Assessment of Salivary Adipokines Resistin, Visfatin, and Ghrelin as Type 2 Diabetes Mellitus Biomarkers

Type 2 diabetes mellitus (T2DM) is emerging as a metabolic epidemic worldwide. Pathologically, dysregulation of many biological pathways precedes hyperglycemia and the clinical diagnosis of T2DM. Changing trajectories along the process of T2DM development necessitates frequent measurement of biomarkers for early identification of at-risk individuals and successful prevention. Increase in circulating inflammatory adipokines has been suggested as predictive of T2DM. Human saliva is an easily accessible biospecimen amenable for painless frequent collection and possesses nearly 50% of serum proteome. In this study, we measured the adipokines resistin, visfatin, TNF-α, and ghrelin as markers for T2DM in unstimulated whole saliva (UWS) using specific assay kits. Resistin and visfatin concentrations were significantly higher in T2DM saliva. Although the concentration of acylated or unacylated ghrelin was lower in diabetic saliva, the decrease was not significant. Since resistin and visfatin are biomarkers integral to T2DM pathology, their salivary assessments may receive clinical acceptance.


Introduction
e World Health Organization estimated that globally 422 million adults were living with diabetes in 2014 [1]. Type 2 diabetes mellitus (adult-onset/noninsulin-dependent diabetes: T2DM) accounts for 90-95% of all diabetes [2]. e disease develops insidiously through periods of increased insulin secretion, insulin resistance, impaired glucose tolerance, and β-cell dysfunction [3]. Consistently, the most acceptable markers for T2DM diagnosis are based on measurements of blood glucose and glycosylated hemoglobin c (HbA1c), an indicator of average glycemic control [4]. However, research elucidating the disease pathogenesis suggests that multiple mechanisms including chronic inammation, obesity, lipotoxicity, and oxidative stress contribute to the glucose dysregulation in T2DM [5,6]. Hence, several hypothesis-based nonglycemic biomarkers have been assessed as risk factors for diabetes [7,8].
Adipokines are polypeptides secreted by adipocytes, in ammatory cells, and other cells. ey regulate multiple physiological functions including energy balance, insulin sensitization, appetite regulation, and in ammatory response [9]. It has been suggested that activation of the adipokine resistin in the islet cells of the pancreas inhibits cell surface glucose transporters and thereby insulin signaling [10,11]. Visfatin, also known as pre-B-cell colonyenhancing factor (PBEF), has been described as an adipokine with a potential glucose-lowering e ect due to its nicotinamide phosphoribosyltransferase (NAMPT) activity [10,12,13]. Ghrelin, originally identi ed as a growth hormone secretagogue with orexigenic and lipogenic e ects, has also been shown to play signi cant roles in glucose regulation. While acylated ghrelin has been shown to exert hyperglycemic e ects leading to insulin resistance, the unacylated ghrelin counters hyperglycemia and enhances insulin sensitivity [14,15]. Biomarker studies showed that the circulating levels of resistin and visfatin are upregulated in T2DM [13,[16][17][18]. On the other hand, the plasma concentration of acylated ghrelin has been shown to be lower in T2DM individuals as well as in their healthy o spring [19][20][21].
Since monitoring of serological parameters typically involves invasive techniques with associated pain and distress, e orts are directed at identifying noninvasive measures for frequent monitoring of diabetes. Some of the alternative methods evaluated include assessing skin autouorescence for accumulation of advanced glycation endproducts and measuring analytes in exhaled breath, urine, or saliva [5,[22][23][24]. Human saliva is a rich reservoir of analytes consisting of over 3652 proteins and 12,562 peptides and shares nearly 51% of proteins and 79% of peptides with the serum proteome and peptidome, respectively [25,26]. Alterations in the salivary ow and composition in diabetes are well documented [27,28]. Both glucose and immunoreactive insulin are increased in saliva and are correlated with plasma levels in T2DM patients [29][30][31][32]. Circulating biomolecules are thought to reach saliva by either active (e.g., sIgA) or passive transportation (e.g., steroids) or ultra ltration (e.g., creatinine) or from crevicular uid [26,33]. e objective of this study is to compare the salivary levels of two proinammatory adipokines, namely, resistin and visfatin, and that of the anti-in ammatory adipokine ghrelin between healthy and T2DM individuals.

Study Population.
All participants were recruited from patients attending the Indiana University School of Dentistry after obtaining informed consent in full accordance with the Indiana University Institutional Review Board. e study population consisted of twenty periodontally healthy individuals with self-reported T2DM and HbA1c values. Twenty gender-matched individuals with no known oral or systemic condition were recruited as control group. It was estimated that this sample size will be su cient at 80% power to detect a di erence in means of 0.91, assuming a common standard deviation of 1 using a two group t-test with a two-sided signi cance level of 0.05.

Collection and Processing of Unstimulated Whole Saliva (UWS).
UWS was collected at approximately the same time of day by the drooling method as described [34,35]. Brie y, subjects were asked to refrain from eating or drinking for 2 h prior to saliva collection. At least 2 mL of UWS was collected by passively drooling into a chilled centrifuge tube for 5-10 min. e tubes were codi ed and transferred on ice to the laboratory for processing. Each sample was clari ed by centrifuging at 3500 rpm at 4°C for 10 min and stored in Complete ™ Protease Inhibitor Cocktail (Roche, Mannheim, Germany). e supernatant-clari ed saliva was stored at −80°C until further analysis.

