Serum proteins associated with periodontitis relapse post-surgery: A pilot study

DOI: 10.1002/JPER.21-0089

O R I G I N A L A R T I C L E

Serum proteins associated with periodontitis relapse post-surgery: A pilot study

Anders Esberg

Catrine Isehed

Anders Holmlund

Susanne Lindquist

Pernilla Lundberg

1Department of Odontology, Umeå University, Umeå, Sweden

2Department of Molecular

Periodontology, Umeå University, Umeå, Sweden

3Gävle County Hospital, Department of Periodontology, Public Dental Health County Council of Gävleborg, Gävle, Sweden

4Center for Research and Development Uppsala University/Region Gävleborg, Gävle, Sweden

Correspondence

Anders Esberg, Department of Odontol- ogy, Faculty of Medicine, Umeå University, SE-901 87 Umeå (Sweden).

Email:anders.esberg@umu.se

Abstract

Background: The knowledge of which genes and proteins that are connected to the susceptibility to gingivitis with subsequent local tissue degradation seen in periodontitis is insufficient. Changes of serum proteins associated with recur- rence of bleeding on probing (BOP) and increased periodontal pocket depths (PPD) after surgical treatment of periodontitis could reveal molecules that could be early signals of tissue destruction and/or of importance for systemic effects in other tissues or organs.

Methods: We performed a longitudinal pilot study and followed 96 inflammation-related proteins over time in serum from patients who underwent surgical treatment of periodontitis (n = 21). The samples were taken before (time 0), and then at 3, 6, and 12 months after surgery. Changes in protein levels were analysed in relation to the clinical outcome measures, that is, proportion of surfaces affected by BOP and PPD.

Results: Changes in treatment outcomes with early signs of relapse in periodon- titis after surgical treatment, for example, increased BOP and PPDs, were during 12-months follow up associated with increased serum levels of high-sensitivity C- reactive protein (hs-CRP) and programmed death-ligand 1 (PD-L1), and reduced serum levels of cystatin-D protein.

Conclusion: This study shows that clinical signs of recurrence of periodontitis after surgery are reflected in serum, but larger studies are needed for verification.

Our novel findings of an association between increased PD-L1- and decreased cystatin D-levels and recurrence in periodontitis are interesting because PD-L1 has been shown to facilitate bacterial infections and chronic inflammation and cystatin D to inhibit tissue destruction. Our results justify mechanistic studies regarding the role of these molecules in periodontitis.

K E Y W O R D S

bone resorption, infection control, inflammation, periodontal diseases, periodontal pocket

This is an open access article under the terms of theCreative Commons Attribution-NonCommercial-NoDerivsLicense, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made.

© 2021 The Authors. Journal of Periodontology published by Wiley Periodicals LLC on behalf of American Academy of Periodontology

J Periodontol.2021;1–9. wileyonlinelibrary.com/journal/jper 1

1 INTRODUCTION

Tooth-associated commensal bacteria can initiate inflam- matory and immune responses, which lead to gingivitis, a reversible process that affects almost all individuals at some point.

In individuals susceptible to periodon- titis, the inflammatory process destroys the periodontal connective tissues, which activates bone-resorbing osteoclast-cells, and the tooth-supporting jawbone is irreversibly reduced.

Periodontitis is the result of a complex interplay between pathobiontic bacteria in the dental biofilm and the host. The role of specific microorganisms and their products in periodontitis initiation and propagation remain unclear.

However, host genetic, inflammatory, immunological, and envi- ronmental factors, especially smoking, have been shown to contribute to periodontitis susceptibility.

The aetio- pathogenic process of periodontitis is not yet fully understood.

Currently, the diagnostic tools used in dental care can only discover periodontitis after the loss of tooth- supporting tissues has occurred. Although the popula- tion’s oral hygiene efforts have increased in most coun- tries and that many societies spend enormous resources on professional cleaning of teeth, ≈10% of the popu- lations’ worldwide is affected by advanced periodontal disease.

Current periodontitis treatment practices involve anti-infectious agents for resolving inflammation and pre- serving tooth-supporting tissues. In severe cases, peri- odontal surgery is required to access root surfaces. In most cases, both non-surgical and surgical treatments are effective for treating periodontitis; however, in a recent study ≈14% of the treated patients were considered to respond poorly to treatment; thus, despite treatment, they experience a disease relapse with tissue degradation.

However, it remains unknown which individuals, despite good oral hygiene, supportive periodontal treatment, and no smoking, will experience periodontitis relapses after treatment.

Clinical and epidemiological data have indicated that periodontitis is associated with other chronic inflam- matory diseases, such as cardiovascular diseases, dia- betes, and rheumatoid arthritis.

