An altered oral microbiota induced by injections of superparamagnetic iron oxide nanoparticle‐labeled periodontal ligament stem cells helps periodontal bone regeneration in rats

Abstract Stem cell injection is good for periodontal regeneration due to the capacity of stem cells to differentiate toward osteogenic direction and to regulate the production of pro‐ and anti‐inflammatory cytokines. However, injected cells are difficult to track in vivo. And there is microbiota in oral cavity, the dysbiosis of which leads to the damage and loss of periodontal tissue. Here, we demonstrated an enhanced periodontal repair was due to an altered oral microbiota. Periodontal defects were surgically prepared in rats, and periodontal ligament stem cells (PDLSCs) labeled by superparamagnetic iron oxide (SPIO) nanoparticles (PC‐SPIO) were injected, with PDLSCs and saline treatments as controls. Detected by magnetic resonance imaging (MRI) and histological staining, PC‐SPIO was major at limited areas in regenerated periodontal tissues. PC‐SPIO‐treated rats achieved better periodontal regeneration than the other two groups. Concurrently, the oral microbiota of PC‐SPIO‐treated rats was changed, presenting SPIO‐Lac as a biomarker. SPIO‐Lac assisted periodontal repair in vivo, inhibited the inflammation of macrophages induced by lipopolysaccharide (LPS) and antibacterial in vitro. Therefore, our study proved that SPIO‐labeled cells can be tracked in periodontal defect and highlighted a potential positive role of an oral microbiota in periodontal regeneration, suggesting the possibility of periodontal repair promotion by manipulating oral microbiota.

PDLSCs possess low immunogenicity and can be obtained noninvasively from periodontal tissue during standard dental scaling and root planning (SRP) procedures. They present a similar phenotypic profile to BMMSCs, indicating a strong osteogenic potential. In addition, they are fast growing, providing the possibility of ex vivo expansion. 10 Moreover, signals from the transplanted PDLSCs might alter the immune microenvironment to enhance periodontal regeneration. 11 When transplanted into periodontal defects, PDLSCs have the potential to form bone, cementum, and periodontal ligament-like structures, while BMMSCs mainly produce bone and bone-marrowlike structures. 12 Compared with other stem cell delivery strategies used to treat periodontal bone defects, such as cell sheets and cell-seeded scaffolds, cell injection has its own advantages. Although not all the injected cells can reach the injured site, injection can avoid the trauma caused by surgery. 8 Moreover, this minimally invasive technique can be repeated until periodontal healing reaches the required condition. 13 Stem cell injection is good for periodontal regeneration. 4,6,8 Underlying mechanisms have been explored from their capacities to differentiate toward the osteogenic direction and to regulate the production of pro-and anti-inflammatory cytokines. 4,6,8 However, periodontal defects heal in the oral cavity, which is colonized by billions of bacteria, fungi, and viruses, known as the oral microbiota. Homeostasis of the oral microbiota plays a crucial role in maintaining the well-being and healthy status of the human host. 14  the microbiota caused by certain factors, such as dental plaque accumulation, which is a sticky film composed mainly of bacteria, can lead to the initiation and development of periodontitis. Correspondingly, dysbiosis, which is the shift of species and functions associated with diseases, is also observed in periodontitis. 15 These changes to the microbiota mainly comprise increases in pathogenic bacteria, which amplify the inflammatory response by activating the host immune response, beginning a cycle of microbiota changes and enhanced inflammation, leading eventually to loss of tissue integrity, attachment, and bone loss. 16,17 Although has been notorious for the damaging effect, it is still possible that a distinct oral microbiota plays a positive role for periodontal regeneration. We previously found that superparamagnetic iron oxide (SPIO)-coated scaffolds induced better palate-bone regeneration than the uncoated controls, and the effect was partly related to alteration of the oral microbiota caused by the antibacterial effects of SPIO. 18 And topical treatment with probiotics has protective effect against periodontitis. 19 These results suggested that certain alteration of oral microbiota may be beneficial to bone regeneration.
For cell function enhancements and cell labeling, SPIO nanoparticles have great potential to modify PDLSCs. SPIO nanoparticles containing an iron oxide core (7 nm in diameter) and a polyglucose sorbitol carboxymethyether (PSC) shell (20 nm in thickness) comprise the FDA-approved nanodrug ferumxytol. 20 They are biosafe and have been applied clinically as an iron supplement and a magnetic resonance imaging (MRI) T2 contrast agent. 21 Our group developed an SPIO nanoparticle, whose structure mimics ferumxytol. 20 Previously, we used gelatin sponge scaffolds labeled with this SPIO nanoparticle to fill mandible defects after the extraction of rat incisors and found a significant increase in bone regeneration and a decrease in signal intensity of T2-weighted MRI in the SPIO-labeled scaffold group. 22 During the healing period, changes in the image intensity of the scaffolds, which indicated scaffold degradation and interaction with host  26 and increase the level of interleukin-10 (IL-10) in liver macrophages, leading to inhibition of inflammation in lipopolysaccharide (LPS)-induced sepsis and liver injury. 27 In this study, SPIO-labeled PDLSCs (PC-SPIO) were repeatedly injected into inflammatory periodontal defect in rats, using injections of PDLSCs and saline as controls. The injected PC-SPIO was tracked by MRI in vivo and histological staining, and the periodontal regeneration by PC-SPIO was detected. Moreover, the oral microbiota of rats injected with PC-SPIO was changed, and the local application of the biomarker from changed oral microbiota could enhance periodontal regeneration. This study represents the first exploration of the positive role of an altered oral microbiota in periodontal regeneration.  In in vitro labeling test, no significant difference was detected between PC and 1 Â 10 5 /ml PC-SPIO, both of which appeared as bright white circles ( Figure 1f). Comparatively, the image of 1 Â 10 6 /ml PC-SPIO was significantly different from the above two and produced good contrast, presenting as a dark black circle ( Figure 1f). T2-mapping results were consistent with the images, showing that the T2 relaxation time of 1 Â 10 6 /ml PC-SPIO was significantly shorter than that of PC and 1 Â 10 5 /ml PC-SPIO (p < 0.01; Figure 1f). SPIO nanoparticles (marked by yellow arrows in Figure 1g) were detected in PC-SPIO by transmission electron microscope (TEM). The results of inductively coupled plasma optical emission spectrometer (ICP-OES) showed that the iron content of PC-SPIO was significantly higher than that of PC (p < 0.01; Figure 1h). Therefore, 1 Â 10 6 /ml PC-SPIO was used for the in vivo injection, using PC and Saline injection as controls.

