.32) and femoral neck (ā0.63 vs. 0.27) were also lower (P<0.05), with a trend toward reduction at the hip (ā0.30 vs. 0.21, P=0.092) (FigureĀ 1A, TablesĀ 1, 2). In the overall sample (pooled across sex), unadjusted linear regression showed that olanzapine use was associated with lower BMD Z-scores at all skeletal sites: femoral neck (β=ā1.16, 95% CI ā1.67 to ā0.65; P<0.001), lumbar spine (β=ā0.94, 95% CI ā1.43 to ā0.46; P<0.001), and hip (β=ā0.72, 95% CI ā1.17 to ā0.27; P=0.003). After multivariable adjustment for covariates, the associations remained statistically significant and were slightly stronger: femoral neck (β=ā1.34, 95% CI ā1.84 to ā0.83; P<0.001), lumbar spine (β=ā1.04, 95% CI ā1.50 to ā0.57; P<0.001), and hip (β=ā0.88, 95% CI ā1.33 to ā0.43; P<0.001) (Supplementary Table S1). These findings suggest bone mass loss in patients taking olanzapine. Therefore, we subsequently explored this hypothesis through in vivo animal experiments and in vitro cell experiments. As shown in FigureĀ 1B, 3-month-old mice were administered normal saline or olanzapine via gavage for 12 weeks, during which the weight of the mice was recorded. After euthanasia, the mandibles of the mice were examined using micro-CT.
FigureĀ 1
TableĀ 1
| Male | Control | Olanzapine | P |
|---|---|---|---|
| Age (years) | 22.79±2.83 | 22.88±3.81 | 0.9306 |
| BMI | 23.03±4.12 | 23.10±2.10 | 0.9477 |
| Lumbar Spine BMD (g/cm2) | 1.02±0.09 | 0.89±0.11 | 0.0001 |
| Femoral Neck BMD (g/cm2) | 0.96±0.12 | 0.77±0.10 | 0.0001 |
| Hip BMD (g/cm2) | 1.01±0.13 | 0.83±0.17 | 0.0004 |
| Lumbar Spine Z-score | 0.91±0.91 | -0.36±1.19 | 0.0004 |
| Femoral Neck Z-score | 1.34±1.24 | -0.49±1.03 | 0.0001 |
| Hip Z-score | 1.06±1.18 | -0.09±0.95 | 0.0019 |
Comparison of bone mineral density indicators between clinical male patients taking olanzapine and healthy controls.
Data are presented as mean±standard deviation (Control Group N=24; Olanzapine Group N=17). Statistical significance was determined by unpaired t-test. P<0.05 was considered significant.
TableĀ 2
| Female | Control | Olanzapine | P |
|---|---|---|---|
| Age (years) | 23.76±2.79 | 26.00±4.12 | 0.0928 |
| BMI | 21.07±2.98 | 22.55±1.46 | 0.1687 |
| Lumbar Spine BMD (g/cm2) | 0.98±0.09 | 0.91±0.09 | 0.0435 |
| Femoral Neck BMD (g/cm2) | 0.82±0.09 | 0.75±0.12 | 0.0738 |
| Hip BMD (g/cm2) | 0.89±0.08 | 0.85±0.09 | 0.1687 |
| Lumbar Spine Z-score | 0.32±0.93 | -0.51±1.01 | 0.036 |
| Femoral Neck Z-score | 0.27±0.98 | -0.63±1.26 | 0.0296 |
| Hip Z-score | 0.21±0.63 | -0.30±0.96 | 0.092 |
Comparison of bone mineral density indicators between clinical female patients taking olanzapine and healthy controls.
Data are presented as mean ± standard deviation (Control Group N=21; Olanzapine Group N=9). Statistical significance was determined by unpaired t-test. P < 0.05 was considered significant.
After 12 weeks of olanzapine administration, mice in the experimental group weighed less than those in the control group (FigureĀ 1C, Supplementary Figure S1). Further micro-CT results showed a decrease in bone mass between the first and second molars in the olanzapine group (FigureĀ 1D), along with a significant reduction in bone volume fraction (BV/TV) and trabecular number (TbN), as well as a significant increase in trabecular separation (TbSp) (FigureĀ 1E).
2.2 After olanzapine treatment, the composition of oral microflora in mice was changed
After olanzapine intervention, saliva from the mouths of mice was extracted for 16S rRNA sequencing (Figure 2A). The dilution curves of different samples eventually tended to be horizontal, indicating that the sequencing depth was sufficient and the sample sequencing volume was reasonable (Supplementary Figure S2A). The richness, evenness, and diversity of the oral microflora in mice were evaluated through α diversity analysis (Figure 2B). The results showed that after olanzapine intervention, the Observed Species index, Chao1 index, and ACE index, which represent the richness of the microbial community, showed no statistical differences. Additionally, the Simpson index, Shannon index, and PD whole tree index, which are closely related to the diversity of oral microbial communities in mice, also showed no statistical differences (Figure 2B).
FigureĀ 2
Based on β diversity, 3D PCoA showed an apparent separation in oral microbial composition between control and olanzapine-treated groups (Figure 2C). Each point represents a sample, with closer distances indicating more similar communities. PERMANOVA (9, 999 permutations) confirmed a significant difference between groups (R²=0.247, pseudo-F=2.62, P=0.021).
2.3 Olanzapine may affect bone mass in mice by affecting the abundance of Enterococcus
Comparing the compositional differences of oral microbial communities in olanzapine-treated and non-treated mice, it was found that Enterococcus, Muribacter, and Bergeyella in olanzapine-treated mice were significantly increased at the genus level (| LDA score | ā„2, P<0.05). Meanwhile, Staphylococcus significantly decreased (FiguresĀ 3A, B; Supplementary Figures S2B, C). Among the significantly increased bacteria, Enterococcus was closely related to bone metabolism. As can be seen from Supplementary Figure S2C, there was almost no Enterococcus in the saliva of olanzapine-untreated mice, while the relative expression of Enterococcus in the oral cavity of olanzapine-treated mice ranged from 0.0001 to 0.0018. Enterococcus was among the top predictive genera identified by the random forest model (Supplementary Figure S2D). In a small ancillary pilot based on human saliva, the relative salivary Enterococcus load was higher in the olanzapine group (17.35±1.15) than in controls (1.00±0.19) (P < 0.05) (Supplementary Figure S2E).
FigureĀ 3
After that, lipoteichoic acid (LTA) from Enterococcus was extracted. After 7 days of treatment with 0, 10, and 20 μg/mL LTA, alkaline phosphatase (ALP) staining showed a significant, concentration-dependent reduction in staining intensity in the LTA-treated hPDLSCs (Figures 3C, D). These results indicate that Enterococcus LTA has an inhibitory effect on the osteogenesis of hPDLSCs, and that the inhibitory effect is stronger with higher concentrations within a certain range.
In previous experiments, we found that olanzapine intervention led to a decrease in bone mass in both humans and mice. At the same time, after olanzapine intervention, Enterococcus appeared in the saliva of mice, which was not previously present. In parallel, its abundance also increased in the saliva of individuals taking olanzapine. Therefore, we suspect that Enterococcus may be the possible cause of olanzapine-induced bone mass change, but the relevant mechanism needs to be verified later.
2.4 After olanzapine treatment, the extracellular matrix of mouse bone tissue showed alterations
We explored a plausible mechanistic hypothesis by which olanzapine causes bone loss from two aspects: inflammation and collagen formation (FigureĀ 4A). Regarding inflammation, we observed immune cell infiltration in the bone tissue of olanzapine-treated mice, indicating local inflammatory activity (FigureĀ 4B).
