Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
http://jdr.sagepub.com/
Published by:
http://www.sagepublications.com
On behalf of:
International and American Associations for Dental Research
Additional services and information for Journal of Dental Research can be found at:
Subscriptions: http://jdr.sagepub.com/subscriptions
Reprints: http://www.sagepub.com/journalsReprints.nav
Permissions: http://www.sagepub.com/journalsPermissions.nav
What is This?
Downloaded from jdr.sagepub.com at UNIVERSITY OF BRIGHTON on July 17, 2014 For personal use only. No other uses without permission.
ABSTRACT INTRODUCTION
Periodontitis and type 2 diabetes are co-morbid
conditions, both characterized by infectious
susceptibility. We investigated procalcitonin
Tsalivary
he simple and non-invasive nature of saliva collection and high-
sensitivity assay development has led to an emphasis on the promise of
biomarkers (Tabak, 2001). Saliva may reflect levels of therapeutic,
(ProCT) levels in the serum and saliva of persons hormonal, and immunologic molecules and can yield markers for infectious
with periodontitis and type 2 diabetes (n = 20), to and neoplastic diseases (Kaufman and Lamster, 2002). For the local
determine if these levels are altered by inflammatory process of periodontitis, salivary diagnostics may promote
periodontitis activity or by hyperglycemia. early diagnosis and aid in the monitoring of treatment (Miller et al., 2006).
Persons with severe periodontitis showed higher For some diagnostic purposes, salivary biomarkers may prove more useful
levels of salivary-ProCT than did those with than serum analysis (Streckfus and Bigler, 2005).
moderate periodontitis (241 ± 71 vs. 77 ± 516 Diabetes has long been theorized to influence the course of
pg/mL, p = 0.02) and higher levels than did periodontitis, and it has been well-established that persons with diabetes
healthy control individuals (118 ± 26 pg/mL, p = have an increased prevalence and severity of periodontitis (Löe, 1993).
0.05). Salivary-ProCT levels were correlated with Although the pathogenesis of periodontitis in diabetes remains unclear, the
bleeding-on-probing (r = 0.45, p = 0.05), as well co-morbidity of type 2 diabetes and periodontitis suggests that the infectious
as with HgbA 1c (r = 0.49, p = 0.03). Salivary up-regulation of the local periodontal inflammatory processes and the low-
levels of ProCT were higher than serum levels for level systemic inflammation involved with diabetic complications may
the periodontitis/diabetes group (152 ± 37 vs. 78 ± interact (Shoelson et al., 2006). Furthermore, the effect of periodontitis on
17 pg/mL, p = 0.02) and the control group (118 ± other disease processes has become a focus of study, with evidence
146 vs. 48 ± 17 pg/mL, p = 0.01). Persons with supporting the connection between periodontitis and poor outcomes in type
periodontitis and type 2 diabetes have salivary- 2 diabetes (Janket et al., 2005; Saremi et al., 2005; Shultis et al., 2007),
ProCT levels that reflect their degree of cardiovascular disease (Scannapieco et al., 2003a), and other systemic
periodontitis activity and hyperglycemia. This pathologies (Slavkin and Baum, 2000; Scannapieco et al., 2003b;
study demonstrates, for the first time, the presence Quagliarello et al., 2005).
of procalcitonin (ProCT), an established serum Procalcitonin (ProCT) circulates at very low levels normally, but rises
marker of infection, in saliva. dramatically in persons with systemic infection. It is used as a biomarker for
the presence and severity of severe infection, often with a sensitivity and
KEY WORDS: Saliva, procalcitonin, periodontitis, persistence unmatched by traditional clinical or routine laboratory tests
diabetes. (Snider et al., 1997; Whang et al., 1998; Becker et al., 2004). In the
individual with sepsis, ProCT, usually produced only in the thyroid to be
processed into calcitonin, assumes tissue-wide expression and secretion
(Müller et al., 2001). Although ProCT is a significant serum marker, to our
knowledge there is no comparable study in saliva. The present investigation
was initiated to study levels of serum and salivary-ProCT and to analyze
how these might be related to periodontal disease indicators and/or to
hyperglycemia. We hypothesized that periodontitis and type 2 diabetes may
act as a stimulus for local ProCT production.
630 Downloaded from jdr.sagepub.com at UNIVERSITY OF BRIGHTON on July 17, 2014 For personal use only. No other uses without permission.
