Identi cation the Synergistic Effect of Immune- Related lncRNA KIAA0125 and Vitamin D Metabolism Related Gene CYP24A1 in Periodontitis

Yang Li Department of Stomatology, Zhongshan Hospital, Fudan University Ruixue Li Department of Stomatology, Zhongshan Hospital, Fudan University Ying Wu Department of Stomatology, Zhongshan Hospital, Fudan University Chan-Juan Gong Department of Stomatology, Zhongshan Hospital, Fudan University XiaoJun Ding (  dingxiaojun051214@163.com ) Department of Stomatology, Zhongshan Hospital, Fudan University

post-transcriptional levels, thus affecting the occurrence and development of diseases [4][5]. Currently, some studies have shown that lncRNA is closely related to the occurrence and development of pulpitis, periodontitis and maxillofacial tumors [6][7][8].
In this study, we explored the phenomenon of immune cell in ltration involved in the pathogenesis of periodontitis, so as to further discover the immune related lncRNAs in periodontitis. In addition, we also constructed the competing endogenous RNAs (ceRNA) network and screened out the key genes by weighted gene co-expression network analysis (WGCNA), transcriptomic sequencing and quantitative real-time PCR (qRT-PCR) results. Through the immune-related lncRNA and key genes in ceRNA network can further comprehensively understand the pathogenesis of periodontitis.

Datasets and data processing
We downloaded the gene expression data of 183 cases of periodontitis and 64 cases of normal tissue from the dataset GSE10334 (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE10334) from the Gene Expression Omnibus (GEO) and preprocessed the original data using the RMA method. Next, the "Limma" package of R software was used to screen out differentially expressed genes (DEGs) and differentially expressed lncRNAs (DElncRNAs) associated with periodontitis and the screening criteria were |log (FC)| > 1 and adjusted P value < 0.05 [9]. We used the"Pheatmap"package of R software to draw the corresponding heatmap. In order to correlate functional genes with each lncRNAs, co-expression analysis was performed on DElncRNAs and DEGs with the correlation coe cient greater than 0.5 with P value < 0.001. Then, Cytoscape software was used to construct the integrated network [10].

Abundance calculations
CIBERSORT is a tool for deconvolving the expression matrix of immune cell subtypes based on the principle of linear support vector regression. The expression data of RNA-Seq can be used to estimate the in ltration of 22 immune cells, thus quantifying the abundance of speci c cell types [11]. Therefore, we used CIBERSORT to calculate differences in immune cell in ltration between normal and periodontitis tissues.

Identi cation of Immune-Related lncRNAs
The ImmPort database (https://www.immport.org/) contains an updated list of IRGs participating in immunizations [12]. We downloaded all the IRGs from it and used R software to take the intersection of IRGs and DEGs in GSE10334, and then we screened the differentially expressed IRGs. The ltering criterion was |log (FC)| > 1, and the adjusted P value < 0.05. Then Pearson co-expression analysis of differentially expressed lncRNAs and IRGs was performed and the absolute value of Pearson correlation coe cient > 0.6 and P < 0.001. The Cytoscape software was used to construct a visual co-expression network. Subsequently, the relationship between immune cell types and immune-related lncRNAs was then calculated by Pearson analysis.
Identi cation and validation of hub genes by weighted gene co-expression network analysis (WGCNA) In order to further verify the accuracy of the lncRNAs, miRNAs and mRNAs above, WGCNA analysis method was used for veri cation. The WGCNA analysis approach aims to nd gene modules that are coexpressed and to explore the association between gene networks and phenotypes of concern, as well as the core genes in the networks. Therefore, we downloaded the gene expression of 69 normal tissue and 241 periodontitis from GSE16134 (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE16134) and selected the top 25% of genes with the largest variance to construct the weighted gene network by using "WGCNA" package of R software. First we set a threshold β to establish an adjacency matrix based on connectivity so that the distribution of genes conforms to a scale-free network. The adjacency matrix and topology matrix were obtained according to the β value, and the genes were clustered with dissimilarity between genes. We used hierarchical clustering to generate the clustering gene dendrogram and set the minimum module size to 50. The modules with correlation coe cient greater than 0.75 were merged and the corresponding sample dendrogram was drawn. Next, we drew the heatmap of correlation between modules and sample traits, with rows representing modules and columns representing traits. In order to nd hub genes related to the trait, we calculated gene signi cance (GS) and module membership (MM) to observe whether these genes were highly correlated with their corresponding modules and traits [17].

