Long non-coding RNA 01126 activates IL-6/JAK2/STAT3 pathway via serving as a sponge of miR-655-3p in periodontitis

Periodontitis is a serious threat to oral quality of life and overall health. Although our previous studies conrmed that long intergenic non-coding RNA 01126 (LINC01126) is aberrantly expressed in periodontitis tissues, there are few reports on the pathogenesis of LINC01126 in periodontitis. Our study investigated the biological functions of LINC01126 in periodontitis and the potential mechanism. An inammatory model of human gingival broblasts (HGFs) was successfully established. LINC01126 silencing can alleviate lipopolysaccharide (LPS) induced cell inammation, reduce cell apoptosis, and promote cell migration. As a "sponge" for miR-655-3p, LINC01126 inhibits its binding to mRNA of IL-6, thereby promoting inammation progression and JAK2/STAT3 pathway activation. qRT- PCR, WB, and IHC results of clinical tissue samples further conrmed that miR-655-3p expression was down-regulated and IL-6/JAK/STAT3 was abnormally activated in periodontitis tissues. new


Background
Periodontitis is a chronic infectious disease with high incidence in the population, which seriously harms the oral health-related quality of life 1 . Porphyromonas gingivalis (Pg) and Actinobacillus actinomycetemcomitans (Aa) are the most dominant bacteria in the lesion or active site during periodontitis development 2 . Most periodontitis patients have a chronic, long-term disease process, accompanied by bleeding gums, loose teeth, weak bite, and other symptoms 3 . Clinically, basic periodontal treatment (including supragingival scaling and root planning) is the key measure for controlling dental plaque and delaying periodontitis progression 3,4 . However, there is currently no cure for periodontitis that results in complete physiological healing of periodontal tissues. At present, studies on periodontitis pathogenesis mainly focus on dental plaque 5 , immune regulation 6 , and genetic susceptibility 7 . Numerous studies have demonstrated a close association between general health and periodontitis, including heart diseases, diabetes, Alzheimer's disease, and adverse pregnancy [8][9][10][11][12] . As a result, elucidating the pathogenic mechanism of periodontitis is critical for preventing and treating periodontitis and systemic diseases.
Gingival tissue, an important part of periodontal tissue, is involved in the earliest periodontal in ammatory response and maintains an in ammatory state throughout the whole process of periodontitis occurrence and development. In ammatory gums are often associated with changes in color and texture, and the in ltration of in ammatory cells is signi cantly increased. Human gingival broblasts (HGFs) are the main cells in gingival tissue, and their physiological function determines the health condition of periodontal tissues 13 . Healthy HGFs can participate in connective tissue matrix synthesis, promote the healing of in ammatory periodontal tissues, and play a crucial role in the reconstruction, regeneration, and physiological function of periodontal support tissue [14][15][16] . Toll-like receptors (TLRs) on the surface of HGFs are activated in response to the invasion of periodontal pathogens and are involved in immune regulation 17 .
Long non-coding RNAs (lncRNAs), with a length of more than 200 ribonucleotides and lacking a complete open reading frame (ORF), have great potential for maintaining stem cell differentiation and pluripotency, and their involvement in various biological processes has been extensively reported, including epigenetic processes, transcriptional regulation, gene silencing, and chromatin modi cation 18,19 . MicroRNA, encoded by endogenous genes, often target to the mRNA sequence of the target gene or are sponged by lncRNAs, thereby regulating the target gene's expression and forming "lncRNAs-microRNAs-mRNAs" network 20 . By increasing researches on lncRNAs, some have been gradually introduced into oral diseases. Researches have demonstrated that lncRNAs regulate osteogenic differentiation, cell apoptosis, and alveolar bone metabolism of periodontal cells. Jia et al. found that lncRNA-ANCR down-regulation could promote the proliferation of periodontal ligament stem cells (PDLSCs) and osteogenic differentiation of PDLSCs by upregulation of osteogenic differentiation-related genes 21 . LncRNA PCAT1 acts as a molecular sponge during PDLSCs osteogenic induction, thereby regulating BMP2 expression 22 . In addition, cell apoptosis can be regulated by lncRNA-POIR/miR-182/FoxO1 regulatory network in in ammatory microenvironment 23 . These results all suggest that lncRNAs may regulate periodontitis progression by regulating bone metabolism, cell proliferation, and apoptosis of periodontal tissues.
Previous microarray analysis has revealed that long intergenic non-coding RNA 01126 (LINC01126, located on 2p21) is a discrepant expressed lncRNA between gingival tissues of periodontitis patients and healthy controls. Additionally, the results con rmed that LINC01126 promotes periodontitis progression through miR-518a-5p/HIF-1α/MAPK pathway 24 . However, the biological role of LINC01126 in the in ammatory process of gingival tissue remains unknown. Therefore, the biological roles of LINC01126 in human HGFs under in ammation microenvironment were studied. As expected, the results showed that IL-6 expression was closely related to LINC01126. Additional studies con rmed that LINC01126 could act as a miR-655-3p sponge, thus regulating IL-6/JAK2/STAT3 signaling pathway in periodontitis progression.

