Transcriptome Analysis of Human Peri-implant and Periodontal Gingiva: A Paired Design Study

doi:10.21203/rs.3.rs-1943596/v1

Download PDF

Research Article

Transcriptome Analysis of Human Peri-implant and Periodontal Gingiva: A Paired Design Study

https://doi.org/10.21203/rs.3.rs-1943596/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 15 Apr, 2023

Read the published version in Clinical Oral Investigations →

You are reading this latest preprint version

Objectives: Limited information is available about the biological characterization of peri-implant soft tissue at the transcriptional level. The aim of this study was to investigate the effect of dental implant on the gingival tissue in vivo by using paired samples and compare the differences between peri-implant and periodontal gingiva at the transcriptional level.

Methods: Paired peri-implant soft tissue and periodontal gingiva tissue from 6 patients were obtained, and the pooled RNAs were analyzed by deep sequencing. Venn diagram was used to further screen out differentially expressed genes in every pair of samples. Annotation and enrichment analysis was performed. Further verification was done by quantitative real-time PCR.

Results: A total of 3549 genes including were found differently expressed between peri-implant and periodontal groups. The Venn diagram further identified 185 genes differentially expressed genes (DEGs) in every pair of samples, of which the enrichment analysis identified significant enrichment for cellular component was associated with external side of plasma membrane, for molecular function was protein binding, for biological process was immune system process, and for KEGG pathway was cytokine-cytokine receptor interaction. Among the DEGs, CST1, SPP1, AQP9 and SFRP2 were verified to be upregulated in peri-implant soft tissue.

Conclusions: Peri-implant gingiva showed altered expressions of several genes related to the cell-ECM interaction.

Clinical Relevance: At the transcriptional level, healthy periodontal is alternative to the healthy peri-implant soft tissue as the control of peri-implantitis to a large extent, except in the research related to cell-ECM interactions.

Peri-implant

Transcriptome Sequencing

Gingiva

Extracellular Matrix

Peri-implant soft tissue, which usually referred to the surrounding keratinized gingiva, is an indispensable part that composing the aesthetic outcome of an implant supported restoration and the protection of underlying bony tissue from resorption.[1] It was reported that collagen fibers were parallelly accumulated to the titanium abutment surface with less gingival fibroblasts (GF) nearby. [2, 3] Thus the gingiva transplant surgeries are often required to achieve extra quality and quantity of tissue, the essential components of which are GF and their extracellular matrix (ECM) biologically. GF compose the majority of cellular component of the gingival tissue, are monitors of the dynamic balance of ECM. While the function of GF is easily neglected due to their generally quiescent status, activated GF is able to recruit immune cells and regulate ECM synthesis.[4] Conversely, the biomechanical property of ECM is also decisive to the phenotype of GF, thus constitutes the cell-ECM interaction. Despite the important functions of gingiva tissue in peri-implant homeostats, a comprehensive evaluation of the in vivo effects of titanium dental implants on the gingival tissue, especially ECM and GF, is still missing.

On the other hand, with increasing research focusing on peri-implantitis, the study control is becoming an issue worth contemplating. Ideally, healthy peri-implant tissue should be the restrict control of peri-implantitis, which is hard to achieve under normal circumstances. Previous studies usually compared peri-implantitis with periodontitis, peri-implant mucositis and periodontal tissue in clinical and pathological aspect[5, 6]. Healthy periodontal tissue is comparatively easy to achieve, however, the detailed difference between healthy peri-implant tissue and periodontal tissue remains obscure. The analysis of transcriptome by microarrays or next-generation sequencing techniques has been a valuable tool to study the changes in gene expression profiles in gingival tissues. Paired samples are obtained from the same subject but under different conditions. Paired-design genomics studies yield higher statistical power due to eliminating various subject-level confounding factors[7].

Hence, the aim of the present transcriptome study is to investigate firstly the effect of dental implant itself on the gingival tissue in vivo by using paired samples, and secondly whether peri-implant and periodontal gingiva were comparable or at least shared similarity to some extent that supporting the periodontal gingiva serving as the control of peri-implantitis sample.