Enzyme-Linked Immunosorbent Assay (ELISA) for
Resistin and Visfatin. All UWS samples were depleted of amylase and immunoglobulins by incubating serially with antihuman amylase mAb (1 : 2500, cat. no. ab8944; Abcam) and protein G beads (Miltenyi Biotec Inc.) at 4°C. Total protein of the precleaned saliva samples was determined by spectrophotometry and ranged between 1.3 and 7.7 mg/mL [36]. Volume equal to 1 µg of total protein in precleaned UWS was assessed for resistin, visfatin, and ghrelin per using speci c sandwich ELISA kits (item no. 10007610 81, part no. 579020-96, and part no. 10006306-96, resp.; Bertin Pharma/Cayman Chemical, Ann Arbor, MI, USA) following manufacturer's instructions. e cytokines TNF-α and IL-6 in saliva were measured using speci c ELISA kits (BD Biosciences, CA, USA) following the manufacturer's instructions.

Statistical Analysis.
For all biomarkers, statistical signi cance between the healthy and diabetes cohorts was determined by two-tailed paired t-tests; p < 0.05 was considered signi cant.

Clinical Characteristics.
e study cohort consisted of twenty individuals with self-reported T2DM and twenty healthy individuals (Table 1). e average age of T2DM cohort was 56.5 yrs and that of healthy group was 48 yrs. e average HbA1c value of the T2DM group was 5.4 ± 1.9%. Since the HbA1c values reported were measured within the past three months, the range is consistent with the diagnostic criteria for T2DM [37].

Salivary Cytokines in T2DM.
We observed that the UWS concentration of IL-6 and TNF-α was not signi cantly di erent between T2DM and healthy individuals with no periodontitis (Figure 1). Similar observation of comparable salivary IL-6 levels between systemically healthy and diabetes individuals with healthy periodontium has been reported by others [38].

Discussion
Escalating global burden of T2DM underscores the need for multipronged screening strategies for early identi cation of individuals at high risk [7,39]. Furthermore, elucidation of the molecular pathogenesis has shown that the processes that lead to T2DM are initiated very early with a long lag phase between the disease onset and the clinical diagnosis [6]. Many cross-sectional studies have evaluated multiple serum proteins as predictive biomarkers for T2DM [5,7,39]. Potential applications of salivary biomarkers for T2DM have gained importance with the establishment of shared characteristics of salivary and serum proteomes [26]. It has been suggested that the increased basement membrane permeability often associated with diabetes is a potential mechanism for the increased passage of proteins and metabolites from the exocrine glands as well as for the enhanced leakage of serum-derived components into whole saliva [26,33,40].
Clinical application of salivary components as potential biomarkers is likely to be better accepted for molecules that correlate with the pathological process of T2DM. Considerable evidence suggests that the T2DM is a multifactorial disease involving dysregulation of various biological pathways such as in ammation, adipokine signaling, and incretin signaling [6].
e prodiabetic e ects of the adipokine resistin have been attributed to inhibition of insulin signaling and a pro-in ammatory mechanism that culminates in β-cell loss [9,10,13,41]. e adipokine visfatin has been shown to exhibit glucose-lowering and insulinmimicking/insulin-sensitizing e ects [9,10,12,13]. Circulating levels of both resistin and visfatin are upregulated in T2DM [12,13,18,42]. Previously, others have shown positive correlation between the serum and salivary levels of these two adipokines [42][43][44]. Here, we observed that the salivary resistin and visfatin concentrations are signi cantly elevated in T2DM. Similar observations of elevated salivary visfatin and resistin have been reported earlier in chronic periodontitis and diabetes [42,43,45]. Ghrelin, the orexigenic peptide hormone also a ects glucose metabolism. Circulating levels of ghrelin rise before and fall after a meal, thereby contributing to appetite and weight gain [15]. Plasma concentration of ghrelin has been negatively correlated with insulin resistance [14,46]. We observed that the salivary concentration of both acylated and unacylated ghrelin was lower in T2DM saliva than that in healthy saliva although the di erence did not reach statistical signi cance. Others have reported signi cant reductions in acylated ghrelin in diabetic saliva [47]. e di erence may be attributed to the time and method of sample collection and preparation and the method of ghrelin assessment.

Conclusions
Population-based long-term studies suggest that the biomarker trajectories along the course of T2DM development diverge over time [48,49].
is suggests that repeated measures of mechanism-based biomarkers will increase the predictive value of diabetes risk scores. e noninvasive nature and the feasibility of frequent sampling for real-time monitoring are signi cant advantages of saliva over peripheral blood as specimen for diagnostic/prognostic applications. In addition to the practical bene ts of eliminating the need for a phlebotomist, reduced transmission of infectious disease by eliminating needle sticks and greater ease of testing of special populations of patients (e.g., institutional and children) make the assessment of biomarkers in human saliva an attractive economic strategy [24,50]. Elevated salivary resistin and visfatin in saliva that have also been shown to correlate with serum levels suggest that the two adipokines could represent potential noninvasive T2DM biomarkers [41,43,44]. However, caution must be exercised since the type of sample (stimulated/unstimulated; whole/glandular), circadian variations, and susceptibility to preprocessing as well as oral health/disease are some of the confounding parameters that should be addressed in the biomarker interpretations and implementation [51,52].

Conflicts of Interest
e authors do not have any con icts of interest to disclose.