| SPIO nanoparticles labeled PDLSCs can be tracked in vivo
The process of the animal experiments included one surgery, three LPS injections, and four injection treatments (Figure 2a). MRI was performed on rats to track the PC-SPIO in vivo (Figure 2b,c). The red boxes, which were used for analysis, represent the area of the periodontal defect including some adjacent buccal soft tissue. The similar position on the contralateral side was used as a control. In PC-SPIO group, the thickness from the molar to the epidermis increased from 1 to 8 days because of the inflammation and swelling from the surgery. It gradually became thinner from 10 days and almost returned to normal thickness at 14 days. The defect area was brighter than the control from 1 to 10 days and presented normal brightness after 10 days. The image was much brighter than the control after surgery and the first injection. Then, the images turned darker after the second injection but became brighter with time. However, the brightness did not change much after the third injection, possibly because of the obvious new bone regeneration. When setting the gray value ratio of the control as 1 for the quantitative assessment, it was found that the ratio was significantly greater than 1 from 1 to 10 days (p < 0.01) and was less than 1 from 14 days (p > 0.05; Figure 2b).
After the fourth injection, the image of PC-SPIO group was darker than that of the Saline group, PC group, and the control side at

| PC-SPIO injection promoted periodontal bone regeneration
The mandibles were collected and evaluated using micro-CT

| PC-SPIO injection induced a decrease in Bacteroidaceae and an increase in Lactobacillaceae
The microbiotas from the Saline, PC, and PC-SPIO groups were compared. PCA (Taxon) showed that their samples were all mixed in the PC1 and PC2 directions (Figure 7a). In addition, there was no significant difference detected by beta analysis (Weighted_ UniFrac) The isolation and identification of Lactobacillus spp. were performed ( Figure S10 and Table S1). The result showed that this SPIO-Lac had strong homology to L. reuteri strain (99.93%) (NCBI accession number CP041676.1). To confirm the involvement of SPIO-Lac, which is a PC-SPIO group biomarker in periodontal regeneration, the Spearman correlation between the abundance of Lactobacillaceae and the