2.1 Study Design

The ethical approval and the informed consent achievement please find in the end of the article. Patients who were older than 18 years, non-smoker, and in general health were eligible. Subjects who fulfilled one of the following criteria were excluded: antibiotics or anti-inflammatory medication history within the previous 3 months; chronic medical disease or condition; smoking; disagree to be enrolled in the study.

Healthy periodontal tissue was obtained from one of the teeth neighboring the implant site during the surgery of implant insertion. Briefly, the flap of implant surgery site was raised up with a ridge incision, and about 2×2×2 mm³ gingiva tissue at the proximal site of the neighboring tooth was obtained with an inverse bevel incision extending to the bottom of the pocket. An implant was inserted into an ideal position, and the flap was closed after connecting a titanium healing abutment, leaving unsubmerged healing. After healing for at least three months, patients were back for the second stage gingiva plastic surgery after radiographic and clinical confirmation of successful osteointegration and abundant keratinized soft tissue. The peri-implant gingiva was obtained from the buccal side of the implant healing abutment about 2×2×2 mm³ in size during the second stage surgery.

2.2 RNA-sequencing

The obtained samples were temporally stored in RNAlater at -80℃. Total RNAs were isolated from the sample by using RNAzol reagent (Invitrogen). The isolated RNAs were sent to Beijing Genomics Institute (BGI) to perform sequencing. Briefly, mRNA for each sample was enriched by OligodT magnetic beads, fragmented by the Fragment Buffer, reverse-transcribed with random N6 primer, repaired end as cDNA followed by the addition of a single ‘A’ base and subsequent ligation of bubble adapters, and PCR amplified to enrich the purified cDNA template using specific primers. The PCR products were denatured and cyclized by the splint oligo into a single strand DNA circle, constructing the final library. DNA nanoballs were generated with the single strand DNA circle by rolling circle replication to enlarge the fluorescent signals at the sequencing process. Clean reads were obtained on the DNBSEQ platform (Wuhan, China) as FASTQ format, yielding 150-bp reads with an average of 44.3 M reads/sample. RNA quality was checked by a Fragment Analyzer and all samples had an RNA integrity value > 8. Clean reads were filtered by the SOAPnuke software (v1.5.2) to remove the reads with adapters, reads with unknown base over 5% and reads with low-quality bases (base index < 10) over 20%. Then the filtered clean reads were aligned to the GCF_000001405.38_GRCh38.p12 human reference genome by using the Hierarchical Indexing for Spliced Alignment of Transcripts (HISAT) v2.0.4 software. Filtered clean reads were also mapped to references using Bowtie2 software (v2.2.5) and the average coverage was 73.53%. Transcript expressions were then standardized and the fragments per kilobase million (FPKM) values were calculated with the Expectation Maximization software package (RSEM, v1.2.12). Finally, the differential expressed genes (DEGs) were identified between the periodontal sample and peri-implant sample from each patient and also between the whole periodontal group and peri-implant group by the PossionDis method, setting the fold change ≥ 2 and the false discovery rate (FDR) ≤ 0.001.

2.3 Bioinformation Analysis

Bioinformation analysis were perform using an on-line tool (https://biosys.bgi.com/). A cluster heatmap was generated based on gene-expression levels using the pheatmap package in R software. Terms with positive FDR (Q value < 0.05) were considered statistically significant. The DEGs among different comparisons were also detected and displayed as a Venn diagram. Further, Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway annotation and enrichment analysis were performed using the phyper function in R software. Terms with positive FDR (Q value < 0.05) were considered significantly enriched. Information of the genes of interested was collated from an online database (www.genecard.org).

2.4 qPCR

RNA from the samples was isolated as upper mentioned, and reverse transcripted using the PrimeScript™ RT Master Mix (Takara, Japan). The quantitative real-time PCR was performed by an ABI QuantStudio7 (USA) with the SYBR Green Master Mix (Vazyme, China). ACTB was used as the internal control of the total RNA. The primer sequences were shown in Supplemental Table 1. The relative quantity (RQ) of genes was calculated using the 2 − ΔΔCt method.

2.5 Statistics

The GraphPad Prism software was used to perform the statistical analyses. In the pooled qPCR analysis, 13 samples per group were compared using the unpaired two-tailed Student’s t test except that AQP9, SPP1 and GAPDH using the Mann-Whitney test due to the uneven of variance. In the paired qPCR analysis, 8 pairs of samples per group were compared using the two-tailed ratio paired t test. P values < 0.05 were considered statistically significant.