Nevertheless, the poten- tial mechanisms underlying this association are poorly understood. Many hypotheses exist to explain how peri- odontitis might be linked to other chronic inflamma- tory diseases.

One hypothesis is that oral inflammatory molecules, autoantibodies, oral bacteria, or bacterial prod- ucts are spread systemically and trigger inflammatory reactions in other parts of the body, distant from the oral cavity.

Another suggestion is that inflammatory molecules released from unresolved periodontal inflam- mation might contribute to chronic inflammatory states

at different parts of the body.

A third option is that the link between periodontitis and other systemic inflam- matory diseases could be explained by a genetic back- ground, with a more pronounced proinflammatory con- stitution among individuals that are exposed to this type of diseases.

Most interestingly, poor response to peri- odontal treatment has been suggested to predict future cardiovascular disease.

These diseases may have a com- mon denominator linked to the inability to resolve inflammation.

In this pilot prospective study, we aimed to eval- uate inflammatory-associated serum proteins related to changes in clinical treatment outcome after surgi- cal therapy associated with early signs of relapse of periodontitis.

2 MATERIALS AND METHODS

This study was approved by the Regional Ethics Commit- tee, Uppsala, Sweden (Dnr 2009/029). Signed informed consent was obtained from all patients. The trial is reg- istered at ClinicalTrials.gov (NCT04663165) and the study conform to STROBE Guidelines

and conducted in accor- dance with the Helsinki Declaration.

2.1 Study design

In this pilot prospective clinical intervention study, we

followed periodontal healing capacity after periodontal

surgery and the protein profile in serum. Twenty-one con-

secutive patients referred to the department of Periodon-

tology (Public Dental Health County Council of Gävle-

borg, Gävle, Sweden) in need of periodontal surgery were

recruited to the study. All participants received a thor-

ough case presentation, instruction in oral hygiene and

non-surgical subgingival treatment consisting of scaling

of the root surfaces using ultrasonic and hand instru-

ments before they were included in the study. The cri-

teria for periodontal surgery in at least two quadrants

were: a minimum of at least two sites that showed peri-

odontal probing depths (PPDs) ≥ 6 mm, combined with

bleeding on probing (BOP) and/or pus; and radiographs

that displayed a marginal bone loss ≥ 4 mm from the

cemento-enamel junction. Periodontal healing was fol-

lowed with clinical measurements (BOP and PPD) and

blood collections at 3-, 6-, and 12-months after surgi-

cal therapy. None of the participants had received any

periodontal treatment or taken antibiotics within three

months prior to surgery, and none had taken an anti-

inflammatory medication within two weeks prior to

surgery.

2.2 Clinical registration, serum

collection, sample preparation, and storage

Blood samples were drawn and handled by the chemical laboratory at Gävle hospital, at the time of therapy and at follow-up visits. For each patient, five 1 mL serum aliquots were labeled and stored at -80

C. At each time point, blood samples were collected before the clinical six sites/tooth measurements using dental probes

and the deepest 4- sites/tooth were registered (BOP and PPD)

2.3 Treatment

A periodontal specialist (AH) performed all surgeries.

Briefly, after an intra-crevicular incision, a flap was raised, and inflamed tissue, dental calculus, and bacterial biofilm were removed with an ultrasonic cleaner

fitted with a PE39 tip and curettes

. The area was rinsed with saline solution, and the flap was closed with nonresorbable suture

. Patients were advised to rinse twice daily with 10 mL of a 2 mg/mL chlorhexidine solution

for 12 weeks and not to chew or brush on the treated side for 2 weeks.

Sutures were removed after 2 weeks. Supportive care was provided at 6 weeks and every third month, for 12 months after surgery.

2.4 Protein analysis

Protein arrays were performed on serum samples col- lected pre-surgery (time 0) and at 3-, 6-, and 12-months after surgical treatment. The concentrations of 92 proteins were assessed with a proximity extension assay

and the inflammation panel provided (Table S1.docx). The assay comprised oligonucleotide-labeled antibody pairs, which allowed pair-wise binding to target proteins, and proteins were quantified with high-throughput real-time PCR. Data were presented as normalized protein expression levels, measured in terms of Olink Proteomics’ arbitrary unit, and plotted on a log2 scale. The V-PLEX Vascular Injury Panel 2 Human Kit

was used to assess concentrations of four other proteins: high-sensitivity C-reactive (hs-CRP), inter- cellular adhesion molecule 1 (ICAM-1), serum amyloid A (SAA), and vascular cell adhesion molecule 1 (VCAM-1).

1Dental Probe PCP12 Hue Friedy 3-6-9-12 USA.