FigureĀ 4
Next, we investigated collagen formation. The immunohistochemical staining results of the COL1A2 antibody showed that the brown staining in the periodontal membrane of the molar area in the olanzapine group was lighter than that in the control group (FigureĀ 4C), with the deepest brown color observed in the periodontal ligament stem cells (PDLSCs) and osteoblasts (OB). Immunohistochemical staining of the OPG antibody showed no significant difference in color between the control group and the olanzapine group. After staining with Sirius red, the dense red bunched fibers in the periodontal membrane of both the olanzapine and non-olanzapine groups were visible under the optical microscope (Supplementary Figure S3). Under polarized light, a large number of bright yellow type I collagen fibers could be seen in the periodontal membrane of mice in the control group, exhibiting a dense and orderly arrangement. In contrast, the yellow type I collagen in the periodontal membrane of mice treated with olanzapine was significantly reduced, darkened in color, and exhibited a more disordered arrangement (FigureĀ 4D).
It is worth mentioning that the staining changes in periodontal ligament stem cells and osteoblasts were most pronounced in the immunohistochemical staining. Considering the representativeness of the oral cavity, periodontal ligament stem cells were selected as experimental cells for in vitro studies.
2.5 Enterococcus LTA induced extracellular matrix changes by up-regulating SAA1 gene
To further investigate the mechanism, human periodontal ligament stem cells (hPDLSCs) were used to conduct an in vitro experiment. Enterococcus LTA was used to interfere with hPDLSCs for 7 days, and RNA was extracted for transcriptome sequencing (FigureĀ 5A). The base quality distribution map predicts the probability of base discrimination errors. The average base error rate of each sample read at each site was found to be less than 0.04%, indicating that the sequencing quality evaluation was satisfactory (Supplementary Figure S4A). It was observed that there was relatively little difference within the same group, but significant differences between different groups (Supplementary Figures S4B, C). Compared with the control group, 153 genes in the hPDLSCs of the LTA group were up-regulated and 178 genes were down-regulated (Supplementary Figure S4D). The top 20 differentially expressed genes are shown in TableĀ 3 and FigureĀ 5B. Among these, the expression of SAA1, NKD1, USP53, DKK1, IRS2, SHISA9, TMEM97, B3GALT2, HMGCS1, and ANKRD1 genes was increased, while the expression of CXCL12, CLDN11, SLC14A1, LOC102724560, COL15A1, FRAS1, PLPP3, PENK, PLAU, and TGFBI genes was decreased. GO and KEGG enrichment analysis showed that the differentially expressed genes were closely related toĀ extracellular matrix-related pathways and functions (FiguresĀ 5C, D).
FigureĀ 5
TableĀ 3
| Gene id | Fold change | p-value | q-value | Regulation |
|---|---|---|---|---|
| SAA1 | 4.283057 | 3.81E-85 | 6.62E-81 | Up |
| CXCL12 | 0.35498 | 5.57E-65 | 4.84E-61 | Down |
| NKD1 | 3.600365 | 9.41E-53 | 5.45E-49 | Up |
| CLDN11 | 0.327735 | 1.10E-49 | 4.76E-46 | Down |
| SLC14A1 | 0.284362 | 1.87E-44 | 6.49E-41 | Down |
| USP53 | 2.104107 | 3.89E-44 | 1.13E-40 | Up |
| DKK1 | 2.172764 | 1.08E-43 | 2.68E-40 | Up |
| IRS2 | 2.061607 | 1.30E-43 | 2.83E-40 | Up |
| LOC102724560 | 0.023067 | 1.52E-38 | 2.94E-35 | Down |
| SHISA9 | 3.661599 | 9.52E-37 | 1.65E-33 | Up |
| COL15A1 | 0.389733 | 2.40E-36 | 3.79E-33 | Down |
| FRAS1 | 0.404886 | 8.87E-33 | 1.19E-29 | Down |
| TMEM97 | 2.45424 | 3.30E-31 | 4.09E-28 | Up |
| B3GALT2 | 3.289264 | 8.74E-31 | 1.01E-27 | Up |
| PLPP3 | 0.491811 | 1.16E-30 | 1.25E-27 | Down |
| HMGCS1 | 2.177893 | 9.90E-30 | 1.01E-26 | Up |
| PENK | 0.452315 | 1.19E-28 | 1.15E-25 | Down |
| ANKRD1 | 2.926935 | 8.55E-27 | 6.75E-24 | Up |
| PLAU | 0.478952 | 7.53E-26 | 5.69E-23 | Down |
| TGFBI | 0.435548 | 5.66E-24 | 4.10E-21 | Down |
The top 20 differentially expressed genes (p-value<0.05, |log2FC| >1, ranked by q-value or p-value).
Among the top 20 differentially expressed genes, SAA1, CXCL12, NKD1, CLDN11, DKK1, COL15A1, FRAS1, TGFBI, and PLAU genes were selected, as they were closely related to bone metabolism. The CLDN11 gene was excluded based on a homology test, and the remaining genes were further validated by qPCR (FigureĀ 5E). Additionally, most of the above gene-related pathways are also associated with the extracellular matrix or pathways with a high KEGG ranking, such as the NF-ĪŗB signaling pathway. The results of the qPCR showed that the expression of SAA1 was most up-regulated, while the expression of COL15A1 was most down-regulated (FigureĀ 5F). SAA1 was the most differentially expressed gene in the whole transcriptome sequencing. Therefore, SAA1 was selected for further experiments.
2.6 Enterococcus LTA may inhibit the expression of COL1A1 by promoting the expression of SAA1, and ultimately inhibit osteogenic differentiation
In order to further explore the role of the SAA1 gene in inhibiting osteogenic differentiation of Enterococcus LTA, the SAA1 gene was inhibited in hPDLSCs, followed by ALP staining and qPCR assays (FigureĀ 6A). In the ALP staining, inhibition of the SAA1 gene in hPDLSCs enhanced the staining intensity and increased the stained area (FigureĀ 6B). qPCR results indicated that after the inhibition of the SAA1 gene, the expression of the COL1A1 gene in hPDLSCs was up-regulated (FigureĀ 6C). Inhibition of the SAA1 gene led to the up-regulation of COL1A1 gene expression, suggesting a possible interaction between the two. The relationship between SAA1 and COL1A1 was analyzed using PDBePISA website and PyMol tool, revealing that the two proteins were likely to bind and had multiple interaction sites (FigureĀ 6D). The interface area between SAA1 (Structure 2) and COL1A1 (Structure 1) is 2689 Ć 2, and the free energy (ĪG) under this docking mode is -27.7 kcal/mol (Supplementary Figure S5A). Supplementary Figure S5B shows the number of atoms and residues, solvent-accessible area and solvation energy. All hydrogen bonds and salt bridges formed on the interaction surface are shown in Supplementary Figure S5C. Therefore, Enterococcus may inhibit COL1A1 at both the gene expression and protein action levels by promoting SAA1 expression, thus inhibiting the osteogenesis of hPDLSCs.
FigureĀ 6
3 Discussion
Through clinical trials and animal experiments, this study found that long-term olanzapine use can cause a reduction in bone mass, accompanied by an increase in Enterococcus abundance. Combined with cell experiments, this study suggests that these changes in Enterococcus are related to the SAA1 gene. Based on this, a reasonable hypothesis can be made that olanzapine may increase the abundance of Enterococcus, and its component (LTA) may inhibit osteogenic differentiation through SAA1-mediated inflammatory and extracellular matrixārelated signaling pathways, suggesting that Enterococcus may be involved in olanzapine-induced bone loss. More in vivo studies are needed to verify this in the future.
Olanzapine is commonly used to treat schizophrenia, but patients with schizophrenia who take olanzapine experience a decrease in bone mineral density, which is one of the common adverse effects of atypical antipsychotic drugs (Wyszogrodzka-Kucharska and Rabe-Jablonska, 2005). We found that olanzapine-treated mice also had lower bone mass and body weight than controls. Consistent with this, Langlais etĀ al. (2023) and Li and Nagy (2013) used olanzapine to intervene in mice for 3ā4 weeks and found that the trabecular bone of the femur in these mice was lost, with the volume and connectivity of the trabecular bone lower than that of the control group. The mechanism of weight change is complex and is related to factors such as gender, breed and environment. Previous studies have shown that olanzapine induces weight gain in female rats but has minimal effects on males, while some strains such as A/J mice even exhibit weight loss (Albaugh etĀ al., 2006; Davey etĀ al., 2012a).