Results
Proin ammatory genes were up-regulated in in ammatory model in vitro, while LINC01126 silencing signi cantly alleviates LPS-induced cellular in ammation HGFs were cultured with in ammatory medium containing LPS with a concentration of 1 ng/mL. All proin ammatory genes were signi cantly up-regulated with statistical signi cance (Fig. 1A-D). IL-6 expression increased tenfold in LPS-stimulated HGFs than in the control group. IL-4 expression, an antiin ammatory cytokine, decreased after in ammatory stimulation (Fig. 1E). Meanwhile, it was con rmed that LINC01126 expression increased under in ammatory microenvironment (Fig. 1F). These results indicated a successful establishment of a cellular in ammatory model of HGFs.
Further, we screened the siRNA sequences of LINC01126. As depicted in Fig. 1G, si-LINC01126_002 had the best silencing effect and thus was used in our experiments. Then, the expression of proin ammatory genes following LINC01126 silencing was veri ed. The results indicated that expressions of IL-1β, IL-6 and MMP-1 were statistically decreased after silencing, particularly on the second day ( Fig. 1H-J). However, expression of TNF-α in HGFs with LPS stimulation was not affected by LINC01126 silencing (Fig. 1K). Among all proin ammatory genes, it was found that IL-6 expression decreased most signi cantly. However, LINC01126 silencing had no obvious effect on IL-4 in in ammatory microenvironment (Fig. 1L). These ndings imply that silencing LINC01126 can reduce the expression level of proin ammatory genes, particularly IL-6. LINC01126 silencing has no signi cant effect on cell proliferation, but promotes cell migration and reduces LPS-induced apoptosis To further elucidate the biological function of LINC01126 in HGFs, Cell viability assay manifested that LINC01126 silencing had no statistical signi cance on HGFs proliferation ( Fig. 2A). However, wound healing experiments and ow cytometry showed that silencing LINC01126 could increase the migration of HGFs and inhibited LPS-induced cell apoptosis ( Fig. 2B-D).
LINC01126 silencing inhibits the activation of IL-6/JAK2/STAT3 pathway by down-regulating IL-6 expression under in ammatory microenvironment Given that silencing LINC01126 can signi cantly decreased IL-6 expression under in ammatory stimulation, additional studies were conducted to determine whether LINC01126 can affect IL-6-related signaling pathway, called IL-6/JAK2/STAT3 pathway. RT-qPCR results manifested that expression of IL-6, JAK2, and STAT3 in HGFs under in ammatory stimulation were decreased with statistical signi cance after LINC01126 silencing ( Fig. 3A-C). Figure 3D demonstrated that LINC01126 was successfully silenced. In addition, WB further con rmed that protein expressions of IL-6, p-JAK2, and p-STAT3 were dramatically down-regulated in the silencing groups (Fig. 3E). These results indicated that silencing LINC01126 inhibited the activation of JAK2/STAT3 pathway by reducing IL-6 expression.
LINC01126 acts as a sponge for miR-655-3p It was predicted that miR-655-3p contains a binding site of LINC01126 (Fig. 4A). Therefore, LINC01126 was predicted to be a "sponge" for miR-655-3p. FISH manifested that LINC01126 was expressed in the cytoplasm and nucleus of HGFs (Fig. 4B). LINC01126 expression in cytoplasm provided additional support for our hypothesis. Dual-luciferin reporter assay was used to further con rm the direct binding of miR-655-3p to LINC01126. The data indicated that miR-655-3p mimics transfection decreased the relative luciferase activity of the LINC01126-WT groups, but had no signi cant difference in LINC01126-MUT group (Fig. 4C). In addition, as the duration of in ammatory culture increased, miR-655-3p expression was signi cantly down-regulated in HGFs (Fig. 4D). After LINC01126 silencing, miR-655-3p expression returned to the level of the control group (Fig. 4E). Meanwhile, LINC01126 expression decreased after miR-655-3p overexpression (Fig. 4F). The successful miR-655-3p mimics transfection was further con rmed (Fig. 4G). All these results indicate that miR-655-3p could be sponged by LINC01126.
MiR-655-3p targets to mRNA of IL-6 and inhibits IL-6/JAK2/STAT3 signaling pathway under in ammatory stimulation IL-6 expression was signi cantly down-regulated in HGFs following LINC01126 silencing. Consequently, it was hypothesized that miR-655-3p could target mRNA of IL-6 and thus regulate its expression.
MiR-655-3p is down-regulated in in ammatory gingival tissues, while IL-6/JAK2/STAT3 pathway is abnormally activated Clinical gingival tissue samples were collected (n = 12). As expected, miR-655-3p expression was markedly down-regulated in in ammatory periodontal tissues than that of normal periodontal tissues (Fig. 6A). LINC01126, IL-6, JAK2, and STAT3 expressions were statistically up-regulated ( Fig. 6B-E). LINC01126 expression was consistent with previous lncRNA microarray analysis. WB further con rmed that protein expressions in in ammatory periodontal tissues were markedly up-regulated (Fig. 6F). H&E staining revealed that in ammatory gingival tissues were in ltrated by a mass of in ammatory cells, which were rarely seen in healthy controls (Fig. 6G). IHC staining of gingival tissues also con rmed that expression levels of IL-6, T-STAT3, p-STAT, T-JAK2, and p-JAK2 in in ammatory gingival tissues were much higher than those in healthy tissues (Fig. 6H-L). Combined with previous results, it was manifested that miR-655-3p has anti-in ammatory effect in periodontitis, whereas IL-6/JAK2/STAT3 pathway is abnormally activated during periodontitis progression.