RNA sequence
Total RNA was extracted from 1 gingival tissues with lesions from patients diagnosed with periodontitis and 2 healthy gingival tissues from patients with tooth extraction for orthodontic treatment. The study was approved by Ethical Committee at Zhongshan Hospital. Total RNA was extracted from the tissue using TRIzol® Reagent (Invitrogen) according the manufacturer's instructions and genomic DNA was removed using DNase I (TaKara). RNA-seq transcriptome librariy was prepared following TruSeqTM RNA sample preparation Kit from Illumina (San Diego, CA) using 5 µg of total RNA. Paired-end RNA-seq sequencing library was sequenced with the Illumina HiSeq 4000 (2 × 150 bp read length). To identify DEGs between different samples, the expression level of each transcript was calculated according to the fragments per kilobase of exon per million mapped reads (FRKM) method. RSEM (http://deweylab.biostat.wisc.edu/rsem/) [18] was used to quantify gene abundances. R statistical package software edgeR [19] was utilized for differential expression analysis. Venn diagram was performed by online tool jvenn (http://bioinformatics.psb.ugent.be/webtools/Venn/) to overlap the DEGs.

Human tissue specimens
In order to verify whether immune-related lncRNA KIAA0125 and CYP24A1 are highly expressed in periodontitis tissue, we obtained diseased gingival tissue of 12 patients with periodontitis (6 females, 6 males) and healthy gingival tissue of 13 patients with tooth extraction orthodontic (6 females, 7 males) and analyzed the expression of related genes. Informed consent was given to all participants. The study was approved by Ethical Committee at Zhongshan Hospital (NO. B2020-128R).

Isolation and cultivation of human periodontal ligament cells (hPDLCs)
Scrape the healthy periodontal ligament tissues at the one-third of the root of premolars that removed during orthodontic treatment. Tissues were transferred to a 6 cm culture dish with DMEM supplemented with 20% fetal bovine serum (CSFBS1050, Cellsera Rutherford, Australia), and cultured at 37 °C in a 5% CO2 humidi ed atmosphere. hPDLCs crawled out about a week later and were transferred to a 25 m 2 culture ask. Cells at passages 4 to 6 were used for later experiments.
Cell viability assay Cell activity was measured by a CCK-8 assay (KitkGA317S-500, KeyGEN BioTECH, China). hPDLCs were cultured in 96-well plates and treated with Pg-LPS (InvivoGen, San Diego, USA) at different concentrations (0, 0.1, 1, 5, 10, and 20 µg/ml) for 24 hours. CCK-8 solution was added to each well for 10 µL and incubated at 37℃ for 3 hours for the cell viability assay. Cell viability was determined by reading absorbance at 450 nm using a microplate reader.

Statistical analysis
We analyzed the experimental data and mapped the experimental results using GraphPad Prism 8.0. All results are expressed as the mean ± standard deviation (SD). Comparison between the two groups of samples using Student T test and differences among more than two groups were analyzed by one-way ANOVA test. Results with a P < 0.05 were considered statistically signi cant.

Identi cation of DEGs and DElncRNAs in periodontitis
After setting the threshold as |log2FC| > 1 and adjusted P value < 0.05, we identi ed 104 DEGs and 4 DElncRNAs (KIAA0125, RUNX1-IT1, LOC100130476, LOC101929272). The heatmap of the DEGs and DElncRNAs showed that the periodontitis clustered separately from the healthy tissues ( Fig. 1A-B). At the threshold value of Pearson correlation coe cient greater than 0.5, DElncRNAs and DEGs expressed jointly were predicted and the integrated networks ware visualized. The co-expression between upregulated DElncRNAs (KIAA0125, RUNX1-IT1, LOC101929272) and DEGs could be seen in Fig. 1C-1E while the co-expression between down-regulated DElncRNAs ( LOC100130476) and DEGs could be seen in Fig. 1F. Among them, the positive regulatory relationship between DElncRNAs and DEGs was marked as red, while the negative regulatory relationship was marked as blue.