2Amdent Ultrasonic Scaler, Type No 5601432.

3Langer 1-2, LM 281-282, Colombia 2L-2R, LM 217-218 and Syntette mini LM 215-216, LM-Instruments Oy, Parainen, Finland.

4Ethilon 5-0, 661H, Ethicon US, LCC, USA.

5Hexident, Meda AB, Solna, Sweden.

6Olink Bioscience, Sweden.

7Meso Scale Diagnostics, LLC, USA.

2.5 Statistics

Descriptive statistics were performed with SPSS

including the mean and 95% confidence interval (95% CI) estimates.

Spearman’s correlation method was performed to evalu- ate correlations between variables. Jonckheere–Terpstra test was used for evaluating trends. All tests were two- sided and corrected for multiple comparisons with the Benjamini–Hochberg false discovery rate (FDR < 0.05) when appropriate. Generalized estimated equation (GEE) was used for repeated measurements, that is, 3-, 6-, and 12- months samples, as it allows the use of non-normal dis- tributed data and includes all available data for each sub- ject. GEE approach was used to evaluate if gender, smok- ing, age, other diseases, or number of teeth associated with BOP or PPD over 3 to 12 months post-surgery. Generalized linear regression model (GLM) was performed to evalu- ate variables at individual time points. Multivariate partial least square regression (PLS) model was performed to gen- erate an overview of the independent inflammatory mark- ers and the dependent BOP variable

. PLS regression was chosen because it allows co-variation among the indepen- dent x-variables. GLM was used for time point specific, and GEE models for repeated measurements, to evaluate asso- ciations between serum proteins and BOP or PPD, after adjusting for gender, age, smoking, number of teeth, and other diseases.

3 RESULTS

3.1 Pre- and post-surgery comparison

At time point 0, the mean (95% CI) patient age was 57.5 years (53.2-61.7). Among the 21 patients, 52.4% were female and 57.1% were smokers. Patients had an average (95%

CI) of 25.7 (24.0-27.4) teeth, BOP was observed in 39.5%

(31.7-47.3) of tooth surfaces, and the PPD was > 4 mm in 41.2% (34.4-48.0) and > 6 mm in 10.5% (7.7-13.4) of tooth surfaces (Table S2.docx). Five patients displayed poten- tial confounding medical symptoms, including: diabetes (n = 2), high blood pressure (n = 2) and cardiovascular dis- ease (n = 1), which were adjusted for in all models. One patient did not complete the 12 months follow-up exami- nation.

Before surgery (time 0), we found no associations between gender, smoking, other diseases, number of teeth, or plaque index (PLI), with BOP or PPD (P > 0.05). In con- trast, at time 0, age was negatively correlated with the num- ber of teeth (r = −0.578, P = 0.006) and PLI (r = −0.453,

8IBM SPSS Statistics for Windows, Version 26.0. Armonk, NY: IBM Corp.

9SIMCA 15; Sartorius Stedim Data Analytics AB, Malmö, Sweden.

F I G U R E 1 Correlations between clinical outcomes and indicated BOP and PPD relapse after surgery. Scatterplots show correlations between participants age, number of teeth, plaque index (PLI), the percentage of sites with bleeding on probing (BOP), and the percentage of sites with peridonal probing depths (PPDs) larger than 4 or 6 mm, measured at (A) time point 0 and (B) for time point 3, 6, and 12 months post-surgery. Trend lines show Spearman correlations, with the 95% CI. The correlation coefficient (r), and P-value are shown and significant correlations are colour indicated based on strength. Box plots show the percent sites (%-sites) with (C) BOP and (D) PPD> 4 mm measured at time points, 3-, 6-, and 12-months post-surgery

P = 0.039). Correlations were also observed between BOP and PPD > 4 mm (r = 0.777, P < 0.001) and BOP and PPD > 6 mm (r = 0.669, P = 0.001). In addition, corre- lations were observed between PLI and BOP (r = 0.604, P = 0.004), PLI and PPD > 4 mm (r = 0.549, P = 0.010), and PLI and PPD > 6 mm (r = 0.442, P = 0.045) (Figure 1A).

Similar trends were observed post-surgery throughout the 12-month follow-up (3 to 12 months; Figure 1B) and at the individual time points of 3-, 6-, and 12-months (Figure S1.docx).

Surgical treatment reduced the frequencies of BOP (P < 0.0001) and PPD (P < 0.0001) (data not shown).

