Matrix Metalloproteinases as a Pleiotropic Biomarker in ...Disease Markers Matrix Metalloproteinases as a Pleiotropic Biomarker in Medicine and Biology Guest Editors: Jacek Kurzepa,

Disease Markers

Matrix Metalloproteinases as a Pleiotropic Biomarker in Medicine and Biology

Guest Editors Jacek Kurzepa Fatma M El-Demerdash and Massimiliano Castellazzi

Matrix Metalloproteinases asa Pleiotropic Biomarker inMedicine and Biology

Disease Markers


Guest Editors Jacek Kurzepa Fatma M El-Demerdashand Massimiliano Castellazzi

Copyright copy 2016 Hindawi Publishing Corporation All rights reserved

This is a special issue published in ldquoDisease Markersrdquo All articles are open access articles distributed under the Creative Commons At-tribution License which permits unrestricted use distribution and reproduction in any medium provided the original work is properlycited

Editorial Board

Silvia Angeletti ItalyElena Anghileri ItalyPaul Ashwood USAFabrizia Bamonti ItalyBharati V Bapat CanadaValeria Barresi ItalyJasmin Bektic AustriaRiyad Bendardaf FinlandLuisella Bocchio-Chiavetto ItalyDonald H Chace USAKishore Chaudhry IndiaCarlo Chiarla ItalyMassimiliano Romanelli ItalyBenoit Dugue FranceHelge Frieling GermanyPaola Gazzaniga ItalyGiorgio Ghigliotti ItalyAlvaro Gonzaacutelez SpainMariann Harangi HungaryMichael Hawkes CanadaAndreas Hillenbrand GermanyHubertus Himmerich UKJohannes Honekopp UKShih-Ping Hsu TaiwanYi-Chia Huang Taiwan

Chao Hung Hung TaiwanSunil Hwang USAGrant Izmirlian USAYoshio Kodera JapanChih-Hung Ku TaiwanDinesh Kumbhare CanadaMark M Kushnir USATaina K Lajunen FinlandOlav Lapaire SwitzerlandClaudio Letizia ItalyXiaohong Li USARalf Lichtinghagen GermanyLance A Liotta USALeigh A Madden UKMichele Malaguarnera ItalyHeidi M Malm USAUpender Manne USAFerdinando Mannello ItalySerge Masson ItalyMaria Chiara Mimmi ItalyRoss Molinaro USAGiuseppe Murdaca ItalySzilaacuterd Nemes SwedenDennis Nilsen NorwayEsperanza Ortega Spain

Roberta Palla ItalySheng Pan USAMarco E M Peluso ItalyRobert Pichler AustriaAlex J Rai USAIrene Rebelo PortugalAndrea Remo ItalyGad Rennert IsraelManfredi Rizzo ItalyIwona Rudkowska CanadaMaddalena Ruggieri ItalyVincent Sapin FranceTori L Schaefer USAAnja Hviid Simonsen DenmarkEric A Singer USAHolly Soares USATomaacutes Sobrino SpainClaudia Stefanutti ItalyMirte Mayke Streppel NetherlandsMichael Tekle SwedenStamatios Theocharis GreeceTilman Todenhoumlfer GermanyNatacha Turck SwitzerlandHeather Wright Beatty Canada

Contents

Matrix Metalloproteinases as a Pleiotropic Biomarker in Medicine and BiologyJacek Kurzepa Fatma M El-Demerdash and Massimiliano CastellazziVolume 2016 Article ID 9275204 2 pages

Interplay between Matrix Metalloproteinase-9 Matrix Metalloproteinase-2 and Interleukins inMultiple Sclerosis PatientsAlessandro Trentini Massimiliano Castellazzi Carlo Cervellati Maria Cristina ManfrinatoCarmine Tamborino Stefania Hanau Carlo Alberto Volta Eleonora Baldi Vladimir Kostic Jelena DrulovicEnrico Granieri Franco Dallocchio Tiziana Bellini Irena Dujmovic and Enrico FainardiVolume 2016 Article ID 3672353 9 pages

A Tale of Two Joints The Role of Matrix Metalloproteases in Cartilage BiologyBrandon J Rose and David L KooymanVolume 2016 Article ID 4895050 7 pages

Expressions of Matrix Metalloproteinases 2 7 and 9 in Carcinogenesis of Pancreatic DuctalAdenocarcinomaKatarzyna Jakubowska Anna Pryczynicz Joanna Januszewska Iwona Sidorkiewicz Andrzej KemonaAndrzej Niewiński Łukasz Lewczuk Bogusław Kedra and Katarzyna Guzińska-UstymowiczVolume 2016 Article ID 9895721 7 pages

Serum Gelatinases Levels in Multiple Sclerosis Patients during 21 Months of NatalizumabTherapyMassimiliano Castellazzi Tiziana Bellini Alessandro Trentini Serena Delbue Francesca EliaMatteo Gastaldi Diego Franciotta Roberto Bergamaschi Maria Cristina Manfrinato Carlo Alberto VoltaEnrico Granieri and Enrico FainardiVolume 2016 Article ID 8434209 7 pages

Association of Common Variants in MMPs with Periodontitis RiskWenyang Li Ying Zhu Pradeep Singh Deepal Haresh Ajmera Jinlin Song and Ping JiVolume 2016 Article ID 1545974 20 pages

EditorialMatrix Metalloproteinases as a Pleiotropic Biomarker inMedicine and Biology

Jacek Kurzepa1 Fatma M El-Demerdash2 and Massimiliano Castellazzi3

1Department of Medical Chemistry Medical University of Lublin Chodzki 4a 20-093 Lublin Poland2Environmental Studies Department Institute of Graduate Studies and Research University of Alexandria 163 Horrya AvPO Box 832 El Shatby Alexandria Egypt3Department of Biomedical and Specialist Surgical Sciences Section of Neurological Psychiatric and Psychological SciencesUniversity of Ferrara 44124 Ferrara Italy

Correspondence should be addressed to Jacek Kurzepa kurzepayahoocom

Received 18 September 2016 Accepted 13 October 2016

Copyright copy 2016 Jacek Kurzepa et al This is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

The group of matrix metalloproteinases (MMPs) calcium-and zinc-dependent proteolytic enzymes is responsible forextracellular protein degradation Acting together supportedby intracellular processes they are able to digest any phys-iological extracellular protein However the biochemistryof extracellular matrix (ECM) is very complex and prote-olytic enzymes located in this compartment exert numerouspleiotropic effects beyond the characteristic for the degrada-tion of structural elements Therefore MMPs are involvedinto several physiological and pathological processes [1]

Because of the ECM componentsrsquo ability tomodel as wellas the influence on the activity of some biologically activecompounds such as tumor necrosis factor 120572 chemokineCXCL-8 and transforming growth factor 120573 MMPs affectthe pathogenesis of numerous diseases mostly primarilyassociated with inflammation [2] Therefore the elevatedlevel of particular MMP cannot be associated with failureof specific organ or tissue In that case the MMPs canbe biomarkers of disease MMPs are sensitive and easilymeasurable but due to their prevalence they are not specificfor any tissue For example MMP-9 serum level is elevated inpatients with relapsing remitting and secondary progressivemultiple sclerosis (MS) compared to controls [3] and theMMP-9TIMP-1 ratiomay predictmagnetic resonance image(MRI) activity during interferon-beta therapy [4] Howeverdespite the acknowledged involvement of some MMPs inMS pathogenesis and progression the evaluation of these

enzymes is not routinely recommended for MS diagnosisbecause their elevation is observed in numerous other dis-eases as stroke and bacterial and viral infections and even insmokers [5] Nevertheless the higher activity of individualMMPs in connection with patientsrsquo clinical status can helpto predict the risk diagnosis or progress of the disease Forexample the MMP-9 serum level does not correlate with therisk of stroke but MMP-9 C(-1562)T polymorphism seemsto be significantly associated with risk of stroke in patientswith and without type 2 diabetes mellitus [6] Also remainingMMPs possess the ability to predict the clinical status Theoverexpression of MMP-7 MMP-10 and MMP-12 in coloncancer patientsrsquo sera correlates with a dismal prognosis [7]and high serumMMP-1 level showed a trend for short overallsurvival in non-small cell lung cancer patients [8]

The low tissue specificity of isolated MMPs causes thatsingle enzyme may not play a role of a good biomarkerHowever some MMPs could be useful constituents ofbiomarker panels but only in combination with other bio-chemical parameters The multiplex panel composed ofMMP-7 CA125 CA72-4 and human epididymis protein 4is suitable for the early detection of ovarian cancer [9] Thesimultaneous evaluation of MMP-1 TIMP-1 CD40 ligandandmyeloperoxidase seems to be a novel promising diagnos-tic panel in timely diagnosis of acute aortic dissection [10]Also some products of MMPs catalysis were considered asthe potential biomarkers Citrullinated and MMP-degraded

Hindawi Publishing CorporationDisease MarkersVolume 2016 Article ID 9275204 2 pageshttpdxdoiorg10115520169275204

2 Disease Markers

vimentin (VICM) simultaneously and in combination withothers markers revealed good potential to differentiate ulcer-ative colitis form noninflammatory bowel diseases [11]

Finally last but not least preanalytical conditions mustbe taken into account before starting MMPs analysis in bodyfluids In fact if in one hand the release of MMPs duringclotting could affect their concentrations [12] on the otherhand the use of some calcium-chelating anticoagulants couldinterfere with MMPs activity [13]

In conclusion the enzymes from amongMMPs evaluatedindividually cannot be considered as the specific biomarkersof the particular disease or pathological processHowever thesudden change in their body fluid level can act as an alarmsiren informing on the upcoming threat which combinedwith clinical state of the patient may help in the diagnosistreatment or prognosis

Jacek KurzepaFatma M El-DemerdashMassimiliano Castellazzi

References

[1] A Tokito and M Jougasaki ldquoMatrix metalloproteinases innon-neoplastic disordersrdquo International Journal of MolecularSciences vol 17 no 7 p 1178 2016

[2] V Lemaitre and J DrsquoArmiento ldquoMatrix metalloproteinasesin development and diseaserdquo Birth Defects Research Part CEmbryo Today Reviews vol 78 no 1 pp 1ndash10 2006

[3] E Waubant ldquoBiomarkers indicative of blood-brain barrierdisruption in multiple sclerosisrdquo Disease Markers vol 22 no4 pp 235ndash244 2006

[4] C AvolioM Filippi C Tortorella et al ldquoSerumMMP-9TIMP-1 andMMP-2TIMP-2 ratios in multiple sclerosis relationshipswith different magnetic resonance imaging measures of diseaseactivity during IFN-beta-1a treatmentrdquo Multiple Sclerosis vol11 no 4 pp 441ndash446 2005

[5] S Yongxin D Wenjun W Qiang S Yunqing Z Limingand W Chunsheng ldquoHeavy smoking before coronary surgicalprocedures affects the native matrix metalloproteinase-2 andmatrix metalloproteinase-9 gene expression in saphenous veinconduitsrdquoThe Annals of Thoracic Surgery vol 95 no 1 pp 55ndash61 2013

[6] K Buraczynska J Kurzepa A Ksiazek M Buraczynska and KRejdak ldquoMatrix Metalloproteinase-9 (MMP-9) gene polymor-phism in stroke patientsrdquo NeuroMolecular Medicine vol 17 no4 pp 385ndash390 2015

[7] F Klupp L Neumann C Kahlert et al ldquoSerumMMP7MMP10and MMP12 level as negative prognostic markers in coloncancer patientsrdquo BMC Cancer vol 16 article 494 2016

[8] H J An Y Lee S A Hong et al ldquoThe prognostic role of tissueand serumMMP-1 andTIMP-1 expression in patients with non-small cell lung cancerrdquoPathology-Research and Practice vol 212no 5 pp 357ndash364 2016

[9] A R Simmons C H Clarke D B Badgwell et al ldquoValidationof a biomarker panel and longitudinal biomarker performancefor early detection of ovarian cancerrdquo International Journal ofGynecological Cancer vol 26 no 6 pp 1070ndash1077 2016

[10] E Vianello E Dozio R Rigolini et al ldquoAcute phase of aorticdissection a pilot study on CD40L MPO and MMP-1 -2 9

and TIMP-1 circulating levels in elderly patientsrdquo Immunity ampAgeing vol 13 article 9 2016

[11] J H Mortensen L E Godskesen M D Jensen et al ldquoFrag-ments of citrullinated andMMP-degraded vimentin andMMP-degraded type III collagen are novel serological biomarkers todifferentiate Crohnrsquos disease from ulcerative colitisrdquo Journal ofCrohnrsquos amp Colitis vol 9 no 10 pp 863ndash872 2015

[12] F Mannello ldquoEffects of blood collection methods on gelatinzymography of matrix metalloproteinasesrdquo Clinical Chemistryvol 49 no 2 pp 339ndash340 2003

[13] M Castellazzi C Tamborino E Fainardi et al ldquoEffects of anti-coagulants on the activity of gelatinasesrdquo Clinical Biochemistryvol 40 no 16-17 pp 1272ndash1276 2007

Research ArticleInterplay between Matrix Metalloproteinase-9Matrix Metalloproteinase-2 and Interleukins in MultipleSclerosis Patients

Alessandro Trentini1 Massimiliano Castellazzi2 Carlo Cervellati1

Maria Cristina Manfrinato1 Carmine Tamborino2 Stefania Hanau1 Carlo Alberto Volta3

Eleonora Baldi4 Vladimir Kostic5 Jelena Drulovic5 Enrico Granieri2 Franco Dallocchio1

Tiziana Bellini1 Irena Dujmovic5 and Enrico Fainardi6

1Section of Medical Biochemistry Molecular Biology and Genetics Department of Biomedical and Specialist Surgical SciencesUniversity of Ferrara 44121 Ferrara Italy2Section of Neurology Department of Biomedical and Specialist Surgical Sciences University of Ferrara 44121 Ferrara Italy3Section of Orthopedics Obstetrics and Gynecology and Anesthesia and Resuscitation Department of MorphologySurgery and Experimental Medicine University of Ferrara 44121 Ferrara Italy4Neurology Unit Department of Neurosciences and Rehabilitation Azienda Ospedaliera-UniversitariaArcispedale S Anna 44124 Ferrara Italy5Neurology Clinic Clinical Centre of Serbia School of Medicine University of Belgrade Dr Subotica 6 11000 Belgrade Serbia6Neuroradiology Unit Department of Neurosciences and Rehabilitation Azienda Ospedaliera-UniversitariaArcispedale S Anna 44124 Ferrara Italy

Correspondence should be addressed to Alessandro Trentini alessandrotrentiniunifeit

Received 24 March 2016 Revised 17 June 2016 Accepted 19 June 2016

Academic Editor Michael Hawkes

Copyright copy 2016 Alessandro Trentini et al This is an open access article distributed under the Creative Commons AttributionLicense which permits unrestricted use distribution and reproduction in any medium provided the original work is properlycited

Matrix Metalloproteases (MMPs) and cytokines have been involved in the pathogenesis of multiple sclerosis (MS) However nostudies have still explored the possible associations between the two families of molecules The present study aimed to evaluatethe contribution of active MMP-9 active MMP-2 interleukin- (IL-) 17 IL-18 IL-23 and monocyte chemotactic proteins-3 to thepathogenesis of MS and the possible interconnections between MMPs and cytokines The proteins were determined in the serumand cerebrospinal fluid (CSF) of 89 MS patients and 92 other neurological disorders (OND) controls Serum active MMP-9 wasincreased in MS patients and OND controls compared to healthy subjects (119901 lt 0001 and 119901 lt 001 resp) whereas active MMP-2 and ILs did not change CSF MMP-9 but not MMP-2 or ILs was selectively elevated in MS compared to OND (119901 lt 001)Regarding the MMPs and cytokines intercorrelations we found a significant association between CSF active MMP-2 and IL-18(119903 = 03 119901 lt 005) while MMP-9 did not show any associations with the cytokines examined Collectively our results suggestthat active MMP-9 but not ILs might be a surrogate marker for MS In addition interleukins and MMPs might synergisticallycooperate in MS indicating them as potential partners in the disease process

1 Introduction

Multiple sclerosis (MS) is a disease of the central nervous system(CNS) of supposed autoimmune origin characterized byinflammation demyelination and neurodegeneration [1]Although the pathological features of the disease are hetero-geneous a common event is thought to be the reactivation

within theCNSof infiltratingmyelin-specificT cells which inturn trigger the recruitment of innate immunity cellsmediat-ing demyelination and axonal loss [2]The perivascular trans-migration and accumulation of inflammatory cells within theCNS are mainly mediated by two events the production ofleukocyte-attracting chemokines and the blood-brain barrier


2 Disease Markers

Table 1 Demographic and clinical characteristics of healthy controls and OND and RRMS patients

Healthy controls (119899 = 40) OND (119899 = 92) MS (119899 = 89)Age 370 plusmn 75 355 (303ndash440) 424 plusmn 137 415 (330ndash490) 392 plusmn 109 370 (305ndash480)Sex femalemale 2416 5735 5336Disease duration (yrs) mdash mdash 59 plusmn 71 3 (1ndash77)EDSS mdash mdash 38 plusmn 19 35 (25ndash44)Clinically active MS 119899total () mdash mdash 3240 (80)Clinically stable MS 119899total () mdash mdash 840 (20)EDSS expanded disability status scale MS multiple sclerosis RRMS relapsing-remitting multiple sclerosis OND other neurological disorders

(BBB) breakdown [3] The production of chemokines maybe important for the regulation of the inflammatory cellsinflux to sites of tissue damageWithin the chemokine familyparticularly studied members in the course of MS are themonocyte chemotactic proteins (MCPs) with MCP-1 andMCP-2 being selectively expressed at high levels in activelesions whileMCP-3wasmostly observed in the extracellularmatrix surrounding the vascular elements [4] In additionto the establishment of a chemokine gradient the BBB hasto be disrupted in order for the leukocytes to infiltratewithin the CNS [5] This event is mediated by the actionof matrix metalloproteinases (MMPs) a family of Zn2+-dependent and Ca2+-requiring endopeptidases involved inthemodeling of the extracellularmatrix in both physiologicaland pathological conditions Among all MMPs MMP-9 andMMP-2 have been extensively studied in MS given theirability to degrade the components of the basal lamina and tomediate BBB damage [6ndash8]

Notably growing experimental evidence suggests theinvolvement of MMP-9 in the pathogenesis of MS whereits circulating levels in serum and cerebrospinal fluid (CSF)were found to be upregulated in MS patients comparedwith noninflammatory neurological disorders (NIND) andhealthy controls [9ndash13] On the contrary the implication ofMMP-2 in the pathogenesis of MS is more controversialsince this enzyme had demonstrated both protective [7]and detrimental actions [14] Besides MMPs inflammatorycytokines in particular the interleukins belonging to theTh17axis IL-23 and IL-17 might also play a role in MS [15]

IL-23 a member of the IL-12 cytokine family is a het-erodimeric proteinwith the ability to support the polarizationand expansion of T cells toward aTh17 phenotype [16 17] Itsinvolvement in the pathogenesis of MS has been suggestedby evidence from the animal model of the disease theexperimental autoimmune encephalomyelitis (EAE) Indeedthis cytokine has proven to be essential for the developmentof EAE [18] and the transfer of Th17 cells polarized andexpanded by IL-23 was able to induce the disease in animals[19] Th17 cells are strictly connected to the pathogenesis ofMS through but not limited to the production of severalproinflammatory cytokines including IL-17 (A and F) whichhas been found upregulated in chronic lesions of MS patients[20] and in the serum of Interferon-120573 (IFN-120573) nonrespond-ing patients [21] In addition to the abovementioned factorsIL-18 another cytokine important in Th1 response in thecourse of MS [22] has been found increased in serum and

CSF of MS patients compared to noninflammatory controlswith the levels of the molecule being higher in those withMRI gadolinium enhancing lesions [23] Nonetheless severalanimal and in vitro evidence connected both MMPs to IL-18[24] and to the IL-17IL-23 axis [25] demonstrating a generalstimulating effect on the enzymes production whereas otherreports suggested a regulation of MMP-2 on MCP-3 activity[26] showing an anti-inflammatory effect [27] However tothe best of our knowledge none of the previous studies evalu-ated the possible interrelationships between the active formsof MMP-9 and MMP-2 and the most common cytokinesinvolved in MS pathogenesis Therefore in the present studyour aim was to measure the levels of active MMP-9 andMMP-2 IL-17 IL-18 IL-23 and MCP-3 in the serum andCSF of MS patients and controls in order to investigatethe contribution of these molecules to MS pathogenesisMoreover we aimed to explore possible interrelationshipsbetween cytokines MMPs and clinical variables

2 Material and Methods

21 Patients Selection For this study we recruited 89 con-secutive patients affected by definite relapsing-remitting MS(RRMS) according toMcDonald criteria [28] who presentedat the Neurology Clinic of the University of Belgrade Evi-dence of a relapse at admissionwas considered clinical diseaseactivity [29] The data were available for a total of 40 patientsout of 89 Accordingly 32 patients were clinically activewhereas 8 patients were clinically stable Patient diseaseseverity was measured by Kurtzkersquos Expanded DisabilityStatus Scale (EDSS) [30] Disease duration was scored andexpressed in years At the time of sample collection noneof the patients had fever or other signs of acute infectionnor had they been receiving any disease-modifying therapies(DMTs) during the 6 months before the study A total of 92controls with other neurological disorders (OND) were alsoincluded in the study (Table 1) OND patients were free ofimmunosuppressant drugs including steroids at the time ofsample collection In addition a total of 40 age- and sex-matched healthy controls (HC) were used Informed consentwas given by all patients before inclusion in the study and thestudy design was approved by the Ethics Committee of theSchool of Medicine University of Belgrade

22 CSF and SerumSampling Cerebrospinal fluid and serumsamples were collected under sterile conditions and stored

Disease Markers 3

in aliquots at minus80∘C until assay ldquoCell-freerdquo CSF sampleswere obtained after centrifugation at room temperature ofspecimens taken by lumbar puncture performed for diagnosispurposes Serum sampleswere derived fromcentrifugation ofblood specimens withdrawn by puncture of an anterocubitalvein at the same time of CSF extraction Paired CSF andserum samples from RRMS and OND patients were storedand measured under exactly the same conditions For thehealthy controls only the serum was available

23 Assay of Interleukins in Serum and CSF IL-17A IL-23 and MCP-3 levels were simultaneously measured in seraand CSF of patients twofold diluted with dilution bufferor undiluted respectively by a multiplex sandwich enzyme-linked immunosorbent assay (ELISA) system based onchemiluminescence detection (Aushon SearchLight chemi-luminescent assay kits Tema Ricerca Italy) according to themanufacturerrsquos recommendations All samples were analyzedin duplicateThe interleukin levels are reported as pgmLThelower concentration of each standard curve was 078 pgmLfor IL-17A 195 pgmL for IL-23 and 078 pgmL for MCP-3

IL-18 was measured in CSF and serum samplestwofold diluted with commercially available ELISA (BosterImmunoleader cod EK0864) Samples were assayed induplicate A standard curve was generated in each plate andthe lower standard concentration was 156 pgmL