Immune cells and differentially expressed IRGs in periodontitis
Through the CIBERSORT database, we predicted the abundance of immune cells in the GSE10334 dataset, and plotted the histogram, heatmap, and box map of the corresponding immune cell content ( Fig. 2A-C). The results showed that there was a clear difference in immune cell content between normal tissue and periodontitis. In addition, the number of B cells, especially plasma cells, increased in periodontitis affected tissues. Besides, we drew the corresponding heatmap of the relationship between the abundance values of 22 immune cells (Fig. 2D). Using a cut-of threshold of |log2 FC| > 1 and adjusted P value < 0.05, we identi ed 26 IRGs (Fig. 2E). In addition, we conducted and visualized a co-expression network between 26 IRGs and 4 DElncRNAs (Fig. 2F). The results showed that 4 DElncRNAs have strong associations with CD19 and CD79A (P < 0.0001), which the markers of B cells (Table 1 and  Supplementary Table 1). Therefore, we calculated the Pearson correlation to verify the relationship between the abundance of three types of B cells (B cells naive, B cells memory and plasma cells) and the gene expression of DElncRNAs. The results showed that 4 DElncRNAs were strongly associated with plasma cells ( P < 0.0001; Fig. 3), and lncRNA KIAA0125 showed the highest correlation with plasma cells (correlation = 0.7726). Construction of a ceRNA network for periodontitis The gene expression pro les of periodontitis were integrated to nd the differentially expressed lncRNAs, miRNAs and mRNAs in periodontitis, so as to construct ceRNA network, identify the role of ceRNA network in periodontitis and better understand the role of DElncRNAs on mRNAs mediated by miRNAs. The results showed that only 1 DElncRNA (KIAA0125) interacted with 64 DEmiRNAs through the miRcode database (Supplementary Table 2). We searched for differentially expressed mRNAs (DEmRNAs) from 64 DEmiRNAs in the miRDB, TargetScan and miRTarBase databases. Finally we obtained 1 DElncRNA (KIAA0125), 3 DEmiRNAs (miR-449c-5p, miR-125a-5p and miR-125b-5p) and 2 DEmRNAs (BTG2 and CYP24A1) interacting with each other and established the corresponding ceRNA network (Fig. 4).