Post-surgery, change in BOP was not associated with gen- der, smoking, age, the number of teeth, or other diseases (P > 0.05), however, change in PPD after surgery was asso- ciated with other diseases (P = 0.044), but not with gender, smoking, age, or the number of teeth (all P > 0.05; based on GEE of data from 3 to 12 months). Notably, both BOP and PPD > 4 mm increased over-time post-surgically, indi- cating early signs of relapse (GEE, 3 to 12 months samples, adjusted for gender, smoking, age, and other diseases) (Fig- ure 1C and 1D).

3.2 Multivariate analysis of associations between serum inflammatory markers and BOP

We performed a PLS regression to generate an overview of the 96 serum proteins and their potential associations

with changes in clinical treatment outcome at different time points after surgery. Due to the highly correlated links between BOP, PPD, and PLI, we primarily used BOP as the dependent variable. The model results (R

= 0.89, Q

= 0.41) for samples collected at time point 0 indi- cated that hs-CRP and cytokines IL8, IL20RA, ENRAGE, IL12B, and CCL4 were positively associated with BOP (Figure 2A). At 12 months after surgery, model results (R

= 0.86, Q

= 0.14) showed that CXCL5, MMP10, IL8, OPG, CCL11, hs-CRP, and MCP4 were positively associated with BOP (Figure 2B). For samples taken at 3, 6, and 12 months after surgery, model results (R

= 0.77, Q

= 0.46) showed that BOP associated with increased serum lev- els of CXCL5, TNFβ, DNER, MMP10, CXCL6, CD6, OPG, IL10, IL8, and hs-CRP, and reduced level of cystatin-D (Figure 2C).

3.3 Inflammatory markers associated with BOP and PPD

Next, we used GLM and GEE models to evaluate associ-

ations between serum proteins and BOP, after adjusting

for potential confounders (i.e., gender, smoking, age, and

other diseases). Before surgery, hs-CRP, CXCL5, IL8, and

HGF were positively associated with BOP (Table 1). At 12

months after surgery, hs-CRP and CXCL5 were positively

associated with BOP, and cystatin-D was negatively asso-

ciated with BOP (Table 1). A relapse of BOP between 3

F I G U R E 2 Multivariate partial least square (PLS) regression analysis show associations between serum proteins and BOP. PLS models included BOP as the dependent variable (y) and levels of 96 serum proteins as the independent x-variables. Samples were collected at (A) time point 0 (pre-surgery), (B) at 12 months post-surgery, or (C) at 3, 6, and 12 months post-surgery (combined). (Left) Score plots summarize the relationship between the serum proteins and the dependent variable, BOP. (Right) Bar plot shows the variable importance for the projection (VIP), which summarizes the importance of the variable in explaining the correlation between X and Y. Serum proteins were considered significant, when VIP> 1.0 and 95% CI was above zero. Significant factors are indicated in red. w* describes the PLS weights from the combination of the original variables in the X-swarm; c[1] and c[2] are components 1 and 2, respectively

and 12 months after surgery was associated with elevated levels of hs-CRP, PD-L1, CXCL5, CD6, and TNFβ, and reduced levels of cystatin-D (Table 1).

Due to concern about the confounding effects of miss- ing teeth, we constructed models that were adjusted for gender, smoking, age, other diseases, and the number of teeth. Those results suggested that elevations in hs- CRP and PD-L1, and a reduction in cystatin-D, were asso- ciated with a relapse of BOP after surgical treatment (Figure 3A-C). Serum proteins that were associated with

BOP were then analyzed for relationships to PPD. Con- sistent with the highly correlated structure between BOP and PPD, we found that elevated levels of hs-CRP were associated with both PPD > 4 mm (P < 0.0005) and PPD > 6 mm (P < 0.0005); we also found that reduced levels of cystatin-D were associated with elevations in PPD > 4 mm (P = 0.041) and PPD > 6 mm (P = 0.001). Ele- vated levels of PD-L1 were associated with an elevation in PPD > 4 mm (P = 0.026), but not PPD > 6 mm (P = 0.188;

Figure 3D-I).

T A B L E 1 Association of inflammatory markers with bleeding on probing Post-surgery

Factor Pre-surgery 12 months 3, 6, and 12 months

β (95%CI) P Value β (95%CI) P Value β (95%CI) P Value

hs-CRP 33427 (11394–55460) 0.003 191641 (105699–277582) <0.0005 171484 (100081–242888) <0.0005 Cystatin–D −0.005 (−0.015–0.005) 0.30 −0.023 (−0.035– −0.011) <0.0005 −0.032 (−0.046– −0.019) <0.0005 PD–L1 0.006 (−0.002–0.013) 0.14 0.027 (0.001–0.052) 0.038 0.030 (0.015–0.044) <0.0005 CXCL5 0.018 (0.006–0.029) 0.002 0.058 (0.030–0.085) <0.0005 0.044 (0.020–0.067) <0.0005 TNFβ 0.004 (−0.003–0.012) 0.27 0.013 (−0.011–0.037) 0.30 0.020 (0.008–0.033) 0.002

