Correlation between different genotypes of human cytomegalovirus and Epstein-Barr virus and peri-implant tissue status
Associate Professor Sasha Jankovic
Department of Periodontics
School of Dentistry
The University of Belgrade
Dr Subotica 4 Street
Background: The purpose of this study was to estimate the prevalence of different genotypes of human cytomegalovirus (HCMV) and Epstein-Barr virus (EBV) in peri-implantitis and mucositis sites, and to evaluate the correlation between herpesvirus presence and clinical parameters.
Methods: A total of 80 dental implants (mean time of loading, 4.16 ± 1.8 years) were evaluated during the course of the study (30 peri-implantitis, 25 mucositis and 25 healthy peri-implant sites). The following clinical parameters were assessed: visible plaque index, bleeding on probing, suppuration and probing depth. A polymerase chain reaction (PCR) assay was used to identify the presence of different HCMV and EBV genotypes in peri-implant tissue plaque samples.
Results: HCMV-2 was detected in 53.3% and EBV-1 in 46.6% of the 30 peri-implantitis sites evaluated. By contrast, HCMV-2 was not detected in healthy periodontal sites and EBV-1 was detected in one healthy site. A statistically significant correlation was found between the presence of HCMV-2 and EBV-1 genotypes and clinical parameters of peri-implantitis.
Conclusions: The results from the present study confirmed the high prevalence of HCMV-2 and EBV-1 in the peri-implant tissue plaque of peri-implantitis sites and suggests a possible active pathogenic role of the viruses in peri-implantitis.
Abbreviations and acronyms:
bleeding on probing
polymerase chain reaction
restriction fragment length polymorphism
Numerous data in the literature suggest that bacterial infection plays the most important role in the failure of dental implants.1–3 Peri-implantitis is characterized by soft tissue inflammation, as well as bone loss around an osseointegrated implant in occlusal function.4–6 Mucositis is characterized by the appearance of an inflammation limited to the peri-implant mucosa and reversible with appropriate treatment. Infections of the implant-bearing soft and hard tissues initiate peri-implantitis, which shares with periodontitis a similar bacterial flora. High amounts of periodontal pathogens such as Porphyromonas gingivalis, Aggregatibacter actinomycetemcomitans, Tanerella forsythia, Treponema denticola and Prevotella intermedia in supra and subgingival biofilms7–9 are also detected in peri-implantitis sites.
Growing evidence suggests that certain viruses may play a role in the pathogenesis of periodontitis. In particular, DNA from herpesviruses such as human cytomegalovirus (HCMV) and Epstein-Barr virus (EBV) has been detected in high percentages of subgingival plaque samples from periodontitis patients.10–12 Since periodontitis is primarily initiated by microbial accumulation, the interaction of herpesviruses (HCMV, EBV) with periodontopathic microorganisms presents an important area of investigation. It has been proposed that periodontal herpesvirus activation could destabilize local host immune responses, which would promote subgingival colonization and the proliferation of periodontal bacterial pathogens, resulting in periodontal tissue destruction.11 HCMV infects periodontal monocytes/macrophages and T-lymphocytes, and EBV infects periodontal B-lymphocytes.13 HCMV and EBV infected inflammatory cells produce tissue-destroying cytokines and may harm the ability to defend against periodontopathogen bacterial challenge.11,13–15 Periodontal sites demonstrating HCMV and EBV presence also have a tendency to exhibit elevated levels of Porphyromonas gingivalis and Actinobacillus actinomycetemcomitans, important periodontopathogens.14,15 If herpesviruses possess the biological mechanisms to promote periodontal tissue destruction, then the role and influence of herpesviruses in peri-implantitis has to be carefully evaluated. A recent pilot study16 confirmed the high prevalence of HCMV and EBV in subgingival plaque of peri-implantitis sites and suggested a possible active pathogenic role of the viruses in peri-implantitis.