3.1 Study participants

Six patients aged 35 ± 13.45 were included in this study. Each patient has provided one sample for the periodontal group and one for the peri-implant group. The average probing depth of either periodontal group or peri-implant group was 2.0 mm.

3.2 Bioinformation comparison between periodontal and peri-implant group

A total of 3549 genes were identified as DEGs between the periodontal group and peri-implant group, and averagely 2230 ± 793 DEGs were identified between the periodontal sample and peri-implant sample from each patient. Cluster heatmap of the gene expression shown that, overall, all 12 samples shared similar transcriptome profiles (Fig. 1A). Almost all of differential expressed transcripts were mRNA, only a tiny minority of transcripts were long non-coding RNAs (lncRNAs) and pseudo RNAs.

It was noteworthy that one transcript of GAPDH gene (GenBank NM_001289746.1) was found in the list of differently expressed RNAs, with log2 (I / T) > 0.8 and Q value (I / T) < 0.05. Therefore, ACTB was used as the internal control of the total RNA in the following qPCR analysis, and the gene expression of GAPDH was also included in the genes of interested.

3.3 Screening the genes of interested from the DEGs by Venn diagram

To further screen out the genes cardinally changed in peri-implant gingiva, the intersection of the DEGs of every sample pair in Venn diagram was of interested (Fig. 1B). A total of 185 genes were identified. The cluster heatmap of the fold change of the 185 genes was presented (Fig. 1C). Overall, all these genes were mRNAs, and most genes were upregulated.

The 185 DEGs were annotated by GO Classification and KEGG pathway respectively. In the biological processes catalogue, the number of genes sorted into the cellular process ranked first (166 genes). In the cellular components catalogue, the greatest number of genes were cell (168 genes) and cell part (168 genes). Binding (162 genes) has the commanding lead in the whole molecular function catalogue (Fig. 2). In the KEGG pathway annotation, transport and catabolism (39 genes), signal transduction (40 genes), infectious diseases: viral (25 genes), global and overview maps (6 genes) and immune system (53 genes) ranked first in their respect catalogue (Fig. 3).

Enrichment analysis of GO Classification and KEGG pathway were further performed. In the cellular components catalogue, the most enriched genes were classified as external side of plasma membrane (28 genes, rich ratio 0.08). Noticing the T cell receptor complex genes has a high rich ratio (6 genes, rich ratio 0.33) although ranked low (Fig. 4A). In the molecular function catalogue, the result of enrichment analysis was consistent with that in the annotation analysis, showing the protein binding genes (111 genes, rich ratio 0.01) were prominently enriched. The high rich ratio of inerleukin-2 receptor activity and inerleukin-2 binding (3 genes: IL2RA, IL2RB and IL2RG, rich ratio 0.75-1) also worth noticing in this catalogue (Fig. 4B). In the biological processes catalogue, immune response (34 genes, rich ratio 0.06), immune system process (24 genes, rich ratio 0.05) and cell adhesion (29 genes, rich ratio 045) ranked top three. T cell activation (11 genes, rich ratio 0.15) and positive regulation of T cell proliferation (10 genes, rich ratio 0.10) also have comparatively high rich ratio (Fig. 4C). Consistently, KEGG pathway enrichment analysis shown that the top-ranking pathways were cytokine-cytokine receptor interaction (18 genes, rich ratio 0.05) and T cell receptor signaling pathway (12 genes, rich ratio 0.08). Noticing that several T helper cell related pathways were presented, implying that T helper cell might significantly activated after the implant placement (Fig. 4D).

The top 20 up-regulated/down-regulated genes were listed in detail (Table 1), with at least 3 times of fold change. Most of these genes were annotated with extracellular region or membrane in the GO cellular components catalogue. Protein binding stills annotated most genes in the molecular function catalogue. The GO biological process of these genes was characterized by immune related GO terms. Likewise, these genes involving in diverse pathways whereas lots of them were related to immune system pathways. The top 3 genes have even higher fold change (SULT1A4 -8.530 ± 0.942, CST1 7.939 ± 1.393, LOC100996747 6.577 ± 1.185) than rest genes (less than 4.262). SULT1A4 was the only down-regulated gene among the top 20, the biological process of which was related to xenobiotic metabolic process. CST1 was the most up-regulated gene among the top 20 and was found involved in the pathway of salivary secretion. LOC100996747 was also highly up-regulated, however, little has been known about this gene other than a protein coding gene since reported in 2015[8].

