aera of Bichemisry and Molecar Biology, Vek 40,No. 1, Jarry 2007, pp. 29-35 Biochemistry — mi Molecular Biology A Comparison of Ghrelin, Glucose, Alpha-amylase and Protein Levels in Saliva from Diabetics Suleyman Aydin* Deparonent of Biochemistry and Clinical Biochemikary, Firat Livers, ‘Maxlcal School (Frat Madical Canter), 23119 Ekvig, Turkey Received 23 Je 2006, Accerred 23 Augrst 2006 During the past decade, many salivary parameters have ‘been used to characterize disease states, Ghrelin (GAH) is ‘rom the stomach but also produced in a number of other ‘téssues including salivary glands. The aim of this work was ‘to examine the relationship between active (@GAH) and inactive (GAH) ghredin in the saliva and other salivary Parameters in type II diabetic patients and healthy controk. Salivary parameters were assessed in a single ‘measurement of unstimulated whole saliva from 20 obese and 20 norobes type II diabetes patients, and in 22 healthy contro. Total protein and alphwamylase were determined by colorimetric methods, and glucose by the slucose-axidase method. Saliva aGAH and dGAH levels ‘were measured using a commercial radioimmunoassay (RIA) kit. Salivary concentrations of GAH and dGAH ghrelin were more markedly decreased in obese diabetic subjects than in the two other groups. Ghucese and alpha amylase levels were higher in diabetic subjects than in contro. Furthermore, there were correlations between GAH levds and BML, and between GAH and blood pressure. However; there was no marked variability in saliva flow rates among the groups. These results indicate ‘that measurement of salivary GAH and its relationship to other salivary parameters might help to provide insight into the role of ghrelin in diabetes. Keywords: Activelinactive ghrelin, Alphoamylese, Diabetes, Salva © To whom comespondence should be adresse. Tel: S0S3 HOMES; Fan: A779 18 Eamnil: syd! @hotmal.com Introduction GGinelin (Ghrelin Appetite Hormone, GAH this abbreviation ‘was proposed by Aydin, 2006a), an endogenous ligand forthe szowth hormane (GH) secretagosue receptor (GHSR), was originally discovered in extracts of rat and human stomach, ‘where it is localized in the endocrine X/Aclke cells of the gastric mucosa (Kojima ef als 1999). This homone is composed of 28 amino acid residues, of which the thi, uwsully a serine butin some species a threonine, isacylated by a fatty acids this modification is essential for GAH activity, Ghrelin is the first known example of a bioactive peptide ‘moafied by acylation (Kojima eta, 2005). Two major forms of GAHare present in tissues and blood: desracylated [GAH (inactive); this abbreviation was proposed by Aydin, 2006a] and ctanoylatad [aGAH (active); this abbreviaion was proposed by the same author in 2006ai, Both forms of ghrelin, especially aGAH, are_now known to have importnt physiological kes (Kojima et a, 2005; Aydin, 2006), the only difference being that desacylated ghrelin can arss the blood brain turier (Banks er al, 2002); but in tems of isbetes and insulin resistance the “incive™ form may be ‘more physiologically important, GGAHis predominantly prodicad inthe siomach (Kojima er ai, 1999; Kojima ef a, 2005), bux it is expressa in many other orzars inching Lowel, pancreas, myocardium, kidney, pituitary and hypothalamus (Kojima era, 2005). It promotes appetite. GAH mRNA has also been found in other tisues (Gnanapavan et al, 2002) and GAH is present in human milk (Aydin er al, 20060). We have alo reported a substance in plnt parenchyma cals dit is strongly immunoreactive against human ghrelin (Aydin ef aly 206d), More recent ‘work in cur kboratory (Aydin er al, 2005), and others (Groschl ea, 2005) has shown that GAH is produced and secreted by salivary glands and exhibits biological rhythm in humans (Aydin er a, 20062). In humans, GAH affects cardiovascular activity by acting, asa vasodilator, glucose metabolism by modulating insulin30 ‘Sukyman Ayelin secretion, amino acid uptake, bone formation (Fukashima er /., 2005) and appetite (increases food intake: Wren ef al, 2001; Kojima a all. 2005), It contributes 10 loss of appetite in szstric canoers (Aydin ea, 20050), Plasma GAH levels are inawesed under negative energy balance conditions such as fasting, anorexia nanos and cachexia (Oto et als 2005), and decreased in obesity (Tschop et as 2001; Bellone er a 2002; Hansen et af, 2002; Hag et all, 2008; Rosicka et al, 2008; Groschl et al, 2005) and after fod intake (Cumming et als 2002; Aydin er aly 20062). Mavaullo ancl co-workers have reported thatthe circulating level of GAH decreases in obese ations and that it is significantly comelaied with BMI (earaillo a als 2004), Circulating levels of GAH are diminished in patients with type 2 diabetes mellitus and are inversely conelaed with BMI in nondiabetic and type 2 Antti subjects (Kats tal 200 Cele 25) 1D milion rants workin the year 200. Is prevalence is epi increasing (Wilder a, 2004). Its clinically complex andl associated with many serious complications inchuding kidney failure, lindhessand cardiovascular disease Babligna et al, 2006). The chronic nature of DM entails a substantial decrease in quality of life and life expectancy and accounts for billions of dollar in health care spending (Coffey era 20025 Sephenser a 20), tn pevions (Hamson OU ef al 1986. Pee ‘Anjeh ef aly 1988; Suckfits ef aly 1994; Beli etal, 1998; Data ef al, 2004) and experimental animals (Reuterving, 1986}, However, there has been no oonemsas about saliva compesition in diabetic patients. Far example, total salivary protein concenivation has been found to be similar in diabetic and consol groups in some studies (Haris ef as 1981: TTenowo ef al, 1986), while othas have found salivary protein concentrations fiom diabetics to be either lower (Sueckfius ef als 1994) or higher (Beretych ef al 1988), Dambirski er af, (2003) found a dramatic impaiment of anrylase activity in diabetic animals as compared to controls. ‘Zhang, a al. 2005) showed dex ghrelin inhibits possum stimulated amylase secretion in incubated pancreatic lobules, and there is a comelation between ghrelin and glucose (Hirsh cet al 2005). Ithas ban shown that pretreatment with ghrelin redhees pancreatic damage in caerulein-induced pancreatitis and inhibits the development of gastric lesions inkduaed by ethanol (Dembinski ef as 2008), Tothe bestofmy knowkdee, although several stes have examined serum ghrelin levels inpatients with diabetes (Katsuki et ail, 2004; Celi et ay 2005), measurements of saliva ghrelin levels, alpha-amylase, total protein and glucose in type 2 diabetes have not been undattaken to date. The Present sudy was therefore undertaken 10 invessigate: () ‘whether the salivary levels of dGAH and aGAH are decreased, or increased in diabetic suhjocts; Gi) the relationship between ghrelin and alphoamylas, total protein and ghuxose in diabetic Patients, The results may provide important information concerning the activity of saliva aGAH/AGAH in diabetic patients and its association with alphasamylase, total protein eval and glucose. Materials and Methods Subjects. The group of petientsinclusad in this study corsisted of | 20 obese (BMI>30 kgin?) and 20 nonscbese (BMI25 ken) Patiens with type 2 ciebotes mel aeed ffor 41 10 66 years, ‘who wee refiread by the Endocrinology Service of the Firat Matical Center, Fkvig, The dizenoss of type 2 diabetes melts ‘was tose on the criteria of the American Diabetes Asxciation (ihe Expert Committee on the Dizenosis and Chssification of Diabetes Melltas, 1997) Type I dicbetc subjects were excluded. ‘The exnitol group constied of 2 clinically healthy humans (11 ffamake and 11 male) between 38 and 59 years of ae. Exclusion criteria fer the contol goup were: pregnancy, akobel ccrsumption, ‘olnoco prods (former and current, BMI>25 keh’, chronic ‘meal ines, hisory of drug treament or therepy within the Previous months, and history of dicbetes Subjects were asked not to eat, smoke or