(PD) of > 6 mm and moderate periodontitis showing PD of 4-6 vortex for 5 sec, centrifuged at 220 x g for 5 min, and divided into
mm on more than 30% of available sites (Wiebe and Putnins, 1-mL aliquots stored in microcentifuge tubes at -27°C, then
2000). The examination recorded a full-mouth charting, including thawed for assay.
bleeding-on-probing (BOP), and PD percentages, measured as the Written informed consent for participating individuals was
ratio of either BOP or a PD at a periodontal site and the total included in the study protocols, approved by the Institutional
number of periodontal sites per person (6 sites per tooth). The Review governing the Washington, DC, Veterans Affairs Medical
individual’s number of teeth, age, and smoking status were Center.
recorded. One-stage, full-mouth periodontal scaling and root-
planing therapy, an efficient and effective treatment for chronic Analysis for ProCT
periodontitis, was performed on each person (Wennström et al., The salivary-ProCT samples were analyzed by an
2005). Follow-up visits at 3 mos after the initial visit were aminoprocalcitonin assay (molecular weight 6222 Da). This
scheduled for all persons; 13 persons reported for this secondary ELISA assay (EIA) was performed with a R2B7 antiserum
visit and had saliva and serum samples taken. The individuals who (developed in rabbits against synthetic human aminoprocalcitonin
returned for the second visit were similar in initial HgbA1c (9.7 ± (Bachem, Torrance, CA, USA) at a final dilution of 1:80,000. The
2.3 vs. 10.6 ± 2.3%, p = 0.3) and BOP (31 ± 20 vs. 23 ± 25%, p = antiserum (25 L of 1:20,000) was pre-incubated with standards or
0.4) to the group lost to follow-up; average time to follow-up was unknowns (20-50 L) in 0.1 mL 0.13 M borate buffer (containing
84 ± 27 days. 0.2% gelatin and 0.05% Tween-20; pH = 7.5) at 4°C for 24-72 hrs
The serum of non-diabetic control individuals was analyzed to (depending on desired sensitivity) in a Reacti-BindGoat antirabbit
provide a control level for serum-ProCT (n = 34, 45 ± 12 yrs old). coated, 96-well microtiter plate (Pierce, Rockford, IL, USA). After
A separate group of control individuals without periodontal disease the plate was rinsed 5 times with PBS containing Tween-20 in an
provided salivary samples for salivary-ProCT analysis (n = 8, aged ELP-40 plate washer (Bio-Tek Instruments, Winooski, VT, USA),
42 ± 5 yrs old). These control groups were comprised of both a 2-ng quantity of biotin-human calcitonin (AnaSpec, San Jose,
males and females. CA, USA) in 0.1 mL borate buffer was added and incubated at
Unstimulated whole saliva was collected from the individuals room temperature in an orbital shaker (150 rpm) for 1 hr. After the
described above and was considered an admixture from salivary plate was rinsed 5 times, a 100- L quantity of 1:10,000
gland secretion, crevicular fluid, mucosal seepage, resident strepavidin peroxidase (Ultrasensitive, Sigma-Aldrich, St. Louis,
microflora, and debris. Saliva collection was performed by MO, USA) was added to the plate, and the plate was incubated in
individuals expectorating into 50-mL polyethylene centrifuge tubes an orbital shaker (150 rpm) for 60 min. The plate was again rinsed
containing one drop of Tween® 20 (Fisher Scientific, Fairlawn, NJ, 5 times, and a 0.1-mL quantity of a solution of o-
USA) until 5 mL had been collected (range of 4-5 min, phenylenediamine (0.75 mg/mL) in phosphate-citrate buffer
approximately 1 mL/min). Saliva samples were then agitated by containing perborate (Sigma-Aldrich, St. Louis, MO, USA) was
Table. Characteristics of Persons with Periodontitis/Diabetes before and after Periodontal Treatment
Periodontitis/Diabetes
Initial vs.
Periodontitis/Diabetes Periodontitis/Diabetes
Follow up Initial vs.