Discussion
In recent years, increasing studies have focused on long non-coding RNAs, which may be aberrantly expressed in many diseases and are involved in epigenetic and transcriptional regulation 25,26 . According to the 4th National Oral Health Survey of China, the periodontal health rate of Chinese adults is less than 10%, and the prevalence of periodontitis is signi cantly correlated with age groups 27 . Due to severe situation of periodontitis in the population, it is urgent to reveal pathogenic mechanisms underlying lncRNAs in periodontal diseases. Existing researches indicate that lncRNA-ANCR, PCAT1, and POIR may regulate periodontal cell proliferation, osteogenic differentiation, and bone metabolism [21][22][23] . In previous studies, clinical tissue samples were obtained for lncRNA microarray analysis, which identi ed a long non-coding RNA with the most signi cant expression difference between in ammatory periodontal tissue and healthy controls, namely LINC01126. It has demonstrated that LINC0116 is critical in promoting periodontitis pathogenesis under hypoxia 24 . However, whether LINC01126 regulates proin ammatory gene expression and its speci c mechanism has not been investigated.
Periodontitis aggravated gingival tissue disorder and cellular function impairment 28 . The biological function of HGFs is an important indicator of physical condition of gingival tissues, and their migration is critical for gingival tissue restoration under in ammation microenvironment 29,30 . CCK-8 assay and wound healing experiments revealed that LINC01126 silencing had no obvious effect on cell viability of HGFs. However, cell migration results indicated that LINC01126 expression inhibited the migration of HGFs in the in ammatory microenvironment, thus reducing its resistance to in ammation and tissue repair ability. In addition, numerous studies have demonstrated that lncRNAs, such as lncRNA ATB 31 , MALAT1 32 , and GAS5 33 , have regulatory effects on apoptosis, promoting apoptosis by activating apoptosis-related signaling pathways. Generally, the greater the number of apoptotic cells, the fewer cells can perform normal biological functions, disrupting the body's self-repair and immune defenses 34 . In terms of apoptosis detection, it was found that LINC01126 silencing could reduce LPS-induced apoptosis of HGFs, implying that lncRNAs play a role in apoptosis.
LncRNAs have been shown to regulate the occurrence and development of in ammatory diseases 35,36 . Maninjay K et al. found that LINC-EPS regulated in ammatory response at the transcriptional level, inhibiting in ammation progression 37 . LncRNA-Mirt2 inhibits the NF-κB and MAPK pathways activation and limits the production of pro-in ammatory cytokines through negative feedback 38 . In addition, lncRNAs can participate in in ammatory progression by regulating macrophage oxidative stress and metabolism 39,40 . Our study found that LINC01126 silencing signi cantly reduced expression of proin ammatory cytokines in LPS-induced HGFs, particularly IL-6. As a result, additional studies on IL-6 were conducted to elucidate the role of LINC01126 in in ammation. Through online database comparison, it was found that miR-655-3p contains binding sites for LINC01126 and IL-6, which is an important