24 Assay of Active MMP-9 in Serum and CSF Serum andCSF levels of circulating active MMP-9 were determinedusing a commercially available activity assay system (HumanActive MMP-9 Fluorokine E Kit RampD systems Cat NumberF9M00) following the manufacturerrsquos instructions All thereagents were included in the kit For the determinationsa standard curve in the range of 16ndash0125 ngmL was usedserum and CSF samples were diluted 100 times and 2 timesrespectively with the calibrator diluent (RD5-24) included inthe kit According to the manufacturerrsquos data the minimumdetectable dose was 0005 ngmL and the range of intra-assayand interassay coefficient of variation (CV) was 39ndash48 and80ndash93 respectively

25 Assay of Active MMP-2 in Serum and CSF Serum andCSF levels of circulating active MMP-2 were determinedusing a commercially available activity assay system (MMP-2Biotrak Activity Assay System GE Healthcare Cat NumberRPN2631) following the manufacturerrsquos instructions All thereagents were included in the kit For the determinationsa standard curve in the range of 4ndash0125 ngmL was usedserum and CSF samples were diluted 25 times and 2 timesrespectively with the assay buffer included in the kit Accord-ing to themanufacturerrsquos data the sensitivitywas 0190 ngmLand the range of intra-assay and interassay coefficient ofvariation (CV) was 44ndash70 and 169ndash185 respectively

26 Statistical Analysis Normality of distribution waschecked by Shapiro-Wilk test Since the variables were notnormally distributed group comparisons were performedusing Kruskal-Wallis followed by Mann-Whitney U testswith Bonferroni correction for multiple comparisons Bivari-ate correlations were performed by Spearmanrsquos rank test and

frequency distributions were examined using the Chi-squaretest To assess the association between abnormal MMPs andILs values measured in serum or CSF and the MS pathologya binary logistic regression analysis was performed A valueof 119901 lt 005 was considered statistically significant

3 Results

31 Active MMP-9 and MMP-2 in Serum and CSF of MSPatients and Controls Active MMP-9 and MMP-2 weredetectable in 100 of serum samples and in 100 and in 93(85 OND and 84 MS) of CSF samples for MMP-9 and MMP-2 respectively As reported in Figure 1(a) the levels of activeMMP-9 were different among the groups In particular wefound a higher concentration of active MMP-9 in the serumof both MS patients and OND controls compared to healthysubjects (119901 lt 0001 and 119901 lt 001 resp) On the contraryactive MMP-2 serum levels were similar in MS OND andhealthy subjects (Figure 1(b) Kruskal-Wallis 119867(2) = 1009119901 = 0604) Then we compared the amounts of both activegelatinases measured in the CSF of MS patients and ONDcontrols As depicted in Figure 1(c) MS patients showedalmost a doubled concentration of active MMP-9 comparedto OND controls (119901 = 0009) whereas the levels of activeMMP-2 did not differ (Figure 1(d) median (interquartilerange) 47 (23ndash114) and 51 (27ndash104) for OND and MSpatients resp 119901 = 0713) When patients were groupedaccording to clinical disease activity there were no statisticaldifferences between MS patients with and without clinicalevidence of disease activity for both serum and CSF activeMMP-9 and MMP-2 (data not shown)

32 Interleukin Levels in Serum and CSF of MS Patients andControls The levels of IL-17 IL-18 IL-23 and MCP-3 in theserum of OND and MS patients were detectable in 21 ofsamples for IL-17 (16 OND and 22 MS) 86 for IL-18 (79OND and 77 MS) 71 for IL-23 (65 OND and 64 MS) and61 for MCP-3 (55 OND and 55 MS) In the CSF the valueswere detectable in 35 of samples for IL-17 (35 OND and27 MS) 59 for IL-18 (50 OND and 56 MS) 19 for IL-23 (18 OND and 17 MS) and 53 for MCP-3 (46 OND and50 MS) As reported in Figures 2(a)ndash2(d) we did not findany significant difference in the serum concentration of themeasured cytokinesThe same result was observed in theCSFwhere the levels of the cytokines were not different betweenthe OND controls and the MS patients (Figures 2(e)ndash2(h))When patients were grouped according to clinical diseaseactivity we did not find any statistical differences betweenMSpatients with and without clinical evidence of disease activityfor both serum and CSF IL-17 IL-18 IL-23 andMCP-3 levels(data not shown)

33 Correlations between Interleukin Levels and Active MMP-9 and MMP-2 in Serum and CSF of MS Patients andwith Clinical Outcomes We evaluated possible correlationsbetween the levels ofMMPs and interleukinsmeasured in theserum of MS patients As reported in Table 2 we observedsignificant positive correlations between MCP-3 and IL-17between MCP-3 and IL-23 and between IL-17 and IL-23

4 Disease Markers

Seru

m ac

tive M

MP-

9 (n

gm

L)

HC MS patients OND patients0

500

1000

1500

2000

2500p lt 001

p lt 0001

(a)

OND patients

Seru

m ac

tive M

MP-

2 (n

gm

L)

HC MS patients0

100

200

300

500

600

700

(b)

CSF

activ

e MM

P-9

(ng

mL)

MS patients OND patients000

025

050

075

100

200

300

400p lt 001

(c)

CSF

activ

e MM

P-2

(ng

mL)


10

20

30

40

(d)

Figure 1 Median of serum total active MMP-9 and MMP-2 in RRMS patients OND controls and healthy donors and CSF active MMP-9andMMP-2 in RRMS patients and OND controls Serum levels of total active MMP-9 were statistically different among the groups (Kruskal-Wallis H(2) = 1545 119901 lt 00001) and CSF active MMP-9 levels were elevated in RRMS patients compared to OND patients (a) Serumconcentrations of total activeMMP-9were not different amongRRMS (median (IQR) 552 (318ndash841) ngmL) andOND(492 (330ndash737) ngmL)patients whereas they were higher (Mann Whitney 119901 lt 0001 and 119901 lt 001) in RRMS and OND patients when compared to HC (363(216ndash482) ngmL) (b) Serum levels of active MMP-2 were not different between RRMS patients (252 (131ndash481) ngmL) OND patients(262 (158ndash525) ngmL) and HC (258 (218ndash307) ngmL) (c) CSF amounts of active MMP-9 were more increased in RRMS (0084 (0040ndash0165) ngmL) than in OND (0046 (0027ndash0113) ngmL) patients (Mann Whitney 119901 = 0009) (d) CSF levels of active MMP-2 were notdifferent between RRMS (51 (27ndash104) ngmL) and OND (47 (23ndash114) ngmL) controls IQR interquartile range HC healthy controlsMMP matrix metalloproteinase RRMS relapsing-remitting MS OND other neurologic disorders CSF cerebrospinal fluid

Of note we did not find any relation between the activeforms of MMPs and the interleukins although there was atendency toward a significant negative correlation betweenserum active MMP-9 and IL-18 (119901 = 0076)

Thenwe evaluated the correlations between activeMMPsand interleukins measured in the CSF of patients The results

are summarized in Table 3 Notably we found a positivecorrelation between IL-18 and activeMMP-2 andMCP-3 andIL-17 and between IL-18 and IL-23

There were no significant correlations between diseaseseverity scored by EDSS disease duration and serum andCSF levels of the measured proteins

Disease Markers 5

Seru

m IL

-17

(pg

mL)

MS patients0

200

400

600

800

1000

1500

2000

2500

OND patients

(a)

Seru

m IL

-18

(pg

mL)

0

5000

10000

15000

20000

MS patients OND patients

(b)

Seru

m IL

-23

(pg

mL)

0

2000

4000

6000

800050000

100000

150000


(c)

Seru

m M

CP-3

(pg

mL)

0

50

100

150

200


(d)

CSF

IL-1

7 (p

gm

L)

0

10

20

30

40

50


(e)

CSF

IL-1

8 (p

gm

L)

0

2000

4000

6000

8000


(f)

Figure 2 Continued

6 Disease Markers

CSF

IL-2

3 (p

gm

L)

0

50

100

150

200

2000

4000


(g)

CSF

MCP

-3 (p

gm

L)

0

5

10

15

20

25


(h)

Figure 2Median of serumandCSF IL-17 IL-18 IL-23 andMCP-3 concentrations inRRMSpatients andONDcontrolsNone of the examinedcytokineschemokines was different between RRMS patients and OND patients in either the serum or CSF (a) Serum levels of IL-17 inRRMS (median (IQR) 337 (42ndash3350) pgmL) and OND (448 (41ndash4680) pgmL) patients (b) Serum levels of IL-18 in RRMS (2259 (1232ndash3505) pgmL) and OND (2266 (1509ndash3766) pgmL) patients (c) Serum levels of IL-23 in RRMS (2123 (649ndash6253) pgmL) and OND (1489(546ndash7749) pgmL) patients (d) Serum levels of MCP-3 in RRMS (63 (26ndash184) pgmL) and OND (75 (35ndash147) pgmL) patients (e) CSFlevels of IL-17 in RRMS (70 (20ndash111) pgmL) and OND (71 (23ndash161) pgmL) patients (f) CSF levels of IL-18 in RRMS (218 (49ndash551) pgmL)and OND (269 (71ndash644) pgmL) patients (g) CSF levels of IL-23 in RRMS (134 (59ndash659) pgmL) and OND (182 (114ndash571) pgmL) patients(h) CSF levels of MCP-3 in RRMS (26 (15ndash78) pgmL) and OND (43 (13ndash91) pgmL) patients IQR interquartile range CSF cerebrospinalfluid IL interleukin MCP Monocyte Chemoattractant Protein RRMS relapsing-remitting MS OND other neurological disorders

Table 2 Correlation matrix of active MMP-9 active MMP-2 and interleukins measured in the serum of MS patients

Variables (1) (2) (3) (4) (5) (6)(1) Active MMP-9 mdash(2) Active MMP-2 minus0166 (89) mdash(3) IL-17 minus0207 (22) 0290 (22) mdash(4) IL-18 minus0204 (77) 0132 (77) 0387 (22) mdash(5) IL-23 minus0196 (64) minus0017 (64) 0466 (22)lowast minus0138 (63) mdash(6) MCP-3 minus0128 (55) 0028 (55) 0922 (21)lowastlowast 0212 (55) 0468 (48)lowastlowast mdashValues in brackets represent the degrees of freedom lowast119901 lt 005 lowastlowast119901 lt 001

34 Evaluation of Abnormal MMPs and Interleukin Levels inSerum and CSF of MS Patients and Controls Based on themedian values of the activeMMPs and cytokinesmeasured inthe serum or CSF from the whole population we determinedin all subjects whether the active MMP-9 or cytokines wereincreased or active MMP-2 was decreased These values wereconsidered abnormal In addition the cytokine abnormalvalues were merged in one category (Table 4 combinedinterleukins) including subjects with at least one abnormalvalue of IL-17 IL-18 IL-23 or MCP-3

As shown in Table 4 we did not find any difference in thefrequency of abnormal levels of either interleukins or MMPsin serum The same result was observed when we analyzedthe frequency of patients with abnormal CSF levels of theconsidered proteins with the exception of the active MMP-9 Indeed there was a higher proportion of MS patients withabnormally increased levels of the enzyme compared toOND

controls (Pearson Chi-square (1) 9335 119901 = 0002) Then wesearched for possible association of abnormal levels ofMMPsand interleukins with MS by employing a binary logisticregression analysis considering the diagnosis (MS or OND)as the outcome variable and entering the abnormal levelsof MMPs or cytokines alone or in combination (combinedinterleukins) as predictors From this analysis no associationemerged between serum active MMP-9 active MMP-2 orthe cytokines and MS pathology On the contrary when weanalyzed the proteins measured in the CSF we found thatonly the abnormal values of active MMP-9 were associatedwith an increased likelihood of being affected by MS (OddsRatio 252 95 Confidence Interval 139ndash459 119901 = 0002)Of note the inclusion of the other covariates did not improvethe reliability of the model (data not shown)

Finally we compared the levels of serum or CSF activeMMP-9 and MMP-2 measured in MS patients grouped

Disease Markers 7

Table 3 Correlation matrix of active MMP-9 active MMP-2 and interleukins measured in the CSF of MS patients

(1) (2) (3) (4) (5) (6)(1) Active MMP-9 mdash(2) Active MMP-2 0244 (84) mdash(3) IL-17 minus0306 (27) minus0129 (25) mdash(4) IL-18 minus0076 (56) 0300 (53)lowast 0437 (19) mdash(5) IL-23 0075 (17) minus0344 (16) 0450 (7) 0248 (10) mdash(6) MCP-3 minus0127 (50) 0237 (47) 0485 (20)lowast 0454 (33)lowastlowast 0575 (13)lowast mdashValues in brackets represent the degrees of freedom lowast119901 lt 005 lowastlowast119901 lt 001

Table 4 Percentage of MS patients and OND controls withabnormal serum and CSF values

OND () MS ()Serum

High active MMP-9 467 539Low active MMP-2 533 494High IL-17 87 124High IL-18 435 427High IL-23 326 382High MCP-3 304 303Combined interleukins 630 697

CSFHigh active MMP-9 402 629lowastlowast

Low active MMP-2 518 476High IL-17 207 135High IL-18 283 303High IL-23 98 90High MCP-3 283 247Combined interleukins 457 449

The cut-off values used for the determination of the frequency of abnormalvalues were as followsSerum active MMP-9 534 ngmL active MMP-2 252 ngmL IL-17336 pgmL IL-18 2262 pgmL IL-23 181 pgmL MCP-3 72 pgmLCSF activeMMP-9 0055 ngmL activeMMP-2 506 ngmL IL-17 7 pgmLIL-18 237 pgmL IL-23 159 pgmL MCP-3 3 pgmLlowastlowast119901 lt 001

according to the abnormal level of cytokines alone or incombinationThis analysis did not show any difference in theconcentrations of the two MMPs between the patients withnormal or high values of ILs either in serum or in CSF

4 Discussion

There is evidence connecting both MMPs and Th17Th1-related cytokines to the pathogenesis of MS Indeed bothfamilies of proteins have been advocated asmarkers of diseaseactivity [31ndash33] for therapeutic response [34] and as activeplayers in theMS disease course [15 35] In particular MMPsare involved in both BBB disruption and formation of MSlesions [36] whereas cytokines and chemokines may playimportant roles in the recruitment of leukocytes into the CNS[4] and in the initiation of the autoimmune tissue inflam-mation [15] Nevertheless MMPs and cytokineschemokinesmay also cooperate in the opening of the BBB a key event that

can further support the leukocyte migration within the CNS[37] Notwithstanding the accumulating evidence that mightsuggest a possible interplay between MMPs and cytokinesthere is still a lack of clinical studies exploring possibleassociations between the mentioned molecules in the serumand CSF of MS patients

In light of the above considerations we set out thepresent study with the aim to evaluate the contributionof MMPs namely active MMP-9 and MMP-2 and thecytokineschemokines IL-17 IL-18 IL-23 and MCP-3 to thepathogenesis of MS More importantly for the first timewe evaluated the possible intercorrelations involving thesetwo classes of molecules In agreement with previous studies[31 38] we found that serum active MMP-9 was higher inpatients with MS and OND compared to healthy controlswhereas the CSF active MMP-9 was selectively elevated inMS patients On the contrary our finding of the lack ofassociation between serum and CSF levels of active MMP-2 and MS disagrees with previous observations [32] Inour view divergences in patient selection or genetic andenvironmental factors [39] might partially explain theseconflicting results

The lack of difference we found in the serum and CSFcytokine concentrations also appears in contradiction withprevious reports showing higher serum levels of IL-23 andIL-18 in MS patients compared to healthy controls [23 3440] Moreover other studies showed increased [40] but alsounchanged [34] levels of IL-17 in the serum or PBMC [41] ofMS patients compared to controlsThis apparent discrepancymight be due to a different selection in the control groupsince at variance of ours most of the studies compared MSpatients with heathy donors and to the low detectable rateof cytokines in both serum and CSF Of note to the best ofour knowledge there are no data in literature about MCP-3circulating levels

The observed strong correlations between IL-17 IL-23and MCP-3 in the serum and between MCP-3 IL-17 IL-18and IL-23 in the CSF of MS patients suggest that thoughnot massive the MS pathology might be characterized bya general overproduction of cytokines and chemokinesHowever if the overproduction occurs it remains within thenormal values measured in OND controls since we did notfind any difference in the proportion of abnormal cytokinelevels between the two groups Collectively these resultssuggest that at least in our cohort cytokines might representpoor surrogate markers of the disease

8 Disease Markers

On the contrary active MMP-9 which was selectivelyelevated in the CSF of MS patients could be consideredas appropriate indicator of ongoing inflammation in MSConsistently we found a higher proportion of MS patientswith abnormal CSF levels of active MMP-9 compared toOND patients with an increased likelihood of being affectedby MS (Odds Ratio 252) However the missed correlationbetween active MMP-9 and cytokines in either the serumor CSF (although likely due to the low detection rate ofcytokines) suggests that the activation cascade of the enzymemight not act in concert with these soluble proinflammatoryfactors in the course of MS

On the other hand the significant positive correlationbetween active MMP-2 and IL-18 suggests that this proin-flammatory cytokinemight be able tomodulate the activationcascade of MMP-2 In line with this hypothesis a recentin vitro study on neuron-like cells reported an upregulationof MMP-14 the physiological activator of MMP-2 upontreatment with increasing amounts of IL-18 [42] Howeverthis finding seems in contradiction with the supposed majorrole of active MMP-2 in the resolution phase of the diseasewhere its levels were lower in patients with MRI evidence ofdisease activity [32] indicating a possible anti-inflammatoryaction [26] Of note we did not find any difference inboth MMPs and cytokine levels between clinically activeand inactive MS patients suggesting that these moleculesdo not seem correlated to clinical exacerbations However itis well known that MRI is superior on clinical examinationin measuring MS disease activity [43] and thus we cannotexclude that the real contribution of these proteins to thedisease pathogenesis has been underestimated Indeed inprevious reports [32 38] we observed differences in activeMMPs only when patients were categorized in active orinactive disease based on MRI findings

This study was not without its limitations First thesmall sample size may have weakened the consistency of ourdata making it difficult to draw any definitive conclusionabout the possible use of the analyzed molecules as reliablebiomarkers of the disease Second the design of the study wascross-sectional thereby precluding our ability to establishany real causeeffect relationship between the cytokines andMMPs A longitudinal approach could be more suitableThird the lack of complete data on the clinical activity of thedisease may have mined the ability to detect real differencesbetween clinically active and stable MS patients Howeverin previous studies we did not find significant differenceswhen patients were grouped according to clinical evidence ofdisease activity [31 32] Fourth the lack ofMRI examinationsin our study could have affected our findings Finally thenumber of patients and controls with detectable levels ofcytokines was low limiting the reliability of our resultsConsequently a replication of the data in larger cohorts ofMS patients and controls is warranted

In conclusion although with limitations our studyconfirms that active MMP-9 could be a potential surrogatemarker for monitoring MS disease whereas cytokinesand chemokines seem not able to discriminate betweenMS patients and controls Nonetheless our results alsohighlighted that MMPs and cytokines might synergistically

cooperate in MS indicating them as potential partners inthe disease processes Further studies in a larger number ofpatients are needed to verify the effective nature and roleof this cooperation in the modulation of the inflammatoryresponses operating in MS

Competing Interests

The authors declare that they have no conflict of interests

Acknowledgments

This work has been supported by Research ProgramRegione Emilia Romagna-University 2007ndash2009 (InnovativeResearch) entitled ldquoRegional Network for Implementing aBiological Bank to Identify Biological Markers of DiseaseActivity Related to Clinical Variables in Multiple Sclerosisrdquo

References

[1] A Compston and A Coles ldquoMultiple sclerosisrdquoThe Lancet vol372 no 9648 pp 1502ndash1517 2008

[2] J Goverman ldquoAutoimmune T cell responses in the centralnervous systemrdquo Nature Reviews Immunology vol 9 no 6 pp393ndash407 2009

[3] E H Wilson W Weninger and C A Hunter ldquoTrafficking ofimmune cells in the central nervous systemrdquo The Journal ofClinical Investigation vol 120 no 5 pp 1368ndash1379 2010

[4] C McManus J W Berman F M Brett H Staunton M Farrelland C F Brosnan ldquoMCP-1 MCP-2 and MCP-3 expression inmultiple sclerosis lesions an immunohistochemical and in situhybridization studyrdquo Journal of Neuroimmunology vol 86 no1 pp 20ndash29 1998

[5] G R Dos Passos D K Sato J Becker and K FujiharaldquoTh17 cells pathways in multiple sclerosis and neuromyelitisoptica spectrum disorders pathophysiological and therapeuticimplicationsrdquo Mediators of Inflammation vol 2016 Article ID5314541 11 pages 2016

[6] AMinagar and J S Alexander ldquoBlood-brain barrier disruptionin multiple sclerosisrdquo Multiple Sclerosis vol 9 no 6 pp 540ndash549 2003

[7] V W Yong ldquoMetalloproteinases mediators of pathology andregeneration in the CNSrdquo Nature Reviews Neuroscience vol 6no 12 pp 931ndash944 2005

[8] L W Lau R Cua M B Keough S Haylock-Jacobs and V WYong ldquoPathophysiology of the brain extracellular matrix a newtarget for remyelinationrdquo Nature Reviews Neuroscience vol 14no 10 pp 722ndash729 2013

[9] D Leppert J FordG Stabler et al ldquoMatrixmetalloproteinase-9(gelatinase B) is selectively elevated in CSF during relapses andstable phases of multiple sclerosisrdquo Brain vol 121 no 12 pp2327ndash2334 1998

[10] MA Lee J Palace G Stabler J Ford A Gearing andKMillerldquoSerum gelatinase B TIMP-1 and TIMP-2 levels in multiplesclerosis A longitudinal clinical andMRI studyrdquo Brain vol 122no 2 pp 191ndash197 1999

[11] E Waubant D E Goodkin L Gee et al ldquoSerum MMP-9 andTIMP-1 levels are related to MRI activity in relapsing multiplesclerosisrdquo Neurology vol 53 no 7 pp 1397ndash1401 1999

Disease Markers 9

[12] GM Liuzzi M TrojanoM Fanelli et al ldquoIntrathecal synthesisof matrix metalloproteinase-9 in patients with multiple sclero-sis implication for pathogenesisrdquo Multiple Sclerosis vol 8 no3 pp 222ndash228 2002

[13] Y Benesova A Vasku H Novotna et al ldquoMatrix metallopro-teinase-9 and matrix metalloproteinase-2 as biomarkers ofvarious courses in multiple sclerosisrdquoMultiple Sclerosis vol 15no 3 pp 316ndash322 2009

[14] V W Yong R K Zabad S Agrawal A Goncalves DaSilva andL MMetz ldquoElevation of matrix metalloproteinases (MMPs) inmultiple sclerosis and impact of immunomodulatorsrdquo Journalof the Neurological Sciences vol 259 no 1-2 pp 79ndash84 2007

[15] A Amedei D Prisco andMMDrsquoElios ldquoMultiple sclerosis therole of cytokines in pathogenesis and in therapiesrdquo InternationalJournal of Molecular Sciences vol 13 no 10 pp 13438ndash134602012

[16] L Yang D E Anderson C Baecher-Allan et al ldquoIL-21 andTGF-120573 are required for differentiation of human TH17 cellsrdquoNature vol 454 no 7202 pp 350ndash352 2008