CD6 0.004 (−0.008–0.015) 0.51 0.013 (−0.006–0.033) 0.18 0.025 (0.008–0.041) 0.003

IL8 0.017 (0.007–0.028) 0.001 0.027 (−0.008–0.061) 0.13 0.012 (−0.010–0.034) 0.29 HGF 0.009 (0.003–0.014) 0.001 0.012 (−0.011–0.036) 0.31 0.009 (−0.006–0.024) 0.25 To evaluate associations between the 96 serum proteins and bleeding on probing, we constructed a general linear model to examine measurements at each individual visit, and a generalized estimated equation to evaluate the 3- to 12-month repeated measurements; β: Beta values; 95%CI: 95% confidence interval;

all models were adjusted for age, sex, smoking, and other diseases. Bold font indicates proteins that showed significant associations after adjusting for multiple comparisons with the Benjamini-Hochberg equation, at a false discovery rate of 5%. .

4 DISCUSSION

This pilot prospective study focused on the early signs of periodontitis relapses over 12 months after surgical inter- vention. The levels of 96 serum proteins were evaluated to determine associations with BOP and PPD. Our main find- ing revealed that increases in BOP and PPD after surgery were associated with increased serum levels of hs-CRP and PD-L1, and reduced levels of cystatin-D.

CRP is a highly conserved, type I acute-phase protein, and its plasma concentration can increase up to 1000- fold in response to a stimulus, like an infection or tis- sue damage; thus, it is an important clinical diagnos- tic tool. Minor elevations in serum CRP levels can be detected with hs-CRP assays. Studies have shown that hs- CRP is a reliable marker of systemic low-grade inflam- mation associated with a variety of conditions and sev- eral genetic polymorphisms.

In this study relapse of peri- odontitis after surgery were associated with increased lev- els of hs-CRP in accordance with other studies showing that increase of hs-CRP associates with both inflamma- tion, tissue degradation,

and periodontitis.

In patients with both coronary vascular disease and peri- odontitis, hs-CRP was elevated above the levels observed in either condition alone.

Consistently, periodontal treat- ment was associated with a reduction in CRP levels, along with improvements in surrogate measurements of cardio- vascular health.

These observations have provided a potential mechanism that could link periodontal disease to an increased risk of atherosclerotic diseases.

Moreover, we previously reported that a poor response to periodon- tal treatment could predict future cardiovascular disease.

In this study we followed treament outcome at different time points for 12 months. Signs of relaps of periodontitis

regarding the clinical parameters PPD and BOP were asso- ciated to changes of some serum proteins of interest for the healing process. That clinical signs of relaps of peri- odontitis could be monitored in serum further strength- ens the possibility that periodontitis could contribute to the systemic inflammatory burden. Together, these find- ings suggested that serum proteins could be early indica- tors, of increased risk for inflammatory associated diseases, present in individuals that do not respond well to peri- odontal treatment or experience relapses of the disease.

The balance between bone resorption and bone for-

mation is central in bone metabolism-related diseases

and conditions, such as periodontitis and autoimmune

arthritis. Here, we showed that reduced serum cystatin-

D levels were associated with post-surgery periodontitis

relapses. In this context, it is interesting that up-regulation

of cystatin-D is shown to inhibit the activation of osteo-

clasts and bone resorption.

Cystatin-D is a type-2 cysteine

protease inhibitor, predominantly found in saliva.

Thus,

cystatin-D inhibits proteases, such as cathepsin S

that

degrade extracellular matrix proteins.

Cathepsin S is

produced by for example macrophages, lymphocytes, den-

dritic cells, and periodontal ligament cells.

Inflamma-

tory and microbial stimuli can affect the levels of cathep-

sin S, which suggests that this protease might play a role

in oral inflammatory diseases.

Gene expression analy-

sis of gingival biopsies from periodontitis sites identified

cathepsin S as a central hub protein, which suggested

that it was involved in periodontitis.

Cathepsin S was

also shown to affect bone remodeling. When cathepsin

S expression was disrupted in knockout mice, the bal-

ance between adipocyte and osteoblast differentiation was

altered, bone turnover increased, and bone microarchitec-

ture changed.