However, the mechanism by which olanzapine causes abnormal bone metabolism remains unclear. Our 16S rRNA sequencing results showed that olanzapine could cause changes in oral microflora, but there was no significant difference in the diversity and richness of the microflora. Pelka Wysiecka et al. (2019) also found that after olanzapine treatment, there was no significant change in the α diversity of the microbiota of individuals with schizophrenia. However, Davey et al. found in a rat model that the diversity of the microbial community decreased (Davey et al., 2012b; Davey et al., 2013). This may be related to the strain of the animal and the way it is raised. We found that the abundance of Enterococcus increased after olanzapine intervention. We also found in the human samples that patients taking olanzapine had more Enterococcus in their saliva. Enterococci are Gram-positive bacteria that belong to the gastrointestinal flora of mammals (Hendrickx et al., 2009). Previous studies have shown that olanzapine has direct antibacterial effects on Enterococcus in the intestinal tract (Morgan et al., 2014). In the oral cavity, Enterococcus has been identified as a common root canal filling pathogen (Zehnder and Guggenheim, 2009). Intestinal flora may spread through the faecal-oral route and affect oral flora (Shaffer and Lozupone, 2018). In addition, diseases closely related to intestinal flora (such as inflammatory bowel disease and colitis) may also affect oral flora (Said et al., 2014; Rautava et al., 2015; Park et al., 2021), indicating that the intestinal flora may affect the oral flora through various ways. Olanzapine may further cause a compensatory increase of Enterococcus in the oral cavity after reducing Enterococcus in the intestinal tract. LTA is one of the main virulence factors of Enterococcus (Jett et al., 1994; Kayaoglu and Orstavik, 2004; Zehnder and Guggenheim, 2009). The literature indicates that Enterococcus is rare in the oral cavity of healthy individuals and is an opportunistic pathogen. When detected in the oral cavity, its abundance is often extremely low, yet it can exhibit significant pathogenicity and environmental tolerance (Zehnder and Guggenheim, 2009; Parga et al., 2025). This is consistent with our research. We did not detect Enterococcus in the saliva of the control group mice, while in the olanzapine group, Enterococcus was detectable but remained at a low abundance. However, even at such a low abundance level, Enterococcus could still cause significant local tissue damage (Zehnder and Guggenheim, 2009; Parga et al., 2025). We confirmed that Enterococcus LTA could inhibit osteogenic differentiation. Wang et al. (2016) used Enterococcus to interfere with osteoblasts in vitro and also found that osteogenic differentiation was inhibited. In vivo studies also indicate that Enterococcus is associated with reduced bone mass. For instance, Enterococcus infection has been proven to cause periapical inflammation in mice and to result in significant bone destruction (Dai et al., 2022). Beyond the oral cavity, Balish et al. discovered that Enterococcus can cause inflammatory bowel disease in mice (Balish and Warner, 2002), and such inflammation is closely related to systemic bone loss. In our research, the use of olanzapine occurred simultaneously with an increase in the abundance of Enterococcus in the oral cavity. Combined with the above evidence, this raises a hypothesis that Enterococcus may be one of the potential factors contributing to bone loss, but the mode of its influence still needs further exploration.
We found that Enterococcus LTA caused changes in the inflammatory pathway (NF-ĪŗB signaling pathway) of hPDLSCs. Consistent with this, the literature suggests that Enterococcus can destroy the epithelial barrier, promote intestinal inflammation (Steck etĀ al., 2011), and may also be associated with different stages of root canal infection and periapical inflammation (Kayaoglu and Orstavik, 2004). We also found that the up-regulation of the SAA1 gene was most significant in the whole transcriptome sequencing. The serum amyloid A1 gene (SAA1) encodes a member of the serum amyloid A apolipoprotein family. This is a major acute-phase protein highly expressed in response to inflammation and tissue damage. Previous studies have shown that microbial communities can alter the level of SAA1 expression. After fecal microbiota transplantation, Saa1 expression increased in mice (Huang etĀ al., 2023). In mice treated with low protein microbiota (LPM) and high lipid microbiota (HFM), Saa1 gene expression increased in the latter (Busnelli etĀ al., 2020). Additionally, Mendelian randomization (MR) analysis found that increasing the abundance of Roseburia would decrease the level of SAA1 (Liu etĀ al., 2024).
In addition, the extracellular matrix refers to the complex network of multiple macromolecules around the cell. Collagen is the main component of the extracellular matrix, in which type I collagen is composed of two α1 chains and one α2 chain. The extracellular matrix not only supports and connects, but also contains a large number of signaling molecules, which are closely related to osteogenesis. Our previous studies have shown that cytokine changes in extracellular matrix affected by microRNA can further affect the expression of collagen I α2 and cause osteogenic changes (Guo et al., 2016; Guo et al., 2017; Li et al., 2020). This time, we found a decrease in type I collagen in the extracellular matrix by bone tissue staining. Combined with the results of whole transcriptome sequencing, we hypothesized that extracellular matrix related pathway may be one of the main pathways that Enterococcus LTA inhibits osteogenic differentiation. Previous studies have also shown that Enterococcus can significantly down-regulate the expression level of collagen type 1 (COL1) (Wang et al., 2016), activate tissue matrix metalloproteinase 9 (MMP9) (Shogan et al., 2015), and degrade collagen and glycosaminoglycan (the main components of the extracellular matrix) (Shogan et al., 2015; Kawai et al., 2018).
We further inhibited the SAA1 gene and found that COL1A1 expression was up-regulated, enhancing osteogenic differentiation. Wang WS. etĀ al. (2019) detected the expression level of COL1A1 in amniotic fibroblasts, showing no significant difference. The reason for the difference with our results may be related to cell types and treatment methods. We also predict that the proteins encoded by SAA1 and COL1A1 may be bound by hydrogen bonds and salt bridges. ĪiG is less than zero, indicating that the docking results are meaningful. Specifically, amino acid 17ā108 in SAA1 may interact with amino acid 88ā1384 in COL1A1. Our molecular docking experiments show that residues such as Glu92, Asp78, and Gln98 in SAA1 and Gly183, Gln1380, and Glu1384 in COL1A1 may be important binding sites. Meanwhile, SAA1 has been shown to increase the abundance of collagenases MMP-1, MMP-8, and MMP-13 while decreasing the abundance of lysine oxidase-like 1, an enzyme that cross-links collagen (Han etĀ al., 2016; Wang YW. etĀ al., 2019; Lin etĀ al., 2023). At the same time, Wang WS. etĀ al. (2019) suggested that SAA1 might degrade type I collagen through autophagy and MMP pathways. Therefore, SAA1 may inhibit osteogenic differentiation through extracellular matrix-related pathways by influencing the expression of the COL1A1 gene, directly acting on COL1A1-encoded proteins, or indirectly affecting type I collagen through autophagy or enzyme-related pathways.
This study has some limitations. First, the in vivo model did not fully rule out other potential confounding factors induced by olanzapine, such as alterations in prolactin levels or metabolic pathways, which may also influence bone metabolism. Second, the relatively small sample size may limit the generalizability and statistical power of the findings. As such, the roles of Enterococcus and SAA1 should be interpreted as exploratory signals rather than established mechanisms. Future research should increase the sample size, adopt standardized animal models, and, where feasible, employ colonization or clearance models to directly assess the effects of Enterococcus on bone phenotypes.
In conclusion, we propose a plausible hypothesis that olanzapine may increase the relative abundance of oral Enterococcus, promote the expression of the SAA1 gene, and thus inhibit the expression of the COL1A1 gene. It may also affect the protein interaction between amino acids 17ā108 in SAA1 and amino acids 88ā1384 in COL1A1, ultimately inhibiting bone formation. These findings suggest a possible link between oral microbiota alterations and olanzapine-induced bone metabolism abnormalities, supporting further investigation into microbiota-informed strategies for reducing skeletal complications and maintaining oral health.
4 Materials and methods
4.1 Experimental design
The purpose of this study was to investigate the mechanism of olanzapine aggravating abnormal bone metabolism and causing bacterial imbalance in patients with schizophrenia. This study verified the effect of olanzapine on bone metabolism through clinical trials. A mouse model of olanzapine administration was established to investigate the changes of bone tissue and oral flora. The effect and mechanism of related bacteria on osteogenesis were further explored by cell experiment and prediction of protein interaction.