| DISCUSSION
In the present study, PDLSCs labeled by SPIO (PC-SPIO) were applied as injections for periodontal regeneration. And we injected PC-SPIO once a week for 4 weeks. Stem cell injection is usually injected once and achieves acceptable promoting effects for periodontal regeneration. 4,6,8 Hu et al. found that a single injection of DPMSCs had lower efficiency than their sheet form. 6 It is easy to imagine that the number of injected cells will decrease immediately following a single dose injection, as confirmed by Prussian blue staining after first injection in this study (Figure 2d). Therefore, it suggests that periodic injections are necessary to maintain the level of injected cells over the long term to improve periodontal healing.
SPIO nanoparticles were used for labeling and functional modification of PDLSCs, rather than PKH26 or luciferase. 28 The abundance of Bacteroidaceae was decreased, while the abundance of Lactobacillaceae was increased in PC-SPIO group. We hypothesized that this altered microbiota was not only a consequence of the changed environment but also a driving factor for tissue regeneration. Disturbance of the oral microbiota-ecology balance in the host usually causes a series of oral infectious diseases, including periodontitis. Meanwhile, balanced oral pathogenic bacteria and probiotics can promote wound healing by maintaining MSC homeostasis. 33 On the one hand, a reduction in its abundance benefits periodontal regeneration. The typical pathogenic bacterium of Bacteroidaceae is Bacteroides forsythus, whose name has been changed to Tannerella forsythia, and belongs to the red complex of periodontal pathogens. 34 It is frequently associated with P. gingivalis colonization and is detected significantly more often in sites with bleeding on probing when compared with nonbleeding sites. 35 As a result, it is related to progressive periodontal infections, which present as loss of connective tissue and severe resorption of alveolar bone.
On the other hand, the high abundance of SPIO-Lac, which had oral probiotic properties, such as reducing oral pathogenic bacteria counts and inhibiting their adherence to hGFs ( Figure S11) should also be good for periodontal regeneration. SPIO-Lac was identified as L. reuteri, which is a well-known probiotic used to treat oral infections. L. reuteri plays a role in periodontal regeneration by producing a broad-spectrum antimicrobial substance, known as reuterin, 36 and protecting against periodontal bone loss through its anti-inflammatory and immunomodulatory effects. Oral treatment of patients with chronic periodontitis with tablets containing probiotic strain of L. reuteri induced a significant reduction in TNFα, IL-1β, and IL-17 levels. 37 The increase in Lactobacillus spp.
in the SPIO group might lead to the regulation of macrophage polarization and the Th17/Treg balance to promote repair. [38][39][40] Clinically, parameters including the sulcus bleeding index, the periodontal probing depth, and the clinical attachment level were improved. 41 The adjunctive use of L. reuteri along with SRP achieved significantly more pocket depth reduction and attachment gain, with more P. gingivalis reduction. 41 In addition, L. reuteri can prevent bone loss by decreasing the abundance of Bacteriodales. 36 Therefore, our study indicated the existence of a distinct oral microbiota in response to a specific periodontal treatment such as PC-

| Animal experimental design and injection therapy
Periodontal defects were surgically prepared following the method of King et al. 42 with minor modifications using specific-pathogenfree (SPF) male SD rats (180-200 g weight, Animal Center of Nanjing Medical University, n = 6). All animal procedures were reviewed and approved by the Institutional Animal Care and Use Committee of Nanjing Medical University (IACUC-1908036; Nanjing, China). Generally, the rats were maintained in a SPF facility under controlled temperature and dark cycle, with free access to water and normal rodent food. The defect was approximately 2 mm in width, 3.5 mm in length, and 1.5 mm in depth and was about 2 mm below the alveolar crest at the buccal side of the second molars in the left mandible. After surgery, 0.1 ml saline contained 1 Â 10 6 PC or 1 Â 10 6 PC-SPIO were injected into periodontal bone defects, using 0.1 ml saline as control. 43  the middle of the bone defect, and drawn a little back when significant resistance was encountered. Therefore, the tip of the needle stayed at the bottom of the bone defect beneath the periosteum. Then the cell mixture was injected. It was injected slowly for 3-5 s and the tip of the needle was kept for 5-10 s after injection. 43 The needle was withdrawn without compression. The injections were repeated once a week for four times.
Lactobacillus from SPIO was isolated and purified (term as SPIO-Lac). To confirm its repair promoting effect, it was applied in INF.
Saline group and INF.PC group.

| Collection of oral microbiotas and bioinformatic analysis
The oral microbiota of periodontal defect area was collected before and after surgery. After anesthesia, the aseptic sampling swab was

| Isolation of Lactobacillus spp. from the microbiota
Lactobacillus spp. from PC-SPIO group was isolated and purified using MRS medium (Oxoid, UK) in an anaerobic incubator (90% N 2 , 5% H 2 , and 5% CO 2 , 37 C). The round-shape, off-white, clear-edged, smoothsurfaced mono colony was picked and inoculated on a new MRS plate.
The purification was repeated until the obtained colonies were single and consistent.

| Statistical analysis
All data were analyzed using the Prism 7 program (GraphPad Software Inc., La Jolla, CA, USA). Independent unpaired two-tailed Student's t tests were used when comparing two groups. When more than two groups were compared, one-way analysis of variance (ANOVA) was applied. p values less than 0.05 were considered statistically significant.

| CONCLUSION
Although more conclusive research is needed, an altered oral microbiota induced by injections of PC-SPIO helps periodontal bone regeneration in rats due to its probiotic properties, including antiinflammation and anti-bacterial. Moreover, SPIO labeling can be used to track injected cells.

CONFLICT OF INTEREST
The authors have no conflicts of interest to declare.

DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request.