(matched data) Control
n 20 13
Age (yrs) 56 ± 2* 55 ± 2
Sex
(male/female) 20/0 13/0
Smoking status
(yes/no) 7/13 4/9
Number of teeth 23 ± 1 22 ± 1
P value P value
Downloaded from jdr.sagepub.com at UNIVERSITY OF BRIGHTON on July 17, 2014 For personal use only. No other uses without permission.
added to each well. Following an additional 30 min of incubation not significantly higher than those for the healthy control
in the orbital shaker (150 rpm), the results were read at 450 nm on individuals (152 ± 37 vs. 118 ± 3, p = 0.51); serum-ProCT
an Elx808 ELISA reader (Bio-Tek Instruments, Winooski, VT, values for the periodontitis/diabetes group were significantly
USA). After the enzyme reaction was stopped by the addition of 50 higher than those found in control serum (78 ± 17 vs. 48 ± 3
L 3 M H2SO4 to each of the wells, the plate was read again at pg/mL, p = 0.04; Table). Individuals with severe periodontitis
490 nm. Usually, the results obtained at 490 nm were used in had significantly higher levels of salivary-ProCT than did those
calculating the assay curves and results. For this assay, the with moderate periodontitis (241 ± 71 vs. 77 ± 516 pg/mL, p =
functional sensitivity was 20 pg/mL and the 50% B/Bo (initial 0.02), and also significantly higher levels than did healthy
binding of the labeled peptide to the antibody in the absence of control individuals (118 ± 26 pg/mL, p = 0.05). The difference
standards) was 240 pg/mL. Intra-assay reliability for this ELISA in serum-ProCT between persons with severe periodontitis and
was < 10%, and inter-assay reliability was < 18%. The serum those with moderate periodontitis did not reach significance
levels of ProCT were analyzed for ProCT (molecular weight, (107 ± 31 vs. 57 ± 6 pg/mL, p = 0.16), though that between
12,741 Da) by means of the KRYPTOR-TRACE technology those with severe periodontitis and healthy control individuals
(BRAHMS, Henningsdorf, Germany). Intra-assay reliability for did (102 ± 20 vs. 48 ± 3, p = 0.01; Fig. 1). The HgbA1c values
serum-ProCT was < 7%, and inter-assay reliability was < 10% were not different between these groups separated by
(data provided by the manufacturer and tested by the authors). The periodontal severity. Salivary-ProCT levels were not correlated
functional sensitivity for this assay was 60 pg/mL. Assays were with matched serum ProCT levels.
run in duplicate. Positive correlations exist between salivary-ProCT and BOP
Statistical Analysis (r = 0.45, p = 0.05, Fig. 2A), and between salivary-ProCT and
Statistical analysis was performed with the R statistical analysis HgbA1c values (r = 0.50, p = 0.03, Fig. 2B). Neither of these
package (R Development Core Team, 2006). Normally distributed parameters was correlated with PD. Possible confounding
continuous variables were expressed as means ± SEM. We used parameters were analyzed for data from this initial visit, with no
Student’s t tests to analyze differences in groups and performed further correlations being seen between salivary-ProCT or
Pearson’s correlation on data to indicate a linear relationship serum-ProCT and teeth number, age, smoking, WBC, or
between variables. P values < 0.05 were considered statistically glucose levels. Further, periodontal indicators were not
significant. correlated with HgbA1c values. Serum-ProCT was not correlated
with periodontal indicators, or with glycemic control.
Levels of salivary-ProCT were higher than matched serum-
RESULTS ProCT levels for the periodontitis/diabetes group (152 ± 37 vs.
Salivary-ProCT levels for the diabetes/periodontitis group were 78 ± 17 pg/mL, p = 0.02). Salivary-ProCT in the control group
Downloaded from jdr.sagepub.com at UNIVERSITY OF BRIGHTON on July 17, 2014 For personal use only. No other uses without permission.
DISCUSSION Figure 3. Levels of salivary-ProCT (■) were higher than matched serum-
This study showed the heretofore-unreported presence of ProCT levels (■) for the periodontitis/diabetes group (n = 20, 152 ± 37
vs. 78 ± 17 pg/mL, p = 0.02). Salivary-ProCT in the control group (■,
ProCT in saliva and its correlation with the degree of n = 8) was also higher than the serum-ProCT of the healthy control
periodontitis in type 2 diabetes. We hypothesized that individuals (■, n = 38, 118 ± 146 vs. 48 ± 17 pg/mL, p = 0.01).