3.4 qPCR analysis

To verify the expressions of the top-5 genes and GAPDH (except LOC100996747), we performed a qPCR test of an enlarged sample size including 8 paired samples and another 5 independent samples per group, using ACTB as the internal control. When pooled samples were analyzed, SPP1 and SFRP2 in the genes of interested showed statistical difference between the periodontal group and the peri-implant group. The great varies of genes expression levels could be seen in the violin plot, implying the diverse expression of the tested genes between individuals (Fig. 5A). In order to improve the power of the test and eliminate the influence of individual variance, the 8 paired samples were further analyzed. The expression of CST1, SPP1, AQP9 and SFRP2 showed statistical difference between the two groups in the ratio paired t test (P < 0.05) (Fig. 5B). Of notice, although GAPDH showed no statistic difference in the pooled analysis, down-regulation trends in the peri-implant group could be seen to certain extend when the paired analysis was performed. Together, the up-regulation of CST1, SPP1, AQP9 and SFR2 was confirmed in the peri-implant gingiva compared with the periodontal gingival.

The current study identified 185 DEGs and verified the upregulation of CST1, SPP1, AQP9 and SFRP2 using paired samples in peri-implant gingiva. The biological function and associated pathway of these 185 DEGs displayed close connection with cell-ECM adhesion and immune system activity, especially T cell activities.

Several top-ranking DEGs encode proteins that might function in the cell-ECM interaction, including CST1, SPP1, SFRP2, ADAM12, FNDC1, GPC6 and GZMK. Within the limitation of a small sample size, the upregulation of CST1, SPP1, SFRP2 were verified. Moreover, the enrichment analysis of GO Classification in the molecular function and the biological process showed DEGs were enriched in the ECM binding, the ECM structural constituent, and the ECM organization catalog. In fact, it is proven recently that sensing the biomechanical property of the surrounding ECM via the mechanosensor (focal adhesion) is a cardinal physical activity of cells.[16] Particularly, gingival fibroblasts not only produce but also be affected by the gingival ECM tremendously.[9, 10] The current result implied that, the ECM surrounding a healed implant might change the cellular phenotype of GF in the cell-ECM adhesion aspect, probably due to the change of biomechanical property of ECM. Whether this “prime” status of the peri-implant cells increase the susceptibility of peri-implantitis deserves further investigation.

It is reported that gingiva-resident memory T cells play important roles in the immune surveillance network of periodontal tissue[11]. During the 2–12 weeks of healing period of peri-implant gingiva, CD3 positive T cells were also observed in the connective tissue at apical front of peri-implant pocket[12]. Material of healing abutment may also affect T cell activities during the healing of the peri-implant gingiva[13]. All 3 chains of Interleukin-2 receptor (IL2RA, IL2RB and IL2RG) were up regulated 2–3 folds in the peri-implant gingiva in our data, serving as an evidence of T helper cell activation. Our data implying the active T cells response at the observed time point (3 months after implant placement), moreover, T helper cells activities may feature this process.

GAPDH showing up in the DEGs list was unexpected. As GAPDH is extensively used as the internal reference in the gene expression analysis, it is of great importance to confirm its expression changes in the peri-implant group. The qPCR data in the present study showed no statistical difference between the periodontal group and the peri-implant group in both pooled and paired analysis, but a down-regulation trend was still speculated in the paired samples. It is known that GAPDH as a housekeeping gene that key to glycolysis, is an inappropriate internal reference gene for the cellular metabolism studies, whereas emerging research found that GAPDH expression is also affected by cancer, cell death, and inflammatory diseases.[14, 15] Our result also sobered us to select the internal reference gene with caution when performing qPCR analysis of gingiva tissue.

Nevertheless, at the overall transcription level, periodontal and peri-implant gingiva still shared great similarities. Thus, healthy periodontal gingiva is one of the reasonable candidates as the control tissue for peri-implantitis tissue to some extent. Also, our data would provide reference when detailed analysis of some specific genes were interested when choosing healthy periodontal gingiva as the control tissue for peri-implantitis tissue.