drink (exept water for an overnight fast pri to collection of sativa samples. Ther ets were similar With respect Protein content ard uptake of fat and carbohydrates, Socior ceconnmical sats wax similar for both groups (survey data). ‘Saliva collection and conservation. Unstimulated saliva (Sml) fiom the diabetic and contol groups was collctad using the reviusly publhed meta (Aydin al, 2005; Ayetin ef aly 200) and the results chained were confined by Groshle er a (2006), Briefly: in the moming, at 8 o’clock (before breakfast), the subjects were asked to rinse their mouths thoecughly with wate, then to bend their heads fonward ane allow saliva to flow ino an ive-chilled sterile container bearing the appropriate preservatives. ‘The axntainars were brought immediately to the Iborstry from the Endocrinology Clinic. Coecton tock Smin and the salivary ‘flow rate rate was defined as the volume of saliva secreted per min, ‘Onxe the saliva was collect, itwas centrifitged at 4000 rpm for 15 min to remove any panioulae meteral, Fach sipematant was divided into three aliquots and stored at 70°C until analysis, To exche the possibly of effacs of the mmsiual ee cn the huclin shyt, the data for each subject were obtained ding the Tueal phase, Ghveln was measned in urstimubste teal salva using the same human-ghreinRIA kit, whichis designed to test any’ biological ‘uid With sufficient Ives of the peptide 10 be dxemnined, Validation of the GANT radpinmunoasay in whole human saliva has been reported elsewhere (Ayein er cy 2005e, Grech era, 2008; Avelin ef al, 2006e). All Samples were read with a gamma ‘counter. Salivary dGAH was calculated as follows: inactive GAH otal GAH concentrate GAH concentatcn. Active salivary GAH was measured using the humangtrelin-RIA kit. Active saliva GAH measurements were first validated as follows,Saliva Gira in Dice a Sensitivity: The lowest concentration that could be distinguished ‘ftom the evo standard was 15 pei. Precision: The interassay (withineday) variation was determined {far two different salva (S) and two different plasma (P) spk Using the means of 2 replicates of each, The celicient of variation (CV) iscakeulated ax CV = Standard Deviation (SDyMaan. Sample Number Mean (pail) SD (pg/ml) CV C4) SI 3 u 36 105 2 2 31 42 BS Pl 2 4 81 23 P2 2 CG 66 97 ‘The interes (betweenstys) variation was abbo determined for ‘vo different saliva (S) and two ferent plasma (P) samples using ‘the means of several 2 replicates of each, Sample Number Mean (pginl) SD (pg/ml) CV @) SI 3 28 45 16 2 3 2 66 157 PL 2 ™ 84 na P2 2 6 1A 12 Linearity: Two saliva (S) and plasma (P) sempkes were eluted ‘with disilled water and assayed. (Concentrations in pen) Undilued V2 v4 18 SI 34 40 28 38 0) 18%) AID) 2 2 6 0 2 Or) (8) IH BHD Pl 6 2 a B 00) 100%) (IM) AIS) Pr 14 8% 0 2 oer) AH) (108%)__ 24D Recovery: Two saliva (S) and plasma (P) samples were enriched ‘with incteesing amounts of active ghielin, The peroentage recovery ‘was caleubted as follows: observed valuo-bescline valusfmount ackied 100. The concentrations ae given in pet. Tnital Amount Amount Amount Recovery concenimtion “adled recovered expected 4) ss a % 8 2 Ds 10 Im ID 6 oo B61 104 P TB 128 210 206 02 ‘These resus verified that the kit detec ghrelin quantitatively insaliva The lonest sensitivity of salivary active ghrelin was found to be [Spin The intre and intenarsary porcentage coefficients of \atiation were found tobe 23 and 160, respectively. “Total protein and alpheamylase were determined immediately fier collection in order to avoid daily variations caused by endogenous Proteolytic activity, Total protein concentration was measured by the Bradfoed method add et al, 1976) ard amylee avty was determined by the colorimetric method of Winn-Deen and 0o- Workers (Winn-Deen ef al., 1988). Glucose was determined in 200 lof salva by the gumooxidre method (Reljc er al, 1952) Salivary total proteins @min), amylase activity (Cit ane gacose evel (mg/ml/min) were