4.2 Clinical trials
A total of 71 participants were enrolled in this study. The olanzapine group consisted of 26 patients (17 males and 9 females), while the healthy control group included 45 individuals (24 males and 21 females), with age-matching performed between the two groups. Inclusion Criteria: Patients in the olanzapine group met the diagnostic criteria for schizophrenia according to the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV), and their diagnosis was independently confirmed by two or more psychiatrists with attending or senior titles. Eligible patients were between 20 and 35 years old (inclusive), received monotherapy with olanzapine at doses ranging from 5 to 25 mg/day, and had been on continuous and regular olanzapine treatment for more than 6 months. The healthy control group consisted of adults aged between 20 and 40 years with no history of psychiatric illness. Exclusion Criteria: Participants were excluded if they had taken medications affecting bone metabolism for more than 6 months, such as bisphosphonates, glucocorticoids, parathyroid hormone, calcitonin, or active vitamin D analogs. Other exclusion criteria included diseases that may affect bone metabolism, such as uncontrolled diabetes mellitus (HbA1c > 7%), hyperthyroidism or hypothyroidism, primary parathyroid disorders, osteoarthritis, unresolved skeletal disorders, or a history of radiation therapy or chemotherapy within the past 6 months. Subjects with contraindications for dual-energy X-ray absorptiometry (DXA) were also excluded.
This study protocol was approved by the Human Research and Ethics Committee of the Second Xiangya Hospital of Central South University. The clinical study was registered on the ClinicalTrials.gov platform of the U.S. National Institutes of Health (Registration Number: NCT06123897). All procedures strictly adhered to the ethical guidelines outlined in the Declaration of Helsinki. All experiments were conducted in accordance with relevant guidelines and regulations. Written informed consent was obtained from all participants after a thorough explanation of the study procedures. All procedures associated with this study were performed strictly after obtaining written informed consent from each participant.
Bone mineral density (BMD) of the lumbar spine, femoral neck, and hip bone was measured using a dual-energy X-ray absorptiometry (DEXA) scanner (Hologic, USA). Given the significant gender- and age-dependent variation in physiological bone density, Z-scores were calculated to minimize related influences. The Z-score was defined as follows: Z-score=(Measured BMD - Mean BMD of age-, sex-, and ethnicity-matched reference population)/Standard deviation of BMD in the reference population.
An ancillary saliva pilot (n=3) was conducted. Exclusions: antibiotics within 3 months; antiseptic mouthwash on the sampling day; dental/periodontal procedures within 3 months; active oral infection or acute respiratory illness; current smoking; major systemic disease or immunosuppression. Unstimulated whole saliva was collected by passive drooling after ā„30-min abstinence from eating, drinking, or oral hygiene, immediately aliquoted, and stored at ā80 °C.
4.3 Olanzapine animal modeling
3-month-old male C57BL/6J mice (Hunan Slake Jingda Laboratory Animal Co., LTD., China) weighing 20-24g were selected. The mice were quarantined and acclimatized for 1 week, then randomly assigned to different treatment groups using a computer-generated randomization schedule. Weight before administration<18 or > 26 g or mice in poor physical condition were excluded from the experiments. All mice were maintained under specific pathogen-free conditions in an animal facility with controlled temperature (22 ± 2 °C), humidity (50 ± 10%), and a 12-hour light/dark cycle. Mice had ad libitum access to autoclaved standard chow and sterile water. The mice were divided into olanzapine and control groups, with 10 mice in each group. Olanzapine group: 3-month-old mice were treated with olanzapine (4 mg/kg/day) by gavage for 12 weeks. Mice were given solvent using the same treatment protocol served as healthy controls. Group allocation and outcome assessment were performed in a blinded manner to reduce bias. Changes in body weight were recorded weekly. The dose of olanzapine was adjusted for each mouse. All animals were monitored daily for signs of distress, and predefined humane endpoint were established but not triggered during the study.
The animal study was reviewed and approved by the Experimental Animal Ethics Committee of the Second Xiangya Hospital. All experiments were performed in accordance with relevant named guidelines and regulations. The authors complied with the ARRIVE guidelines.
4.4 16S RNA identification
(1) Sample extraction: Take a sterile cotton ball, place it in the mouseās mouth, allow it to be fully chewed, then remove the fully wetted cotton ball and place it into an EP tube. Preserve it at -80 °C, transport it in liquid nitrogen, and send it to Shanghai Ouyi Biotechnology Co., Ltd. for 16S rRNA sequencing.
(2) DNA extraction: We use MagPure Soil DNA LQ Kit based on the instructions provided by the reagent vendor (Magen China). The kit separates DNA from each sample. Using the NanoDrop 2000 Spectrophotometer (Thermo Fisher Scientific, MA, USA) and Agarose Gel Electrophoresis (Depending on the situation) DNA yield, purity, and integrity were assessed.
(3) PCR Augmentation and Library Construction: Use universal primer 343F: 5'-TACGGRAGGCAGCAG-3' and 798R: 5'-AGGGTATTAATCCT-3') PCR amplification was performed on V3-V4 high-variation regions of 16S rRNA gene. Both primers are connected to the Illumina sequencing adapter. Gel electrophoresis is used to visualize the quality of amplified products. The Agencourt AMPure XP bead (Beckman Coulter Co., USA) was used to purify PCR products and quantify them with Qubit dsDNA assay kits. The concentration is then adjusted for sequencing.
(4) High throughput sequencing, bioinformatics: Sequencing is using EE (OE Biotech Shanghai, China) the Illumina NovaSeq6000 high-throughput sequencing platform, Follows its standard procedures. The original data is in FASTQ format, and after the data is disembarked, the pair of end readings are pre-processed using cutadapt software to detect and disconnect the adapter. After trimming, use DADA2 and QIME2 (2020.11) The default parameter filters low-quality sequences, de-noises, merges, and detects and cuts chimeric readings for pairing end readings to obtain representative sequences and ASV abundance tables. Use the QIIME 2 package to select the representative sequences for each ASV and compare all the representative sequences with the database. 16S uses the Silva (version138) database comparison. Species comparison annotations are analyzed using the q2 feature classifier default parameters Statistical analysis of microbial composition and function was carried out at species and genus levels. The differential expression of bacterial genes was identified by comparing relative abundance. The Wilcoxon rank and test are analyzed to assess the differences in the relative abundance of microbial communities. Bacterial abundance and diversity among samples were assessed using the α index (ACE, Chao1, Shannon, and Simpson). Based on relative abundance, the t-test is used to compare the α diversity index. We use Primary Coordinate Analysis (PCoA) to map similarities or differences in sample community composition. In addition, a similarity analysis (Adonis) was conducted to assess the significance of the differences in community structure between the two groups using the β diversity index. Use the default parameter (α parameter with a significant threshold set to 0.05 and the logarithmic LDA score cut-off set to 2.0) to perform a differential taxon analysis using the linear discriminant analysis effect size (LEfSe). We used the random forest model as an exploratory predictive method to assess the contribution of Enterococcus to group discrimination.
4.5 μCT analysis
After 12 weeks of continuous oral gavage with olanzapine (4Ā mg/kg/day), the mandibular alveolar bone was collected and placed in polyformaldehyde for two days. Following this, the alveolar bone was rinsed with tap water 10ā12 times, dehydrated in 50% alcohol for 30 minutes, and then dehydrated in 70% alcohol for another 30 minutes at 4 °C.
Treated mandibles from all experimental groups were analyzed with micro-computed tomography (micro-CT) (Aoyin AX2000ct, Shanghai, China). X-ray imaging was conducted at a tube voltage of 90 kV and a tube current of 80 μA, with an image acquisition time of 2 minutes and a pixel size of 9.5 μm/100 μm. VGstudio MAX 3.4.4 (Volume Graphics, Inc., Heidelberg, Germany)was used to analyze the bone mass between the roots of the first and second molars in mice.