periodontitis in diabetes may act as a stimulus for ProCT
production, since endotoxin is a potent stimulator for the
production of ProCT and can promote the systemic release of
calcitonin precursors from nearly all tissues of the body
(Dandona et al., 1994; Preas et al., 2001). Periodontitis is rationale for the evidence-based aggressive treatment of
initiated and promoted by pathogenic Gram-negative, periodontitis in the uncontrolled diabetic population. Mounting
endotoxin-producing bacterial infections (Mealey and Oates, evidence that systemic inflammation, particularly in
2006), and may promote a local up-regulation of ProCT in cardiovascular disease, is modulated with periodontitis severity
saliva. We found significantly higher mean levels of salivary- and improves with periodontal therapy supports this
ProCT for persons with severe periodontitis when compared interventional approach (Tonetti et al., 2007). The use of
with those with moderate periodontitis. As a confirmation of ProCT as a local as well as a systemic biomarker for
this effect, salivary-ProCT was found to be significantly inflammation and infection may prove useful for future
correlated with the periodontal indicator BOP, a measure of research in this field.
active periodontal inflammation. No correlation, however, was Periodontal therapy did not reduce BOP% or improve PD
found between salivary and serum levels of ProCT, which of the 13 individuals who returned for a follow-up visit. Poorly
argues against a simple reflection of serum components in the controlled diabetes is a risk factor for aggressive periodontitis
salivary fluid. Future work may determine the extent to which and contributes to treatment resistance. This may explain the
variations in stimulated and unstimulated salivary-ProCT might lack of a reduction in salivary-ProCT values for persons with
be related to salivary circadian rhythms, the fasting state, and treatment. Matrix metalloproteinase (MMP)-8, a putative
flow rate. Larger population samples, with age- and gender- salivary biomarker of periodontitis (Miller et al., 2006), is
matched controls, will be needed to compensate for the reduced with periodontal therapy (Choi et al., 2004; Emingil et
limitations of this present study. al., 2004) and does exist in higher levels in the saliva of an
The correlation between salivary-ProCT and HgbA1c values individual with diabetes (Collin et al., 2000; Kumar et al.,
may be another supportive indication of the underlying 2006), but has not clearly been shown to be reduced in a
connection between the local and systemic inflammatory states diabetic population with periodontal treatment. If salivary-
of periodontitis and type 2 diabetes (Temelkova-Kurktschiev et ProCT can be postulated as a salivary biomarker for
al., 2002; Syrenicz et al., 2006). Since serum-ProCT was not periodontal disease activity, its reduction with treatment could
linked with hyperglycemia in this study, an increasing proximal be used as an objective end-point and therapeutic goal for
expression of ProCT with worsening diabetic status or guided intervention, especially for the individual with diabetes.
periodontitis activity may be postulated. In this regard, Another member of the procalcitonin gene family,
procalcitonin levels have recently been measured in extremity adrenomedullin, has wide tissue distribution in epithelial
wound effluent and have been correlated with dehiscence in surfaces, and both it and calcitonin gene-related peptide
wartime extremity injuries, suggesting a local production of (CGRP) have antimicrobial activities, implicating this family of
ProCT at the extremity wound site (Forsberg et al., 2008). Our gene proteins in a role in mucosal defense (Marutsuka et al.,
findings show the comparable analogy of a local up-regulation 2001; Lopez and Martinez, 2002). Adrenomedullin has been
of ProCT found in saliva and its correlation with the diabetic found in saliva at higher concentrations than serum and is
complication of periodontitis. expressed in salivary tissue, and CGRP is found in saliva
Our results show that serum-ProCT was higher in our (Kapas et al., 2004; Lundy and Linden, 2004; Bellamy et al.,
periodontitis/diabetes group than in our control group, and that 2006). We have shown that salivary-ProCT levels were higher
it was further influenced by periodontitis severity, but not by than serum-ProCT levels for both our periodontitis/diabetes
glycemic control. This may form the basis for further research group and for the healthy control individuals. It is therefore
on the systemic influence of periodontitis on systemic disease. possible that our suggestion of the local production of ProCT in
The impact of the local chronic infection of periodontitis on the oral environment may have yet to determine the local
glycemic control and systemic inflammation may add to the functions in the gingival tissues’ response to be bacterial
Downloaded from jdr.sagepub.com at UNIVERSITY OF BRIGHTON on July 17, 2014 For personal use only. No other uses without permission.
Downloaded from jdr.sagepub.com at UNIVERSITY OF BRIGHTON on July 17, 2014 For personal use only. No other uses without permission.