expressed in the units shown for ease of comparscn with previcusstdies on ahi abet (Reka, 1958), (Chemicals. Total ghrelin (Phonix-Genmany) and active ghrelin ‘were obtained fic Linco Rearch, INC. USA. Otber chemicals, ‘were purchased fom Sigma-Aldrich, ‘Statistical analysis. Statistical analysis was performed using SPSS 100 fer Windows software. All values ae reponed as mean SD. ‘The statistical significance of differences in BMI, GAH, GAH, amybse, glucese, ofa protins, and systolic and diastolic bleed presure betwen the stkly ggoups and the controls was estimated by cnefctor analysis of variance (ANOVA) with or without “Tukey's posthoc test values smaller than 005 were accepted nificant. Results ‘The demographic characteristics of the subjects are shown in Table 1. There were no significant intergroup differences in age or duration of diabetes (Table 1). Total protein did not differ among the soups (Table 2). Salivary glucose 00%) and alpha-amylase 27) levels were significantly higher in obese disbeticsubjects than in contols (p < 008) and salivary ucose (19224) and alpheramylase 24%) levels in novrobese icbetic subjects were also significantly higher than those of control. Salivary glucose and alphacmylse kevels in the obese diabetic subjects were almost the seme (Table 2) Salivary aGAH (52%) and AGAH 24%) levels were lower than contol in obese subjects with type 2 diabetes (Table 2) Saliva aGAH G96) and dGAH (19%) level were also lower than control in nonobese type 2 diabetes subjects (Table 2). Saliva aGAH (22%) and dGAH (6.5%) level were lower in the obese than in the norrobese patients (Table 2). This means that type 2 diabetes is associated with a decrease in both aGAH and dGAH. The aGAH and dGALH saliva kxels were also found to be correlated with body mass index (BMD in both obese (r= 048, p<0,05) and nonobese (F=0.34, (p<0.06) type 2 diabetic patie an contol groups. When saliva eGAH kevels were compared with salivary dGAH concentrations on an individual basis, a kage variance was observed, with values ranging fiom 16 10 59 pes for aGAH and fiem 74 t0 289 pe/ml for (GAH, ‘The diastolic and systolic blood pressures were higher in type 2 diabetes patients than normal subjects (Table 1; P'<0.05) but there was no marked difference between obese and norrobese2 ‘Sukyman Ayelin ‘Table 1. Clinical characteristics of type 2 diabetic patients and contols. Data are expressed as arithmetic means stack deviations SD Parmees (Obese =) Nancbese (1=20) Coral n=2 Age Gums? 4727 8285 9288 Sex (WEP om 1010 1on2 DBP (nm Fey" 9742 +76 S468 SBP (am Hay 17464 136482 12666 Duration of diabetes Gears) 942031 98205 none DBP; Diasolic blood pressure, SBP; systolic blood presure. “Age and sex were not different among the study groups p <005 obese ‘ys contol, 7p <(06 obese ¥s nonobese. “Duration of diabetes did noe differ bewvoen obese and non-cbese subjects. Table 2 Mean salivary values of glucose, total protein, alphramylase, and aGAH and dGAH levels in cbesehomobese diabetic and control groups Pacmeters Obese (F=20) Nomobese patients (n=20) Control n= 2 Glucose (ginning 38406 ‘Total protein (g/min)? 1.68 +02 Amykase (UimlF 612457 GAN (pein? 27446 GAH (pgimlF 156+47, Flow rate (mliminy 1.0940. p<0001 obese vs con, “p<0105 obese vs nor-cbese ard non-obese vs conto, "> <005 obese ¥S canto nonobese vs CANE, obese vs nonchese, $<0001 cbse vs contol, < 0.05 cbese vs norechese and nonchese vs contol, $p Aydin, S, Dogry, 04 Celebi, Sand Gwvel, SP. @005e) Ghrelin immunchisochemisey of gzsvic adenocarcinoma, and mucepidemoid cacinama of salivary sland, Biotech, HisocTem, 80 163-168, Baliga, BS. and Weinberze, J. 2006) Diabetes and stoke: pat one—tisk eeeS and pathophsiobow. Cex Cre. Rep. & DR, Banks, W. A, Echop, M, Robinson, §. M, and Heiman, MLL. QUO) Extent and direction of ghietin tensport acres the blood-brain banier is determined by iss unique primays inte. J: Pharmac. Ege Ther M2, 922507, Boku M, Aj Gall-Tsimopoulow, A. Drakoulckos, D., Fla, A. and