4.6 HE
(1) Burying and Slices
Fresh mouse mandible tissue was immediately placed in polyformaldehyde fixation solution for more than 24h, followed by EDTA decalcification for 28 days, Changing the solution every 7 days, The tissue was then trimmed and places into a embedding box for dehydration and infiltration with wax for paraffin embedding.
Finally, the wax block was placed in the slicer, with the mouse molarās oriented from near to far as the axis, and cut into 4 μm slices. The slices were baked in an oven at 60°C. After drying, the wax blocks were removed and stored at room temperature.
(2) HE
a) Dewaxing and Hydration: Paraffin slices are dewaxed in xylene solution for 20 minutes, followed by 5 minutes in a gradient ethanol solution, and then placed in distilled water for 3 minutes. b) Hematoxylin staining: The slices are stained with hematoxylin solution for about 3ā5 minutes, washed with tap water for about 10 minutes, treated with 1% hydrochloric acid differentiation solution for about 3 seconds, and finally placed in flowing water to restore the blue color. c) Eosin staining: The slices are stained in 0.5% eosin Y solution 5min. d) Dehydration and Sealing: Slices are placed in gradient ethanol and xylene solution 2min, then sealed with neutral gum.
4.7 IHC
The slides were placed in an oven at 65 °C for 4ā6 hours, followed by dewaxing with turpentine, gradient ethanol, distilled water, and PBS. Next, the tissue was repaired with 0.25% trypsin. After cooling to room temperature, a circle was drawn around the tissue with an immunohistochemical pen at a distance of about 2 mm. The slides were incubated with 3% H2O2 to avoid light exposure. Afterward, 50 μL of anti-Collagen I alpha2 (Affinity USA) and OPG (Abcam USA) were added to each slide, and they were stored at 4 °C overnight. For the negative control, PBS or antibody diluent was added instead.
After overnight incubation, bring to room temperature, discard the antibody fluid, rinse with PBS, add 50ā100 μL of HRP-marked goat anti-rabbit IgG (Servicebio China), incubate at room temperature for 30ā60 min. Following this, 50-100μL DAB working solution (Servicebio China)was applied to each slide. Distilled water was used to terminate color reaction, followed by counterstaining with hematoxylin. The slides were rinsed with tap water to restore the blue color, subjected to gradient dehydration (80%, 95%, 100%), sealed with neutral gum, and observed under a dry lens.
4.8 Sirius red staining
The slices were put into turpentine, gradient ethanol, distilled water, and PBS for dewaxing. Slices were stained with Sirius red dye for 8min, dehydrated with anhydrous ethanol; After that, the slices were placed in a clean xylene solution for 5min, and sealed with neutral gum. Microscope examination, and image acquisition were performed for analysis.
Results: Collagen fiber appear red under an optical microscope, with a yellow background. In polarized light, type I collagen appears as orange or bright red coarse fiber, and type III collagen appears as green fine fiber.
4.9 Acquisition and culture of PDLSCs
(1) Enzyme preparation:
60mg of type I collagenase (Abiowell China)was placed in a 15mL centrifugal tube; 80mg of dispase (SIGMA-ALDRICH USA)was placed in a 15ml centrifugal tube, dissolved in 10 mL PBS with 1% P/S, and filtered, stored at 4°C.
(2) PDLSCs extraction:
After extraction(it is best to harvest cells as soon as possible, within 4 hours extraction, need to be transported with an ice box), add the free teeth directly into the 15ml centrifugal tube containing PBS with 1% P/S, and stored at 4°C.
Place the extracted teeth upside down on the dish (Root up) in the clean bench, rinse the root from top to bottom with PBS containing 1% penicillin/streptomycin to remove blood and other substances until rinsed clean, put teeth into another dish;
Use a sterile scalpel to scrape the periodontal ligament from the middle third of the root, rinse 3 times with 1% penicillin/streptomycin-PBS, transfer it to another dish, and cut it into pieces;
Add 1ml each of type I collagenase and dispase, seal the tube, and digest 45-60min in 37° incubator, shaking every 10ā15 minutes, Until there is no distinct tissue blocks or flakes, transfer the mixture into a 15ml centrifugal tube, centrifuge at 1100rpm, 7min.
Remove the upper layer, add 5mL conditioned medium, which consisted of αMEM (Gibco USA) containing 10% FBS (Gibco USA) and 1% penicillin/streptomycin (Gibco USA), transfer the mixture to a T25 flask, and change the medium every 3ā4 days.
The attached cells can be seen under the microscope and should reach 70ā80% confluency before passaging.
(3) Cultivation of PDLSC: change the medium every 3ā4 days with conditioned medium.
4.10 LTA intervention
Enterococcus hirae LTA (Sigma-Aldrich, USA), is dissolved in a phosphate buffer to a concentration of 5 mg/mL as the stock solution and sterilized using a 0.22μm filter. Before application, the LTA solution is diluted with a conditioned medium to the desired concentration. All solutions are prepared at room temperature and pH is adjusted to 7.4.
Conditioned Medium Configuration: The conditioned medium consists of 10% bovine serum, 1% penicillin/streptomycin in α-MEM.10 and 20 μg/ml Enterococcus LTA is applied to hPDLSCs cells in a pore plate and induces osteogenesis for 7 days, with the medium changed every 3 days.
4.11 ALP
The cells were plated in six-well plates and treated with 0 μg/mL, 10 μg/mL and 20 μg/mL of Enterococcus LTA osteogenic induction medium, which contains αMEM, 10% FBS, 100 μg/mL ascorbic acid, 10 mM β-glycerophosphate, dexamethasone, and 1% penicillin/streptomycin. The medium containing LTA was replaced every 3 days. After 7 days of intervention, the fixation solution was prepared in a 9:1 ratio of methanol to formaldehyde. The 0.2g of pentobarbital sodium, 0.2g of calcium chloride, 0.3g of β-glycerol, and 0.4g of magnesium sulfate were dissolved in 10mL of distilled water. The solution was then incubated and stored at 4°C. After washing cell-cultured plates with PBS, the fixation solution was added for 30 seconds, followed by a quick rinse with PBS. Fresh incubation fluid a then added, and the plates were incubated at 37°C for 4 hours. The fluids were removed, and the plates were rinsed before soaking in a 2% cobalt nitrate solution for 5 minutes. After rinsing, 1.25% ammonium sulfide was added for 10 seconds, followed by another rinse. Finally, the cells were stained with 2% sand yellow dye for 10 seconds, rinsed, and dried. Microscopically, a positive reaction was indicated by a gray-black granular or flaky precipitate located in the cytoplasm.
4.12 Total RNA sequencing
Sample collection: For the Enterococcus LTA 10 μg/mL intervention and osteogenesis after 7 days of hPDLSCs treatment, add 1 mL of TRIzol per 10 cm² of cells. Then move the frozen storage tube into -80 °C standby.
RNA extraction and library construction: Total RNA was extracted using TRIzol reagents according to the specification. The purity of the RNA was identified using the NanoDrop 2000 Spectrophotometer (Thermo Scientific, USA), and RNA integrity was assessed using the Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA). The VAHTS Universal V5 RNA-seq Library Prep kit was used to construct a transcriptome library based on the instructions. Transcriptome sequencing and analysis are carried out in Shanghai, China.
RNA Sequencing and Differential Expression Gene Analysis: the library was sequenced using the Illumina NovaSeq 6000 sequencing platform generating 150bp paired-end reads. Fastp software were used to process the raw reads in FASTQ format, removing low-quality reads to obtain clean reads for subsequent data analysis. HISAT2 software was employed to align the reads to the reference genomes, calculate gene expression levels (FPKM), and obtain gene counts using HTSeq-count. Principal Component Analysis (PCA) was conducted using R (v 3.2.0) to assess biological duplication of the samples. Differential expression gene analysis was performed using DESeq2 software, with genes defined as differentially expressed (DEGs) if they met the thresholds of q<0.05 and fold change > 2 or fold change<0.5. Hierarchical clustering of DEGs was carried out using R (v 3.2.0) to demonstrate gene expression patterns across different groups and samples. An R package was used to visualize the top 30 genes, showing changes in the expression of upregulated and downregulated genes. Subsequently, Gene Ontology (GO), KEGG Pathway enrichment analyses were performed based on hypergeometric distribution algorithms to filter significant enrichment entries. R (v 3.2.0) was used to create column diagrams, or enrichment analysis circles for significant enrichment feature entries.