Identi cation of key genes by WGCNA analysis
We performed WGCNA analysis on the gene expression pro les of 69 normal tissues and 241 periodontitis tissues to nd the modules most relevant to clinical traits and explore the key genes among them. In this study, we chose the power of β = 13 to construct a scale-free network (Fig. 5A). Finally, we obtained 9 gene co-expression modules by combining dynamic tree cutting (Fig. 5B). After combining the module with clinical traits, we found that the turquoise module was closely related to the periodontitis (P = 8e-38; Fig. 5C), and it also contained the genes BTG2, CYP24A1 and KIAA0125 (Fig. 5C an 5D). In addition, we drew the scatter diagram of GS vs MM in the turquoise module with the periodontitis (correlation = 0.7, P = 1.4e-104) (Fig. 5D).
Gene expression pro les of healthy tissues and periodontitis We analyzed the mRNA expression levels measured in 1 gingival tissue from periodontitis patient and 2 normal gingival tissues, which divided into normal gingival group 1 (N1), normal gingival group 2 (N2) and periodontitis group (CP). The clinical characteristics of the periodontitis patient and normal people can be referred to Supplementary Table 3. Through the RNA sequence data, we found that the gene expression patterns showed signi cantly different expression levels among periodontitis and healthy tissues. The differential scatter diagrams in Fig. 6A was created to identify differences among mRNAs using the edgeR algorithm (fold changes > 2 and adjusted P value < 0.05). The clustered heatmap and Venn diagram of DEGs are shown in Figs. 6B and 6C. There were 3461 differentially expressed genes between N1 and CP, among which 1050 genes were up-regulated and 2411 genes were down-regulated (Supplementary Table 4). There were 2318 differentially expressed genes between N2 and CP groups, among which there were 1710 up-regulated genes and 608 down-regulated genes (Supplementary Table 5). In total, there were 808 identical genes in the two groups, including gene CYP24A1. Figure 1A have shown a positive regulatory relationship between lncRNA KIAA0125 and CYP24A1. Therefore, we further veri ed the expression and functional relationship between CYP24A1 and KIAA0125 in periodontitis.
Validation of Immune-related lncRNA KIAA0125 and CYP24A1 in gingival tissues with periodontitis For further veri cation, we used qRT-PCR to measure the expression of KIAA0125, CYP24A1, CD19, CD79A, IL1B and IL6 in 13 periodontitis gingival tissues and 12 normal tissues and the clinical characteristics of all patients were shown in Supplementary Table 6. The KIAA0125 and CYP24A1 gene expression were signi cantly up-regulated in periodontitis gingival tissues compared with healthy tissues (P = 0.0024, P < 0.0001; Fig. 7A and 7B). Furthermore, as shown in Fig. 7C-7D, CD19 and CD79A were also over-expressed in periodontitis compared to the normal tissues (P < 0.001). In addition, the signi cant increase of IL1B and IL6 also indicated the higher in ammation level in the periodontitis group (P < 0.0001; Fig. 7E and 7F). The above results suggested that KIAA0125 was positively regulated with CD19 and CD79A, so we analyzed the correlation between the expression levels of CD79A and CD19 and the expression levels of KIAA0125 and CYP24A. From Fig. 7G-7H, we found KIAA0125 have the highest correlation with CD19 and CD79A (correlation = 0.8760, P = 0.0002; correlation = 0.6953, P = 0.0121). In addition, the results also showed that CYP24A1 was also highly correlated with CD19 and CD79A (correlation = 0.8115, P = 0.0014; correlation = 0.7032, P = 0.0107; Fig. 7I and 7J). Next, we analyzed the correlations between the expression of KIAA0125 and CYP24A1 and clinicopathological features in all the patients respectively (Fig. 8A-8H). In addition to the gingival index (GI), the other clinical features such as plaque index (PLI), probing depth (PD) and clinical attachment level (CAL) were signi cantly correlated with the expression of KIAA0125 and CYP24A1 (P < 0.05). Especially, the expression of KIAA0125 was signi cantly correlated with PD (correlation = 0.9361, P < 0.0001; Fig. 8C).
Validation the synergistic effect of KIAA0125 and CYP24A1 CCK-8 was used to detect the change of hPDLCs proliferation rate under the in uence of different concentrations of LPS, and the results showed that there were statistically signi cant differences between the experimental groups and the control group. The low concentration of LPS promoted the proliferation of hPDLCs, which reached the peak at 1 µg/ml, and then the proliferation rate decreased with the increase of LPS concentration (Supplementary Fig. 1). Therefore, the concentration of LPS used in the test was nally determined to be 1 µg/mL. The results of qRT-PCR showed that under LPS stimulation, the relative expression levels of KIAA0125, CYP24A1, CD19, CD79A, IL1B and IL6 were increased. Compared with the control group, KIAA0125, CD79A and IL1B showed a statistically signi cant difference (P < 0.05; Fig. 9).
CYP24A1 is a key enzyme in vitamin D metabolism, which enables 1,25D to produce extremely inactive metabolites, thus maintaining stable blood calcium concentrations. To clarify the effect of 1,25D on hPDLCs in an in ammatory environment, we added 10 nM 1,25D to treat hPDLCs for 24 hours. The results showed that the relative expression levels of in ammatory factors IL1B and IL6 decreased compared with the LPS group, and the difference of IL1B was statistically signi cant (P < 0.001; Fig. 10E).
In addition, the expression levels of CD19 and CD79A also decreased compared with the LPS group, but the results were not statistically signi cant. Contrary to this result, the expression levels of KIAA0125 and CYP24A1 increased signi cantly compared with the LPS group, especially the increase of KIAA0125 was statistically signi cant (P < 0.05; Fig. 10A).
The cytochrome P450 inhibitor ketoconazole was used to block the effect of 24-hydroxylase (encoded by CYP24A1 gene) on 1,25D catabolic activity. Therefore, we investigated the possibility that ketoconazole could suppress the response of hPDLCs to 1,25D in the in ammatory environment. The results showed that 10 µM ketoconazole plus 10 nM 1,25D and 1 µg/mL LPS signi cantly increased the level of CYP24A1 mRNA transcription in hPDLCs compared to LPS group (P < 0.001; Fig. 10B). With the increase of CYP24A1, we found that KIAA0125, CD19, CD79A, IL1B and IL6 were also increased, especially the increase of KIAA0125 and IL1B was statistically signi cant (P < 0.01; P < 0.05; Fig. 10A and 10E). In contrast, 100 µM ketoconazole combined with 1,25D (10 nM) and 1 µg/mL LPS reduced the expression of CYP24A1 in hPDLCs, relative to the cells treated with 10 µM ketoconazole (P < 0.001; Fig. 10B). With the decrease of CYP24A1, the expression levels of in ammatory cytokines IL1B and IL6 were signi cantly decreased compared with the LPS group (P < 0.001; P < 0.01; Fig. 10E and 10F). In addition, the expression levels of KIAA0125, CD19, and CD79A were also decreased compared to hPDLCs treated with 10 µM ketoconazole (P < 0.01; Fig. 10A, 10C and 10D). Through the above experiments, we found that the expression trend of KIAA0125 and CYP24A1 was consistent in the in ammatory environment.