premise for our study on "endogenous competitive RNA". According to current reports, miR-655-3p has primarily focused on cancer-related elds, where miR-655-3p plays an anti-tumor role by targeting related genes 41,42 . To our knowledge, miR-655-3p has not been reported in mechanistic studies of in ammation-related diseases. Through dual luciferin report assay, it was con rmed that miR-655-3p had a targeted effect with LINC01126 and IL-6. Moreover, it was con rmed that miR-655-3p expression in in ammatory group was dramatically lower than that of controls and was negatively correlated with LINC01126. Additionally, these ndings indicate that there exists a mechanism of "competing endogenous RNAs" between LINC01126/miR-655-3p/IL-6, which is an important pathway for in ammation regulation in periodontitis.
IL-6 is mainly expressed in broblasts and activated T cells, and has received extensive attention, especially in the study of chronic in ammation, autoimmune disease, and tumors [43][44][45] . During periodontitis progression, HGFs are stimulated by bacteria-derived LPS, resulting in increased expression of various proin ammatory genes, particularly IL-6. Increased IL-6 leads to IL-6 receptor activation, which then activates IL-6-related cell signaling pathways, especially IL-6/JAK2/STAT3 pathway 46 . According to most current studies, aberrant activation of JAK2/STAT3 signaling pathway promotes disease progression and is detrimental to the control of diseases, including liver cancer 47 , in ammation 48 and neurological disease 49 . In our study, it was observed that the expressions of IL-6, p-JAK2, and p-STAT3 were abnormal up-regulated in HGFs after LPS induction, and the effect of LPS stimulation was partially reversed following transfection with miR-655-3p mimics. In addition, IL-6, p-JAK2, and p-STAT3 expression levels in periodontitis tissues were far higher than those in normal tissues through IHC and WB. These data indicate that IL-6/JAK2/STAT3 signaling pathway is abnormally activated in in ammatory periodontal tissues, thus exerting a pro-in ammatory effect and regulating various cell biological behaviors as previous reports 50 .
Our results revealed an important mechanism of LINC01126 in periodontitis regulation. As a "sponge" of miR-655-3p, LINC01126 obstructs the binding between miR-655-3p and IL-6, thus promoting the activation of IL-6/JAK2/STST3 pathway (Fig. 7). It is worth mentioning that the regulatory mechanism of LINC01126 in the nucleus of HGFs remains unclear. Recently, Chen et al. are the rst to identify that lncRNA-SLERT regulates protein conformation and polymer state through RNA chaperone mechanism, thus maintaining normal nucleolus morphology and function 51 . For further speculation, LINC01126 may directly or indirectly serve as a molecular chaperone to regulate target genes through a whole new mechanism, in addition to performing its biological function in the nucleus. Moreover, studies in vivo should be conducted to corroborate our current ndings. In conclusion, our study proposes a new mechanism of LINC01126 in regulating periodontitis, which is also the rst report on the biological role of miR-655-3p in in ammatory periodontal diseases and may provide a new strategy for preventing and treating periodontitis.