[17] G L Stritesky N Yeh and M H Kaplan ldquoIL-23 promotesmaintenance but not commitment to the Th17 lineagerdquo Journalof Immunology vol 181 no 9 pp 5948ndash5955 2008

[18] D J Cua J Sherlock Y Chen et al ldquoInterleukin-23 ratherthan interleukin-12 is the critical cytokine for autoimmuneinflammation of the brainrdquo Nature vol 421 no 6924 pp 744ndash748 2003

[19] C L Langrish Y Chen W M Blumenschein et al ldquoIL-23drives a pathogenic T cell population that induces autoimmuneinflammationrdquo Journal of ExperimentalMedicine vol 201 no 2pp 233ndash240 2005

[20] C Lock G Hermans R Pedotti et al ldquoGene-microarrayanalysis of multiple sclerosis lesions yields new targets validatedin autoimmune encephalomyelitisrdquo Nature Medicine vol 8 no5 pp 500ndash508 2002

[21] R C Axtell B A De Jong K Boniface et al ldquoT helper type 1and 17 cells determine efficacy of interferon-Β in multiple scle-rosis and experimental encephalomyelitisrdquo Nature Medicinevol 16 no 4 pp 406ndash412 2010

[22] S Alboni D Cervia S Sugama and B Conti ldquoInterleukin 18 inthe CNSrdquo Journal of Neuroinflammation vol 7 article 9 2010

[23] J Losy and A Niezgoda ldquoIL-18 in patients with multiplesclerosisrdquoActaNeurologica Scandinavica vol 104 no 3 pp 171ndash173 2001

[24] Y Ishida K Migita Y Izumi et al ldquoThe role of IL-18 inthe modulation of matrix metalloproteinases and migration ofhuman natural killer (NK) cellsrdquo FEBS Letters vol 569 no 1ndash3pp 156ndash160 2004

[25] J Song C Wu E Korpos et al ldquoFocal MMP-2 and MMP-9 activity at the blood-brain barrier promotes chemokine-induced leukocyte migrationrdquo Cell Reports vol 10 no 7 pp1040ndash1054 2015

[26] G A McQuibban J-H Gong J P Wong J L Wallace IClark-Lewis and C M Overall ldquoMatrix metalloproteinaseprocessing of monocyte chemoattractant proteins generatesCC chemokine receptor antagonists with anti-inflammatoryproperties in vivordquo Blood vol 100 no 4 pp 1160ndash1167 2002

[27] D Westermann K Savvatis D Lindner et al ldquoReduced degra-dation of the chemokine MCP-3 by matrix metalloproteinase-2 exacerbates myocardial inflammation in experimental viralcardiomyopathyrdquo Circulation vol 124 no 19 pp 2082ndash20932011

[28] W I McDonald A Compston G Edan et al ldquoRecommendeddiagnostic criteria for multiple sclerosis guidelines from theinternational panel on the diagnosis of multiple sclerosisrdquoAnnals of Neurology vol 50 no 1 pp 121ndash127 2001

[29] C M Poser D W Paty L Scheinberg et al ldquoNew diagnosticcriteria for multiple sclerosis guidelines for research protocolsrdquoAnnals of Neurology vol 13 no 3 pp 227ndash231 1983

[30] J F Kurtzke ldquoRating neurologic impairment in multiple sclero-sis an expanded disability status scale (EDSS)rdquo Neurology vol33 no 11 pp 1444ndash1452 1983

[31] E Fainardi M Castellazzi T Bellini et al ldquoCerebrospinal fluidand serum levels and intrathecal production of active matrixmetalloproteinase-9 (MMP-9) as markers of disease activity inpatients with multiple sclerosisrdquoMultiple Sclerosis vol 12 no 3pp 294ndash301 2006

[32] E Fainardi M Castellazzi C Tamborino et al ldquoPotentialrelevance of cerebrospinal fluid and serum levels and intrathecalsynthesis of active matrix metalloproteinase-2 (MMP-2) asmarkers of disease remission in patients withmultiple sclerosisrdquoMultiple Sclerosis vol 15 no 5 pp 547ndash554 2009

[33] A P Kallaur S R Oliveira A N C Simao et al ldquoCytokineprofile in relapsing-remittingMultiple sclerosis patients and theassociation between progression and activity of the diseaserdquoMolecular Medicine Reports vol 7 no 3 pp 1010ndash1020 2013

[34] J S Alexander M K Harris S R Wells et al ldquoAlterationsin serum MMP-8 MMP-9 IL-12p40 and IL-23 in multiplesclerosis patients treatedwith interferon-1205731brdquoMultiple Sclerosisvol 16 no 7 pp 801ndash809 2010

[35] G Opdenakker and J Van Damme ldquoProbing cytokineschemokines and matrix metalloproteinases towards betterimmunotherapies of multiple sclerosisrdquo Cytokine and GrowthFactor Reviews vol 22 no 5-6 pp 359ndash365 2011

[36] F Romi G Helgeland and N E Gilhus ldquoSerum levels ofmatrix metalloproteinases implications in clinical neurologyrdquoEuropean Neurology vol 67 no 2 pp 121ndash128 2012

[37] C Larochelle J I Alvarez and A Prat ldquoHow do immune cellsovercome the blood-brain barrier in multiple sclerosisrdquo FEBSLetters vol 585 no 23 pp 3770ndash3780 2011

[38] A Trentini M C Manfrinato M Castellazzi et al ldquoTIMP-1resistant matrix metalloproteinase-9 is the predominant serumactive isoform associated with MRI activity in patients withmultiple sclerosisrdquoMultiple Sclerosis vol 21 no 9 pp 1121ndash11302015

[39] F M Goncalves A Martins-Oliveira R Lacchini et al ldquoMatrixmetalloproteinase (MMP)-2 gene polymorphisms affect circu-lating MMP-2 levels in patients with migraine with aurardquoGenevol 512 no 1 pp 35ndash40 2013

[40] Y-C Chen S-D Chen L Miao et al ldquoSerum levels ofinterleukin (IL)-18 IL-23 and IL-17 in Chinese patients withmultiple sclerosisrdquo Journal of Neuroimmunology vol 243 no1-2 pp 56ndash60 2012

[41] DMatusevicius P Kivisakk B He et al ldquoInterleukin-17mRNAexpression in blood andCSFmononuclear cells is augmented inmultiple sclerosisrdquo Multiple Sclerosis vol 5 no 2 pp 101ndash1041999

[42] EM SutinenMA Korolainen J Hayrinen et al ldquoInterleukin-18 alters protein expressions of neurodegenerative diseases-linked proteins in human SH-SY5Y neuron-like cellsrdquo Frontiersin Cellular Neuroscience vol 8 article 214 2014

[43] D H Miller F Barkhof and J J P Nauta ldquoGadoliniumenhancement increases the sensitivity of MRI in detectingdisease activity in multiple sclerosisrdquo Brain vol 116 part 5 pp1077ndash1094 1993

Review ArticleA Tale of Two Joints The Role of Matrix Metalloproteases inCartilage Biology

Brandon J Rose and David L Kooyman

Department of Physiology and Developmental Biology Brigham Young University (BYU) LSB 4005 Provo UT 84602 USA

Correspondence should be addressed to Brandon J Rose brose04008gmailcom

Received 26 March 2016 Accepted 12 June 2016

Academic Editor Fatma M El-Demerdash

Copyright copy 2016 B J Rose and D L KooymanThis is an open access article distributed under theCreative CommonsAttributionLicense which permits unrestricted use distribution and reproduction in anymedium provided the originalwork is properly cited

Matrix metalloproteinases are a class of enzymes involved in the degradation of extracellular matrix molecules While thesemolecules are exceptionally effective mediators of physiological tissue remodeling as occurs in wound healing and duringembryonic development pathological upregulation has been implicated in many disease processes As effectors and indicatorsof pathological states matrix metalloproteinases are excellent candidates in the diagnosis and assessment of these diseases Thepurpose of this review is to discuss matrix metalloproteinases as they pertain to cartilage health both under physiologicalcircumstances and in the instances of osteoarthritis and rheumatoid arthritis and to discuss their utility as biomarkers in instancesof the latter

1 Introduction

Matrix metalloproteinases are a family of zinc-dependentendopeptidases collectively capable of degrading all compo-nents of the extracellularmatrixThe actions of these enzymesare potent and highly catabolic and as such physiologicexpressions of the genes coding formatrixmetalloproteinasesare strictly regulated and reserved for instances where dra-matic tissue remodeling is required as occurs during woundhealing [1] and embryonic development [2] Their versatilityand efficacy also render them potent effectors of pathologicalprocesses and this is where much interest in their activityis garnered Ectopic overexpressionmatrixmetalloproteinaseactivity has been implicated in a wide array of disease statesincluding tumor initiation and metastasis atherosclerosisosteoarthritis and rheumatoid arthritis The purpose of thepresent review is to discuss matrix metalloproteinases as theyrelate to articular cartilage homeostasis

2 The Role of Matrix Metalloproteinases inHealthy Cartilage

Seven matrix metalloproteinases have been shown tobe expressed under varying circumstances in articularcartilagemdashmatrix metalloproteinase-1 (MMP-1) matrix

metalloproteinase-2 (MMP-2) matrix metalloproteinase-3(MMP-3) matrix metalloproteinase-8 (MMP-8) matrixmetalloproteinase-9 (MMP-9) matrix metalloproteinase-13(MMP-13) and matrix metalloproteinase-14 (MMP-14)Of those seven four have been found to be constitutivelyexpressed in adult cartilage presumably serving roles intissue turnover and upregulated in diseased statesmdashMMP-1MMP-2 MMP-13 and MMP-14 [3] The presence of theMMP-3 MMP-8 and MMP-9 in cartilage appears to becharacteristic of pathologic circumstances only

MMP-1 (interstitial collagenase) is involved in the degra-dation of collagen types I II and III In embryonic develop-ment its expression is restricted to areas of endochondral andintramembranous bone formation and is especially abundantin the metaphyses and diaphysis of long bones During thattime it is expressed in hypertrophic chondrocytes (imme-diately preceding terminal differentiation in endochondralossification) and osteoblasts only [4] Expression levels arelow under healthy circumstances but significant upregula-tion is observed in arthritic cartilage and may play an activerole in collagen degradation in this tissue but is evidentlyabsent in the instance of synovitis [5]

MMP-2 (gelatinase A) is involved in the breakdown oftype IV collagen and ismost commonly expressed early in theprocess of wound healing [6] Expression in adult cartilage is


2 Disease Markers

weak and attributable to normal (very low) collagen turnoverand similar toMMP-1 it is upregulated in arthritic states [7]

MMP-3 (stromelysin-1) is capable of degrading a widearray of extracellular molecules including collagen typesII III IV IX and X fibronectin laminin elastin andvarious proteoglycans In addition it has been found to havetranscription factor-like activity apparently being able toupregulate the expression of other matrix metalloproteinases[8] It is involved in wound healing expression being typicalin fibroblasts and epithelial cells following expression toinflammatory compounds [9] possibly explaining the pres-ence of high MMP-3 levels in osteoarthritic cartilage and thesynovium in osteoarthritis [10] and absence in normal jointtissues and showing promise for this enzyme as a candidatemarker for osteoarthritis [11]

MMP-8 (neutrophil collagenase) is the principal colla-genase found in human dentin being involved in turnoverand remodeling in that tissue [12] and it is expressed in awide array of cell types including neutrophil precursors andepithelial cells [13] Consistent with most other matrix met-alloproteinases it is involved principally in wound healingmostly in wounds of an acute character [14] Its expressionin arthritic tissue is clearly beneficial genetic deficienciesof MMP-8 exacerbate inflammation in arthritis throughdownregulation of neutrophil apoptosis and clearance sub-sequently causing hyperinfiltration of joints with neutrophils[15]

MMP-9 (gelatinase B) similar to MMP-1 is most activeduring embryonic development being essential to angio-genesis in the growth plate and apoptosis of hypertrophicchondrocytes in utero [16] It has also been demonstratedto be highly expressed in the early stages of wound healing[17] perhaps due to its involvement in angiogenesis In thejoint capsule it is produced by monocytes and macrophagesproduction by chondrocytes appears minimal [18] thoughthese cells do appear to play a considerable regulatory role inleukocyte MMP-9 expression Inhibition of leukocyte releaseby chondrocytes appears to be lifted in an arthritic stateapparently in response to increasedMMP-3 activity (possiblyvia transcriptional regulation) and MMP-13 expression [19]

MMP-13 is by far the most studied of the matrix met-alloproteinases in terms of its role in cartilage as it isconsidered the major catabolic effector in osteoarthritis andother forms of arthritis owing to its robust ability to cleavethe type II collagen that predominates in articular cartilageHaving such a unique and dramatic effect on said tissue it isobvious why MMP-13 is frequently employed as the matrixmetalloproteinase of choice in the detection and study ofosteoarthritis Particularly telling and supporting its utilityas a biomarker in osteoarthritis is the tendency of externalinfluences to act upon the joint through upregulation ofMMP-13 specifically as will be discussed in further detaillater in this review Though as is mentioned it is involvedprincipally in the degradation of type II collagen the enzymealso targets other matrix molecules such as types IV and IXcollagen perlecan osteonectin and proteoglycan [20] and itis likely involved in matrix turnover in healthy cartilage

MMP-14 (membrane-type 1 matrix metalloproteinase)is involved in aggrecan degradation and cadherin cleavage

and has been shown to be involved in inhibition of tumorangiogenesis [21] It has been shown to have a significant rolein postnatal bone formation through promotion of osteo-genesis and chondrogenesis [22] MMP-14 is upregulatedin arthritic cartilage and furthermore appears to have theability to activate MMP-2 [23] and MMP-13 [24] potentiallycompounding its influence in arthritis

The careful modulation of matrix metalloproteinases isrequired for maintenance of cartilage health The presenceof these enzymes alone does not constitute pathology asindicated deficiencies in MMP-8 are deleterious to jointhealth rather a careful balance is required for maintenanceof the anaboliccatabolic balance and it is dysregulation thatbrings about the catabolism of articular cartilage in arthriticdisease

Having discussed the role of matrix metalloproteinasesin healthy cartilage it is now pertinent to discuss theiroverexpression and the relation of this phenomenon todisease states beginning with osteoarthritis and followingwith rheumatoid arthritis

3 The HTRA1-DDR2-MMP-13 Axis

As indicated several matrix metalloproteinases are upreg-ulated and known to play a role beneficial or deleteriousin osteoarthritis and as such are candidate biomarkers inosteoarthritis Nevertheless we look to one in particularMMP-13 as the principal effector of cartilage degradation inosteoarthritis and the most obvious and useful candidate formatrix metalloproteinases as a biomarker in the disease

Common to the pathogenesis of osteoarthritis in knownprocesses culminating in the condition is the activation of theHTRA1-DDR2-MMP-13 axis High temperature requirementA1 (HTRA1) a serine protease is strongly expressed inthe presence of stressors in murine osteoarthritis modelsHTRA1 is responsible for degradation of pericellular matrixcomponents including fibronectinmatrilin 3 collagen oligo-metric matrix protein biglycan fibromodulin and type VIcollagen [25] Breakdown of the pericellular matrix exposesthe chondrocyte membrane to the type II collagen char-acteristic of articular cartilage activating and augmentingthe expression of the transmembrane discoidin-containingdomain receptor 2 (DDR2) [25] Heightened expression ofDDR2 results in excess binding of the receptor to its ligand[26] in turn stimulating high levels of expression of theMMP-13 gene culminating in the extracellular matrix andleading to the destruction of articular cartilage [27 28] Asidefrom the wide array of evidence showing HTRA1-DDR2-MMP-13 activity in osteoarthritis of multiple modalitiesthe convergence of diverse noxious stimuli upon this axisis further supported in the protective effects exerted inDDR2 hypomorphic strains of mice which when subjectto the DMM procedure demonstrated a significant decreasein the progression of osteoarthritis compared to wild typelittermates [29] comparable to the ablation of MMP-13which has the effect of protecting cartilage from degrada-tion in osteoarthritis induced through destabilization of themedial meniscal ligament (DMM-induced osteoarthritis) inmurine models though interestingly enough chances to the

Disease Markers 3

chondrocytes themselves and other cells in response to theprocedure persisted [17]

4 Molecular Pathways Associated withTranscriptional Regulation of MMP-13Gene Expression

Gene expression of MMP-13 appears to occur through anumber of molecular pathways that work through eitherinflammation or primary cilia That is not to say there arenot some common themes Stress-inducible nuclear protein1 (Nupr1) has been shown to regulate MMP-13 expressionin vitro [30] Yammani and Loeser showed that Nupr1expressed in cartilage is required for expression of MMP-13via IL-1120573 This might be a pathway for the catabolic effects ofOA to be mediated through inflammation This is especiallyinteresting in light of the study done by Xu et al 2015in which they analyzed differential expression of genes incartilage involved inOA and RA [31]While these researchersidentified multiple genes associated with the regulation ofMMPs the predominant ones were associated with inflam-mation This might give greater credence for the role of earlyinflammatory signals (ie AGEs and IL-1) in the initiationand progression of OA Meanwhile more obviously a similarrole for inflammation appears to be present in RA Araki etal 2016 reported that histone methylation and the bindingof signal transducer activator of transcription 3 (STAT3) wereassociated with RA and OA [32] They report that histoneH3 methylation is associated with elevated expression ofMMP-1 MMP-3 MMP-9 and MMP-13 However STAT3was shown to increase expression either spontaneous or IL-6activated of MMP-1 MMP-3 and MMP-13 but not MMP-9 As previously indicated primary cilia appear to also beinvolved in OA Sugita et al 2015 reported that transcriptionfactor hairy and enhancer of split-1 (Hes1) is involved in theupregulation of expression of MMP-13 [33] Normally Hes1acts as a transcriptional repressor but under the influence ofcalciumcalmodulin-dependent protein kinase 2 (CaMK2) itbecomes a transcriptional activator thus upregulatingMMP-13 expression [34] Thus Hes1 acts to increase expression ofMMP-13 It is of particular interest to note that Hes1 actsthrough Notch signaling pathway [35] Notch has previouslybeen shown to modulate sonic hedgehog signaling and workthrough primary cilia [36 37] In an apparent unrelatedmechanismNiebler et al 2015 showed that the transcriptionfactor AP-2120598 is intimately involved in the upregulation ofMMP-13 as OA progresses [38]

5 Metabolic Syndrome and Upregulation ofMatrix Metalloproteinases

An area of study in the field of rheumatology that presentsample opportunity for research and great promise for ther-apeutic intervention involves the interaction between artic-ular cartilage and metabolic (insulin resistance) syndromeThe comorbidity between osteoarthritis and metabolicsyndrome and its individual manifestations is strikingmdashepidemiological data reveals that 49 of individuals with

heart disease 47with diabetes 44with hypertension and31 of obese individuals also have some form of arthritis[39] suggestive of a commonor overlapping etiology betweenosteoarthritis and these conditions Further incriminating aload-bearing hypothesis of osteoarthritis has been the deter-mination that hand osteoarthritis is more strongly correlatedwith body mass index than hip osteoarthritis the latter ofwhich was found to have such a weak relationship as to notbeing statistically significant [40]

The opportunities for interaction between metabolicsyndrome and osteoarthritis are vast and cross a temporalspectrum spanning from an initial hyperinsulinemic state[41] to proper insulin resistance [42] to the downstreameffects of metabolic syndrome including but not limited totype II diabetes mellitus [43] a decrease in circulating HDLparticles [44] and high circulating levels of adipokines

While hyperinsulinemia is implicated in the chondrocyteapoptosis facet of osteoarthritis [41] expression of MMP-13 has not been documented to occur until the point ofinsulin resistance in the joint capsule In one study mice feda high fat diet to generate the obesetype II diabetes mellitus(obt2d) phenotype showed considerably increased levels oftumor necrosis factors (TNFs) in the synovial fluid of theknee joint and studies comparing obt2d with TNF knockoutmice demonstrated that TNF species were linked to increasedexpression of MMP-1 MMP-13 and ADAMTS4mdasha mousehomologue of matrix metalloproteinases Supplementationwith insulin in these mice inhibited these effects by 50 [42]

Leptin a peptide hormone involved in maintaininginsulin sensitivity and contributing to the sensation of satietyis expressed at very high levels in obese individuals It appearsto be correlated with osteoarthritis as well with interventionat the level of MMP-13 expression occurring Downregu-lation of leptin mRNA translation via small interferenceRNA molecules inhibits MMP-13 expression in culturedosteoarthritic chondrocytes [45] The situation appears to bemost exacerbated in cases of extreme obesity there exists astrong positive correlation between the responsiveness of theMMP-13 gene to leptin and the BMI of osteoarthritic individ-uals [46] Its effects are not limited to MMP-13 alone MMP-1 expression and MMP-3 expression are strongly upregu-lated in osteoarthritic cartilage and leptin levels stronglycorrelated with the presence of these two matrix metallo-proteinases in osteoarthritic synovial fluid [47] Adiponectinalso appears to have some ability to stimulate expression ofMMP-13The presence of adiponectin is positively correlatedwith the presence of membrane-associated prostaglandin E2synthase (mPGES) andMMP-13 [48] Furthermore culturedosteoarthritic chondrocytes treated with adiponectin showedincreases in production of nitric oxide via inducible nitricoxide synthase (iNOS) expression of MMP-1 MMP-3 andMMP-13 and levels of collagenase-cleaved type II collagenneoepitope These effects were attenuated in the presenceof AMP-activated protein kinase (AMPK) c-Jun N-terminalkinase and iNOS inhibitors implicating these as mediatorsof adiponectin-induced insult [49]

Hyperglycemia is linked to the presence of high levelsof circulating advanced glycation end products particularlyof the S100 family of proteins [50] This phenomenon is

4 Disease Markers

linked to an array of diabetes-induced complications includ-ing atherosclerosis and a deficiency in the receptor foradvanced glycation end products (RAGE) proves to haveprotective effects implicating this receptor in the pathwayAblation of RAGE in osteoarthritic murine models likewiseshows a protective effect wild-type mice on the otherhand demonstrate increased expression of proinflammatorymarkers and catabolic mediators including MMP-13 [51]RAGE is known to act through a host of proinflammatorymeans including NF-120581B TNF IL-1 and TGF-120573 [52] andthough it has not been conclusively demonstrated there isabundant potential for a catabolic role of RAGE in metabolicosteoarthritis

6 Matrix Metalloproteinases in Response toMechanical Insult

Activation of matrix metalloproteinases especially MMP-13is known to occur in response to mechanical injury to thejoint and factors leading to its expression are far more clear-cut than in metabolic osteoarthritis In response to injuryto the joint the first response that appears to be elicitedfrom chondrocytes and synoviocytes secretes interleukin-1120573This in turn activates the cell surface receptor IL-1RI whichinitiates a cascade of intracellular events involving NF-120581120573MAPK p38 and JNK This results in increased expressionof TNF-120572 which further potentiates inflammatory eventsalready in play [53] and initiates chondrocyte expression ofMMP-1 [54] MMP-2 [55] and MMP-13 [56]