Cathepsin S has also been implicated in

F I G U R E 3 Generalized estimated equation results show hs-CRP, cystatin-D, and PD-L1 associations with bleeding on probing (BOP) and deep periodontal probing depths (PPDs). Generalized estimated equation models, including measurements from time points 3, 6, and 12 months post-surgery, were used to evaluate associations between the indicated proteins and percent surfaces with (A-C) BOP, (D-F) PPD> 4 mm, or (G-I) PPD > 6 mm. All models were adjusted for gender, age, smoking, other diseases, and the number of teeth. Each graph includes trend lines with 95% CIs

osteoimmunological diseases, such as rheumatoid arthri- tis, and in vascular and metabolic complications related to obesity, like diabetes mellitus,

diseases shown to asso- ciate with periodontal diseases.

Studies have also shown that cathepsin S knockout mice had a reduced disease bur- den in experimental models of arthritis.

However, a potential direct or indirect role of cystatin-D in periodontal diseases needs to be further evaluated.

In addition, we found that PD-L1 was associated with BOP and PPD. PD-L1 is present on macrophages and

antigen-presenting cells. PD-L1 can also be induced in

other cell types, including endothelial, epithelial, and

B cells, during inflammation.

When PD-L1 inter-

acts with the programmed death protein, it can adjust

the immune response by down-regulating active T-cells,

which protects the host from uncontrolled immune

responses and inflammation-induced tissue damage.

However, elevated PD-L1 expression might lead to a PD-

L1-dependent attenuation of the immune system during

inflammation or even during acute inflammation, which

could be life threatening; for example, during sepsis.

Periodontitis-associated bacteria (e.g., Porphyromonas gin- givalis) can induce elevations in PD-L1 expression.

This activity down-regulates the immune response, attenu- ates bacterial clearance, and thus, facilitates a chronic bacterial infection. Indeed, persistently elevated PD-L1 levels can drive a chronic inflammatory disease, like periodontitis.

This study had some strengths and limitations that should be recognized and considered when evaluating the results. The study strengths included the prospective study design with frequent follow-ups and the focus on the recurring frequency and progression of the disease for 12 months after surgery. Furthermore, serum was col- lected at each time points and screened against a wide panel of inflammatory markers. A major limitation was the modest clinical sample size. Still, this small sample size facilitated relatively frequent recalls. The selected study design did not include a healthy control group. There- fore, we might potentially have overlooked inflammatory markers that were elevated in a population affected by periodontitis. However, that was not the focus of this study as it is changes in protein levels after treatment that we are investigating. In future studies, the set of serum proteins that we found were associated with peri- odontitis should be validated in an independent sam- ple, which preferably includes a large biobank of mate- rial and reliable periodontitis scores. Further studies are also needed to evaluate if serum levels of these biomark- ers could be mirrored in the local environment such as gingival crevicular fluid or saliva, which are more easily accessible.

5 CONCLUSION

This pilot study showed that clinical signs of periodontal disease relapse after surgery were associated with changes in several inflammatory associated serum proteins. In particular, relapses were associated with elevated PD-L1 and hs-CRP concentrations and reduced cystatin-D con- centrations. Increased PD-L1 expression can lead to T- cell inactivation, local immune suppression, and attenu- ated bacterial clearance, which facilitate chronic bacte- rial infections and can potentially drive chronic inflamma- tion. The increase in hs-CRP might indicate the involve- ment of a pathway in common with other diseases or a response to unresolved PD-L1-driven inflammation. The reduced cystatin-D levels could lead to bone resorption by osteoclasts and/or increased cathepsin activity, which leads to tissue destruction and can induce auto-immune processes.

A C K N O W L E D G M E N T S

This work received financial support from the the County Council of Västerbotten, Sweden, Spjutspetsmedel, grant number RV 396172146, and the County Council of Väster- botten, Sweden, Internal Research Foundation, grant number RV 396172134 and the Public Dental Health County Council of Gävleborg, Gävle, Sweden, and from the Centre for Research and Development, Uppsala Univer- sity/Region Gävleborg, Gävle, Sweden. None of the fund- ing bodies had any influence on the study design, the data collection, analysis, or interpretation, or the writ- ing of the manuscript. We also want to acknowledge SciLifeLab and the Clinical Biomarkers facility, Uppsala, Sweden.

C O N F L I C T S O F I N T E R E S T

The authors have no conflicts of interest to declare.

A U T H O R S C O N T R I B U T I O N S

Anders Esberg: data analysis and interpretation; drafting and critical revision of the manuscript

Catrine Isehed: study conception and design; data anal- ysis and interpretation; drafting and critical revision of the manuscript.

Anders Holmlund: study conception and design; data interpretation; and critical revision the manuscript.

Susanne Lindquist: data analysis and interpretation;

drafting and critical revision of the manuscript.

Pernilla Lundberg: data acquisition, analysis, and inter- pretation; drafting, and critical revision of the manuscript.