Methods
Gingival tissues collection and cell culture Gingival tissues were obtained from patients in the clinic. HGFs were isolated from healthy gingival tissues collected from teeth during tooth extraction. The obtained gingival tissue was rinsed with PBS solution at least three times. After cutting up the tissues in EP tubes, they were digested with Type I collagenase (Sigma, USA) in a 37 ℃ water bath for 30 min and were shaken every ve minutes. Followed by centrifugation, precipitated tissue blocks were evenly spread on Petri dish for 12 h before adding culture medium. Cell culture medium of HGFs was composed of 89% (v/v) α-MEM (Gbico, Shanghai,

The establishment of cell in ammation model
For establishing a model of in ammatory HGFs, a complete culture medium containing a 1 ng/mL concentration of lipopolysaccharide (LPS) (Sigma, USA) was used as the in ammatory medium. The in ammatory medium was added to the 6-well plate when cell density reaches about 60%. Before harvesting, HGFs were cultured with in ammatory medium for 1, 2, and 3 days.

The silencing of LINC01126 in HGFs
Three di erent siRNAs (si-LINC01126_001, si-LINC01126_002 and si-LINC01126_003) of LINC01126 were designed and synthesized by Guangzhou RiboBio Co.,LTD (Table S1). Transfection experiments were performed in 6-well plate when cell density reached 80%. HGFs were transfected with siRNA or si-NC at a concentration of 50 nM using lipo3000 (Thermo Fisher Scienti c, USA). HGFs were harvested 24 h later after transfection of synthesized sequences. LINC01126 silencing was veri ed using quantitative real-time PCR.

Cell transfection
HGFs were seeded in 6-well plates with 2×10 5 cells per well. HGFs were transfected with small interfering RNA from negative control (si-NC) or screened small interfering RNA from LINC01126 (si-LINC01126). MiR-655-3p mimics or mimics-NC (RiboBio, Shanghai, China) were transfected in the same manner with a concentration of 50 nM. And complete medium was substituted by an in ammatory medium 24 h later after cell transfection. Cell samples were harvested at the appointed time.

Cell viability assay
Cell Counting Kit-8 (CCK-8) kit (Bioss, Beijing, China) was utilized to perform cell viability assay. HGFs were seeded in 96-well plates with 10 3 cells per well. Following cell attachment, the selected si-LINC01126 sequence was utilized to transfect HGFs with a concentration of 50 nM. Cell viability was measured at 0, 1, 3, and 5 days after transfection, and OD values were measured at 450 nm using a SpectraMax ID5 Multi-Mode Microplate Reader (Molecular Devices, USA).
Wound healing assay 2×10 5 cells were seeded in 6-well plates. Brie y, simulated wounds were created with a 200 µ sterile tip on a single layer of HGFs when cells reached approximately 90% con uency. Following that, HGFs were rinsed with sterile PBS three times to rinse off the scraped cells. Finally, HGFs were cultured with fresh opti-MEM medium (Gbico, Shanghai, China). After 24 h, the healing of scratches was observed and photographed for analysis.

Flow cytometry
When the density reached approximately 80%, cell transfection was performed as before. After 24 h of si-LINC01126 transfection, HGFs were cultured with in ammatory medium for 48 h. HGFs were then collected for apoptosis detection through ow cytometry. All operations were performed following manufacturer's instructions.