Over the course of this process chondrocytes begin toexpress the cell proliferant transforming growth factor-120573(TGF-120573) [57] perhaps in response to its ability to mitigatesome of the activities of IL-1120573 and produce additionalchondrocytes to cope with stressors placed on the jointOverexpression of this factor however appears to somehowbe involved in initiating the osteoarthritic process [58] Thiselevation in TGF-120573 corresponds to an increase in HTRA-1[59] perhaps in consequence of the ability of the latter tocleave the former [60] and consequentiallyDDR2 is activatedandMMP-13 is highly expressedThe body of evidence impli-cates MMP-13 as a major effector of mechanically mediatedjoint destruction in osteoarthritis

Consistent with previously discussed studies DDR2 isshown to be a major effector of MMP-13 expression inDMM models such that almost complete protection fromosteoarthritis is shown in DDR2 hypomorphic strains It isalso telling to note that genetic ablation of the receptor foradvanced glycation end products (RAGE) corresponds to adecreased expression of MMP-13 in DMM models thoughnot as dramatic as DDR2 hypomorphism and that thisdecreased expression corresponds with decreased progres-sion and severity of osteoarthritis This suggests a contribut-ing role of RAGE in the pathogenesis of osteoarthritis andcertainly a target for therapeutic intervention Converselyand for reasons that are not yet completely understood phar-macological antagonism of the proinflammatory transmem-brane protein Toll-like receptor 4 (TLR-4) results in height-ened MMP-13 expression and a corresponding dramaticincrease in the severity of osteoarthritis [61] Further research

is required to elucidate a rationale for this phenomenon andthis information must be considered in devising therapeuticintervention for acute joint injury with the goal of delaying orpreventing osteoarthritis development later in life

7 Genetic Anomalies Resulting in MatrixMetalloproteinases Overexpression andOsteoarthritis

As stated a number of disease states involve overexpressionof matrix metalloproteases and there exist diseases in whichmutations result in overexpression of MMP-13 and prema-ture development of osteoarthritis One such abnormalityis spondyloepiphyseal dysplasia congenita (SEDC) whichserves as an experimental model of osteoarthritis In SEDCa mutation occurs in the COL2A1 gene resulting in theformation of superfluous disulfide bridges in type II collagenadversely affecting association between individual collagenmolecules and thus the triple helix that is characteristic of thetypical collagenmolecule [62] It is possible that this failure ofcollagen monomers to form proper interactions predisposesthe individual to premature osteoarthritis rendering themolecules ideal targets for the binding and activity of HTRA1and DDR2 and downstream to these MMP-13 explainingthe dramatically increased levels of each characteristic of thesyndrome [4]

Another disease that has been the focus of osteoarthritisresearch as of late is the ciliopathy Bardet-Biedl syndrome(BBS) BBS may result from mutations in a number ofthe known genes that are involved in ciliary formationand transport [63] resulting in a number of congenitalabnormalities including polydactyly cognitive impairmentcardiac and renal malformation obesity hypertension andtype II diabetes mellitus In addition mouse models of BBSmanifest early onset osteoarthritis whether this is a directresult of ciliary malformation or secondary to osteoarthritisrisk factors such as obesity and type II diabetes mellitus ispresently unclearWhat is known is that in BBS osteoarthriticcartilage TGF-120573 is downregulated HTRA1 is upregulatedand MMP-13 is strongly expressed in the absence of DDR2This finding strongly suggests a role of primary cilia in DDR2activity andor signal transduction and further research isclearly warranted to elucidate the link between cilia andDDR2

8 Matrix Metalloproteinases andRheumatoid Arthritis

Rheumatoid arthritis is a form of arthritis in which the hostimmune systemmounts an attack on connective tissue in thejoint MMPs are associated with rheumatoid arthritis [64]In their study Ahrens et al demonstrated that MMP-9 iselevated in the synovial fluid of patients with rheumatoidarthritis Indeed they indicated that SF levels of MMP-9were higher in rheumatoid arthritis patients compared toosteoarthritis It is of interest to note that they also speculatedthat elevated MMP-9 was associated with connective tissueturnover in rheumatoid arthritis patients

Disease Markers 5

These observations led to the intriguing possibility ofdetermining the severity of rheumatoid arthritis by exami-nation of matrix metalloproteinases in blood Keyszer et alwere the first to show a correlation between blood circulatingmatrix metalloproteinases and the clinical manifestation ofrheumatoid arthritis [65]They demonstrated that circulatinglevels of MMP-3 were a better predictor of rheumatoidarthritis severity or activity than cytokine level These obser-vations led to the thinking that perhaps clinicians couldseparate the immunological and inflammatory mechanismsassociated with RA from the actual erosion of articularsurface Uncoupling these two pathophysiological pathwaysmay aid in new treatment regimens and strategies IndeedCunnane et al were the first to make this connection [66]They showed that the degree of articular surface erosioncorrelated with levels ofMMP-1 andMMP-3They concludedthat treatments for rheumatoid arthritis which specificallytarget MMP-1 may limit the number of new joint erosionfoci and thus improve the overall functional outcome for RApatients

Gender can be a complicating issue for both osteoarthri-tis and rheumatoid arthritis In a recent study in whichmatrix metalloproteinases associated with tuberculosis wereexamined in relation to gender Sathyamoorthy et al foundthatwhile plasmaMMP-8 concentrations inversely correlatedwith body mass index they were significantly higher inmales than in females [67] The authors pointed out that thissignificant difference in gender expression of MMP-8 wasnot associated with or due to disease severity Indeed theyconcluded that plasma analysis of MMP-1 and MMP-8 was abetter discriminator for tuberculosis in men than in womenIn a more recent study it was shown that plasma MMP-3was significantly higher in men compared to women in anumber of clinical conditions that included both infectiousand noninfectious diseases including those rheumatic innature [68]

9 Conclusion

Expression of MMPs is ubiquitous characteristic of devel-opment Adherently high expression of MMPs is associatedwith a host of diseasesmdashinfectious and noninfectious Theyare particularly significant in the progression and severityof osteoarthritis regardless of etiology This attests to theirutility as major biomarkers for osteoarthritis and mediatorsof joint destruction It is worthy to note that MMP-13 is mostnotably associated with osteoarthritis and once its expressionis elevated in the joint significant damage is imminent andthe progression to joint destruction is rapid Present medicalknowledge does not provide a way to reverse damage tocartilage caused by osteoarthritis regardless of the insultingmodality It is highly likely that pharmacologic interventionof osteoarthritis will target upstream of MMP-13 expression

The present short-term treatment for osteoarthritis ispain management typically in the form of opiates andnonsteroidal anti-inflammatory drugs While effective atimproving the quality of life for osteoarthritis sufferers fora time the long-term health consequences are severe andin spite of such interventions joint replacement is almost

invariably necessary eventually There is much opportunityfor research leading to an understanding of the processesupstream of the expression of MMP-13

Competing Interests

The authors declare that there are no competing interestsregarding the publication of this paper

References

[1] N Hattori S Mochizuki K Kishi et al ldquoMMP-13 plays a role inkeratinocytemigration angiogenesis and contraction inmouseskin wound healingrdquo The American Journal of Pathology vol175 no 2 pp 533ndash546 2009

[2] J Shi M-Y Son S Yamada et al ldquoMembrane-type MMPsenable extracellular matrix permissiveness and mesenchymalcell proliferation during embryogenesisrdquo Developmental Biol-ogy vol 313 no 1 pp 196ndash209 2008

[3] S Chubinskaya K E Kuettner and A A Cole ldquoExpressionof matrix metalloproteinases in normal and damaged articularcartilage from human knee and ankle jointsrdquo Laboratory Inves-tigation vol 79 no 12 pp 1669ndash1677 1999

[4] S Gack R Vallon J Schmidt et al ldquoExpression of intersti-tial collagenase during skeletal development of the mouse isrestricted to osteoblast-like cells and hypertrophic chondro-cytesrdquo Cell Growth amp Differentiation vol 6 no 6 pp 759ndash7671995

[5] H Wu J Du and Q Zheng ldquoExpression of MMP-1 in cartilageand synovium of experimentally induced rabbit ACLT trau-matic osteoarthritis immunohistochemical studyrdquo Rheumatol-ogy International vol 29 no 1 pp 31ndash36 2008

[6] T Salo M Makela M Kylmaniemi H Autio-Harmainen andH Larjava ldquoExpression of matrix metalloproteinase-2 and-9during early human wound healingrdquo Laboratory InvestigationA Journal of Technical Methods and Pathology vol 70 no 2 pp176ndash182 1994

[7] S Duerr S Stremme S Soeder B Bau and T Aigner ldquoMMP-2gelatinase A is a gene product of human adult articular chon-drocytes and is increased in osteoarthritic cartilagerdquo Clinicaland Experimental Rheumatology vol 22 no 5 pp 603ndash6082004

[8] T Eguchi S Kubota K Kawata et al ldquoNovel transcriptionfactor-like function of human matrix metalloproteinase 3 reg-ulating the CTGFCCN2 generdquoMolecular and Cellular Biologyvol 28 no 7 pp 2391ndash2413 2008

[9] R L Warner N Bhagavathula K C Nerusu et al ldquoMatrixmetalloproteinases in acute inflammation induction of MMP-3 and MMP-9 in fibroblasts and epithelial cells following expo-sure to pro-inflammatory mediators in vitrordquo Experimental andMolecular Pathology vol 76 no 3 pp 189ndash195 2004

[10] Y Okada M Shinmei O Tanaka et al ldquoLocalization of matrixmetalloproteinase 3 (stromelysin) in osteoarthritic cartilage andsynoviumrdquo Laboratory Investigation vol 66 no 6 pp 680ndash6901992

[11] E Kubota H Imamura T Kubota T Shibata and K-IMurakami ldquoInterleukin 1120573 and stromelysin (MMP3) activityof synovial fluid as possible markers of osteoarthritis in thetemporomandibular jointrdquo Journal of Oral and MaxillofacialSurgery vol 55 no 1 pp 20ndash28 1997

[12] M Sulkala T Tervahartiala T Sorsa M Larmas T Salo and LTjaderhane ldquoMatrixmetalloproteinase-8 (MMP-8) is themajor

6 Disease Markers

collagenase in human dentinrdquo Archives of Oral Biology vol 52no 2 pp 121ndash127 2007

[13] P Van Lint and C Libert ldquoMatrixmetalloproteinase-8 cleavagecan be decisiverdquo Cytokine amp Growth Factor Reviews vol 17 no4 pp 217ndash223 2006

[14] E Pirila J T Korpi T Korkiamaki et al ldquoCollagenase-2 (MMP-8) and matrilysin-2 (MMP-26) expression in human wounds ofdifferent etiologiesrdquoWoundRepair and Regeneration vol 15 no1 pp 47ndash57 2007

[15] J H Cox A E Starr R Kappelhoff R Yan C R Roberts andC M Overall ldquoMatrix metalloproteinase 8 deficiency in miceexacerbates inflammatory arthritis through delayed neutrophilapoptosis and reduced caspase 11 expressionrdquo Arthritis andRheumatism vol 62 no 12 pp 3645ndash3655 2010

[16] T H Vu J M Shipley G Bergers et al ldquoMMP-9gelatinase Bis a key regulator of growth plate angiogenesis and apoptosisof hypertrophic chondrocytesrdquo Cell vol 93 no 3 pp 411ndash4221998

[17] C B Little A Barai D Burkhardt et al ldquoMatrix metallopro-teinase 13-deficient mice are resistant to osteoarthritic cartilageerosion but not chondrocyte hypertrophy or osteophyte devel-opmentrdquo Arthritis and Rheumatism vol 60 no 12 pp 3723ndash3733 2009

[18] S Soder H I Roach S Oehler B Bau J Haag and T AignerldquoMMP-9gelatinase B is a gene product of human adult articularchondrocytes and increased in osteoarthritic cartilagerdquo Clinicaland Experimental Rheumatology vol 24 no 3 pp 302ndash3042006

[19] R Dreier S Grassel S Fuchs J Schaumburger and P BrucknerldquoPro-MMP-9 is a specific macrophage product and is acti-vated by osteoarthritic chondrocytes via MMP-3 or a MT1-MMPMMP-13 cascaderdquo Experimental Cell Research vol 297no 2 pp 303ndash312 2004

[20] T Shiomi V Lemaıtre J DrsquoArmiento and Y Okada ldquoMatrixmetalloproteinases a disintegrin and metalloproteinases anda disintegrin and metalloproteinases with thrombospondinmotifs in non-neoplastic diseasesrdquo Pathology International vol60 no 7 pp 477ndash496 2010

[21] L J A C Hawinkels P Kuiper E Wiercinska et al ldquoMatrixmetalloproteinase-14 (MT1-MMP)-mediated endoglin shed-ding inhibits tumor angiogenesisrdquo Cancer Research vol 70 no10 pp 4141ndash4150 2010

[22] Q Yang M Attur T Kirsch et al ldquoMembrane-type 1 matrixmetalloproteinase controls osteo-and chondrogenesis by aproteolysis-independent mechanism mediated by its cytoplas-mic tailrdquo Osteoarthritis and Cartilage vol 23 article A64 2015

[23] J Esparza C Vilardell J Calvo et al ldquoFibronectin upregulatesgelatinase B (MMP-9) and induces coordinated expression ofgelatinase A (MMP-2) and its activator MT1-MMP (MMP-14)by human T lymphocyte cell lines A process repressed throughRASMAP kinase signaling pathwaysrdquo Blood vol 94 no 8 pp2754ndash2766 1999

[24] V Knauper HWill C Lopez-Otin et al ldquoCellular mechanismsfor human procollagenase-3 (MMP-13) activation Evidencethat MT1-MMP (MMP-14) and gelatinase A (MMP-2) are ableto generate active enzymerdquoThe Journal of Biological Chemistryvol 271 no 29 pp 17124ndash17131 1996

[25] I Polur P L Lee J M Servais L Xu and Y Li ldquoRoleof HTRA1 a serine protease in the progression of articularcartilage degenerationrdquo Histology and Histopathology vol 25no 5 pp 599ndash608 2010

[26] H Xu N Raynal S Stathopoulos JMyllyharju RW Farndaleand B Leitinger ldquoCollagen binding specificity of the discoidin

domain receptors binding sites on collagens II and III andmolecular determinants for collagen IV recognition by DDR1rdquoMatrix Biology vol 30 no 1 pp 16ndash26 2011

[27] J Su J Yu T Ren et al ldquoDiscoidin domain receptor 2 is associ-ated with the increased expression of matrix metalloproteinase-13 in synovial fibroblasts of rheumatoid arthritisrdquoMolecular andCellular Biochemistry vol 330 no 1-2 pp 141ndash152 2009

[28] L Xu H Peng DWu et al ldquoActivation of the discoidin domainreceptor 2 induces expression of matrix metalloproteinase 13associated with osteoarthritis in micerdquoThe Journal of BiologicalChemistry vol 280 no 1 pp 548ndash555 2005

[29] L Xu J Servais I Polur et al ldquoAttenuation of osteoarthritisprogression by reduction of discoidin domain receptor 2 inmicerdquo Arthritis and Rheumatism vol 62 no 9 pp 2736ndash27442010

[30] R R Yammani and R F Loeser ldquoStress-inducible nuclearprotein 1 regulates matrix metalloproteinase 13 expression inhuman articular chondrocytesrdquo Arthritis amp Rheumatology vol66 no 5 pp 1266ndash1271 2014

[31] Y Xu Y Huang D Cai J Liu and X Cao ldquoAnalysis ofdifferences in themolecularmechanism of rheumatoid arthritisand osteoarthritis based on integration of gene expressionprofilesrdquo Immunology Letters vol 168 no 2 pp 246ndash253 2015

[32] Y Araki T T Wada Y Aizaki et al ldquoHistone methylation andSTAT-3 differentially regulate interleukin-6- induced matrixmetalloproteinase gene activation in rheumatoid arthritis syn-ovial fibroblastsrdquo Arthritis amp Rheumatology vol 68 no 5 pp1111ndash1123 2016

[33] S Sugita Y Hosaka K Okada et al ldquoTranscription factorHes1modulates osteoarthritis development in cooperationwithcalciumcalmodulin-dependent protein kinase 2rdquo Proceedingsof the National Academy of Sciences of the United States ofAmerica vol 112 no 10 pp 3080ndash3085 2015

[34] B-G Ju D Solum E J Song et al ldquoActivating the PARP-1sensor component of the groucho TLE1 corepressor complexmediates a CaMKinase II120575-dependent neurogenic gene activa-tion pathwayrdquo Cell vol 119 no 6 pp 815ndash829 2004

[35] R Kageyama TOhtsuka andTKobayashi ldquoTheHes gene fam-ily repressors and oscillators that orchestrate embryogenesisrdquoDevelopment vol 134 no 7 pp 1243ndash1251 2007

[36] E J Ezratty N Stokes S Chai A S Shah S E Williams andE Fuchs ldquoA role for the primary cilium in notch signaling andepidermal differentiation during skin developmentrdquo Cell vol145 no 7 pp 1129ndash1141 2011

[37] J H Kong L Yang E Dessaud et al ldquoNotch activity modulatesthe responsiveness of neural progenitors to sonic hedgehogsignalingrdquo Developmental Cell vol 33 no 4 pp 373ndash387 2015

[38] S Niebler T Schubert E B Hunziker and A K Bosser-hoff ldquoActivating enhancer binding protein 2 epsilon (AP-2120576)-deficient mice exhibit increased matrix metalloproteinase 13expression and progressive osteoarthritis developmentrdquoArthri-tis Research andTherapy vol 17 article 119 2015

[39] K E Barbour C G Helmick K A Theis et al ldquoPrevalenceof doctor-diagnosed arthritis and arthritis-attributable activitylimitationmdashUnited States 2010ndash2012rdquoMorbidity and MortalityWeekly Report vol 62 no 44 pp 869ndash873 2013

[40] M Grotle K B Hagen B Natvig F A Dahl and T KKvien ldquoObesity and osteoarthritis in knee hip andor hand anepidemiological study in the general population with 10 yearsfollow-uprdquo BMC Musculoskeletal Disorders vol 9 article 1322008

Disease Markers 7

[41] M Ribeiro P Lopez de Figueroa F Blanco A Mendes and BCarames ldquoInsulin decreases autophagy and leads to cartilagedegradationrdquoOsteoarthritis andCartilage vol 24 no 4 pp 731ndash739 2016

[42] D Hamada R Maynard E Schott et al ldquoInsulin suppressesTNF-dependent early osteoarthritic changes associated withobesity and type 2 diabetesrdquo Arthritis amp Rheumatology vol 68no 6 pp 1392ndash1402 2016

[43] N Yoshimura S Muraki H Oka et al ldquoAccumulation ofmetabolic risk factors such as overweight hypertension dys-lipidaemia and impaired glucose tolerance raises the risk ofoccurrence and progression of knee osteoarthritis A 3-yearFollow-Up of the ROAD Studyrdquo Osteoarthritis and Cartilagevol 20 no 11 pp 1217ndash1226 2012

[44] I-E Triantaphyllidou E Kalyvioti E Karavia I Lilis KE Kypreos and D J Papachristou ldquoPerturbations in theHDL metabolic pathway predispose to the development ofosteoarthritis inmice following long-term exposure to western-type dietrdquo Osteoarthritis and Cartilage vol 21 no 2 pp 322ndash330 2013

[45] D Iliopoulos K N Malizos and A Tsezou ldquoEpigeneticregulation of leptin affects MMP-13 expression in osteoarthriticchondrocytes possible molecular target for osteoarthritis ther-apeutic interventionrdquoAnnals of the Rheumatic Diseases vol 66no 12 pp 1616ndash1621 2007

[46] S Pallu P-J Francin C Guillaume et al ldquoObesity affectsthe chondrocyte responsiveness to leptin in patients withosteoarthritisrdquo Arthritis Research and Therapy vol 12 no 3article R112 2010

[47] A Koskinen K Vuolteenaho R Nieminen T Moilanen andE Moilanen ldquoLeptin enhances MMP-1 MMP-3 and MMP-13 production in human osteoarthritic cartilage and correlateswith MMP-1 and MMP-3 in synovial fluid from OA patientsrdquoClinical and Experimental Rheumatology vol 29 no 1 pp 57ndash64 2011

[48] P-J Francin A Abot C Guillaume et al ldquoAssociation betweenadiponectin and cartilage degradation in human osteoarthritisrdquoOsteoarthritis and Cartilage vol 22 no 3 pp 519ndash526 2014

[49] E H Kang Y J Lee T K Kim et al ldquoAdiponectin is a potentialcatabolicmediator in osteoarthritis cartilagerdquoArthritis ResearchandTherapy vol 12 no 6 article R231 2010

[50] A Soro-Paavonen A M D Watson J Li et al ldquoReceptor foradvanced glycation end products (RAGE) deficiency attenuatesthe development of atherosclerosis in diabetesrdquo Diabetes vol57 no 9 pp 2461ndash2469 2008

[51] D J Larkin J Z Kartchner A S Doxey et al ldquoInflamma-tory markers associated with osteoarthritis after destabilizationsurgery in youngmice with and without Receptor for AdvancedGlycation End-products (RAGE)rdquo Frontiers in Physiology vol4 article 121 Article ID Artisle 121 2013

[52] J A Roman-Blas and S A Jimenez ldquoNF-120581B as a potentialtherapeutic target in osteoarthritis and rheumatoid arthritisrdquoOsteoarthritis and Cartilage vol 14 no 9 pp 839ndash848 2006

[53] A R Klatt G Klinger O Neumuller et al ldquoTAK1 downregu-lation reduces IL-1120573 induced expression of MMP13 MMP1 andTNF-alphardquo Biomedicine and Pharmacotherapy vol 60 no 2pp 55ndash61 2006

[54] Z Fan H Yang B Bau S Soder and T Aigner ldquoRole ofmitogen-activated protein kinases and NF120581B on IL-1120573-inducedeffects on collagen type II MMP-1 and 13 mRNA expression innormal articular human chondrocytesrdquo Rheumatology Interna-tional vol 26 no 10 pp 900ndash903 2006

[55] Z Tang L Yang Y Wang et al ldquoContributions of differentintraarticular tissues to the acute phase elevation of synovialfluidMMP-2 following rat ACL rupturerdquo Journal of OrthopaedicResearch vol 27 no 2 pp 243ndash248 2009

[56] A Liacini J Sylvester W Q Li et al ldquoInduction of matrixmetalloproteinase-13 gene expression by TNF-120572 is mediated byMAPkinases AP-1 andNF-120581B transcription factors in articularchondrocytesrdquo Experimental Cell Research vol 288 no 1 pp208ndash217 2003

[57] A Scharstuhl H L Glansbeek H M van Beuningen E LVitters P M van der Kraan and W B van den Berg ldquoInhibi-tion of endogenous TGF-120573 during experimental osteoarthritisprevents osteophyte formation and impairs cartilage repairrdquoTheJournal of Immunology vol 169 no 1 pp 507ndash514 2002

[58] G Zhen C Wen X Jia et al ldquoInhibition of TGF-120573 signalingin mesenchymal stem cells of subchondral bone attenuatesosteoarthritisrdquoNatureMedicine vol 19 no 6 pp 704ndash712 2013

[59] K Andersen C Black P Crepeau et al ldquoTGF-1205731 and HtrA1interactive pathway in themolecular progression of osteoarthri-tis (11399)rdquoTheFASEB Journal vol 28 no 1 supplement article11399 2014