4.13 qPCR
The Quick RNA Extraction Kit (Accurate Bio-Medical Hunan, China) is utilized to extract mRNA from the PDLSCs. Next, 1 μg mRNA of satisfactory quality is selected for reverse transcription to obtain cDNA (reverse transcription kit Cat# AG11728 Accurate Bio-Medical Hunan, China), which is subsequently subjected to qPCR (SYBR green qPCR kit Cat#AG11701Accurate Bio-Medical Hunan, China). RocheLightCyclerĀ®480, The PCR cycle conditions are 95 °C, 15min, followed by 40 two-step cycles at 94 °C, 10s and 60°C, 45s, using a comparison cycle threshold (Cycle threshold, Ct). The primer sequences are provided in TableĀ 4. The relative mRNA abundance of the target genes was normalized to β-actin and was then calculated using the 2-ĪĪCt method.
TableĀ 4
| Target | Direction | Primers (5ā²ā3ā²) |
|---|---|---|
| SAA1 | Forward | CAGACAAATACTTCCATGCTCG |
| Reverse | TCTCTGGATATTCCTCTGGCA | |
| TGFBI | Forward | ACTCAGCCAAGACACTATTTGA |
| Reverse | CTTGTATGGGCATCAATTGGAG | |
| PLAU | Forward | AATTTCAGTGTGGCCAAAAGAC |
| Reverse | GTCCTCCTTCTTTGGGTAATCA | |
| FRAS1 | Forward | CCAAGTATACGAGCATGGTGAG |
| Reverse | TACCGCACAACTCCATCATAG | |
| COL15A1 | Forward | GATCATCCTCTACTACACGGAG |
| Reverse | CTTCCTCACCCTGGACAATCAT | |
| DKK1 | Forward | TACCAGACCATTGACAACTACC |
| Reverse | TCCATTTTTGCAGTATTCCCG | |
| NKD1 | Forward | ACCATTGCGTAGATGAGAACAT |
| Reverse | AAATTGGGACGTGTAGTTTT | |
| CXCL12 | Forward | CCAACGTCAAGCATTCAAAAT |
| Reverse | CACACTTGTCTGTTGTTGTTCT | |
| COL1A1 | Forward | AGGCGAACAGGGCGACAGAG |
| Reverse | GGACCAACAGGACCAGCATCAC | |
| ALPL | Forward | ACCGCACGGAACTCCTGACC |
| Reverse | TGCCTGCCTGCCCGATGG | |
| BGLAP | Forward | GCCAGGCAGGTGCGAAGC |
| Reverse | GCCGATGTGGTCAGCCAACTC | |
| RUNX2 | Forward | AGCAGCAGCAGCAGCAACAG |
| Reverse | GCAGCACCGAGCACAGGAAG |
Nucleotide sequences of primers used for qPCR.
The HiPure Bacterial DNA Kit (Magen, China) was utilized to extract DNA from human saliva. 100 ng of DNA was subsequently subjected to qPCR (2ĆHS Taq Universal SYBR Green qPCR Master Mix, Richamp, Hunan, China). The PCR was performed on a CFX Opus 96 Real-Time PCR System (Bio-Rad, USA) under the following cycling conditions: 95°C for 30 s, followed by 40 three-step cycles at 95°C for 10 s, 50°C for 10 s, and 72 °C for 30 s, using the comparative cycle threshold (Ct) method. SYBR Green qPCR quantified salivary Enterococcus using 23S rRNA primers ECST748F (5ā²ā3ā²: GAGAAATTCCAAACGAACTTG) and ENC854R (5ā²ā3ā²: CAGTGCTCTACCTCCATCATT); total bacterial load for normalization was measured using universal 16S rRNA primers EUB338 (5ā²ā3ā²: ACTCCTACGGGAGGCAGCAG) and EUB518 (5ā²ā3ā²: ATTACCGCGGCTGCTGG), and relative abundance was calculated using the 2^āĪĪCt method.
4.14 Transfection
Cell transfection: The 4th to 6th generations of hPDLSCs were inoculated in 12-well plates and divided into 3 groups: positive control group, negative control group, and knockdown group. Transfection was performed when the cell confluence reached 60-80%. siRNA was purchased from Guangzhou Ribobio Co., Ltd. (Guangzhou, China). According to the instructions, Lipofectamine⢠2000 was used to transfect the three groups of hPDLSCs.
4.15 Protein interaction
Protein PDB files for SAA1 and COL1A1 were downloaded from the Uniprot database (P05366Ā·SAA1_MOUSE; P11087Ā·COL1A1_MOUSE). Predictions of potential interaction interfaces between SAA1 and COL1A1 were obtained from the GRAMM website. Then the data of binding energy, hydrogen bond and salt bridge were predicted by PDBePISA website. Finally, PyMol tool (https://pymol.org) was used to visualize the prediction results.
4.16 Statistical analysis
Data are the mean±standard deviation. Statistical analyses were carried out using one-way analysis of variance. A comparison of two groups was assessed by Tukeyās post-hoc comparisons. Some samples are two groups of data, and unpaired t-test is used. P <0.05 was considered significant.
For clinical research, based on exclusion criteria, prior literature and a priori causal framework, we considered the following potential confounders: age, sex and BMI. Continuous covariates were mean-centered; categorical covariates were represented with indicator variables. We used linear regression (separate models for each skeletal site). We report β coefficients as mean differences in Z-score for olanzapine users versus non-users with 95% confidence intervals. Two-sided P values<0.05 were considered statistically significant. Analyses were performed in R (version 4.4.0).
Statements
Data availability statement
The data presented in the study are deposited in the NCBI Sequence Read Archive (SRA) under accession numbers PRJNA1370146 and PRJNA1370464, and in the Gene Expression Omnibus (GEO) under accession number GSE311844.
Ethics statement
The studies involving humans were approved by the Human Research and Ethics Committee of the Second Xiangya Hospital of Central South University. The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study. The animal study was approved by the Experimental Animal Ethics Committee of the Second Xiangya Hospital. The study was conducted in accordance with the local legislation and institutional requirements.
Author contributions
HY: Writing ā original draft, Methodology, Visualization, Investigation. MW: Writing ā original draft, Visualization, Investigation, Methodology. RL: Writing ā review & editing, Methodology, Visualization. SS: Writing ā review & editing. QL: Writing ā review & editing. LT: Writing ā review & editing. YZ: Writing ā review & editing. MZ: Writing ā review & editing. QZ: Writing ā review & editing. XS: Writing ā review & editing. NC: Writing ā review & editing. MW: Writing ā review & editing. YY: Writing ā review & editing. JL: Writing ā review & editing. HZ: Writing ā review & editing. JC: Writing ā review & editing, Supervision. YF: Writing ā review & editing, Supervision. FZ: Supervision, Conceptualization, Writing ā review & editing. YG: Methodology, Conceptualization, Supervision, Writing ā review & editing.
Funding
The author(s) declared that financial support was received for this work and/or its publication. This study was supported by the National Natural Science Foundation of China (81800788 and 81773339), Department of Science and Technology Hunan Province, China (2017WK2041 and 2018SK52511), Scientific Research Project of Hunan Provincial Health Commission (202208043514), Health research project of Hunan Provincial Health Commission of China (W20243054), Natural Science Foundation of Changsha City (kq2202403 and kq2202412), Fund for the Xiangya Clinical Medicine Database of Central South University (2014-ZDYZ-1-16), Education and Teaching Reform Research Project of Central South University (2020jy165-3), Research Project on Postgraduate Education and Teaching Reform of Central South University(2021JGB072), and the Fundamental Research Funds for the Central Universities of Central South University (2024ZZTS0521, 2024ZZTS0277, 2024ZZTS0891).