Recent literature has demonstrated that different genotypes of HCMV and EBV might exhibit different pathogenic abilities in other non-oral diseases.17–19 According to the sequence of the gB gene, which encodes a glycoprotein in the outer membrane of HCMV, the virus could arbitrarily be divided into four genotypes (gB genotype I [gB-I to gB-IV]).19 HCMV gB-I does not infect T lymphocytes, whereas types II and III do.19 Two types of EBV exist, based on the allelic polymorphisms in the latent gene sequences encoding EBV nuclear antigen 2 (EBNA2). Wu et al.20 confirmed that different HCMV and EBV genotypes are associated and correlated with severity of periodontal disease. These findings clearly indicated that HCMV gB-II (HCMV-2) and EBV-1 are the dominant virus genotype detected in subgingival samples in aggressive and chronic periodontitis. HCMV-2 co-infection with EBV-1 is closely associated with periodontal tissue inflammation and destruction. HCMV-1 and EBV-2 genotypes were more often detected in healthy individuals.
In view of the fact that our understanding of the aetiopathogenesis of peri-implantitis is still deficient, it is critical to identify the predominant genotypes of HCMV and EBV present in subgingival samples to further clarify their roles in the aetiology of peri-implantitis.
Careful assessment of the prevalence of HCMV and EBV is required because some geographical and population differences might have significant impact on the prevalence of herpesviruses in periodontitis and peri-implantitis. Imronito et al.21 reported that in a Brazilian population EBV (40%) was prevalent compared with HCMV (20%) in chronic periodontitis.
Therefore, the aim of this pilot study was to evaluate the presence of HCMV and EBV viruses in peri-implantitis sites and investigate the correlation of virus presence within clinical parameters. Virus screening was also performed on mucosal sites as mucositis may be the clinical prequel to peri-implantitis. A polymerase chain reaction (PCR) assay was used to identify the presence of different HCMV and EBV genotypes in subgingival plaque samples from 30 peri-implantitis, 25 mucositis and 25 healthy peri-implant sites.
Materials and Methods
Eighty patients from the Clinic for Periodontology, School of Dentistry, University of Belgrade, Serbia were enrolled for this clinical study. The University Ethics Committee approved the protocol for human subjects (EAp1089). Inclusion criteria were subjects treated with at least one Biohorizons® (Birmingham, AL, USA) tapered implant in function for at least one year. Subjects were systemically healthy non-smokers. Patients were periodontally healthy or, if they presented a history of periodontal disease, they should have been periodontally treated and engaged in a supportive periodontal therapy. Exclusion criteria were use of systemic antibiotics, anti-inflammatory drugs and oral antimicrobial agents within the preceding six months and the presence of herpetic infection (self-reported) during the last six months. Implant exclusion criteria were mobility, previous peri-implant treatment and unscrewed prostheses. Subjects were informed of the characteristics of the study and gave their written consent to the described procedures.
All clinical examinations were performed by the same trained and calibrated operator (PMD). The following parameters were assessed at six peri-implant sites (mesio, medio, disto/buccal and lingual) using a periodontal probe (Hu-Friedy, Chicago, IL, USA): (1) visible plaque index (PI) – presence or absence of plaque along the mucosal/gingival margin; (2) bleeding on probing (BOP) – presence or absence of bleeding of up to 15 seconds after gentle probing; (3) suppuration (SUP) – presence or absence of spontaneous or after probing SUP; and (4) probing depth (PD) – distance (mm) between the mucosal/gingival margin and the bottom of the sulcus/pocket.
According to clinical and radiographic data, the implants were divided into one of the following groups in accordance with Máximo et al.22: (1) healthy – PD <4 mm without BOP, SUP and radiographic bone loss; (2) mucositis – BOP, absence of radiographic bone loss or SUP; and (3) peri-implantitis: PD ≥5 mm with BOP and/or SUP and concomitant radiographic bone loss (bone loss ≥3 threads until the half of implant length). If the same subject had healthy implants and implants with mucositis or peri-implantitis, he/she was included in only one group based on the worst diagnosis as follows: peri-implantitis, mucositis and healthy. In addition, all the implants within the same clinical diagnosis per subject were included in the study.