Discussion
Periodontitis is a common and frequent occurring disease in human oral cavity, which can cause gingival in ammation, alveolar bone resorption and teeth loosening and falling out. Plaque bio lm is the initiating factor that causes periodontitis, and excessive immune in ammatory response of the host also plays an important role in the occurrence and development of periodontitis [20]. At present, some studies have con rmed that lncRNA may change the function of periodontal related cells in the microenvironment of periodontitis. Li et al. Found lncRNA SNHG1 associated with osteoblastic dysfunction in periodontitis can regulate the osteogenic differentiation of in ammatory periodontal ligament stem cells through EZH2mediated H3K27me3 methylation of KLF2 promotor [21]. In this study, 4 DElncRNAs (KIAA0125, Runx1-IT1, LOC100130476, LOC101929272) were screened out from the dataset GSE10334. The results showed that the expressions of KIAA0125, Runx1-IT1, and LOC10191929272 were signi cantly increased in periodontitis, while the expressions of LOC100130476 were decreased.
Next, we investigated the in ltration of immune cells in the periodontitis group and the normal group, and found that plasma cell in ltration was dominant. Besides, we found that four DElncRNAs have high correlation to B cell surface Markers CD19 and CD79A. qRT-PCR results also showed that CD19 and CD79A were up-regulated in periodontitis. Some studies have shown that T lymphocytes and B lymphocytes are dominant in the chronic in ammatory period of periodontitis, and up to 50% of B lymphocytes were detected in gingival tissue [22]. B cells/plasma cells play a protective role in periodontitis. The antibody response against bacteria is bene cial to control the imbalance of microbial ora in the periodontal pocket and prevent bacteria from entering the connective tissue of the gingiva, thus limiting the in ammation and disease. While B cells secrete TNF-α, IL-1β, IL-17, MMPS or plasma cells secrete IgA, IgG, IgM and other antibodies to kill pathogens, they also promote the expression of RANKL to activate osteoclasts and accelerate alveolar bone absorption [23,24]. Studies have shown that B lymphocytes are one of the main sources of RANKL in periodontal lesion tissues. Up to 90% of B lymphocytes in in ammatory tissues of periodontitis express RANKL [25]. Currently, Suzuki et al.
identi ed CD19 and CD79A as the molecular biomarker candidates and pathogenesis of chronic periodontitis [26]. The above evidence suggested that B lymphocytes play an important role in the development of periodontitis.
In order to explore the regulatory network of DElncRNAs, we constructed the ceRNA network and a total of 1 lncRNA (KIAA0125), 3 miRNAs (miR-449c-5p, miR-125a-5p and miR-125b-5p) and 2 mRNAs (CYP24A1, BTG2) were involved in establishing the ceRNA network for periodontitis. KIAA0125 is a long non-coding RNA gene, which is located on chromosome 14q32.33. KIAA0125 was rst reported in 1995, the mechanism of which under human physiological and pathological conditions has been rarely explored [27]. Studies have found that lncRNA KIAA0125 was down-regulated in colorectal cancer and inhibited epithelial-mesenchymal transition (EMT) through Wnt/β-catenin signaling [28]. However, Diniz et al.
detected that the expression level of KIAA0125 transcript in the ameloblastoma group was higher than that of dental follicles, which may be involved in the pathobiological process of ameloblastoma [29]. Few studies on KIAA0125 have been conducted in periodontitis. However, Guzeldemir-akcakanat et al. found that KIAA0125 was highly expressed in the chronic periodontitis group by whole-genome Transcriptomic [30]. The speci c mechanism of KIAA0125 involved in the pathogenic process of periodontitis needs to be further explored.
Through WGCNA analysis and transcriptomic sequencing results, we further found that CYP24A1 was the key mRNA. 1,25D is the main bioactive form of vitamin D, and its role in the prevention and treatment of osteoporosis, diabetes and periodontitis has been attracting people's attention. The main role of 1,25D is to regulate calcium and phosphorus metabolism. However, in recent years, as an immunomodulator, its role in anti-in ammatory and immunomodulatory has become a focus of research, especially its role in innate immunity. 