All authors gave final approval of the manuscript and agree to be accountable for all aspects of the work.

D A T A AVA I L A B I L I T Y S T A T E M E N T The data that support the findings of this study are available from the corresponding author upon reasonable request.

O R C I D

Anders Esberg https://orcid.org/0000-0002-4430-8125

R E F E R E N C E S

1. Loe H, Theilade E, Jensen SB. Experimental gingivitis in man. J Periodontol. 1965;36:177-187.

2. Yucel-Lindberg T, Bage T. Inflammatory mediators in the patho- genesis of periodontitis. Expert Rev Molecular Med [Electronic Resource]. 2013;15:e7.

3. Dahlen G, Basic A, Bylund J. Importance of virulence factors for the persistence of oral bacteria in the inflamed gingival crevice and in the pathogenesis of periodontal disease. J Clin Med.

2019:8.

4. Pihlstrom BL, Michalowicz BS, Johnson NW. Periodontal dis- eases. Lancet. 2005;366:1809-1820..

5. Bernabe E, Marcenes W, Hernandez CR, et al. Global, regional, and national levels and trends in burden of oral conditions from 1990 to 2017: a systematic analysis for the global burden of dis- ease 2017 study. J Dental Res. 2020;99:362-373.

6. Holmlund A, Lampa E, Lind L. Poor response to periodontal treatment may predict future cardiovascular disease. J Dental Res. 2017;96:768-773..

7. Hajishengallis G. Periodontitis: from microbial immune subver- sion to systemic inflammation. Nat Rev Immunol. 2015;15:30-44.

8. Lundberg K, Wegner N, Yucel-Lindberg T, Venables PJ. Peri- odontitis in RA-the citrullinated enolase connection. Nat Rev Rheumatol. 2010;6:727-730.

9. Preshaw PM. Host modulation therapy with anti-inflammatory agents. Periodontology 2000. 2018;76:131-149.

10. Kushner I, Samols D, Magrey M. A unifying biologic expla- nation for “high-sensitivity” C-reactive protein and “low- grade” inflammation. Arthritis Care Res (Hoboken). 2010;62:

442-446.

11. Hasturk H, Kantarci A. Activation and resolution of periodon- tal inflammation and its systemic impact. Periodontology 2000.

2015;69:255-273.

12. Loos BG, Van Dyke TE. The role of inflammation and genetics in periodontal disease. Periodontology 2000. 2020;83:26-39.

13. von Elm E, Altman DG, Egger M, et al. The strengthen- ing the reporting of observational studies in epidemiology (STROBE) statement: guidelines for reporting observational studies. Lancet. 2007;370:1453-1457.

14. Bostrom EA, Kindstedt E, Sulniute R, et al. Increased eotaxin and MCP-1 levels in serum from individuals with periodontitis and in human gingival fibroblasts exposed to pro-inflammatory cytokines. Plos One. 2015;10:e0134608.

15. Buhlin K, Gustafsson A, Pockley AG, Frostegard J, Klinge B.

Risk factors for cardiovascular disease in patients with periodon- titis. Euro Heart J. 2003;24:2099-2107.

16. Buhlin K, Hultin M, Norderyd O, et al. Risk factors for atherosclerosis in cases with severe periodontitis. J Clin Peri- odontol. 2009;36:541-549.

17. Lopez NJ, Quintero A, Casanova PA, Ibieta CI, Baelum V, Lopez R. Effects of periodontal therapy on systemic markers of inflam- mation in patients with metabolic syndrome: a controlled clini- cal trial. J Periodontol. 2012;83:267-278.

18. Demmer RT, Trinquart L, Zuk A, et al. The influence of anti- infective periodontal treatment on C-reactive protein: a system- atic review and meta-analysis of randomized controlled trials.

Plos One. 2013;8::e77441.

19. Koppolu P, Durvasula S, Palaparthy R, et al. Estimate of CRP and TNF-alpha level before and after periodontal therapy in cardio- vascular disease patients. Pan Afr Med J. 2013;15:92.

20. Sanz M, Del Castillo AM, Jepsen S, et al. Periodontitis and car- diovascular diseases. Consensus report. Glob Heart. 2020;15:1.

21. Wang F, Zhang C, Ge W, Zhang G. Up-regulated CST5 inhibits bone resorption and activation of osteoclasts in rat models of osteoporosis via suppression of the NF-kappaB pathway. J Cell Mole Med. 2019;23:6744-6754.

22. Freije JP, Balbin M, Abrahamson M, et al. Human cystatin D.

cDNA cloning, characterization of the Escherichia coli expressed inhibitor, and identification of the native protein in saliva. J Biol Chem. 1993;268:15737-15744.