Clinical parameters (PI, BOP, SUP, PD) and frequency of detection of HCMV and EBV were compared between the groups using the Mann-Whitney test. Statistical correlation between the presence of different genotypes of HCMV or EBV subgingivally and the clinical parameters of peri-implantitis were computed using Fisher’s Exact test. For all of the above, the significance level was set at 0.05.
After removing the supragingival plaque with sterile cotton pellets, the sites were isolated with cotton rolls and gently dried with an air syringe. Subgingival material was collected by means of a sterile curette. After gently inserting the curette into the bottom of the periodontal site, a single stroke was taken to remove subgingival debris.
Polymerase chain reaction procedures
DNA extraction and PCR for the detection of HCMV and EBV was carried out according to Saygun and Slots,23 and Wu et al.20 A nested PCR method was used to detect the DNA of HCMV, EBV-1 and EBV-2. The sequences of the HCMV-specific outer primers were 5′GGC ATC AAG CAA AAA TCT-3′ (forward primer) and 5′-CAG TTG ACG GTA CTG CAC-3′ (reverse primer). The inner primers for HCMV were 5′-TGG AAC TGG AAC GTT TGG G-3′ (forward primer) and 5′-GAA ACG CGC GGC AAT CGG-3′ (reverse primer). The sequences of the EBV-specific outer primers were 5′-AGG GATG CCT GGA CAC AAG A-3′ (forward primer) and 5′-TGG TGC TGC TGG TGG TGG CAA-3′ (reverse primer). The inner primers for EBV-1 were 5′-TCT TGA TAG GGA TCC GCT AGG ATA-3′ (forward primer) and 5′-ACC GTG GTT CTG GAC TAT CTG GAT C-3′ (reverse primer). The inner primers for EBV-2 were 5′-CAT GGT AGC CTT AGG ACA TA-3′ (forward primer) and 5′-AGA CTT AGT TGA TGC TGC CCT AG-3′ (reverse primer).
Samples were initially denatured at 94 °C for 5 minutes, followed by 30 cycles, which included denaturation for 30 seconds at 94 °C, annealing for 30 seconds at 59 °C, and extension for 30 seconds at 72 °C, with a final extension at 72 °C for 5 minutes. Controls included HCMV and EBV positive and negative cell lines. Specificity was confirmed by determining the size of the amplicons and retesting of positive samples. PCR products were detected by electrophoresis in a 1.5% agarose gel containing 0.5 μg/ml ethidium bromide. Gels were analysed using Quantity One software (BioRad Laboratories, Hercules, CA, USA).
Restriction endonuclease digestion analysis
The amplicons were further identified by restriction fragment length polymorphism (RFLP) analysis. The HCMV gB gene fragments were digested with endonucleases RsaI and Hinf I (TaKaRa). After RsaI digestion, the HCMV products were cleaved into two fragments (gB-I, 239 and 66 bp; gB-II, 239 and 63 bp) or three fragments (gB-III, 195, 63, and 41 bp; gB-IV 195, 66, and 44 bp). The HCMV products were separated into another two fragments (gB-II, 202 and 100 bp; gB-III, 202 and 97 bp) or three fragments (gB-I and gB-IV, 202, 67, and 36 bp) with endonuclease HinfI. For further identification of EBV, AfaI was used to digest the 497-bp amplicon of EBV-1 into 355- and 142-bp fragments, while StuI was used to turn the 165-bp amplicon of EBV-2 into 118- and 47-bp fragments. The restriction enzyme digests were resolved in an 2.5% agarose gel containing 1 μg/ml of ethidium bromide.
Blood samples were collected by venipuncture in tubes without anticoagulant. Blood samples were taken on the same day as subgingival sampling. Serum was separated after centrifugation and stored at −20 °C. Specific anti-VCA (Viral Capsid Antigen) IgG and IgM antibodies to EBV and IgG and IgM class antibodies to cytomegalovirus in patients’ serum samples were detected by enzyme-linked immunosorbent assays using commercial kits (Dia.Pro Diagnostic Bioprobes Srl, Milan, Italy).