1,25D can up-regulate the expression of antimicrobial peptides, urge phagocytes to kill pathogenic microorganisms, down-regulate the expression of in ammatory factors, and reduce the in ammatory response [31]. Vitamin D 24-hydroxylase can add hydroxyl to the 24-bit carbon atom of 1,25D, thus greatly reducing the activity. It is an inhibitor of 1,25D activity in vivo and negatively regulates the biological effects of 1,25D, the gene encoded by which is CYP24A1 [32]. Liu et al. found that the mRNA expression of CYP24A1 and RANKL in human gingival broblasts (hGF) and PDLCs signi cantly increased after treatment 1,25D [33]. In addition, most studies suggested that the abnormal expression of CYP24A1 leads to excessive or insu cient 1,25D, which is associated with the occurrence and development of most cancers. Studies have found that the risk of colorectal cancer was negatively correlated with the status of vitamin D in patients, and vitamin D supplementation can reduce the incidence of colorectal cancer. The expression level of CYP24A1 increased during the occurrence of colorectal cancer, and the higher the level of CYP24A1, the higher the malignancy of colorectal cancer. At the same time, the expression level of Ki-67, a marker for cancer cell proliferation, also increased, suggesting that the overexpression of CYP24A1 reduced the local practicability and anti-tumor effect of 1,25D [34]. Wang et al. demonstrated that knockdown of CYP24A1 can aggravate 1,25D to suppress EMT, proliferation and invasion, increase the expression of E-cadherin, and reduce the expression of Ncadherin, Vimentin, β-catenin and Snail in mouse ovarian epithelial cells [35]. These above ndings suggested that inhibition of CYP24A1 may activate the vitamin D pathway in prevention and treatment of diseases.
qRT-PCR results by us showed that both KIAA0125 and CYP24A1 were highly expressed in periodontitis and correlated with the clinical characteristics of PLI, PD and CAL. In addition, we found that both KIAA0125 and CYP24A1 were positively correlated with CD19 and CD79A expression levels. When treated with 1 μg/mL LPS for 24 h, KIAA0125 and CYP24A1 expression in hPDLCs were increased, and the results of KIAA0125 were statistically signi cant. Next, we explored the effects on the expression levels of key genes in an in ammatory environment. When treated with 1 μg/mL LPS and 10 nM 1,25D, the expressions of KIAA0125 and CYP24A1 were signi cantly increased, while the in ammatory factors IL1B and IL6 were relatively decreased, suggesting that 1,25D could alleviate in ammation. When we added low-dose ketoconazole, we found that the expression levels of KIAA0125 and CYP24A1 were further increased. With the increase of the expression levels, CD19, CD79A, IL1B and IL6 were relatively increased, which also indicated that CYP24A1 and KIAA0125 could reduce the anti-in ammatory effect of 1,25D. When we used high doses of ketoconazole, KIAA0125 and CYP24A1 were signi cantly inhibited, and as their expression decreased, so did the expression of in ammatory factors. Through vitro experiments, we found that 1,25D could alleviate the in ammation of LPS-induced hPDLCs, while the increased expression of KIAA0125 and CYP24A1 would antagonize the anti-in ammatory effect.
This study has some limitations. First of all, we screened out the key gene CYP24A1 through WGCNA analysis and transcriptional sequencing and made subsequent veri cation, but we did not conduct subsequent experiments on BTG2 and miRNAs in the ceRNA network. In addition, the speci c mechanism of the co-expression of KIAA0125 and CYP24A1 remains to be further explored.

Conclusions
Through this study, we identi ed differentially expressed immune related lncRNA KIAA0125 may have important signi cance in the immune microenvironment and pathogenesis of periodontitis. In addition, we found that KIAA0125 have the closely correlation with vitamin D metabolism related gene CYP24A1 in periodontitis, which needs to be further explored the speci c mechanism by later experiments.

Declarations
The tissue collection process, cell culture, and RNA isolation were conducted according to the principles expressed in the Declaration of Helsinki and approved by Ethical Committee at Zhongshan Hospital (NO. B2020-128R). Written informed consent for the gingival tissues used in this study was obtained from all patients prior to surgery.

Consent for publication
Not applicable.    Map of the immune-related lncRNA-miRNA-mRNA network generated using Cytoscape software. Note: The designations employed and the presentation of the material on this map do not imply the expression of any opinion whatsoever on the part of Research Square concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. This map has been provided by the authors.

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