[60] S Launay E Maubert N Lebeurrier et al ldquoHtrA1-dependentproteolysis of TGF-120573 controls both neuronal maturation anddevelopmental survivalrdquo Cell Death and Differentiation vol 15no 9 pp 1408ndash1416 2008

[61] M Siebert S K Wilhelm J Z Kartchner et al ldquoEffect ofpharmacological blocking of TLR-4 on osteoarthritis in micerdquoJournal of Arthritis vol 4 article 164 2015

[62] D W Macdonald R S Squires S A Avery et al ldquoStructuralvariations in articular cartilage matrix are associated withearly-onset osteoarthritis in the spondyloepiphyseal dysplasiacongenita (sedc) mouserdquo International Journal of MolecularSciences vol 14 no 8 pp 16515ndash16531 2013

[63] H-J Yen M K Tayeh R F Mullins E M Stone V CSheffield andD C Slusarski ldquoBardet-Biedl syndrome genes areimportant in retrograde intracellular trafficking and Kupfferrsquosvesicle cilia functionrdquoHuman Molecular Genetics vol 15 no 5pp 667ndash677 2006

[64] D Ahrens A E Koch R M Pope M Stein-Picarella andM J Niedbala ldquoExpression of matrix metalloproteinase 9 (96-kd gelatinase B) in human rheumatoid arthritisrdquo Arthritis andRheumatism vol 39 no 9 pp 1576ndash1587 1996

[65] G Keyszer I Lambiri R Nagel et al ldquoCirculating levels ofmatrix metalloproteinases MMP-3 andMMP-1 tissue inhibitorof metalloproteinases 1 (TIMP-1) andMMP-1TIMP-1 complexin rheumatic disease Correlation with clinical activity ofrheumatoid arthritis versus other surrogate markersrdquo Journal ofRheumatology vol 26 no 2 pp 251ndash258 1999

[66] G Cunnane O Fitzgerald C Beeton T E Cawston and BBresnihan ldquoEarly joint erosions and serum levels of matrixmetalloproteinase 1 matrix metalloproteinase 3 and tissueinhibitor of metalloproteinases 1 in rheumatoid arthritisrdquoArthritis amp Rheumatism vol 44 no 10 pp 2263ndash2274 2001

[67] T Sathyamoorthy G Sandhu L B Tezera et al ldquoGender-dependent differences in plasma matrix metalloproteinase-8elevated in pulmonary tuberculosisrdquo PLoS ONE vol 10 no 1article e0117605 2015

[68] J Collazos V Asensi G Martin A H Montes T Suarez-Zarracina and E Valle-Garay ldquoThe effect of gender and geneticpolymorphisms on matrix metalloprotease (MMP) and tissueinhibitor (TIMP) plasma levels in different infectious and non-infectious conditionsrdquo Clinical and Experimental Immunologyvol 182 no 2 pp 213ndash219 2015

Research ArticleExpressions of Matrix Metalloproteinases 2 7 and 9 inCarcinogenesis of Pancreatic Ductal Adenocarcinoma

Katarzyna Jakubowska1 Anna Pryczynicz2 Joanna Januszewska2

Iwona Sidorkiewicz3 Andrzej Kemona2 Andrzej NiewiNski4 Aukasz Lewczuk2

BogusBaw Kwdra5 and Katarzyna GuziNska-Ustymowicz2

1Department of Pathomorphology Comprehensive Cancer Center 15-027 Białystok Poland2Department of General Pathomorphology Medical University of Białystok 15-269 Białystok Poland3Department of Reproduction and Gynecological Endocrinology Medical University of Białystok 15-276 Białystok Poland4Department of Rehabilitation Medical University of Białystok 15-276 Białystok Poland52nd Department of General and Gastroenterological Surgery Medical University of Białystok 15-276 Białystok Poland

Correspondence should be addressed to Katarzyna Jakubowska kathianwppl

Received 22 March 2016 Revised 13 May 2016 Accepted 31 May 2016

Academic Editor Massimiliano Castellazzi

Copyright copy 2016 Katarzyna Jakubowska et al This is an open access article distributed under the Creative Commons AttributionLicense which permits unrestricted use distribution and reproduction in any medium provided the original work is properlycited

Pancreatic ductal adenocarcinoma (PDAC) is a highly fatal disease usually diagnosed in an advanced stage which gives a slightchance of recovery Matrix metalloproteinases (MMPs) are a family of zinc-dependent endopeptidases that participate in tissueremodeling and stimulate neovascularization and inflammatory response The aim of the study was to evaluate the expression ofMMP-2MMP-7 andMMP-9 in normal ducts tumor pancreatic adenocarcinoma cells and peritumoral stroma in correlation withclinicohistopathological parametersThe studymaterial was obtained from 29 patients with pancreatic ductal adenocarcinomaTheexpressions of MMP-2 MMP-7 and MMP-9 were performed by immunohistochemical technique Microvessel density (MVD)was visualized by special immunostaining The expressions of MMP-2 MMP-7 and MMP-9 were mainly observed in tumorcells and peritumoral stroma MMP-2 expression in cancer cells was correlated with female gender stronger inflammation andhistopathological type of cancer (119877 = 0460 119901 = 0013 119877 = 0690 119901 = 00001 119877 = minus0440 119901 = 0005 resp) The expression ofMMP-7 in tumor cells was found to positively correlate with the presence of necrosis and negatively correlate withMVD (119877 = 0402119901 = 0031 119877 = minus0682 119901 = 0000) We also showed that positive MMP-9 expression in tumor cells was associated with MVD(119877 = 0368 119901 = 0084) however it was not statistically significant Our results demonstrate that MMP-2 MMP-7 and MMP-9expressions correlate with various morphological features of the PDAC tumor such as inflammation necrosis and formation ofthe new blood vessels

1 Introduction

Pancreatic ductal adenocarcinoma (PDAC) is a highly fataldisease usually diagnosed in an advanced stage which givesa slight chance of recovery Pancreatic cancer is characterizedby a rapid course poor prognosis and high mortality sincemost patients are diagnosed with metastases to lymph nodesand distant organs [1] This increases with age and is closelyassociated with genetic factors [2]

Matrix metalloproteinases (MMPs) are a family of zinc-dependent endopeptidases [3] MMPs are produced by con-nective tissue cells (fibroblasts) leukocytes macrophages

and endothelial cells [4] They are involved in degradation ofthe extracellular matrix and have been implicated in physi-ological processes MMPs are involved in supporting tissueremodeling and are required for the skeletal developmentThey stimulate neovascularization both in physiological andin pathological conditions for example in tumors [3] MMPscan regulate vascular stability and permeability in responseto tissue injury [5] Metalloproteinases are responsible forproper migration of cells involved in the inflammatoryresponse [6] They allow the cells to move into the damagedtissue and to release cytokines and their receptors through


2 Disease Markers

their ability to destroy the basement membrane It hasbeen shown that metalloproteinases destroy interleukin-2receptors on T cells whichmute the immune response againsttumor cells The imbalance between the activity of MMPsand their tissue inhibitors (TIMPs) has been attributed tothe ability of cancer cells to migrate [7] Recent studies haveconfirmed that the overexpression of MMPs in tumor andstromal cells in various cancers was associated with tumorinvasion and progression [8]MoreoverMMPs released fromdistant organs along with tumor cell growth factors canparticipate in premetastatic niche formation and metastasis[9 10]

Therefore the aim of our study was to evaluate theimmunohistochemical expressions of MMP-2 MMP-7 andMMP-9 in the normal pancreas pancreatic adenocarcinomatumor cells and stromal cells in correlation with clinico-pathological features


21 Materials The study material was obtained from 29patients (23 men 6 women 14 patients aged le60 and 15aged gt60) with pancreatic ductal adenocarcinoma (PDAC)operated on in the 2nd Department of General Surgery andGastroenterology Medical University of Białystok Patientsreceived neither preoperative cancer therapy nor inflamma-tion therapy

The study was performed in conformity with the Decla-ration of Helsinki for Human Experimentation and receivedapproval by the Local Bioethics Committee of the MedicalUniversity of Białystok (number R-I-0021672013)

22 Histopathological Examination The routine histopatho-logical assessment of the sections (stained with H+E) wasconductedwith reference to the histological typemalignancygrade (G) clinicopathological pT status regional lymphnode involvement and the presence of distant metastasesInflammation was defined as mild when it consisted oflt10 immune cells per 10 hpf under 100x magnification asmoderate when it consisted of 10 to 50 immune cells andas severe when it consisted of gt 50 immune cells [11 12]Desmoplasia was classified as poor when it occupied lt25of tumor area observed in 10 hpf under 100x magnificationwhereas it was predominant for gt25 Hemorrhage wasmeasured under 400x magnification and defined as single(one focus) or numerous (more than one hemorrhagic focus)Necrosis in the central tumor was graded as absent (none)weakfocal (lt10 of tumor area) moderate (10ndash30) orstrongextensive (gt30) [13]

23 Assessment of Vessel Formation Microvessel density(MVD)was visualized by special immunostaining of collagentype IV Since protein can be expressed on the mesenchymalcomponents and epithelial basal laminae to prevent misdi-agnosis we analyzed only vascular structures with distinct(slot-like or tubular) lumens showing positively stainedendotheliocyte layers They were counted as a number ofintratumoral microvessels per unit area of the tumor subjec-tively selected from the most vascularized areas (5 hpf under

200x magnification) [14] The PDAC tumors were dividedinto two groups with low or high MVD The cutoff valuewas the meanMVDThemeanMVD of the tumor tissue was528 vessels (range 0ndash21) confirming the histopathologicalfeatures of hypovascular pancreatic tumors

24 Immunohistochemical Analysis Formalin-fixed and par-affin-embedded tissue specimens were cut on a microtomeinto 4 120583m sections The sections were deparaffinized inxylenes (CHEMlowast115208603 Chempur Poland) and hydratedin alcohols To visualize the antigens of MMP-2 MMP-7 and MMP-9 the sections were heated in a microwaveoven for 20min in EDTA buffer (pH = 90) (EDTAbuffer Antigen Retriever E1161 Sigma-Aldrich Co MOUSA)Then they were incubated with 3 hydrogen peroxidesolution for 20min in order to block endogenous perox-idase Next incubation was performed with anti-humanantibodies against monoclonal antibody of matrix metallo-proteinase 2 (clone 17B11 Novocastra UK dilution 1 60)mouse monoclonal antibody of matrix metalloproteinase7 (clone 111433 RampD Systems USA dilution 1 75) andmouse monoclonal antibody of matrix metalloproteinase9 (clone 15W2 Novocastra UK dilution 1 80) during aperiod of 1 hour at room temperature The reaction wascarried out using the streptavidin-biotin system (BiotinylatedSecondary Antibody Streptavidin-HRP Novocastra UK) Acolor reaction for peroxidase was developed with chromogenDAB (DAB Novocastra UK) The negative control sectionwas incubated instead of the primary antibody All sectionslides were counterstained with hematoxylin

Immunohistochemical staining was evaluated by twoindependent pathologists who were blinded to the clinicalinformation A positive reaction was observed in the cyto-plasm of the normal pancreas pancreatic ductal adenocar-cinoma and peritumoral stroma Due to the small studygroup immune cells and fibroblasts were analyzed jointlyThe expression of proteins was evaluated and defined aspositive (reaction present in gt25 of normal ductal cellstumor cells or peritumoral stroma components) or negative(lack of reaction or reaction present inlt25 of normal ductalcells tumor cells or peritumoral stroma components) Thepercentage of the reaction-positive cells was calculated in 500neoplastic cells in each study sample at 400x magnification

25 Statistical Analysis Statistical analysis was conductedusing Statistica 100 (StatSoft Krakow Poland) In order tocompare the two groups the parameters were calculated bythe Chi-quadrate test Correlations between the parameterswere calculated by Spearmanrsquos correlation coefficient test A119901 value of lt005 was considered statistically significant

3 Results

31 Histopathological Findings Pancreatic adenocarcinomaswere moderately differentiated (G2) in 2529 cases andpoorly differentiated (G3) in 429 cases They were classifiedas mucinous in 329 cases and as nonmucinous in 2629cases We noted lymph node involvement in 1229 casesand metastases to distant organs (liver intestine) in 929

Disease Markers 3

(a) (b) (c)

(d) (e) (f)

Figure 1 Immunohistochemical expressions ofMMP-2MMP-7 andMMP-9 in normal ducts tumor cells and peritumoral stroma of PDACThe lack of MMP-2 (a) MMP-7 (b) andMMP-9 (c) expressions was found in normal ducts Positive reaction of MMP-9 (d) was noted in thecytoplasm of pancreatic adenocarcinoma cells MMP-2 (d) overexpression was observed in the peritumoral stroma in the majority of PDACcases and color reaction of MMP-7 was observed in the cytoplasm of both cancer cells and the stroma (e) Positive reaction of MMP-9 wasnoted in the cytoplasm of pancreatic adenocarcinoma cells (f)

cases In addition we assessed the degree of inflamma-tion desmoplasia necrosis and foci of hemorrhage Weobserved a weak inflammatory response in 9 cases mod-erate response in 10 cases and strong response in 10 casesTumors with poor and prominent fibrosis were noted in12 and 17 cases respectively Hemorrhage was numerousin 5 cases and single in 10 cases Pancreatic adenocarcino-mas were associated with weak or moderate necrosis in 13cases

32 Expression of Matrix Metalloproteinases (MMP-2 MMP-7 and MMP-9) in Normal Ducts Pancreatic AdenocarcinomaTumor Cells and Stromal Cells The expression of MMP-2 was predominantly present in tumor cells and stroma(5517 and 7931 resp) in comparison to its absencein the normal pancreas (9655) In contrast the positiveexpression of MMP-7 was observed mainly in tumor cells(9655) less in the stromal cells (5517) as compared tothe normal pancreas (9310) Immunohistochemical assess-ment of MMP-9 in patients with PDAC showed the proteinpresence in tumor cells in 4483 of cases and its absencein normal pancreatic ducts and stroma (100 and 7587resp) (Figure 1) Furthermore statistical analysis showeda significant difference between the expressions of MMP-2and MMP-7 in normal and tumor tissues (119901 = 00001) Inaddition the expression of MMP-7 in tumor cells differedstatistically significantly from that noted in the peritumoral

stroma (119901 = 0005) The differences in MMP expressions(MMP-2 MMP-7 andMMP-9) in normal tissue tumor cellsand stroma are shown in Table 1

33 The Correlation between the Expression of Matrix Met-alloproteinases (MMP-2 MMP-7 and MMP-9) in Tumorsof PDAC and Clinicopathological Features The analysis ofthe expression of matrix metalloproteinases in a variety oftissues and selected anatomical clinical factors demonstrateda correlation between MMP-2 expression in tumor andfemale gender (119877 = 0460 119901 = 0013) A strong positivecorrelation was noted between MMP-2 in tumor tissue andthe presence of stronger inflammation (119877 = 0690 119901 =00001) MMP-2 expression in stromal cells was found tonegatively correlate with the nonmucinous type of PDAC(119877 = minus0440 119901 = 0005) Positive expression of MMP-2 wasmore frequent in patients over 60 years of age although it wasnot statistically significant MMP-7 expression in tumor cellsrevealed a positive associationwithmore frequent occurrenceof necrosis in the main mass of tumor (119877 = 0402 119901 = 0031)and a negative correlation with the lower index of MVD(119877 = minus0681 119901 = 0000) Positive expression of MMP-7 wasaccompanied by frequent changes indicating the presence ofnecrosis The expressions of matrix metalloproteinases werenot correlated with malignancy grade fibrosis degree theincidence of hemorrhage or the presence of metastases tolymph nodes and distant organs (Tables 2 and 3)

4 Disease Markers

Table 1 Expressions of MMP-2 MMP-7 and MMP-9 in normal pancreas cancer cells and stroma

MMP-2 MMP-7 MMP-9Negative Positive Negative Positive Negative Positive

Normal tissue 28 (9655) 1 (345) 27 (9310) 2 (690) 29 (100) 0 (0)Cancer cells 13 (4483) 16 (5517)lowast 1 (345) 28 (9655)lowastlowast 16 (5517) 13 (4483)Tumor stroma 6 (2069) 23 (7931) 13 (4483) 16 (5517)lowastlowastlowast 22 (7587) 7 (2413)lowastMMP-2 in cancer cells versus MMP-2 in normal tissue 119901 = 00001lowastlowastMMP-7 in cancer cells versus MMP-7 in normal tissue 119901 = 00001lowastlowastlowastMMP-7 in cancer cells versus MMP-7 in tumor stroma 119901 = 0005

Table 2 Correlations between MMP-2 MMP-7 and MMP-9 expressions in tumor cells and clinicopathological parameters

Variables Patients119873 () MMP-2 MMP-7 MMP-9R p value R p value 119877 p value

GenderMale 23 (793) 0460 0013 minus0047 0814 minus0042 0804Female 6 (207)Agele60 14 (483)

minus0012 0949 0339 0071 minus0383 0046gt60 15 (517)InflammationAbsent 2 (69)

0690 lt0001 0053 0782 0081 0666Weak 9 (310)Moderate 10 (345)Strong 8 (276)DesmoplasiaPoor 12 (414) 0016 0931 0215 0262 0135 0491Prominent 17 (586)Foci of hemorrhageAbsent 14 (483)

0088 0648 0161 0403 minus0271 0162Single 10 (345)Numerous 5 (172)NecrosisAbsent 16 (552)

minus0007 0968 0402 0031 0084 0668Weak 7 (241)Moderate 6 (207)Strong 0 (00)MVDLow 15 (517) 0072 0738 minus0682 lt0001 0368 0084High 14 (483)Adenocarcinoma typeNonmucinous 26 (897)

minus0193 0314 minus0139 0495 minus0081 0681Mucinous 3 (103)Grade of malignancyG2 25 (862)

minus0156 0417 minus0036 0850 minus0129 0512G3 4 (138)Lymph node involvementAbsent 17 (586) 0024 0899 minus0261 0172 0033 0866Present 12 (414)Distant metastasesAbsent 20 (690) 0014 0940 minus0193 0313 minus0081 0681Present 9 (310)

Disease Markers 5

Table 3 Correlations between MMP-2 MMP-7 and MMP-9 expressions in peritumoral stroma cells and clinicopathological parameters

Variables Patients119873 () MMP-2 MMP-7 MMP-9119877 p value 119877 p value 119877 p value

GenderMale 23 (793)

minus0051 0793 minus0603 0060 minus0237 0215Female 6 (207)Agele60 14 (483) 0199 0299 0029 0879 minus0261 0170gt60 15 (517)InflammationAbsent 2 (69)

0263 0167 0132 0494 minus0042 0826Weak 9 (310)Moderate 10 (345)Strong 8 (276)DesmoplasiaPoor 12 (414) 0033 0864 minus0129 0505 0189 0325Prominent 17 (586)Foci of hemorrhageAbsent 14 (483)

minus0198 0302 minus0010 0958 minus0164 0395Single 10 (345)Numerous 5 (172)NecrosisAbsent 16 (552)

minus0196 0306 minus0050 0796 0121 0529Weak 7 (241)Moderate 6 (207)Strong 0 (00)MVDLow 15 (517)

minus0220 0300 minus0022 0916 minus0046 0829High 14 (483)Adenocarcinoma typeNonmucinous 26 (897)

minus0440 0005 0106 0584 minus0194 0313Mucinous 3 (103)Grade of malignancyG2 25 (862)

minus0130 0500 0125 0518 minus0021 0912G3 4 (138)Lymph node involvementAbsent 17 (586)

minus0021 0910 minus0176 0360 minus0105 0586Present 12 (414)Distant metastasesAbsent 20 (690) 0101 0602 minus0106 0582 minus0224 0241Present 9 (310)

4 Discussion

PDAC has poor prognosis and patientsrsquo survival time is esti-mated to be several months The research into mechanismsof the outstanding malignancy and invasiveness of tumorcells is still being conducted [15] It has been proven thatmetalloproteinases are involved in different phases of tumordevelopment such as extracellular matrix degradation neo-vascularization inhibition of inflammatory cell migrationand metastasis formation in the lymph nodes and distantorgans [3 5 9 15] Their role in the above mechanisms has

been investigated in various types of the digestive systemtumors located in the colorectum the stomach and the liver[16ndash19]

In our study we noted a positive expression of MMP-2 in 5517 of tumor cells and in 7931 of the stromawhich was significantly higher than in the normal pancreas(345) Giannopoulos et al [17] also showed the presenceof MMP-2 in cancer cells in most cases of PDAC In turnGress et al [18] demonstrated that the overexpression ofMMP-2 was significantly higher in tumor cells than in thestroma We also reported a positive correlation between the

6 Disease Markers

expression of MMP-2 in tumor cells and inflammation Thismay suggest that MMP-2 protein is secreted from tumor cellsto the stroma where it modifies the immune response cellsIn our research MMP-2 expression was significantly morefrequent in the nonmucinous type of PDAC Histologicallythe nonmucinous type of pancreatic cancer is characterizedby tubular structures generally located in the rich desmo-plastic stroma Tumor cells of the mucinous type have theability to produce a large amount of mucus that fills in thetumor microenvironment and may limit the developmentof connective tissue The results of our small study suggestthat MMP-2 expression correlates with the histopathologicaltype of PDAC and that its expression may be involved in theregulation of the inflammatory response

MMPs are involved in the degradation of ECM leadingto promotion of cancer cell invasion The smallest familyof MMPs namely MMP-7 shows the greatest proteolyticactivity [19] In our study the positive MMP-7 expressionwas present in most cases in tumor cells in more thanhalf of the cases in the stroma and in only two cases innormal pancreatic ducts Crawford et al [20] and Li et al [21]showed the presence of MMP-7 expression in the cytoplasmof most tumor cells and its absence in normal pancreaticducts Our statistical analysis demonstrated a significantcorrelation between MMP-7 expression in PDAC cells and apositive reaction in stromal cells as well as in normal ductsThese results are consistent with the observations of Joneset al [22] In contrast Yamamoto et al [23] reported anincrease in MMP-7 expression in tumor nests located in thefront of PDAC tumors Tan et al [24] showed that MMP-7 overexpression in the tumor front was present much lessfrequently than in the center of the tumor mass In ourstudy the positive expression of MMP-7 in PDAC tumorswas associated with the presence of necrotic lesions Otherstudies have confirmed that MMP-7 shows proteolytic activ-ity participates in cell dissociation throughdisruption of tightjunction structures determines tumor dissociation from theprimary tumor and stimulates cancer cell invasion [2526] Moreover we observed a strong negative relationshipbetween MMP-7 expression in tumor cells and MVD MMP-7 can cleave collagen IV which constitutes the skeleton ofthe basement membranes on blood vessels and the ECMcomponents In turn MMP-7 may lead to the degradation ofthe tumor stroma decay of current vessels and the inhibitionof neovascularization As a result of these processes necroticlesions were observed in tumor mass and hypovascularfeatures of PDAC tumors were noted Our findings suggestthat MMP-7 expression in PDAC cancer cells may contributeto the degradation of the peritumoral stroma thus facilitatingtumor cell invasion and metastasis