It is noteworthy that, in this study, the healthy peri-implant gingiva was harvested before restoration, therefore it was not the strictly representative peri-implant gingiva in function. The time course of healing and function is another confounding factor that should be considered. Whether the restoration and time course affect the biology of peri-implant gingiva need to be evaluated in the future.

In summary, the data in the current study shown overall similar transcriptome profile between the periodontal gingiva and peri-implant gingiva (3 months after implant placement). Paired analysis identified 185 DEG, which were characterized by protein binding in molecular function aspect, and the immune activities (especially T cell related activities) may be the most affected biological process. Further, CST1, SPP1, AQP9 and SFRP2 were confirmed upregulated in the peri-implant gingiva, suggesting the cell-ECM interaction might be changed in the peri-implant gingiva. Our study characterized the peri-implant gingiva at the transcriptional level and provided evidence for the selection of control group in future peri-implantitis studies.

FUNDING

This study was supported by funding from the National Natural Science Foundation of China (81970975 and 81600914); the Guangdong Basic and Applied Basic Research Foundation, China (No. 2021A1515010821 and 2021A1515110303); the Science and Technology Program of Guangzhou, China (No. 202102021198); and the Guangdong Financial Fund for High-Caliber Hospital Construction, China (No. 174-2018-XMZC-0001-03-0125/D-10).

AUTHOR CONTRIBUTIONS

Danying Chen was responsible for the transcriptomic analysis and the manuscript drafting; Zhixin Li was responsible for the qPCR analysis, Qifan Liu, Yue Sun, Kaidi Chen, Jieting Yang, and Jiaying Song were responsible for sample processing; Huaxiong Cai was responsible for interpretation of results and manuscript polishing; Baoxin Huang and Zhuofan Chen conducted the clinical surgeries and supervised the studies.

CONFLICT OF INTEREST

All of the authors report no conflicts of interest related to this study.

ETHICAL APPROVAL

The current study was conducted in accordance with the World Medical Association Declaration of Helsinki (version, 2013). The ethical approval was obtained from the Ethical Committee of Guanghua School of Stomatology, Hospital of Stomatology, Sun Yat-Sen University (Ethics number: ERC-2016-21).

INFORMED CONSENT

The written informed consent of all the participating subjects was obtained prior to enrolling in this study.