23. Balbin M, Hall A, Grubb A, Mason RW, Lopez-Otin C, Abra- hamson M. Structural and functional characterization of two allelic variants of human cystatin D sharing a characteristic inhi- bition spectrum against mammalian cysteine proteinases. J Biol Chem. 1994;269:23156-23162.

24. Dickinson DP. Cysteine peptidases of mammals: their biologi- cal roles and potential effects in the oral cavity and other tissues in health and disease. Crit Rev Oral Biology Med. 2002;13:238- 275.

25. Pietschmann P, Foger-Samwald U, Sipos W, Rauner M. The role of cathepsins in osteoimmunology. Crit Rev Eukaryotic Gene Express. 2013;23:11-26.

26. Petanceska S, Canoll P, Devi LA. Expression of rat cathepsin S in phagocytic cells. J Biol Chem. 1996;271:4403-4409.https://doi.

org/10.1074/jbc.271.8.4403.

27. Zavasnik-Bergant V, Sekirnik A, Golouh R, Turk V, Kos J. Immunochemical localisation of cathepsin S, cathepsin L and MHC class II-associated p41 isoform of invariant chain in human lymph node tissue. Biological Chem. 2001;382:799- 804.

28. Memmert S, Damanaki A, Nogueira AVB, et al. Role of cathep- sin S in periodontal inflammation and infection. Mediators Inflamm. 2017;2017:4786170.

29. Song L, Yao J, He Z, Xu B. Genes related to inflammation and bone loss process in periodontitis suggested by bioinformatics methods. BMC Oral Health. 2015;15:105.

30. Rauner M, Foger-Samwald U, Kurz MF, et al. Cathepsin S con- trols adipocytic and osteoblastic differentiation, bone turnover, and bone microarchitecture. Bone. 2014;64:281-287.

31. Arnlov J. Cathepsin S as a biomarker: where are we now and what are the future challenges?. Biomarkers Med. 2012;6:

9-11.

32. Hou WS, Li W, Keyszer G, et al. Comparison of cathepsins K and S expression within the rheumatoid and osteoarthritic syn- ovium. Arthritis Rheumatism. 2002;46:663-674.

33. Jobs E, Riserus U, Ingelsson E, et al. Serum cathepsin S is asso- ciated with decreased insulin sensitivity and the development of type 2 diabetes in a community-based cohort of elderly men.

Diabetes Care. 2013;36:163-165.

34. Nakagawa TY, Brissette WH, Lira PD, et al. Impaired invari- ant chain degradation and antigen presentation and diminished collagen-induced arthritis in cathepsin S null mice. Immunity.

1999;10:207-217.

35. Yang H, Kala M, Scott BG, Goluszko E, Chapman HA, Chris- tadoss P. Cathepsin S is required for murine autoimmune myasthenia gravis pathogenesis. J Immunol. 2005;174:1729- 1737.

36. Chen Y, Zhang J, Guo G, et al. Induced B7-H1 expression on human renal tubular epithelial cells by the sublytic termi- nal complement complex C5b-9. Mole Immunol. 2009;46:375- 383..

37. Freeman GJ, Long AJ, Iwai Y, et al. Engagement of the PD-1 immunoinhibitory receptor by a novel B7 family member leads to negative regulation of lymphocyte activation. J Experimental Med. 2000;192:1027-1034.

38. Yamazaki T, Akiba H, Iwai H, et al. Expression of programmed death 1 ligands by murine T cells and APC. J Immunol.

2002;169:5538-5545.

39. Francisco LM, Sage PT, Sharpe AH. The PD-1 pathway in tolerance and autoimmunity. Immunol Rev. 2010;236:219- 242.

40. Patil NK, Bohannon JK, Sherwood ER. Immunotherapy: a promising approach to reverse sepsis-induced immunosuppres- sion. Pharmacological Res. 2016;111:688-702.

41. Groeger S, Jarzina F, Mamat U, Meyle J. Induction of B7-H1 receptor by bacterial cells fractions of Porphyromonas gingi- valison human oral epithelial cells: b7-H1 induction by Por- phyromonas gingivalisfractions. Immunobiology. 2017;222:137- 147.

42. Vakkila J, Lotze MT. Inflammation and necrosis promote tumour growth. Nat Rev Immunol. 2004;4:641-648.

S U P P O R T I N G I N F O R M A T I O N

Additional supporting information may be found online in the Supporting Information section at the end of the article.

How to cite this article: Esberg A, Isehed C, Holmlund A, Lindquist S, Lundberg P Serum proteins associated with periodontitis relapse post-surgery: a pilot study. J Periodontol. 2021;1–10.

https://doi.org/10.1002/JPER.21-0089