MMP-9 is considered to be a strong factor stimulatingthe secretion of proangiogenic factors such as vascularendothelial growth factor (VEGF) which participates in thenew blood vessel formation [26 27] The study of mousemodel has confirmed that tumor cells in MMP-9++ miceproduce bigger pancreatic tumors with high index of MVD[24] Moreover Huang et al [28] also demonstrated anincreased incidence of cancer and tumor size in MMP-9++mice which was associated with a high index of MVD

and increased macrophage infiltration We noted positiveexpression of MMP-9 in the cytoplasm of tumor cells andin the stroma of patients with PDAC Our observations areconsistent with those reported by Giannopoulos et al [17]and Gress et al [18] Statistical analysis confirmed that thepositive expression ofMMP-9 only in the tumor cells showeda growing tendency in MVD These observations suggestthat MMP-9 has an angiogenic property in comparisonwith much stronger desmoplastic activity of MMP-2 andproteolytic activity of MMP-7 in the tumor stroma in PDACtumors

In conclusion our findings confirmed that MMP-2MMP-7 andMMP-9 may take part in various morphologicalfeatures of PDAC tumor MMP-2 is mainly responsible forthe regulation of inflammatory response MMP-7 takes partin the degradation of the peritumoral stroma whereas MMP-9 may have an impact on the formation of the new bloodvessels In our opinion immunohistochemical evaluation ofthe study metalloproteinases in the tissue of patients withPDAC may help to better understand the morphology anddevelopment of PDAC tumors Our observations need to beconfirmed in a larger group of PDAC tumors

Competing Interests

The authors declare that they have no competing interests

Acknowledgments

This research was based on the work supported by theMedical University of Białystok under academic funds (113-37-558-LM)The authors would like to express their gratitudeto all the faculty staff members and lab technicians of theDepartment of General Pathomorphology whose servicesturned this research a success

References

[1] C Li D G Heidt P Dalerba et al ldquoIdentification of pancreaticcancer stem cellsrdquo Cancer Research vol 67 no 3 pp 1030ndash10372007

[2] A B Lowenfels and P Maisonneuve ldquoEpidemiology and riskfactors for pancreatic cancerrdquo Best Practice and Research Clini-cal Gastroenterology vol 20 no 2 pp 197ndash209 2006

[3] R R Baruch H Melinscak J Lo Y Liu O Yeung andR A R Hurta ldquoAltered matrix metalloproteinase expressionassociatedwith oncogene-mediated cellular transformation andmetastasis formationrdquo Cell Biology International vol 25 no 5pp 411ndash420 2001

[4] L A Shuman Moss S Jensen-Taubman and W G Stetler-Stevenson ldquoMatrixmetalloproteinases changing roles in tumorprogression and metastasisrdquo American Society for InvestigativePathology vol 181 no 6 pp 1895ndash1899 2012

[5] N E Sounni K Dehne L L C L van Kempen et al ldquoStromalregulation of vessel stability by MMP9 and TGF120573rdquo DiseaseModels amp Mechanisms vol 3 no 5-6 pp 317ndash332 2010

[6] A Lekstan P Lampe J Lewin-Kowalik et al ldquoConcentrationsand activities of metalloproteinases 2 and 9 and their inhibitors

Disease Markers 7

(TIMPS) in chronic pancreatitis and pancreatic adenocarci-nomardquo Journal of Physiology and Pharmacology vol 63 no 6pp 589ndash599 2012

[7] M D Sternlicht and Z Werb ldquoHow matrix metalloproteinasesregulate cell behaviorrdquoAnnual Review ofCell andDevelopmentalBiology vol 17 pp 463ndash516 2001

[8] CMonteagudoM JMerino J San-Juan L A Liotta andWGStetler-Stevenson ldquoImmunohistochemical distribution of typeIV collagenase in normal benign and malignant breast tissuerdquoAmerican Journal of Pathology vol 136 no 3 pp 585ndash592 1990

[9] M Bond R P Fabunmi A H Baker and A C NewbyldquoSynergistic upregulation of metalloproteinase-9 by growthfactors and inflammatory cytokines an absolute requirementfor transcription factor NF-120581Brdquo FEBS Letters vol 435 no 1 pp29ndash34 1998

[10] M Groblewska B Mroczko M Gryko et al ldquoSerum levels andtissue expression of matrix metalloproteinase 2 (MMP-2) andtissue inhibitor of metalloproteinases 2 (TIMP-2) in colorectalcancer patientsrdquo Tumor Biology vol 35 no 4 pp 3793ndash38022014

[11] J C Nickel L D True J N Krieger R E Berger A H Boagand ID Young ldquoConsensus development of a histopathologicalclassification system for chronic prostatic inflammationrdquo BJUInternational vol 87 no 9 pp 797ndash805 2001

[12] C H Richards K M Flegg C S D Roxburgh et al ldquoTherelationships between cellular components of the peritumouralinflammatory response clinicopathological characteristics andsurvival in patients with primary operable colorectal cancerrdquoBritish Journal of Cancer vol 106 no 12 pp 2010ndash2015 2012

[13] G J Krejs ldquoPancreatic cancer epidemiology and risk factorsrdquoDigestive Diseases vol 28 no 2 pp 355ndash358 2010

[14] S-Z Chen H-Q Yao S-Z Zhu Q-Y Li G-H Guo and JYu ldquoExpression levels of matrix metalloproteinase-9 in humangastric carcinomardquo Oncology Letters vol 9 no 2 pp 915ndash9192015

[15] A Pryczynicz M Gryko K Niewiarowska et al ldquoImmunohis-tochemical expression of MMP-7 protein and its serum level incolorectal cancerrdquo Folia Histochemica et Cytobiologica vol 51no 3 pp 206ndash212 2013

[16] L Ochoa-Callejero I Toshkov S Menne and A MartınezldquoExpression of matrix metalloproteinases and their inhibitorsin the woodchuck model of hepatocellular carcinomardquo Journalof Medical Virology vol 85 no 7 pp 1127ndash1138 2013

[17] GGiannopoulos K Pavlakis A Parasi et al ldquoThe expression ofmatrix metalloproteinases-2 and -9 and their tissue inhibitor 2in pancreatic ductal and ampullary carcinoma and their relationto angiogenesis and clinicopathological parametersrdquoAnticancerResearch vol 28 no 3 pp 1875ndash1882 2008

[18] T M Gress F Muller-Pillasch M M Lerch H Friess HBuchler and G Adler ldquoExpression and in-situ localization ofgenes coding for extracellular matrix proteins and extracellularmatrix degrading proteases in pancreatic cancerrdquo InternationalJournal of Cancer vol 62 no 4 pp 407ndash413 1995

[19] H D Li C Huang K J Huang et al ldquoSTAT3 knock-down reduces pancreatic cancer cell invasiveness and matrixmetalloproteinase-7 expression in nude micerdquo PLoS ONE vol6 no 10 Article ID e25941 2011

[20] H C Crawford C R Scoggins M K Washington L MMatrisian and S D Leach ldquoMatrix metalloproteinase-7 isexpressed by pancreatic cancer precursors and regulates acinar-to-ductal metaplasia in exocrine pancreasrdquo The Journal ofClinical Investigation vol 109 no 11 pp 1437ndash1444 2002

[21] Y-J Li Z-M Wei Y-X-Y Meng and X-R Ji ldquoBeta-cateninup-regulates the expression of cyclinD1 c-myc and MMP-7 inhumanpancreatic cancer relationshipswith carcinogenesis andmetastasisrdquoWorld Journal of Gastroenterology vol 11 no 14 pp2117ndash2123 2005

[22] L E Jones M J Humphreys F Campbell J P Neoptolemosand M T Boyd ldquoComprehensive analysis of matrix metallo-proteinase and tissue inhibitor expression in pancreatic cancerincreased expression of matrix metalloproteinase-7 predictspoor survivalrdquo Clinical Cancer Research vol 10 no 8 pp 2832ndash2845 2004

[23] H Yamamoto F Itoh S Iku et al ldquoExpression of matrixmetalloproteinases and tissue inhibitors of metalloproteinasesin human pancreatic adenocarcinomas clinicopathologic andprognostic significance of matrilysin expressionrdquo Journal ofClinical Oncology vol 19 no 4 pp 1118ndash1127 2001

[24] X Tan H Egami S Ishikawa et al ldquoInvolvement of matrixmetalloproteinase-7 in invasion-metastasis through inductionof cell dissociation in pancreatic cancerrdquo International Journalof Oncology vol 26 no 5 pp 1283ndash1289 2005

[25] D-H Zhou A Trauzold C Roder G Pan C Zheng and HKalthoff ldquoThe potential molecular mechanism of overexpres-sion of uPA IL-8 MMP-7 and MMP-9 induced by TRAIL inpancreatic cancer cellrdquo Hepatobiliary and Pancreatic DiseasesInternational vol 7 no 2 pp 201ndash209 2008

[26] S R Harvey T CHurd GMarkus et al ldquoEvaluation of urinaryplasminogen activator its receptor matrix metalloproteinase-9and vonWillebrand factor in pancreatic cancerrdquoClinical CancerResearch vol 9 no 13 pp 4935ndash4943 2003

[27] G Bergers R Brekken G McMahon et al ldquoMatrixmetalloproteinase-9 triggers the angiogenic switch duringcarcinogenesisrdquo Nature Cell Biology vol 2 no 10 pp 737ndash7442000

[28] S Huang M Van Arsdall S Tedjarati et al ldquoContributionsof stromal metalloproteinase-9 to angiogenesis and growth ofhuman ovarian carcinoma in micerdquo Journal of the NationalCancer Institute vol 94 no 15 pp 1134ndash1142 2002

Research ArticleSerum Gelatinases Levels in Multiple Sclerosis Patientsduring 21 Months of Natalizumab Therapy

Massimiliano Castellazzi1 Tiziana Bellini1 Alessandro Trentini1

Serena Delbue2 Francesca Elia2 Matteo Gastaldi3 Diego Franciotta3

Roberto Bergamaschi3 Maria Cristina Manfrinato1 Carlo Alberto Volta4

Enrico Granieri1 and Enrico Fainardi5

1Department of Biomedical and Specialist Surgical Sciences University of Ferrara 44124 Ferrara Italy2Department of Biomedical Surgical and Dental Sciences University of Milano 20133 Milano Italy3Department of General Neurology National Neurological Institute C Mondino 27100 Pavia Italy4Department of Morphology Experimental Medicine and Surgery University of Ferrara 44124 Ferrara Italy5Department of Neurosciences and Rehabilitation S Anna Hospital 44124 Ferrara Italy

Correspondence should be addressed to Massimiliano Castellazzi massimilianocastellazziunifeit

Received 23 March 2016 Accepted 9 May 2016

Academic Editor Hubertus Himmerich

Copyright copy 2016 Massimiliano Castellazzi et alThis is an open access article distributed under theCreativeCommonsAttributionLicense which permits unrestricted use distribution and reproduction in anymedium provided the originalwork is properly cited

Background Natalizumab is a highly effective treatment approved for multiple sclerosis (MS) The opening of the blood-brainbarrier mediated by matrix metalloproteinases (MMPs) is considered a crucial step in MS pathogenesis Our goal was to verifythe utility of serum levels of active MMP-2 and MMP-9 as biomarkers in twenty MS patients treated with Natalizumab MethodsSerum levels of active MMP-2 andMMP-9 and of specific tissue inhibitors TIMP-1 and TIMP-2 were determined before treatmentand for 21 months of therapy Results Serum levels of active MMP-2 and MMP-9 and of TIMP-1 and TIMP-2 did not differ duringthe treatmentThe ratio betweenMMP-9 andMMP-2 was increased at the 15thmonth compared with the 3rd 6th and 9thmonthsgreater at the 18th month than at the 3rd and 6th months and higher at the 21st than at the 3rd and 6th months Discussion Ourdata indicate that an imbalance between activeMMP-9 and activeMMP-2 can occur inMS patients after 15 months of Natalizumabtherapy however they do not support the use of serum active MMP-2 and active MMP-9 and TIMP-1 and TIMP-2 levels asbiomarkers for monitoring therapeutic response to Natalizumab

1 Introduction

Multiple sclerosis (MS) is a chronic inflammatory disease ofthe central nervous system (CNS) of presumed autoimmuneorigin that is characterized by demyelination and axonalloss [1] MS affects young adults women more frequentlythan men and is clinically marked by exacerbations calledrelapses which typically show dissemination in space andtime [2] Brain inflammation is initiated and sustained bylymphocyte migration across the blood-brain barrier (BBB)[3] In particular the interaction of 12057241205731 integrin on thesurface of lymphocytes with vascular-cell adhesion molecule1 (VCAM-1) and on the surface of vascular endothelial cellsin brain and spinal cord blood vessels mediates the adhesion

and migration of lymphocytes in inflamed CNS sites [4]Natalizumab (Tysabri Biogen Idec Inc Cambridge Mas-sachusetts USA) is a humanized anti-1205724 integrinmonoclonalantibody approved for relapsing-remitting multiple sclerosis(RRMS) [5 6]The efficacy of Natalizumabmonotherapy wasdemonstrated in clinical trials by the reduction in relapse rateand the progression of disability [7] Consequently Natal-izumab is used as a second-line treatment inMS patients whohave a suboptimal response to first-line disease-modifyingtherapies or as a first-line therapy in those with a highly activedisease [8] Notwithstanding the unquestionable benefitsanti-1205724 integrin treatment is however associated with JohnCunningham Virus- (JCV-) mediated progressive multifocalleukoencephalopathy (PML) a severe adverse event [9]


2 Disease Markers

Although MS etiology remains largely unknown the migra-tion of immunocompetent cells into the CNS is dependenton several factors but it fundamentally requires the openingof the BBB a mechanism in which matrix metalloproteinases(MMPs) play a crucial role [10 11]These enzymes are a familyof zinc-containing and calcium-requiring endopeptidaseswhich are secreted into extracellular space as a latent inac-tive proform that becomes activated through a proteolyticcleavage [12] Specific tissue inhibitors of metalloproteinases(TIMPs) are molecules capable of binding either activatedMMPs or their preforms and then finally of regulatingMMP activity [13] Due to their ability to degrade type IVcollagen and gelatin which are the main constituents of basallamina MMP-2 (gelatinase A 72 kDa type IV collagenase)and MMP-9 (gelatinase B 92 kDa type IV collagenase) arethe most extensively studied subfamily of MMPs in thecourse of MS In particular previous studies demonstratedthat cerebrospinal fluid (CSF) and serum levels of MMP-9 were higher in RRMS compared to progressive forms[14ndash16] and that elevated CSF and serum concentrations ofMMP-9were associatedwith clinical andmagnetic resonanceimaging (MRI) evidence of disease activity [15 17ndash19] anddisease evolution [20] Moreover CSF levels of MMP-9 werereduced after 12 months of Natalizumab treatment in 7 MSpatients and in the same study CSFMMP-9 mean levels werehigher in MS patients before Natalizumab treatment than inpatients with other neurological diseases [21] On the otherhand the significance of MMP-2 is more controversial inMS In fact while MMP-9 is predominantly upregulated ininflammatory conditions MMP-2 is constitutively expressedin the brain [22] Contradictory results have been reported inprevious studies where MMP-2 levels in acute and chronicdemyelinated lesions [23ndash25] as well as in CSF [26] inserum [15 17 20] and in peripheral blood mononuclear cells(PBMCs) [27ndash29] were increased decreased or representedin equivalent amounts in MS patients and in controls Intwo previous studies we investigated the role of the activeforms of MMP-9 and MMP-2 in MS and our results showeda reciprocal variation in these enzymes compared to theactivity of the disease In particular serum and CSF levelsand the intrathecal synthesis of active MMP-9 forms wereassociatedwith clinical andMRI disease activity [30] whereasCSF levels and intrathecal synthesis of active MMP-2 weremore elevated in MS patients without MRI evidence ofdisease activity [31]

Considering these findings in this study we aimed toinvestigate serum temporal concentrations of active MMP-2 and active MMP-9 in a cohort of RRMS patients during 21months of Natalizumab therapy

2 Materials and Methods

21 Study Design and Sample Handling The study designand the sample population were the same as in an earlierstudy [32] Briefly twenty consecutive RRMS [33] patients(17 female and 3 male) in treatment with Natalizumab wereincluded in the study All the patients were enrolled in theldquoFondazione Istituto Neurologico C Mondinordquo in PaviaSerum samples were collected before starting therapy and

then every three months for 21 months of treatment Allthe samples were withdrawn stored and analyzed underthe same conditions At any time point (a) disease severitywas scored using Kurtzkersquos Expanded Disability Status Scale(EDSS) [34] (b) presence of relapse was recorded as clinicalactivity and (c) anti-JCV antibodies were determined toassess the risk of PML [35] During the treatment disabilityprogression was defined as an increase of one point on EDSSscore from baseline [5] Brain MRI scans were performed atentry and at the end of the study and the occurrence of anew lesion on T2-weighted scans andor a new gadolinium-(Gd-) enhancing lesion on T1-weighted scans was definedas MRI activity [33] The approval of the Committee forMedical Ethics in Research was obtained for experimentsinvolving human subjects Written informed consent wasobtained from all subjects participating in the study

22 MMP-2 and MMP-9 Activity Assays Serum levels ofactive MMP-2 and active MMP-9 were determined usingcommercially available specific activity assay systems aspublished before [30 31] (Activity Assay System BiotrakAmersham Biosciences Little Chalfont UK code RPN2631and code RPN2634 resp) With these methods only circu-lating active forms of MMP-2 and MMP-9 were measuredAll the reagents and standards were included in the kitsBriefly in both activity assays serum samples were appliedin duplicate into 96-microwell microtiter plates precoatedwith anti-MMP-2 or anti-MMP-9 antibodies Human pro-MMP-2 and pro-MMP-9 respectively activated with p-aminophenylmercuric acetate were used in six serial dilu-tions as standard in each plate The detection enzyme wasthe proform of a modified urokinase an enzyme that canbe activated by captured active MMPs in an active detectionenzyme The natural activation sequence in the prodetectionenzyme was replaced using protein engineering with anartificial sequence recognized by specific MMP Activatedurokinases were thenmeasured using a specific chromogenicsubstrate (S-2444) The amount of active MMP-2 or activeMMP-9 in all samples was determined by interpolationfrom a standard curve According to the manufacturerrsquosinstructions for the MMP-2 activity assay the lower limitof quantification was 019 ngmL the range of intra-assaycoefficient of variations (CV) was 44ndash70 and the rangeof interassay CV was 169ndash185 while for the MMP-9activity assay the lower limit of quantification was assumed at0125 ngmL the range of intra-assay CV was 34ndash43 andthe range of interassay CV was 202ndash217

23 TIMP-1 and TIMP-2 Detection Assays As previouslydescribed [30 31] serum levels of TIMP-1 and TIMP-2 were measured using commercially available ldquosandwichrdquoELISA kits (Biotrak Amersham Biosciences Little ChalfontUK codes RPN2611 and RPN2618 resp) according to themanufacturerrsquos instructions All the reagents and standardswere included in the kits The limit of sensitivity in both theassays was 313 ngmL

24 Data Analysis Statistical analysis was performed withGraphPad Prism The normality of each variable was

Disease Markers 3

Table 1 Demographic and clinical characteristics of 20 RRMSpatients stratified according to response to therapy before andduring treatment with Natalizumab

Responders NonrespondersPatients (119899) 15 5Sex (malefemale) 312 05Age at entry years (mean plusmn SD) 351 plusmn 101 316 plusmn 94EDSS at baseline (mean plusmn SD) 10 plusmn 11 23 plusmn 24EDSS after 21 months of therapy 13 plusmn 13 28 plusmn 24Relapses during 21 months oftherapy (mean plusmn SD) 0 16 plusmn 09

Patients with new MRI lesions atthe end of treatment 4 0

EDSS = Expanded Disability Status Scale MRI = magnetic resonanceimaging SD = standard deviation

checked by using the Kolmogorov-Smirnov test Whennormality of data distribution was found in all variablesstatistical analysis was performed by a parametric approachAccordingly ANOVA test was used to compare variablesamong the various groups and when significant differenceswere found Studentrsquos 119905-test was used for the comparisonbetween two groups On the other hand when normalityof data distribution was rejected statistical analysis wasperformed by a nonparametric approach Kruskal-Wallis testwas used to compare variables among the various groups andif significant differences were found Mann-Whitney 119880-testwas then used to compare two different groups In case ofmultiple comparisons a Bonferroni post hoc correction wasapplied A value of 119901 lt 005 was accepted as significant

3 Results

Demographic and clinical characteristics of 20 RRMSpatients treatedwithNatalizumab are listed in Table 1 Duringthe therapy five patients experienced relapses (3 patientshad 1 relapse between baseline and 3 months one had 2relapses between 6 and 9 months and at 12 months andone had 2 relapses between 9 and 12 months and between18 and 21 months) and four patients showed new T2 andorGd-enhancing lesions on the last MRI examination at the21st month No patients showed anti-JCV seroconversionduring the 21 months of Natalizumab treatment The tim-ing of sample collection was not sequential and resultedincomplete for ten patients However we decided to analyzeall the variables in RRMS patients considered as a wholeSerum levels of active MMP-2 active MMP-9 and TIMP-1were detected in all samples while serum levels of TIMP-2 were measured in 145148 (98) of samples As reportedin Figure 1 no differences were found for serum levels ofactive MMP-2 (panel (a) ANOVA ns) and active MMP-9 (panel (b) Kruskal-Wallis ns) and TIMP-2 (panel (c)Kruskal-Wallis ns) and TIMP-1 (panel (d) ANOVA ns)among the various time points The ratios between MMPsand the specific tissue inhibitors and between active MMP-9and active MMP-2 were then calculated for all the patients at

each time point (Figure 2) No differences were found for theMMP-2TIMP-2 (panel (a) Kruskal-Wallis ns) and MMP-9TIMP-1 (panel (b) Kruskal-Wallis ns) ratios while theactive MMP-9active MMP-2 ratio was different at varioustime points (panel (c) Kruskal-Wallis 119901 lt 0001) and inparticular it was higher at the 15th month (Mann-Whitneywith Bonferroni correction) than at the 3rd (119901 lt 001) 6th(119901 lt 001) and 9th months (119901 lt 005) more elevated at the18th month than at the 3rd and 6th (119901 lt 005) and finallymore increased at the 21st month of treatment than at the 3rdand 6th months (119901 lt 005) Afterwards we tried to comparepatients who were free of relapses during the treatmentconsidered as ldquorespondersrdquo with patients who experienced atleast one relapse ldquononrespondersrdquo Despite the small numberof patients in each group we compared all the variablesserum concentrations of active MMP-2 and active MMP-9and TIMP-2 and TIMP-1 and the ratios calculated betweenMMPs and TIMPs and between active MMP-9 and activeMMP-2 No differences were found between the respondersand the nonresponders for all the data analyzed (data notshown)