The study population consisted of 80 individuals: 34 females and 46 males (mean age 46.10 ± 12.5 years). A total of 80 dental implants (mean time of loading, 4.16 ± 1.8 years) were evaluated during the course of the study. Thirty peri-implantitis, 25 mucositis and 25 healthy peri-implant sites were evaluated. No significant differences were observed for demographic and clinical parameters among the three clinical groups at baseline (Table 1).
Table 1. Demographic and clinical characteristics
|Number of patients||30||25||25|
|Age (years; mean ± SD)||46.3 ± 8.6||45.6 ± 11||42.4 ± 10.5|
|Gender; F (female)/M (male)||F 14, M 16||F 9, M 16||F 11, M 14|
|Number of implants||30||25||25|
|Time of loading (years/mean ± SD)||3.6 ± 1.4||4.1 ± 2.0||4.2 ± 1.5|
|Mean probing pocket depth in mm||5.90 ± 0.91*||2.75 ± 0.76*||1.55 ± 0.60|
|% sites showing bleeding upon probing||100%*||100%*||0%|
|% sites showing suppuration upon probing||65%*||0%||0%|
|Mean Plaque Index||1.5 ± 0.6*||0.6 ± 0.3||0.3 ± 0.2|
The mean clinical parameters of all the implants evaluated (mucositis, peri-implantitis and healthy) are also presented in Table 1. The percentage of sites with plaque and BOP were significantly higher around mucositis and peri-implantitis when compared with healthy implants (p < 0.05). The mean PD detected in the peri-implantitis group was 5.90 ± 0.91 mm. The recorded mean PD in the mucositis and healthy group was 2.75 ± 0.76 mm and 1.55 ± 0.60 mm, respectively. The mean PD around the implants was significantly higher in peri-implantitis, followed by mucositis and healthy implants (p < 0.05). Sixty-five per cent of sites showed suppuration upon probing in the peri-implantitis group.
Table 2 shows the distribution of herpesviruses. HCMV-2 was detected in 16 (53.3%), HCMV-1 in 4 (13.33%), EBV-1 in 11 (36.66%) and EBV-2 in 3 (10%) of the 20 peri-implantitis sites evaluated. By contrast, HCMV-2 and EBV-1 were not detected in healthy peri-implant sites. HCMV-1 was recorded in 2 (8%) and EBV-2 was recorded in 2 (8%) healthy peri-implant sites. Co-infection by HCMV and EBV were recorded in 19 (33.3%) of the peri-implantitis sites. Herpesviruses detected in co-infection were EBV-1 and HCMV-2. In the mucositis group, HCMV-2 was detected in 6 (24%), HCMV-1 in 2 (8%), EBV-1 in 8 (36%) and EBV-2 in 1 (4%) of the 25 evaluated sites. No co-infection was detected in the mucositis and healthy group. Differences between distribution of HCMV-2 and EBV-1 in peri-implantitis and healthy sites were significant (p < 0.05). The same relationship was recorded between the peri-implantitis and mucositis group (p < 0.05) when HCMV-2 was evaluated. There was no significant difference between distribution of HCMV-1 and EBV-2 in peri-implantitis and healthy sites (p > 0.05). EBV-1 detection in mucositis sites was significantly higher than in the healthy sites. There was a significantly higher percentage of sites showing absence of HCMV and EBV in the healthy group (76%) compared with 26.6% recorded in the peri-implantitis group. The same relation was detected between sites showing absence of HCMV and EBV in the mucositis (40%) and healthy group.