Acknowledgments
We thank the subjects for their participation and support.
Conflict of interest
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Generative AI statement
The author(s) declare that no Generative AI was used in the creation of this manuscript.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Publisherās note
All claims expressed in this article are solely those of the authorsĀ and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
Supplementary material
The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fcimb.2025.1673931/full#supplementary-material
References
1
AlbaughV. L.HenryC. R.BelloN. T.HajnalA.LynchS. L.HalleB.et al. (2006). Hormonal and metabolic effects of olanzapine and clozapine related to body weight in rodents. Obes. (Silver Spring)14, 36ā51. doi:Ā 10.1038/oby.2006.6
2
BalishE.WarnerT. (2002). Enterococcus faecalis induces inflammatory bowel disease in interleukin-10 knockout mice. Am. J. Pathol.160, 2253ā2257. doi:Ā 10.1016/S0002-9440(10)61172-8
3
BusnelliM.ManziniS.JablaouiA.BruneauA.KriaaA.PhilippeC.et al. (2020). Fat-shaped microbiota affects lipid metabolism, liver steatosis and intestinal homeostasis in mice fed a low-protein diet. Mol. Nutr. Food Res.64, 1900835. doi:Ā 10.1002/mnfr.201900835
4
ChaitY. A.MottaweaW.TompkinsT. A.HammamiR. (2020). Unravelling the antimicrobial action of antidepressants on gut commensal microbes. Sci. Rep.10, 17878. doi:Ā 10.1038/s41598-020-74934-9
5
CharlsonF. J.FerrariA. J.SantomauroD. F.DiminicS.StockingsE.ScottJ. G.et al. (2018). Global epidemiology and burden of schizophrenia: findings from the Global Burden of Disease Study 2016. Schizophr. Bull.44, 1195ā1203. doi:Ā 10.1093/schbul/sby058
6
ChenH.CaoT.ZhangB.CaiH. (2023). The regulatory effects of second-generation antipsychotics on lipid metabolism: potential mechanisms mediated by the gut microbiota and therapeutic implications. Front. Pharmacol.14, 1097284. doi:Ā 10.3389/fphar.2023.1097284
7
DaiX.MaR.JiangW.DengZ.ChenL.LiangY.et al. (2022). Enterococcus faecalis-induced macrophage necroptosis promotes refractory apical periodontitis. Microbiol. Spectr.10, e0104522. doi:Ā 10.1128/spectrum.01045-22
8
DaveyK. J.CotterP. D.OāSullivanO.CrispieF.DinanT. G.CryanJ. F.et al. (2013). Antipsychotics and the gut microbiome: olanzapine-induced metabolic dysfunction is attenuated by antibiotic administration in the rat. Transl. Psychiatry. 3, e309. doi:Ā 10.1038/tp.2013.83
9
DaveyK. J.O'MahonyS. M.SchellekensH.O'SullivanO.BienenstockJ.CotterP. D.et al. (2012a). Gender-dependent consequences of chronic olanzapine in the rat: effects on body weight, inflammatory, metabolic and microbiota parameters. Psychopharmacol. (Berl)221, 155ā169. doi:Ā 10.1007/s00213-011-2555-2
10
DaveyK. J.OāMahonyS. M.SchellekensH.OāSullivanO.BienenstockJ.CotterP. D.et al. (2012b). Gender-dependent consequences of chronic olanzapine in the rat: effects on body weight, inflammatory, metabolic and microbiota parameters. Psychopharmacology. 221, 155ā169. doi:Ā 10.1007/s00213-011-2555-2
11
De HertM.DetrauxJ.StubbsB. (2016). Relationship between antipsychotic medication, serum prolactin levels and osteoporosis/osteoporotic fractures in patients with schizophrenia: a critical literature review. Expert Opin. Drug Saf.15, 809ā823. doi:Ā 10.1517/14740338.2016.1167873
12
DuffulerP.BhullarK. S.WuJ. P. (2024). Targeting gut microbiota in osteoporosis: impact of the microbial based functional food ingredients. Food Sci. Hum. Wellness. 13, 15ā28. doi:Ā 10.26599/FSHW.2022.9250001
13
GuoY.TangC. Y.ManX. F.TangH. N.TangJ.WangF.et al. (2016). Insulin receptor substrate-1 time-dependently regulates bone formation by controlling collagen Iα2 expression via miR-342. FASEB J.30, 4214ā4226. doi:Ā 10.1096/fj.201600445RR
14
GuoY.TangC. Y.ManX. F.TangH. N.TangJ.ZhouC. L.et al. (2017). Insulin-like growth factor-1 promotes osteogenic differentiation and collagen I alpha 2 synthesis via induction of mRNA-binding protein LARP6 expression. Dev. Growth Differ.59, 94ā103. doi:Ā 10.1111/dgd.12342
15
HanS.JinS. P.OhJ. H.SeoE. Y.ParkC. H.YoonH. S.et al. (2016). Serum amyloid A1 secreted from UV-irradiated keratinocytes induces matrix metalloproteinase-1 in fibroblasts through toll-like receptor 4. Exp. Dermatol.25, 526ā531. doi:Ā 10.1111/exd.12979
16
HendrickxA. P. A.WillemsR. J. L.BontenM. J. M.van SchaikW. (2009). LPxTG surface proteins of enterococci. Trends Microbiol.17, 423ā430. doi:Ā 10.1016/j.tim.2009.06.004
17
HuangX. R.ZhangY.HuangJ. S.GaoW. L.YongfangX.ZengC. S.et al. (2023). The effect of FMT and vitamin C on immunity-related genes in antibiotic-induced dysbiosis in mice. PeerJ. 11, e15356. doi:Ā 10.7717/peerj.15356
18
JauharS.JohnstoneM.McKennaP. J. (2022). Schizophrenia. Lancet399, 473ā486. doi:Ā 10.1016/S0140-6736(21)01730-X
19
JettB. D.HuyckeM. M.GilmoreM. S. (1994). Virulence of enterococci. Clin. Microbiol. Rev.7, 462ā478. doi:Ā 10.1128/CMR.7.4.462
20
KawaiK.KamochiR.OikiS.MurataK.HashimotoW. (2018). Probiotics in human gut microbiota can degrade host glycosaminoglycans. Sci. Rep.8, 10674. doi:Ā 10.1038/s41598-018-28886-w
21
KayaogluG.OrstavikD. (2004). Virulence factors of Enterococcus faecalis: relationship to endodontic disease. Crit. Rev. Oral. Biol. Med.15, 308ā320. doi:Ā 10.1177/154411130401500506
22
LanglaisA. L.MountainR. V.KunstR. F.BarlowD.HouseknechtK. L.MotylK. J. (2023). Thermoneutral housing does not rescue olanzapine-induced trabecular bone loss in C57BL/6J female mice. Biochimie. 210, 50ā60. doi:Ā 10.1016/j.biochi.2023.05.002
23
LiJ.GuoY.ChenY. Y.LiuQ.ChenY.TanL.et al. (2020). miR-124-3p increases in high glucose induced osteocyte-derived exosomes and regulates galectin-3 expression: a possible mechanism in bone remodeling alteration in diabetic periodontitis. FASEB J.34, 14234ā14249. doi:Ā 10.1096/fj.202000970RR
24
LiX.NagyT. R. (2013). Atypical antipsychotic drugs inhibit trabecular bone accrual in C57BL/6J mice. Int. J. Body Compos. Res.11, 21ā24.