Chappuis V, Araujo MG and Buser D (2017) Clinical relevance of dimensional bone and soft tissue alterations post-extraction in esthetic sites. Periodontol 2000 73:73-83. doi: 10.1111/prd.12167
Berglundh T, Lindhe J, Ericsson I, Marinello CP, Liljenberg B and Thomsen P (1991) The soft tissue barrier at implants and teeth. Clinical oral implants research 2:81-90. doi: 10.1034/j.1600-0501.1991.020206.x
Moon IS, Berglundh T, Abrahamsson I, Linder E and Lindhe J (1999) The barrier between the keratinized mucosa and the dental implant. An experimental study in the dog. Journal of clinical periodontology 26:658-63. doi: 10.1034/j.1600-051x.1999.261005.x
Chiquet M, Katsaros C and Kletsas D (2015) Multiple functions of gingival and mucoperiosteal fibroblasts in oral wound healing and repair. Periodontol 2000 68:21-40. doi: 10.1111/prd.12076
Schwarz F, Derks J, Monje A and Wang HL (2018) Peri-implantitis. Journal of periodontology 89 Suppl 1:S267-S290. doi: 10.1002/JPER.16-0350
Liu Y, Liu Q, Li Z, Acharya A, Chen D, Chen Z, Mattheos N, Chen Z and Huang B (2020) Long non-coding RNA and mRNA expression profiles in peri-implantitis vs periodontitis. J Periodontal Res 55:342-353. doi: 10.1111/jre.12718
Stevens JR, Herrick JS, Wolff RK and Slattery ML (2018) Power in pairs: assessing the statistical value of paired samples in tests for differential expression. BMC Genomics 19:953. doi: 10.1186/s12864-018-5236-2
Hein MY, Hubner NC, Poser I, Cox J, Nagaraj N, Toyoda Y, Gak IA, Weisswange I, Mansfeld J, Buchholz F, Hyman AA and Mann M (2015) A human interactome in three quantitative dimensions organized by stoichiometries and abundances. Cell 163:712-23. doi: 10.1016/j.cell.2015.09.053
Hinz B (2010) The myofibroblast: paradigm for a mechanically active cell. Journal of biomechanics 43:146-55. doi: 10.1016/j.jbiomech.2009.09.020
Nan L, Zheng Y, Liao N, Li S, Wang Y, Chen Z, Wei L, Zhao S and Mo S (2019) Mechanical force promotes the proliferation and extracellular matrix synthesis of human gingival fibroblasts cultured on 3D PLGA scaffolds via TGFbeta expression. Mol Med Rep 19:2107-2114. doi: 10.3892/mmr.2019.9882
Mahanonda R, Champaiboon C, Subbalekha K, Sa-Ard-Iam N, Yongyuth A, Isaraphithakkul B, Rerkyen P, Charatkulangkun O and Pichyangkul S (2018) Memory T cell subsets in healthy gingiva and periodontitis tissues. Journal of periodontology 89:1121-1130. doi: 10.1002/JPER.17-0674
Tomasi C, Tessarolo F, Caola I, Piccoli F, Wennstrom JL, Nollo G and Berglundh T (2016) Early healing of peri-implant mucosa in man. Journal of clinical periodontology 43:816-24. doi: 10.1111/jcpe.12591
Serichetaphongse P, Chengprapakorn W, Thongmeearkom S and Pimkhaokham A (2020) Immunohistochemical assessment of the peri-implant soft tissue around different abutment materials: A human study. Clinical implant dentistry and related research 22:638-646. doi: 10.1111/cid.12942
Zhang JY, Zhang F, Hong CQ, Giuliano AE, Cui XJ, Zhou GJ, Zhang GJ and Cui YK (2015) Critical protein GAPDH and its regulatory mechanisms in cancer cells. Cancer Biol Med 12:10-22. doi: 10.7497/j.issn.2095-3941.2014.0019
Bednarz-Misa I, Neubauer K, Zacharska E, Kapturkiewicz B and Krzystek-Korpacka M (2020) Whole blood ACTB, B2M and GAPDH expression reflects activity of inflammatory bowel disease, advancement of colorectal cancer, and correlates with circulating inflammatory and angiogenic factors: Relevance for real-time quantitative PCR. Adv Clin Exp Med 29:547-556. doi: 10.17219/acem/118845

Table 1 is available in the Supplementary Files section.

No competing interests reported.

SupplementaryTable1.xlsx
Supplemental Table 1. The primer sequences.
healthydatatable1.xls
Table 1 The top 20 up-regulated/down-regulated genes from the 185 DEGs. The fold change of each gene is shown by the average and standard deviation of the fold change of all 6 paired sample. The information about each gene is collated from an online database ( href="http://www.genecard.org/" rel="noopener noreferrer" target="_blank">www.genecard.org).

Download PDF

Journal Publication

published 15 Apr, 2023

Read the published version in Clinical Oral Investigations →

Editorial decision: Major revision
29 Jan, 2023
Reviews received at journal
23 Jan, 2023
Reviewers agreed at journal
28 Dec, 2022
Reviewers agreed at journal
12 Nov, 2022
Reviewers invited by journal
12 Nov, 2022
Editor assigned by journal
12 Aug, 2022
Submission checks completed at journal
12 Aug, 2022
First submitted to journal
09 Aug, 2022

You are reading this latest preprint version

Transcriptome Analysis of Human Peri-implant and Periodontal Gingiva: A Paired Design Study

Status:

Journal Publication

Version 1

Abstract

Figures

1 Introduction

2 Materials And Methods

2.1 Study Design

2.2 RNA-sequencing

2.3 Bioinformation Analysis

2.4 qPCR

2.5 Statistics

3 Results

3.1 Study participants

3.2 Bioinformation comparison between periodontal and peri-implant group

3.3 Screening the genes of interested from the DEGs by Venn diagram

3.4 qPCR analysis

4 Discussion

5 Conclusions

Declarations

References

Table

Additional Declarations

Supplementary Files

Status:

Journal Publication

Version 1