4 Discussion

To the best of our knowledge this is the first study thatlongitudinally analyzes serum levels of active MMP-2 andactiveMMP-9with sensitive activity assay systems in a cohortof RRMS patients during the treatment with Natalizumab inan attempt to provide further insight into the real significanceof gelatinases in MS pathology and their role in monitoringefficacy of treatmentThe involvement of MMP-2 andMMP-9 in MS pathogenesis and progression has been widelyinvestigated in the past decades There is a large agreementparticularly on the proinflammatory role of MMP-9 andon the protective function of TIMP-1 In particular serumMMP-9 levels were greater in RRMS than in the progressiveforms [14ndash16] in MS patients with MRI evidence of diseaseactivity [15 17 19] and in MS subjects with clinically isolatedsyndromes who developed clinically definite MS [18] Onthe other hand TIMP-1 levels were lower in MS patientsthan in controls [14 15 17] and serum MMP-9TIMP-1 ratiohas been indicated as a potential biomarker of MS diseaseactivity [15 18] The study of the active forms of MMP-9also demonstrated that CSF and serum levels of active MMP-9 could represent a potential biomarker for monitoring MSdisease activity and that serum active MMP-9TIMP-1 ratioseems to be an indicator of ongoing MS inflammation [30]In addition beneficial effects of Natalizumab treatment wereassociated with decreased CSFMMP-9 levels after 12 monthsof therapy and for this reason MMP-9 was proposed asa biomarker for clinical trials on new drugs for MS [21]On the contrary the role of MMP-2 still remains unclearPrevious studies have reported that MMP-2 was elevatedin PBMCs and CD4+ Th1 cells of MS patients and couldcontribute to the homing of these immune cells inside thebrain through the BBB [28 29] In addition while CSFMMP-2 levels appeared to be comparable between MS and controls[14 15 22 25 26] serumMMP-2 concentrations were similaror lower in MS patients than in controls [15 20] The active

4 Disease Markers

Seru

m ac

tive M

MP-2

leve

ls (n

gm

L)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

0

5

10

15

20

(a)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

Seru

m ac

tive M

MP-9

leve

ls (n

gm

L)

0

5

10

15

(b)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

Seru

m T

IMP-

2le

vels

(ng

mL)

0

50

100

150

200

250

(c)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

Seru

m T

IMP-

1le

vels

(ng

mL)

0

200

400

600

800

(d)

Figure 1 Longitudinal fluctuations of serum active MMP-2 (a) and active MMP-9 and (b) TIMP-2 (c) and TIMP-1 (d) in patients withrelapsing-remitting multiple sclerosis (RRMS) treated with Natalizumab for 21 months MMP = matrix metalloproteinases TIMP = tissueinhibitors of metalloproteinases T0 = baseline T3 = 3rd month T6 = 6th month T9 = 9th month T12 = 12th month T15 = 15th month T18= 18th month and T21 = 21st month Horizontal bars indicate medians and error bars correspond to interquartile range The boundaries ofthe box represent the 25thndash75th quartiles The line within the box indicates the median The whiskers above and below the box correspondto the highest and lowest values excluding outliers

form of MMP-2 has been described as a potential markerof MS recovery as detected by MRI suggesting a beneficialfunction that sustains the resolution of the inflammatoryresponse and the remission of the disease [31] In the presentstudy serum levels of active MMP-2 and active MMP-9and of the specific tissue inhibitors TIMP-2 and TIMP-1respectively were found to be stable during the 21 months ofNatalizumab therapy in all patients Surprisingly despite thesmall number of patients included in the study we did notfind differences between patients who experienced relapsesduring the treatment considered as ldquononrespondersrdquo andpatients thatwere free of relapses considered as ldquorespondersrdquofor activeMMP-2 and activeMMP-9 and TIMP-2 and TIMP-1 serum levels at each time point Moreover the ratiosbetween active MMPs and the respective TIMPs did not

appear influenced by the Natalizumab treatment and did notdiffer between responders and nonresponders during the 21months of observation On the one hand this could indicatethat Natalizumab treatments maintain stable serum levels ofMMPs and TIMPs but on the other hand this excludes theuse of these molecules in monitoring the pharmacologicalresponse Previous studies on recombinant interferon beta-1a one of the most used disease-modifying therapies forMS showed that low MMP-9 serum levels were associatedwith a positive outcome while MMP-2 serum levels werestable during treatment [36] and that serum MMP-9TIMP-1 ratio may be regarded as a reliable marker and may bepredictive of MRI activity in RRMS [37] Moreover TIMP-1 has also been suggested as a good indicator of response totherapy [38] Our principal finding was that an imbalance

Disease Markers 5

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

0

200

400

600

800Se

rum

activ

e MM

P-2

TIM

P-2

(times103)

(a)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

0

10

20

30

40

Seru

m ac

tive M

MP-9

TIM

P-1

(times103)

(b)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

00

05

10

15

20

Seru

m ac

tive M

MP-9

act

ive M

MP-2

1 2 3

4 5

6 7

(c)

Figure 2 Longitudinal fluctuations of serum active MMP-2TIMP-2 ratio (a) serum active MMP-9TIMP-1 ratio (b) and serum activeMMP-9active MMP-2 ratio (c) in relapsing-remitting multiple sclerosis (RRMS) patients during 21 months of Natalizumab treatment Nodifferences were found for the MMP-2TIMP-2 (a) and MMP-9TIMP-1 (b) ratios while the active MMP-9active MMP-2 ratio was differentat various time points ((c) 119901 lt 0001) in particular it was higher at the 15th month than at the 3rd (1119901 lt 001) 6th (2119901 lt 001) and 9thmonths (3119901 lt 005) increased at the 18th month than at the 3rd and 6th (45119901 lt 005) and more elevated at the 21st month of treatment thanat the 3rd and 6th months (67119901 lt 005) MMP = matrix metalloproteinases TIMP = tissue inhibitors of metalloproteinases T0 = baselineT3 = 3rd month T6 = 6th month T9 = 9th month T12 = 12th month T15 = 15th month T18 = 18th month and T21 = 21st month Horizontalbars indicate medians and error bars correspond to interquartile rangeThe boundaries of the box represent the 25thndash75th quartiles The linewithin the box indicates the median The whiskers above and below the box correspond to the highest and lowest values excluding outliers

occurred between active MMP-9 and active MMP-2 serumlevels after 15months of Natalizumab therapy without furtherdifferences between responder and nonresponder patientsThe ratio between MMP-9 and MMP-2 was proposed as aserum marker to monitor the progression of liver diseaseand the ratio between MMP-2 and MMP-9 was associatedwith poor response to chemotherapy in osteosarcoma in twoprevious studies [39 40] however this is the first time thatthe ratio between the active forms of MMP-9 and MMP-2was calculated in MS patients and for this reason furtherstudies are required to investigate the biological significanceof the imbalance between serum levels of gelatinases The

main limitations of this study were the small number ofenrolled patients and above all the lack of samples collectedat the time of relapse In conclusion taken together our dataseems to indicate that Natalizumab treatment could eithermaintain serum levels of activeMMP-2 and activeMMP-9 aswell as of the specific tissue inhibitors TIMP-1 and TIMP-2stable ormake themnot affected at all however this influencetends to be reduced after 15 months of therapy resulting inan imbalance between serum active MMP-9 considered asa marker of disease activity [30] and serum active MMP-2described as a marker of disease remission [31] Moreoverthe present study argues against the use of serum levels of

6 Disease Markers

gelatinases for the monitoring of Natalizumab treatments inMS patients Nevertheless more extensive research in a largernumber of patients is advisable for a better understandingof the correlations between Natalizumab therapy and gelati-nases in MS

Competing Interests


Acknowledgments

This work has been supported by Research ProgramRegione Emilia-Romagna University 2007ndash2009 (InnovativeResearch) entitled ldquoRegional Network for Implementing aBiological Bank to Identify Biological Markers of DiseaseActivity Related to Clinical Variables in Multiple SclerosisrdquoThe authors thank Dr Elizabeth Jenkins for helpful correc-tions in the paper

References

[1] M Sospedra andRMartin ldquoImmunology ofmultiple sclerosisrdquoAnnual Review of Immunology vol 23 pp 683ndash747 2005



[4] R A Rudick and A Sandrock ldquoNatalizumab 1205724-integrinantagonist selective adhesion molecule inhibitors for MSrdquoExpert Review of Neurotherapeutics vol 4 no 4 pp 571ndash5802004

[5] C H Polman P W OrsquoConnor E Havrdova et al ldquoA ran-domized placebo-controlled trial of natalizumab for relapsingmultiple sclerosisrdquo The New England Journal of Medicine vol354 no 9 pp 899ndash910 2006

[6] E Havrdova S Galetta M Hutchinson et al ldquoEffect ofnatalizumab on clinical and radiological disease activity inmultiple sclerosis a retrospective analysis of the NatalizumabSafety and Efficacy in Relapsing-Remitting Multiple Sclerosis(AFFIRM) studyrdquo The Lancet Neurology vol 8 no 3 pp 254ndash260 2009

[7] J Chataway andDHMiller ldquoNatalizumab therapy formultiplesclerosisrdquo Neurotherapeutics vol 10 no 1 pp 19ndash28 2013

[8] L Kappos D Bates G Edan et al ldquoNatalizumab treatmentfor multiple sclerosis updated recommendations for patientselection and monitoringrdquoThe Lancet Neurology vol 10 no 8pp 745ndash758 2011

[9] G Bloomgren S Richman C Hotermans et al ldquoRiskof natalizumab-associated progressive multifocal leukoen-cephalopathyrdquo The New England Journal of Medicine vol 366no 20 pp 1870ndash1880 2012



[12] I Massova L P Kotra R Fridman and S Mobashery ldquoMatrixmetalloproteinases structures evolution and diversificationrdquoThe FASEB Journal vol 12 no 12 pp 1075ndash1095 1998

[13] P Henriet L Blavier and Y A Declerck ldquoTissue inhibitorsof metalloproteinases (TIMP) in invasion and proliferationrdquoAPMIS vol 107 no 1ndash6 pp 111ndash119 1999


[15] C Avolio M Ruggieri F Giuliani et al ldquoSerum MMP-2 andMMP-9 are elevated in different multiple sclerosis subtypesrdquoJournal of Neuroimmunology vol 136 no 1-2 pp 46ndash53 2003

[16] J Sastre-Garriga M Comabella L Brieva A Rovira MTintore and X Montalban ldquoDecreased MMP-9 productionin primary progressive multiple sclerosis patientsrdquo MultipleSclerosis vol 10 no 4 pp 376ndash380 2004


[18] E Waubant D Goodkin A Bostrom et al ldquoIFN120573 lowersMMP-9TIMP-1 ratio which predicts new enhancing lesions inpatients with SPMSrdquo Neurology vol 60 no 1 pp 52ndash57 2003

[19] F Sellebjerg H O Madsen C V Jensen J Jensen and PGarred ldquoCCR5 Δ32 matrix metalloproteinase-9 and diseaseactivity in multiple sclerosisrdquo Journal of Neuroimmunology vol102 no 1 pp 98ndash106 2000

[20] J Correale and M D L M Bassani Molinas ldquoTemporalvariations of adhesionmolecules andmatrixmetalloproteinasesin the course of MSrdquo Journal of Neuroimmunology vol 140 no1-2 pp 198ndash209 2003

[21] M Khademi L Bornsen F Rafatnia et al ldquoThe effects ofnatalizumab on inflammatory mediators in multiple sclerosisprospects for treatment-sensitive biomarkersrdquo European Jour-nal of Neurology vol 16 no 4 pp 528ndash536 2009

[22] G A Rosenberg ldquoMatrix metalloproteinases in neuroinflam-mationrdquo Glia vol 39 no 3 pp 279ndash291 2002

[23] D C Anthony B Ferguson M K Matyzak K M MillerM M Esiri and V H Perry ldquoDifferential matrix metallopro-teinase expression in cases of multiple sclerosis and strokerdquoNeuropathology and Applied Neurobiology vol 23 no 5 pp406ndash415 1997

[24] M Diaz-Sanchez K Williams G C DeLuca and M M EsirildquoProtein co-expression with axonal injury in multiple sclerosisplaquesrdquo Acta Neuropathologica vol 111 no 4 pp 289ndash2992006

[25] R L P Lindberg C J A De Groot L Montagne et al ldquoTheexpression profile of matrix metalloproteinases (MMPs) andtheir inhibitors (TIMPs) in lesions and normal appearing whitematter of multiple sclerosisrdquo Brain vol 124 no 9 pp 1743ndash17532001

[26] R N Mandler J D Dencoff F Midani C C Ford W Ahmedand G A Rosenberg ldquoMatrix metalloproteinases and tissueinhibitors of metalloproteinases in cerebrospinal fluid differ inmultiple sclerosis and Devicrsquos neuromyelitis opticardquo Brain vol124 no 3 pp 493ndash498 2001

[27] M Kouwenhoven V Ozenci A Tjernlund et al ldquoMonocyte-derived dendritic cells express and secrete matrix-degradingmetalloproteinases and their inhibitors and are imbalanced inmultiple sclerosisrdquo Journal of Neuroimmunology vol 126 no 1-2 pp 161ndash171 2002

Disease Markers 7

[28] A Bar-Or R K Nuttall M Duddy et al ldquoAnalyses of all matrixmetalloproteinase members in leukocytes emphasize mono-cytes as major inflammatory mediators in multiple sclerosisrdquoBrain vol 126 no 12 pp 2738ndash2749 2003

[29] M Abraham S Shapiro A Karni H L Weiner and A MillerldquoGelatinases (MMP-2 andMMP-9) are preferentially expressedby Th1 vs Th2 cellsrdquo Journal of Neuroimmunology vol 163 no1-2 pp 157ndash164 2005



[32] M Castellazzi S Delbue F Elia et al ldquoEpstein-barr virusspecific antibody response in multiple sclerosis patients during21 months of natalizumab treatmentrdquo Disease Markers vol2015 Article ID 901312 5 pages 2015

[33] C H Polman S C Reingold B Banwell et al ldquoDiagnosticcriteria for multiple sclerosis 2010 revisions to the McDonaldcriteriardquo Annals of Neurology vol 69 no 2 pp 292ndash302 2011


[35] L Gorelik M Lerner S Bixler et al ldquoAnti-JC virus antibodiesimplications for PML risk stratificationrdquo Annals of Neurologyvol 68 no 3 pp 295ndash303 2010

[36] M Trojano C Avolio M Ruggieri et al ldquoSerum solubleintercellular adhesion molecule-1 inMS relation to clinical andGd-MRI activity and to rIFN120573-1b treatmentrdquoMultiple Sclerosisvol 4 no 3 pp 183ndash187 1998


[38] M Comabella J Rıoa C Espejoa et al ldquoChanges in matrixmetalloproteinases and their inhibitors during interferon-betatreatment in multiple sclerosisrdquo Clinical Immunology vol 130pp 145ndash150 2009

[39] T W Chung J R Kim J I Suh et al ldquoCorrelation betweenplasma levels of matrix metalloproteinase (MMP)-9MMP-2ratio and 120572-fetoproteins in chronic hepatitis carrying hepatitisB virusrdquo Journal of Gastroenterology and Hepatology vol 19 no5 pp 565ndash571 2004

[40] P Kunz H Sahr B Lehner C Fischer E Seebach and J Fel-lenberg ldquoElevated ratio of MMP2MMP9 activity is associatedwith poor response to chemotherapy in osteosarcomardquo BMCCancer vol 16 no 1 p 223 2016

Review ArticleAssociation of Common Variants in MMPs withPeriodontitis Risk

Wenyang Li1 Ying Zhu2 Pradeep Singh1 Deepal Haresh Ajmera1 Jinlin Song1 and Ping Ji1

1Chongqing Key Laboratory of Oral Diseases and Biomedical Sciences and College of Stomatology Chongqing Medical UniversityChongqing 400016 China2Department of Forensic Medicine Faculty of Basic Medical Sciences Chongqing Medical University Chongqing 400016 China

Correspondence should be addressed to Ping Ji ckjiping136163com

Received 5 December 2015 Revised 18 February 2016 Accepted 16 March 2016

Academic Editor Jacek Kurzepa

Copyright copy 2016 Wenyang Li et al This is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

BackgroundMatrixmetalloproteinases (MMPs) are considered to play an important role during tissue remodeling and extracellularmatrix degradation And functional polymorphisms in MMPs genes have been reported to be associated with the increased riskof periodontitis Recently many studies have investigated the association between MMPs polymorphisms and periodontitis riskHowever the results remain inconclusive In order to quantify the influence of MMPs polymorphisms on the susceptibility toperiodontitis we performed a meta-analysis and systematic review Results Overall this comprehensive meta-analysis included atotal of 17 related studies including 2399 cases and 2002 healthy control subjects Our results revealed that although studies of theassociation between MMP-8 minus799 CT variant and the susceptibility to periodontitis have not yielded consistent results MMP-1(minus1607 1G2Gminus519AG andminus422AT)MMP-2 (minus1575GAminus1306CTminus790TG andminus735CT)MMP-3 (minus1171 5A6A)MMP-8(minus381 AG and +17 CG)MMP-9 (minus1562 CT and +279 RQ) andMMP-12 (minus357 AsnSer) as well asMMP-13 (minus77 AG 11A12A)SNPs are not related to periodontitis risk Conclusions No association of these common MMPs variants with the susceptibility toperiodontitis was found however further larger-scale and multiethnic genetic studies on this topic are expected to be conductedto validate our results

1 Introduction

Periodontitis being one of the most common forms ofdestructive periodontal disease in adults can be defined asbacterial plaque induced inflammation of the attachmentapparatus of teeth and supporting structures which initiallymanifests as gingivitis and is characterized by extensionof inflammation from the gingiva into deeper periodontaltissues that if left untreated results in destruction of periodon-tium associated with progressive attachment loss and irre-versible bone loss [1] Currently periodontitis is consideredto be multifactorial disease developing as a result of complexinteractions between specific host genes and the environment[2] Although periodontitis is initiated and sustained bybacterial plaque host factors determine the pathogenesis andrate of progression of the disease [3]

Matrix metalloproteinases (MMPs) are a large family ofmetal-dependent extracellular proteinases which are respon-sible for the tissue remodeling and degradation of the extra-cellular matrix (ECM) including collagens elastins gelatinmatrix glycoproteins and proteoglycans [4] To date at least26 members of MMPs have been identified [5] The majorityof MMPs proteins are secreted as inactive proMMPs whichare subsequently processed by other proteolytic enzymes(such as serine proteases furin and plasmin) to generate theactive formsThe proteolytic activities of MMPs are preciselycontrolled during activation from their precursors and inhi-bition by endogenous inhibitors a-macroglobulins and tis-sue inhibitors ofmetalloproteinases (TIMPs) or by nonselect-ive synthetic inhibitors (batimastat BB-94) [6]

Significant evidence suggests that MMPs comprise themost important pathway in the tissue destruction associated


2 Disease Markers

with periodontal disease [7] And based on previous studiesdramatically elevated levels of MMP-1 MMP-2 MMP-3MMP-8 and MMP-9 have been detected in gingival crevic-ular fluid peri-implant sulcular fluid and gingival tissue ofperiodontitis patients [8] Likewise recent studies have alsoshown that mRNA levels of MMPs are significantly increasedin inflamed gingival tissue MMPs activity may be regulatedby interactions with their endogenous inhibitors (TIMPs)and posttranslational modifications as well as at the levels ofgene transcription [9] Consequently it can be hypothesizedthat functional polymorphisms in MMPs genes may affectMMPs expression or activity and thus may predispose toperiodontal disease conditions

According to some genotype analyses of single nucleotidepolymorphisms (SNPs) in MMPs genes they have shownincreased frequency of several common MMPs SNPs inpatients with periodontitis [10ndash13] On the contrary someother studies have demonstrated little or no association ofthese SNPs in MMPs genes with etiopathogenesis of peri-odontitis [14ndash17] Despite comprehensive studies focusing onthe association of gene polymorphismswith the susceptibilityandor severity of periodontitis there exists a high degreeof inconsistency and the results are inconclusive thereforefor the purpose of deriving a more precise estimation ofassociation between theseMMPs SNPs andperiodontitis riskwe performed a meta-analysis and systematic review of alleligible studies


21 Protocols and Eligibility Criteria The meta-analysis andsystematic review reported here are in accordance with thePreferred Reporting Items for Systematic Review and Meta-Analyses (PRISMA) statement (Appendix S1 in the Supple-mentaryMaterial available online at httpdxdoiorg10115520161545974) The research question for this study wasformulated based on the PICO (population interventioncomparison and outcomes) criteriaThe literature searchwaslimited to original studies performed in humans on the asso-ciation of matrix metalloproteinases SNPs with periodontitisrisk

22 Search Strategy Studies addressing the correlations ofMMPs genetic polymorphisms with the risk of periodon-titis were identified by performing an electronic search inPubMed (1966 toMay 2015)Medline (1950 toMay 2015) andWeb of Science databases (1900 to May 2015) by using thefollowing search terms in PubMed (((((((ldquoMatrix Metallo-proteinasesrdquo [Mesh]) OR Matrix Metalloproteinases) ORMatrix Metalloproteinase) OR MMPs) OR MMP)) AND(((((ldquoPolymorphism Geneticrdquo [Mesh]) OR Polymorphism)OR ldquoGenetic Variationrdquo [Mesh]) OR Genetic Variation)OR genetic variant)) AND (((((((((((ldquoPeriodontitisrdquo [Mesh])OR Periodontitis) OR ldquoChronic Periodontitisrdquo [Mesh]) ORChronic Periodontitis)ORCP)OR ldquoAggressive Periodontitisrdquo[Mesh]) OR Aggressive Periodontitis) OR AgP) OR ldquoPeri-odontal Diseasesrdquo [Mesh]) OR Periodontal Diseases) ORPD) Other databases were searched with comparable termssuitable for the specific database Furthermore in order to

identify any additional studies that may have been misseda computer-assisted strategy based on manual searching ofreference lists from potentially relevant reviews and retrievedarticles was performed Full texts of the relevant articles andstudies published in English were retrieved and included toexplore the association between MMPs polymorphisms andthe susceptibility to periodontitis

23 Selection of Studies The studies included in the presentmeta-analysis and systematic review had to meet the follow-ing inclusion criteria (a) studies used validated genotypingmethods (such as PCR-RFLP and TaqMan) to measure theassociation of SNPs in MMP genes with periodontitis risk(b) studies were in an appropriate analytical design includingcase-control cohort or nested case-control (c) studies werepublished in English (d) the full text of studies was availableand (e) the data of studies were not duplicated in anothermanuscript However studies were excluded if they did notprovide enough information on genotype frequency or didnot report sufficient genotype distribution for calculation ofodds ratios (ORs) and its variance Besides studies investigat-ing the mixed population were excluded if they did not pro-vide the detailed information for each ethnicity Moreoverstudies were also excluded if genotype distributions of con-trol subjects were varied from Hardy-Weinberg equilibrium(HWE)

24 Data Extraction To ensure homogeneity of data collec-tion and to rule out the effect of subjectivity in data gatheringdata extraction was performed independently by two investi-gators (Ying Zhu and Pradeep Singh) using a predefined pro-tocol Disagreements were resolved by iteration discussionand consensus A series of items were collected for each trialincluding first authorrsquos surname publication year countryethnicity (Caucasian Asian or mixed (excluding the detailedethnic results of mixed population in the original study))type and severity of periodontitis matching criteria of casesand controls source of controls allelic frequency in bothcases and controls genotyping methods and also the genesand variants genotyped Furthermore the evidence of HWEin controls was verified through the application of an onlinesoftware (httpwwwoegeorgsoftwarehwe-mr-calcshtml)119901 value less than 005 of HWE was considered to be signifi-cant