Parnrayictu, P. J. (1998) Salivary alterations in insiline dependent dietes mellitus. £ Poadiar: Deve. 8, 9-33, Ballons, S, Rapa A, Vivenz, D, Catling No Pet, Ay Bellons J, Me, E5 Broglio, F, Proce, F Ghign, E: andM ‘Sukyman Ayelin Bora, G G2) Gieubating dvelin kvels as finction of sendes pubertal statis and atipsity in childhood. Exiberind, mest. 8, 1315. BarAnels H. Coby M, Kanes, Yo Saas, R. and Lanta D. (1988) Salivary compesition in dicbetic parts. Diatees Conptcat 2.9693. Bradford M. M. (1976) A rapid and sensitive method far the quntittion of micgram quantities of protein utilizing the Principle of proeinedye binding, 4, Biochem. 72, 248254, Celi, Fp Bini, V5 Papi, Fo Sarl, Fp Foret, Aw Mencarti; My Bexioli M, G, De Gio, G and Falomi, A, (2005) Circulating acylated and toed ghvelin and galanin in chien wid insline treated ope 1 dabetes relationship to insulin thecpys ‘metabolic contol and puberakkevelopment, Cin, Brebcrinol, (09.8, 139-145, Claves, E. M, Borrell L. Nz Tayo, GW. and Ship, J. A. 200) A neineinal anaksis of ‘sainary flow in oon subj and okle ads with type 2 dihtes. Ora! Swe, Onul Med Ona Pethed. Oral Rel. Endext 91, 16173. ‘fey, J.T, Brande, M, Zhou, Ha Mariott D, Burke, R, Taboei, BLP, Frgelan, M. M, Kaplan, RM. and Harmen, W.H. (04D) Valuing heltheletad quality of Tite in diabetes. Diabetes Care 25, 2382243. Cummings, D. E, Weigk D. S, Fryo RS. Breen P. Ay Ma M. K,, Dellingsr EP. and Purell, J. Q. G12) Pkesma ghrelin levels afferdistindhoad weight las or gastic bypass sanry. N. Bight Mak, M6, 12-1630. Dambineki, A, Waradha, Z Camnowicz, R, Tomasenska, Re Stochura, J, Kone, S.J, and Kontek, PC. C003) Ghrelin atctes the development of acute pancreatitis in rte Jt Pissiol, Pharmacol, 54 561-573. Dicks M. Ws and Dos A. P. C997) Eifbats of etcemic como! on Sava flow rates and prtein cxmposition in nae insulimdependant diabetes mellitus. Oral Sz, Oral Mri Oral Pehl. Oral Raol. End. 8, 46540. Daxkls, M. W, Johnson, D. A. and Yeh C. K. (005) Health benefits of sliver a review. J, Dent 33, 223233, Fukushima, Nj Hanada, R, Tenis, HL, Fuku, Y> Tachibana, ‘Ty Bhikava, Hy Takecky, S, Takeuchi, Y5) Fudamoto, S5 Kareama, K, Nazi, K: and Kojim, M (2005) Ghrelin lrecly rege bone formation, Bone Minor Res. 20, 790+ PR Grenapavan, $_ Kola B. Bustin, S.A Maris, D. G, McGee, Ps Fairckush, 2 Bhatachaya, S, Carpenter, R Grossman, AL B. and Korboniss M. 2002) ‘The tisue disribution of the MRNA of ghrelin and subypes of iss recep GHSR, in urens Clin Exborind Metab 87, 2988-2991, Greenman Yo, Gotan, No Gilde So Yarons May Limos Re and Stem N. 004) Ginelin secretion & modulated in auxin and gender-specific manner, Clin Endberinol. (Ox). 60, 30- 388 Groschl, Mz Topf, H. G, Boblendet, J, Zenks J, Khusmann, S Dash, J, Rascher, W. and Rath, M. 005) Kdntificaton of sven in human saliva: precution by the salivary glands and Potential rein prolifiaton of eral keratinocytes. Cnn Crem. 1, 99721006. Groschl Mz Topé H. G, Reuh, M, Kurai, M, Rascher, W an Kohler, Fl 2006) Postrrandil rsparee of salivary helin and leptin to carbohyckae uptake. Gi $5, 23-34, Fauwen TK Dall, Ry Hhsada, 1 Kojima, M, Kanga, Ky Cristian, J. S. and Janssen, J. 0. (OD) Weiht loss incre circicting keels of heli in human obesity, Clin, Endborinel (0%), 86, 23206, Hanon, R. and Bowen, W. HL. (1987) Flow rate and onanic consitents of whole sliva in insu diabetic chilen and adolescents Pa: Der, 9, 287-201. Hag A. My Faroogi I. S. ORahilly S, Saxe, D. D, Resend, RG, Prt, K. Ly Laftenchi S. Hand Panel, J Q. 2B) Serum ghrelin keels ae inversely corebed with body mass inde, ase and insulin concerraions in. nonral dhilden and are markedly increased in PradsWill syndrome, Oz Na Ensey, F- ae Yeniay; 1. (1995) The alterations of whole saliva constivents in patents with diabetes melius, use Dent Jey 19197, ‘Zien Wo Chen, M, Chen, X, Seaur B,J are, Mutholand M, W. 2001) Inhibition of pancreatic proven saeson by ehyelin inthe ra. J. Pinsiol $87, BID,