Table 2. Occurrence of HCMV and EBV virus genotypes in 20 peri-implantitis, 18 mucositis and 18 peri-implantitis healthy sites
|Peri-implantitis (n = 30)||4 (13.33%)||16 (53.3%)||11 (36.66%)||3 (10%)||10 (33.3%)||8 (26.6%)|
|Mucositis (n = 25)||2 (8%)|| 6 (24%)|| 8 (36%)||1 (4%)||0||10 (40%)|
|Healthy sites (n = 25)||2 (8%)|| 0|| 1 (4%)||2 (8%)||0||19 (76%)|
|Peri-implantitis/mucositis||0.37|| 0.03*|| 0.49||0.36|| ||0.20|
|Mucositis/healthy sites||0.49|| 0.07|| 0.027*||0.33|| ||0.02*|
|Peri-implantitis/healthy sites||0.37|| 0.00*|| 0.02*||0.45|| ||0.00*|
Table 3 presents the relationship between different genotypes of HCMV, EBV and clinical severity of peri-implantitis. Fifteen peri-implantitis sites with detected HCMV-2 and/or EBV-1 tended to demonstrate more severe clinical symptoms and signs than the seven with no specific virus genotypes (HCMV-1, EBV-2) and seven showing neither HCMV nor EBV presence. The highest level of BOP, SUP and PD was found within peri-implantitis showing HCMV-2 and EBV-1 co-infection. Seven peri-implantitis sites showing no presence of HCMV and EBV demonstrated enhanced results for BOP, SUP and PD. Suppuration was recorded in 100% of peri-implantitis showing HCMV-2 and EBV-1 co-infection and in 16.6% of the group showing absence of HCMV and EBV. BOP was detected in all evaluated peri-implantitis sites.
Table 3. Relationship between herpesviruses and clinical severity of peri-implantitis
The relationship of presence of subgingival HCMV and EBV genotypes with peri-implant tissue status is presented in Table 4. A statistically significant correlation was found between presence of HCMV-2 subgingivally and peri-implantitis (Fisher’s Exact test, p = 0.002). A statistically significant correlation was also recorded between presence of EBV-1 subgingivally and peri-implantitis (Fisher’s Exact test, p = 0.003). This study failed to show statistically significant correlation between presence of HCMV-1 or EBV-2 subgingivally and peri-implantitis (Fisher’s Exact test; p = 0.283, p = 0.350). A statistically significant correlation was recorded between presence of EBV-1 subgingivally and mucositis (p = 0.010). The present study also showed a statistically significant correlation between presence of HCMV-2 subgingivally and mucositis (p = 0.011).
Table 4. Correlation of subgingival HCMV and EBV virus genotypes with peri-implant tissue conditions
Serological findings data were evaluated separately for peri-implantitis, mucositis and healthy sites. A high percentage of participants (93.3% of peri-implantitis patients, 96% of mucositis and 88% healthy) were seropositive for anti-EBV IgG. Only one participant from the peri-implantitis group exhibited seropositivity for anti-EBV IgM (3.33%). Seropositivity for anti-HCMV IgG was recorded in 86.7% of peri-implantitis patients, 88% of mucositis patients and 88% of healthy patients.
Studies on a viral cause for periodontitis mark a turning point in periodontal research, which until recently was centered almost exclusively on a bacterial aetiology. The aetiopathogenic, clinical and microbial characteristics of peri-implantitis are similar to chronic periodontitis as well as mucositis and gingivitis.7–9 Human herpesviruses seem to play an important role in the aetiopathogenesis of severe types of periodontitis.8,13 Of the eight human herpesviruses, cytomegalovirus and Epstein-Barr virus were detected in a majority of deep and active periodontitis sites.24,25 Genomes of the two herpesviruses occur at high frequency in chronic, progressive periodontitis in adults, and localized and generalized aggressive (juvenile) periodontitis.10,11,26,27
In view of the fact that the causal pattern of peri-implantitis is still incomplete, the present study was performed to determine the prevalence of different genotypes of EBV and HCMV in plaque obtained from individual subgingival sites of patients with peri-implantitis and to investigate the hypothesis that peri-implantitis is associated with HCMV and EBV viruses. The latest pilot study16 confirmed the high prevalence of HCMV and EBV in peri-implant plaque of peri-implantitis sites and demonstrated an association and correlation between the presence of HCMV, EBV and the status of peri-implantitis tissues. Also, the recent literature17–19 has demonstrated that different genotypes of HCMV and EBV might exhibit different pathogenicity, even in the periodontal tissues.20
Four types of HCMV gB genes are known but only two types (HCMV gB I, II –HCMV-1 and HCMV-2) were detected in subgingival samples of the present study. Both EBV genotypes were detected. This study found a significantly higher presence of HCMV-2 in subgingival samples from peri-implantitis lesions than from healthy peri-implant sites. Also, the rate of detection for EBV-1 was extensively higher in peri-implantitis patients than mucositis patients and peri-implant healthy individuals, suggesting that EBV-1 and HCMV-2 infection could be associated with the pathogenesis of peri-implantitis. In contrast, the rate of positivity for HCMV-1 and EBV-2 genotypes showed no significant difference between the peri-implantitis, mucositis and healthy groups. Presence of HCMV-2 and EBV-1 significantly correlates with peri-implantitis when compared with healthy peri-implant sites. This study failed to find a statistically significant correlation between the presence of HCMV-1 or EBV-2 at peri-implant sites and peri-implantitis and mucositis. Peri-implantitis sites dominated HCMV-2 (53.3%) and in mucositis sites, the most frequent virus was EBV-1 (36%).