25
LinY. K.ZhangF.LeiW. J.GanX. W.LiM. D.PanF.et al. (2023). Amnion-derived serum amyloid A1 participates in sterile inflammation of fetal membranes at parturition. Inflammation Res.72, 797ā812. doi:Ā 10.1007/s00011-023-01713-3
26
LiuY. J.ZhuQ.GuoG. J.XieZ. P.LiS. L.LaiC. Y.et al. (2024). Causal associations of genetically predicted gut microbiota and blood metabolites with inflammatory states and risk of infections: a Mendelian randomization analysis. Front. Microbiol.15, 1342653. doi:Ā 10.3389/fmicb.2024.1342653
27
LuL. Y.TangM. J.LiJ.XieY.LiY. J.XieJ. W.et al. (2021). Gut microbiota and serum metabolic signatures of high-fat-induced bone loss in mice. Front. Cell Infect. Microbiol.11, 788576. doi:Ā 10.3389/fcimb.2021.788576
28
MaagensenH.LarsenJ. R.JĆørgensenN. R.Fink-JensenA.VilsbĆøllT. (2021). Liraglutide does not change bone turnover in clozapine- and olanzapine-treated schizophrenia overweight patients with prediabetes: randomized controlled trial. Psychiatry Res.296, 113670. doi:Ā 10.1016/j.psychres.2020.113670
29
MorganA. P.CrowleyJ. J.NonnemanR. J.QuackenbushC. R.MillerC. N.RyanA. K.et al. (2014). The antipsychotic olanzapine interacts with the gut microbiome to cause weight gain in mouse. PloS One. 9, e115225. doi:Ā 10.1371/journal.pone.0115225
30
NaidooU.GoffD. C.KlibanskiA. (2003). Hyperprolactinemia and bone mineral density: the potential impact of antipsychotic agents. Psychoneuroendocrinology28, 97ā108. doi:Ā 10.1016/S0306-4530(02)00129-4
31
NehmeH.SaulnierP.RamadanA. A.CassisaV.GuilletC.EveillardM.et al. (2018). Antibacterial activity of antipsychotic agents, their association with lipid nanocapsules and its impact on the properties of the nanocarriers and on antibacterial activity. PloS One13, e0189950. doi:Ā 10.1371/journal.pone.0189950
32
PargaA.MattuJ.BelibasakisG. N.KlineK. A.LeprinceJ. G.ManoilD. (2025). A polymicrobial perspective into the ecological role of Enterococcus faecalis in dental root canal infections. NPJ Biofilms Microbiomes. 11, 83. doi:Ā 10.1038/s41522-025-00722-w
33
ParkS. Y.HwangB. O.LimM.OkS. H.LeeS. K.ChunK. S.et al. (2021). Oral-gut microbiome axis in gastrointestinal disease and cancer. Cancers. 13, 2124. doi:Ā 10.3390/cancers13092124
34
Pelka WysieckaJ.KaczmarczykM.Baba KubisA.LiskiewiczP.WronskiM.Skonieczna ZydeckaK.et al. (2019). Analysis of gut microbiota and their metabolic potential in patients with schizophrenia treated with olanzapine: results from a six-week observational prospective cohort study. J. Clin. Med.8, 1605. doi:Ā 10.3390/jcm8101605
35
PetersonJ.GargesS.GiovanniM.McInnesP.WangL.SchlossJ. A.et al. (2009). The NIH human microbiome project. Genome Res.19, 2317ā2323. doi:Ā 10.1101/gr.096651.109
36
RautavaJ.PinnellL. J.VongL.AkseerN.AssaA.ShermanP. M.et al. (2015). Oral microbiome composition changes in mouse models of colitis. J. Gastroenterol. Hepatol.30, 521ā527. doi:Ā 10.1111/jgh.12713
37
SabeM.PillingerT.KaiserS.ChenC.TaipaleH.TanskanenA.et al. (2022). Half a century of research on antipsychotics and schizophrenia: a scientometric study of hotspots, nodes, bursts and trends. Neurosci. Biobehav. Rev.136, 104608. doi:Ā 10.1016/j.neubiorev.2022.104608
38
SaidH. S.SudaW.NakagomeS.ChinenH.OshimaK.KimS.et al. (2014). Dysbiosis of salivary microbiota in inflammatory bowel disease and its association with oral immunological biomarkers. DNA Res.21, 15ā25. doi:Ā 10.1093/dnares/dst037
39
ShafferM.LozuponeC. (2018). Prevalence and source of fecal and oral bacteria on infant, child and adult hands. mSystems. 3, e00192. doi:Ā 10.1128/msystems.00192-17
40
ShoganB. D.BelogortsevaN.LuongP. M.ZaborinA.LaxS.BethelC.et al (2015). Collagen degradation and MMP9 activation by Enterococcus faecalis contribute to intestinal anastomotic leak. Sci. Transl. Med.7, 286ra68. doi:Ā 10.1126/scitranslmed.3010658
41
SteckN.HoffmannM.SavaI. G.KimS. C.HahneH.TonkonogyS. L.et al. (2011). Enterococcus faecalis metalloprotease compromises epithelial barrier and contributes to intestinal inflammation. Gastroenterology. 141, 959ā971. doi:Ā 10.1053/j.gastro.2011.05.035
42
WangS.DengZ.YeX.GengX.ZhangC. (2016). Enterococcus faecalis attenuates osteogenesis through activation of p38 and ERK1/2 pathways in MC3T3-E1 cells. Int. Endod. J.49, 1152ā1164. doi:Ā 10.1111/iej.12579
43
WangW. S.LiW. J.WangY. W.WangL. Y.MiY. B.LuJ. W.et al. (2019). Involvement of serum amyloid A1 in the rupture of fetal membranes through induction of collagen I degradation. Clin. Sci.133, 515ā530. doi:Ā 10.1042/CS20180950
44
WangY. W.WangW. S.WangL. Y.BaoY. R.LuJ. W.LuY.et al. (2019). Extracellular matrix remodeling effects of serum amyloid A1 in the human amnion: implications for fetal membrane rupture. Am. J. Reprod. Immunol.81, e13073. doi:Ā 10.1111/aji.13073
45
Wyszogrodzka-KucharskaA.Rabe-JablonskaJ. (2005). Decrease in mineral bone density in schizophrenic patients treated with 2nd generation antipsychotics. Psychiatr. Pol.39, 1173ā1184.
46
XuJ. Y.YuL.YeS. R.YeZ. T.YangL. Y.XuX. X. (2024). Oral microbiota-host interaction: the chief culprit of alveolar bone resorption. Front. Immunol.15, 1254516. doi:Ā 10.3389/fimmu.2024.1254516
47
ZehnderM.GuggenheimB. (2009). The mysterious appearance of enterococci in filled root canals. Int. Endod. J.42, 277ā287. doi:Ā 10.1111/j.1365-2591.2008.01537.x
48
ZhaoY. Q.YeQ.FengY.ChenY.TanL.OuyangZ. Y.et al. (2022). Prevotella genus and its related NOD-like receptor signaling pathway in young males with stage III periodontitis. Front. Microbiol.13, 1049525. doi:Ā 10.3389/fmicb.2022.1049525
Summary
Keywords
olanzapine, oral Enterococcus, bone formation, serum amyloid A1 gene, extracellular matrix
Citation
Yuan H, Wang M-Y, Liu R-X, Sreekissoon S, Liu Q, Tan L, Zhao Y-Q, Zhong M-M, Zhang Q, Su X-L, Chen N-X, Wang M, Yang Y-F, Li J-N, Zheng H-Q, Chen J-D, Feng Y-Z, Zhang F-Y and Guo Y (2026) Olanzapine affects bone formation via oral Enterococcus through SAA1 gene and extracellular matrix-related pathways. Front. Cell. Infect. Microbiol. 15:1673931. doi: 10.3389/fcimb.2025.1673931
Received
26 July 2025
Revised
07 November 2025
Accepted
19 November 2025
Published
06 January 2026
Volume
15 - 2025
Edited by
Jochen Mattner, University of Erlangen Nuremberg, Germany
Reviewed by
Gauddy Lizeth Manzanares-Leal, Universidad Autónoma del Estado de México, Mexico
Sivakumar Lingappa, University of Iowa, United States
Updates
Copyright
Ā© 2026 Yuan, Wang, Liu, Sreekissoon, Liu, Tan, Zhao, Zhong, Zhang, Su, Chen, Wang, Yang, Li, Zheng, Chen, Feng, Zhang and Guo.
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Yue Guo, guoyue@csu.edu.cn; Feng-Yi Zhang, kqzhangfengyi@csu.edu.cn
ā These authors have contributed equally to this work and share first authorship
Disclaimer
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.