Quantitative real-time PCR (qRT-PCR) analysis
The purity and concentration of total RNA were detected by NanoDrop-2000 after the extraction of total RNA using RNAiso plus (Takara, Japan). Then, reverse transcription of mRNA and microRNA was performed by using PrimeScript™ Master Mix (Takara, Japan) and miRNA First Strand cDNA Synthesis (Tailing Reaction) (Sangon biotech, China), respectively. GoTaq® qPCR Master Mix (Promega, America) was used for qPCR according to the experiment instruction. The expression level was evaluated using a method of 2 −ΔΔCT . The primers of the target genes were shown in Table S2.

Fluorescence in situ hybridization (FISH)
Fluorescence in situ hybridization Kit speci c for LINC01126 was designed and synthesized by Guangzhou RiboBio Co., LTD. HGFs were washed with PBS for 5 min and xed in 4% paraformaldehyde for 10 min when their cell density reached about 60-70%. HGFs were then permeabilized by sterile PBS containing 0.25% Triton X-100 for 10 min under 4 ℃, and rinsed with sterile PBS at room temperature. Following that, pre-hybridization was performed under 37 ℃ for 30 min. After that, the probes for LINC01126, U6, and 18S were added into a hybridization solution, and the cells were cultured under 37 ℃ overnight in the dark. Finally, the nuclei were stained with DAPI, and photographs were taken using a uorescence microscope. U6 and 18S were acted as internal references for nucleus and cytoplasm, respectively.
Dual-luciferase reporter assay 293T cells were used in this experiment. The wide or mutant type of 3'-UTR of LINC01126 and IL-6 was cloned into pmiR-RB-Report™ vector ( Figure S1) (Ribo, China), a vector speci cally designed for detecting the combining capacity of miRNA. The plasmids, mimics NC or miR-655-3p mimics, were transfected into 293T cells using Lipo3000 reagent (Thermo Fisher Scienti c, USA). Plasmids and mimics NC or miR-655-3p mimics were transfected at 2.5 µg/mL and 50 nM, respectively. The luciferase activity was measured by using Dual-Luciferase Reporter Assay System (Promega, USA) after 48 h. Finally, the activity of re y luciferase gene (hRluc) and renilla luciferase (hLuc+) was detected. The ratio of hRluc activity to hLuc + activity is the relative luciferase activity.

H&E staining and Immunohistochemistry staining
Periodontitis tissues and healthy controls were obtained from department of oral and maxillofacial surgery of Stomatological Hospital a liated to Chongqing Medical University. After xed in 4% paraformaldehyde for 24 h, gradient dehydration and para n embedding were performed. Following that, serial sections were performed and used for H&E and IHC staining. Hematoxylin-Eosin/HE Staining Kit

Statistical analysis
Experimental results were analyzed and processed using CytExpert, Image Lab, GraphPad Prism (8.0), Adobe Photoshop and Image J. One-way ANOVA or the independent-sample T-test was used to compare the differences between groups. Data were presented as mean ± SEM compared with the control groups.
And p < 0.05 was regarded as statistically signi cant. All experiments were repeated at least three times.

Declarations Funding
This study was nancially supported by Young and Middle-aged Senior Medical Talents studio of Chongqing (ZQNYXGDRCGZS2019004) and the National Natural Science Foundation of China (81700982, 82071072).

Availability of data and materials
The corresponding author can provide all the original data in our manuscript upon reasonable request.
Ethics approval and consent to participate

Competing interests
The authors declare no con ict of interest.

Figure 2
LINC01126 silencing has no signi cant effect on cell proliferation, but promotes cell migration and reduces LPS-induced apoptosis. (A) Cell vitality of HGFs after LINC01126 silencing was determined by CCK-8 assay. (B) Cell migration ratio of HGFs. (C-D) Apoptosis ratio of HGFs cultured in in ammatory medium for 48 h after LINC01126 silencing. Data were presented as mean ±SEM (*p < 0.05, **p < 0.01).  mimics transfection and the corresponding controls. Data were presented as mean ±SEM (*p < 0.05, ***p < 0.001, ****p < 0.0001).  A schematic diagram of the pathogenic mechanism of LINC01126 in periodontitis.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.