To our knowledge, no study has yet published evaluating the association between peri-implantitis and the presence of specific genotypes of HCMV and EBV. Therefore, a comparison with the study by Wu et al.20 where the presence of different virus (HCMV and EBV) genotypes was correlated with clinical parameters of the periodontal disease was performed. The detection rate for HCMV and EBV genotypes when nested PCR was used at individual sites with periodontitis and peri-implantitis was similar. Prevalence of HCMV-1/2 and EBV-1/2 in peri-implantitis sites was 13.33%/53.3% and 36.6%/3%, respectively. Prevalence of HCMV-1/2 and EBV-1/2 in periodontitis sites was 10.5%/62% and 43.5%/8.2%, respectively. Significant discrepancy was detected between the positivity rate of HCMV and EBV genotypes recorded in healthy periodontal and peri-implant sites. HCMV-1/2 and EBV-1/2 was detected in the subgingival plaque of periodontally healthy sites in approximately 40/30% and in 27/2.5%, respectively. In the present study, recordings of HCMV-1/2 and EBV-1/2 genotypes in healthy peri-implant sites were 8/0% and 4/8%. The similar prevalence of HCMV and EBV genotypes in subgingival plaque in periodontal disease and peri-implantitis supports the theory of the related role of specific herpesvirus genotypes (HCMV-2 and EBV-1) on the aetiopathogenesis of periodontitis and peri-implantitis.
Assessment of the serological status of patients for anti-EBV and anti-HCMV IgG and IgM levels in serum appears to be a useful procedure in order to avoid the possibility of a clinical and subclinical viral infection that might affect findings from the peri-implant environment.10 Only one participant from the peri-implantitis group exhibited seropositivity for anti-EBV IgM. Elevated IgM titres are indicative of recent, even subclinical infection, with EBV. Seropositivity for anti-HCMV IgM was not recorded. Six out of the 80 participants did not display anti-EBV IgG in serum, therefore our subject sample was 93.5% seropositive. Ten out of 80 participants (12.5%) did not display anti-HCMV IgG in serum.
Careful evaluation of the prevalence of HCMV and EBV is required because some geographical and population differences might have a significant impact on the prevalence of herpesviruses in periodontitis and peri-implantitis.21 Therefore, the results of the present study may not apply to other population groups.
The most important finding of the present study is the high prevalence of HCMV-2 and EBV-1 in subgingival plaque of peri-implantitis sites and the possible active pathogenic role of herpesviruses in peri-implantitis. Our study demonstrates that different HCMV (HCMV-2) and EBV (EBV-1) genotypes are associated with disease, while others appear to be non-pathogenic. Within the limitations of this study population, our findings demonstrated an association and correlation between the presence of HCMV-2, EBV-1 and the status of peri-implantitis tissues. Future studies on the pathogenic mechanisms of these strains of HCMV and EBV should help to elucidate their role in the aetiology of peri-implantitis. Future research should be designed to evaluate whether clinical management of peri-implantitis has an impact on specific virus prevalence.