25 Risk of Bias Methodological quality was indepen-dently evaluated by two researchers (Pradeep Singh andDeepal Haresh Ajmera) according to the recently proposedNewcastle-Ottawa Scale (NOS) criteria for the quality assess-ment of case-control studies To unravel potential systematicbiases a third investigator (Wenyang Li) performed a concor-dance study by independently reviewing all eligible studiescomplete concordance was reached for all variables assessedBriefly the quality of each study was assessed by using thefollowing methodological components (1) subject selection(2) comparability of subject and (3) clinical outcome Table 2illustrates the details of each methodological item NOSscores ranged from 0 to 9 wherein a score of ge5 was regarded

Disease Markers 3

as high-quality study while studies with scores lt5 wereclassified as low-quality studies

26 Heterogeneity A test for heterogeneity (true variance ofeffect size across studies) was performed using a 119876 test (toassess whether observed variance exceeds expected variance)to establish inconsistency in the study results Howeverbecause the test is susceptible to the number of trials includedin the meta-analysis we also calculated 1198682 1198682 directlycalculated from the 119876 statistic indicates the percentage ofvariability in effect estimates because of true heterogeneityrather than sampling error 1198682 ranges from 0 to 100 with0 indicating the absence of any heterogeneity Althoughabsolute numbers for 1198682 are not available values lt50 areconsidered low heterogeneity and the effect is thought to befixed Conversely when 1198682 exceeds 50 then heterogeneityis thought to exist and the effect is random

27 Statistical Analysis The STATA version 110 (Stata CorpCollege Station TX USA) software was used for meta-analysisThe strength of the association betweenMMPs SNPsand periodontitis risk was evaluated by ORs with their 95confidence intervals (CIs) under different geneticmodels theallelemodel (mutant allele versuswild allele) the codominantmodel (homozygous rareheterozygous versus homozygousfrequent and homozygous rare versus heterozygous) thedominant model (heterozygous + homozygous rare versushomozygous frequent) and the recessive model (homozy-gous rare versus heterozygous + homozygous frequent) aswell as the additive model (heterozygous versus homozygousfrequent + homozygous rare) In addition subgroup analyseswere stratified when feasible according to the type of diseaseracial descent severity of chronic periodontitis and smokinghabit respectivelyThe119885-testwas used to estimate the statisti-cal significance of pooledORs and the Bonferroni correctionwas used to account for multiple testing in association analy-ses When all genetic models were tested for each SNP a cor-rected 119901 value lt 001 was considered statistically significant

To estimate the pooled ORs a fixed effects model (theMantel-Haenszel method) was used initially whereas therandom effects model (DerSimonian and Laird method) wasapplied when evidence of significant heterogeneity was foundacross trials (119901 lt 01 and 1198682 gt 50) In order to evaluatethe potential source of heterogeneity a sensitivity analysiswas performed through sequential removal of each includedstudy Publication bias was investigated using funnel plotswherein the standard error of log(OR) was plotted againstlog(OR) for each study Besides funnel plot asymmetry wasassessed with the Begg rank correlation test (Begg test) andthe Egger linear regression approach (Egger test) 119901 valuesof less than 005 from the Eggerrsquos test were consideredstatistically significant In addition the results of the trialswhich could not be pooled through the meta-analysis wereassessed using descriptive statistics

3 Results

The flowchart for the process of includingexcluding arti-cles is shown in Figure 1 After abstracts were screened

for relevance 25 full-text studies comprising chronic peri-odontitis (CP) andor aggressive periodontitis (AgP) werecomprehensively assessed against the inclusion criteriaThreestudies were excluded because they were not in accordancewith HWE [10 18 19] Another four studies were excludedbecause they reported the results of mixed population butdid not provide the detailed information for each ethnicity[15 17 20 21] Besides one more study was excluded dueto insufficient data availability for calculating ORs and theirvariance [7] Finally 17 case-control studies investigating theassociation of MMP-1 (minus1607 1G2G minus519 AG and minus422AT) MMP-2 (minus1575 GA minus1306 CT minus790 TG and minus735CT) MMP-3 (minus1171 5A6A) MMP-8 (minus799 CT minus381 AGand +17 CG) MMP-9 (minus1562 CT and +279 RQ) MMP-12(minus357 AsnSer) and MMP-13 (minus77 AG and 11A12A) withperiodontitis risk were included in this meta-analysis [8 11ndash14 16 22ndash32] And the characteristics and quality assessmentof all included studies are summarized in Tables 1 and 2

31 MMP-1 Table 3 and Figure 2 show the meta-analysisresults of two SNPs in theMMP-1 gene namely minus1607 1G2Gand minus519 AG under various genetic models In Caucasianswe failed to identify any significant association of these twoSNPs with the susceptibility to CP under all comparisonmodels (Table 3 Figure 2) Besides in Asian populationour results also demonstrated that there was no statisti-cally significant association between MMP-1 minus1607 1G2Gpolymorphism and the risk of both CP and AgP (Table 3Figure 2) Furthermore analyses of individual polymorphismrevealed no differences in distribution of MMP-1 minus422 ATvariant between CP and control groups in Caucasians [14]

Considering the influence of disease severity on poly-morphism we also performed stratified analysis by severityof CP Pooled ORs revealed that no significant associationexisted betweenMMP-1 minus1607 1G2G polymorphism and therisk of mild to moderate or severe CP in both Caucasiansand Asians under all comparison models (Table 3) Besidesa study by Pirhan et al [26] reported that the minus519 G allelecarrying genotypes of MMP-1 gene was not suggested to berelated with severe CP in Caucasian population (adjustedOR = 125 119901 = 083) With smoking being one of the majorcontributing factors in the susceptibility of periodontitis wealso performed subgroup analysis according to the smokinghabit of subjects In Caucasians our results revealed thatwhen only nonsmoking or smoking subjects were includedthe difference between MMP-1 minus1607 1G2G polymorphismin CP patients and control population was not significantunder all comparison models (Table 3) Likewise apparentassociation could not be related with the stratified analysisby individual smoking habit in the allelic and genotypefrequencies of MMP-1 minus519 AG polymorphism between CPand control groups in Caucasian population [14] On thecontrary the results by Holla et al [14] also suggested thatthere were significant differences in the distribution ofMMP-1 minus422 AT variant between a subgroup of smoking CPpatients versus smoking controls in Caucasians (119901 = 0017)

4 Disease Markers

Table1MainCh

aracteris

ticso

fincludedstu

dies

Authoryear

Cou

ntry

Ethn

icity

Samples

ize

(casecontrol)

Type

ofperio

dontitis

Matchingcriteria

Genotype

metho

dGene(po

lymorph

ism)amp

HWEin

controls

deSouzae

tal2003

[31]

Brazil

Caucasian

5037

CP(m

oderateo

rsevere)

Smoker

ratio

sPC

R-RF

LPMMP-1(minus1607

1G2G)0

87

Hollaetal2004

[14]

CzechRe

public

Caucasian

133196

CP(m

ildto

mod

erate

tosevere)

Agegender

PCR-RF

LPMMP-1(minus1607

1G2G)0

52MMP-1(minus519AG

)011

MMP-1(minus422AT)0

47

Itagakietal2004

[22]

Japan

Asian

205142

CP(m

ildor

mod

erate

orsevere)

Agegendersm

oker

ratio

sTaqM

anMMP-1(minus1607

1G2G)0

48

MMP-3(minus117

15A6A)0

87

3714

2AgP

Hollaetal2005

[23]

CzechRe

public

Caucasian

149127

CP(m

ildto

mod

erate

tosevere)

Agesmoker

ratio

sPC

R-RF

LP

MMP-2(minus1575

GA

)040

MMP-2(minus1306

CT)

019

MMP-2(minus790TG)0

67

MMP-2(minus735CT)

042

Caoetal2005

[32]

China

Asian

4052

AgP

mdashPC

R-RF

LPMMP-1(minus1607

1G2G)0

78

Caoetal2006

[13]

China

Asian

6050

CP(m

oderateo

rsevere)

mdashPC

R-RF

LPMMP-1(minus1607

1G2G)0

99

Kelese

tal2006

[12]

Turkey

Caucasian

7070

CP(severe)

Agegender

PCR-RF

LPMMP-9(minus1562

CT)

082

Hollaetal2006

[24]

CzechRe

public

Caucasian

169135

CP(m

oderateo

rsevere)

Agegendersm

oker

ratio

sPC

R-RF

LPMMP-9(minus1562

CT)

059

MMP-9(+279RQ)0

25

Chen

etal2007

[16]

China

Asian

7912

8AgP

Agegender

DHPL

CPC

R-RF

LPMMP-2(minus1306

CT)

100

MMP-9(minus1562

CT)

063

Gurkanetal2007

[8]

Turkey

Caucasian

9215

7AgP

Gender

PCR-RF

LPMMP-2(minus735CT)

045

MMP-9(minus1562

CT)

035

MMP-12

(357

AsnSer)0

06

Gurkanetal2008

[25]

Turkey

Caucasian

8710

7CP

(severe)

mdashPC

R-RF

LPMMP-2(minus735CT)

043

MMP-12

(357

AsnSer)0

47

Pirhan

etal2008

[26]

Turkey

Caucasian

10298

CP(severe)

mdashPC

R-RF

LPMMP-1(minus519AG

)079

Ustu

netal2008

[27]

Turkey

Caucasian

12654

CP(m

oderateo

rsevere)

Age

PCR-RF

LPMMP-1(minus1607

1G2G)0

75

Pirhan

etal2009

[28]

Turkey

Caucasian

10298

CP(severe)

mdashPC

R-RF

LPMMP-13

(minus77

AG

)089

MMP-13

(11A

12A)0

92

Chou

etal2011[11]

China

Asian

3611

06CP

(mod

erateto

severe)

Gendersm

oker

ratio

sPC

R-RF

LPMMP-8(minus799CT)

022

9610

6AgP

Hollaetal2012

[29]

CzechRe

public

Caucasian

3412

78CP

(mild

tomod

erate

tosevere)

Agegendersm

oker

ratio

sPC

R-RF

LPMMP-8(+17

CG)0

14MMP-8(minus799CT)

063

Emingiletal2014

[30]

Turkey

Caucasian

100167

AgP

Gender

PCR-RF

LPMMP-8(+17

CG)0

29

MMP-8(minus799CT)

009

MMP-8(minus381A

G)0

87

CPchron

icperio

dontitisAgP

aggressiveperio

dontitisHWE

Hardy-W

einb

ergequilib

riumM

ildchronicperio

dontitispatie

ntsw

ithteethexhibitin

glt3m

mattachmentlossmod

eratechronicperio

dontitis

patientsw

ithteethexhibitin

gge3m

mandlt7m

mattachmentlosssevere

chronicp

eriodo

ntitispatie

ntsw

ithteethexhibitin

gge7m

mattachmentlossA119901valuelessthan005

ofHWEwas

considered

significant

Disease Markers 5

Table 2 Assessing the quality of included studies

Author year Selection Comparability Exposure Scorede Souza et al 2003 [31] f f f f f 5Holla et al 2004 [14] f f f f f f f 7Itagaki et al 2004 [22] f f f f f f 6Holla et al 2005 [23] f f f f f f f 7Cao et al 2005 [32] f f f f f 5Cao et al 2006 [13] f f f f f 5Keles et al 2006 [12] f f f f f f f 7Holla et al 2006 [24] f f f f f f ff f 9Chen et al 2007 [16] f f f f f ff f 8Gurkan et al 2007 [8] f f f f ff f 7Gurkan et al 2008 [25] f f f f ff f 7Pirhan et al 2008 [26] f f f f f f f f 8Ustun et al 2008 [27] f f f f f 5Pirhan et al 2009 [28] f f f f ff f 7Chou et al 2011 [11] f f f f f f f 7Holla et al 2012 [29] f f f f f f f f 8Emingil et al 2014 [30] f f f f f f f 7

Selection

(1) Is the case definition adequate(a) Yes with independent validation f(b) Yes for example record linkage or based on self-reports(c) No description

(2) Representativeness of the cases(a) Consecutive or obviously representative series of cases f(b) Potential for selection biases or not stated

(3) Selection of controls(a) Community controls f(b) Hospital controls(c) No description

(4) Definition of controls(a) No history of disease (endpoint) f(b) No description of source

Comparability(1) Comparability of cases and controls on the basis of the design or analysis(a) Study controls for the most important factor (HWE in control group) f(b) Study controls for any additional factor (eg age gender and smoker ratios) f

Exposure

(1) Ascertainment of exposure(a) Secure record f(b) Structured interview where blind to casecontrol status f(c) Interview not blinded to casecontrol status(d) Written self-report or medical record only(e) No description

(2) Same method of ascertainment for cases and controls(a) Yes f(b) No

(3) Nonresponse rate(a) Same rate for both groups f(b) Nonrespondents described(c) Rate different and no designation

6 Disease Markers

Records identified through database searching(n = 605)

Scre

enin

gIn

clude

dEl

igib

ility

Additional records identified through other sources(n = 8)

Records after duplicates removed(n = 613)

Records screened(n = 33)

Records excludedirrelevant to the topic (n = 580)

Full-text articles assessed for eligibility(n = 25)

Studies included in qualitative synthesis(n = 24)

Studies included in quantitative synthesis (meta-analysis)(n = 17)

Iden

tifica

tion

Studies excluded (n = 7)

HWE n = 3

detailed information for each ethnicity of mixed population n = 4

(ii) Studies did not provide the

(i) Studies not in agreement with

Records excluded (n = 8)

(iii) Non-English studies n = 1

(ii) Studies on cells n = 2

(i) Reviews n = 5

Full-text articles excluded (n = 1)(i) Studies with data not available n = 1

Figure 1 Flow of study identification inclusion and exclusion

32 MMP-2 In the present meta-analysis we failed to asso-ciate MMP-2 minus735 CT polymorphism with CP risk in Cau-casian population under all comparisonmodels (Table 3 Fig-ure 2) Besides a study by Gurkan et al [8] revealed that thisSNP was also not related to AgP risk in Caucasians Similarlyno significant association of MMP-2 minus1575 GA minus1306 CTand minus790 TG SNPs with the susceptibility to periodontitiswas observed in Caucasian and Asian populations [16 23]

As far as the severity of CP was considered the allelicand genotype distributions ofMMP-2 minus735 CT variant weresimilar in severe CP and healthy subjects in Caucasians[25] When stratified by smoking habit we found that thispolymorphism was not linked with the risk of CP in non-smoking Caucasian patients and controls without smokinghistory under all comparison models (Table 3) Likewise theresults of subgroup analysis by Gurkan et al [8] showed thatthere was no significant difference regarding the distributionof this SNP between nonsmoking AgP and nonsmoking

healthy subjects in Caucasian population Besides a similardistribution of other threeMMP-2 variants was also observedbetween CP patients and controls in subgroup analysisaccording to smoking status in Caucasians [23]

33 MMP-9 Ourmeta-analysis results revealed thatMMP-9minus1562 CT SNPmight not contribute to CP risk in Caucasiansunder all comparison models (Table 3 Figure 2) Likewisethe results by Chen et al [16] and Gurkan et al [8] failedto find a significant association of this variant with the riskof AgP in Asian and Caucasian populations respectivelyBesides any significant association of MMP-9 +279 RQpolymorphism with the susceptibility to CP was also absentin Caucasians [24]

When stratified by the severity of CP pooled ORs alsodid not reveal any significant association between MMP-9minus1562 CT variant and severe CP risk in Caucasians underall comparison models (Table 3) Similarly it was reported by

Disease Markers 7

Table3Meta-analysisresults

ofthep

olym

orph

ismsinMMPs

gene

onperio

dontitisrisk

MMP-1

minus1607

1G2G

Stud

ies

(casescon

trols)

2Gversus

1GOR(95

CI)

1198682(

)119901ℎ119901119888

2G2Gversus

1G1G

OR(95

CI)

1198682(

)119901ℎ119901119888

1G2Gversus

1G1G

OR(95

CI)

1198682(

)119901ℎ119901119888

2G2Gversus

1G2G

OR(95

CI)

1198682(

)119901ℎ119901119888

1G2G+2G

2Gversus

1G1G

OR(95

CI)

1198682(

)119901ℎ119901119888

2G2Gversus

1G1G

+1G

2GOR(95

CI)

1198682(

)119901ℎ119901119888

1G2Gversus

1G1G

+2G

2G

OR(95

CI)

1198682(

)119901ℎ119901119888

Type

ofdisease

CP Caucasian

3(309287)

102(067ndash15

6)10

5(044ndash

251)

095

(064ndash

142)

090

(059ndash

137)

091

(062ndash13

2)089

(060ndash

133)

100(072ndash14

0)631

006710

0063000671000

120032

110

0000049010

00499

013610

00382019

810

0000096910

00

Asian

2(30219

2)17

5(077ndash395)

279

(056ndash

1398)

131(074ndash232

)17

5(077ndash394)

196(063ndash609)

201

(072ndash561)

079

(055ndash116)

84500111000

81500201000

27002421000

64600931000

69400711000

797

002710

0000046

710

00Ag

P

Asian

2(77194)

134(048ndash373)

154(028ndash855)

091

(042ndash19

6)16

7(038ndash731)

114(056ndash

232)

164(035

ndash770)

075

(043ndash13

0)84800101000

78400311000

00076310

0081800191000

39002001000

857

000810

00595011610

00Severe

CPCa

ucasian

Mild

tomod

erate

2(6691)

099

(063ndash15

5)099

(039

ndash250)

116(052

ndash260)

085

(039

ndash184)

110(051ndash238)

089

(043ndash18

5)117(062ndash221)

00061310

0000061310

0000066710

0000087410

0000061310

0000075110

0000087610

00

Severe

2(11091)

153(072ndash324)

244

(047ndash1259)

152(070ndash

330)

131(068ndash251)

168(081ndash350)

147(080ndash

273)

098

(056ndash

173)

657

008810

0063400981000

15802761000

00036910

00511

015310

00423018

810

0000084510

00Asian

Mild

tomod

erate

2(16819

2)12

6(093ndash17

2)16

2(084ndash

312)

140(072ndash269)

119(075ndash18

7)15

1(082ndash280)

127(083ndash19

5)097

(063ndash14

8)601

0113098

7627

010110

00438018

210

0000053410

00560013

210

0021002611000

00078410

00

Severe

2(97192)

199(092ndash426)

293

(071ndash1203)

116(053

ndash256)

219

(126ndash

379)

178(086ndash

367)

255

(095ndash685)

058

(035

ndash098)

73500520546

67000820952

00046

810

00489

01620035

381

02040854

697

0069044

1000517028

7Sm

okinghabitinCP

Caucasian

Non

smok

ing

3(200213)

092

(055ndash15

5)090

(033

ndash243)

087

(054ndash

140)

085

(052

ndash141)

096

(045ndash205)

082

(052

ndash130)

098

(066ndash

146)

661

005310

00631

006710

0034902151000

00043810

00569

009810

0040

10118

1000

00057510

00

Smok

ing

2(10974)

110(071ndash17

2)114(043ndash302)

112(054ndash

234)

115(050ndash

264

)112(055ndash229)

119(053

ndash265)

098

(053

ndash184)

19002671000

329

022210

0000097610

00522014

810

0000068210

00540014

010

0000031910

00MMP-1

minus519AG

Stud

ies

(casescon

trols)

Gversus

AOR(95

CI)

1198682(

)119901ℎ119901119888

GGversus

AA

OR(95

CI)

1198682(

)119901ℎ119901119888

AGversus

AA

OR(95

CI)

1198682(

)119901ℎ119901119888

GGversus

AGOR(95

CI)

1198682(

)119901ℎ119901119888

AG+GGversus

AA

OR(95

CI)

1198682(

)119901ℎ119901119888

GGversus

AA+AG

OR(95

CI)

1198682(

)119901ℎ119901119888

AGversus

AA+GG

OR(95

CI)

1198682(

)119901ℎ119901119888

Type

ofdisease

CP Caucasian

2(235293)

103(080ndash

132)

108(064ndash

182)

100(068ndash14

6)10

6(064ndash

175)

102(071ndash14

5)10

6(067ndash16

9)098

(069ndash

139)

00

09841000

00

08061000

00

0811

1000

00

06911000

00

08841000

00

07361000

00077810

00MMP-2

minus735CT

Stud

ies

(casescon

trols)

Tversus

COR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

CTversus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CTOR(95

CI)

1198682(

)119901ℎ119901119888

CT+TT

versus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CC+CT

OR(95

CI)

1198682(

)119901ℎ119901119888

CTversus

CC+TT

OR(95

CI)

1198682(

)119901ℎ119901119888

Type

ofdisease

CP Caucasian

2(236234)

111(079ndash155)

119(043ndash337)

112(074

ndash168)

108(037

ndash313

)10

0(067ndash14

9)116(041ndash324)

110(074

ndash165)

00069510

0000051110

0000094010

0000054110

0000069910

0000052210

0000098910

00Sm

okinghabitinCP

Caucasian

Non

smok

ing

2(13319

8)113(074

ndash172)

115(032

ndash409)

117(070ndash

194)

099

(027ndash366)

116(071ndash18

8)110(031ndash389)

115(069ndash

190)

00033710

00452017

710

0000078410

0032402241000

00053310

00424018

810

0000091710

00

8 Disease Markers

Table3Con

tinued

MMP-9

minus1562

CT

Stud

ies

(casescon

trols)

Tversus

COR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

CTversus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CTOR(95

CI)

1198682(

)119901ℎ119901119888

CT+TT

versus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CC+CT

OR(95

CI)

1198682(

)119901ℎ119901119888

CTversus

CC+TT

OR(95

CI)

1198682(

)119901ℎ119901119888

Type

ofdisease

CP Caucasian

2(239205)

056

(024ndash

135)

036

(011ndash112

)063

(029ndash

136)

051

(015

ndash172)

057

(023ndash13

8)039

(012

ndash124)

072

(047ndash110)

78200321000

35902120553

64000961000

00045910

00739

005010

0020402620777

571

0127092

4Severe

CPCa

ucasian

Severe

2(163205)

063

(020ndash

197)

044

(013

ndash149)

071

(024ndash

209)

053

(014

ndash195)

065

(019

ndash215

)046

(014

ndash157)

075

(028ndash201)

861

000710

00544013

910

0079300281000

00042810

0084200121000

410019

310

0076000411000

MMP-1matrix

metalloproteinase-1M

MP-2matrix


MP-9matrix

metalloproteinase-9C

Pchronicp

eriodo

ntitisAgP

aggressiveg

eneralized

perio

dontitis

119901ℎthe119901valueo

fheterogeneity119901119888the119901valuec

orrected

byBo

nferroni

correctio

nOR

odds

ratio

CIconfi

denceinterval

When119901ℎislt01and1198682exceeds5

0the

rand

omeffectsmod

elisusedC

onversely

the

fixed

effectsmod

elisused

119901119888lt001

isconsidered

statisticallysig

nificant

Disease Markers 9

de Souza et al (2003)

Holla et al (2004)

Ustun et al (2008)

Itagaki et al (2004)

Cao et al (2006)


Cao et al (2005)

Holla et al (2006)

Keles et al (2006)

Study ID

165 (090 303)

075 (055 103)

103 (065 161)

102 (067 156)

119 (087 163)

274 (157 481)

175 (077 395)

080 (048 135)

229 (124 420)

134 (048 373)

084 (055 130)

035 (017 069)

056 (024 135)

OR (95 CI)

2551

4121

3327

10000

5398

4602

10000

5118

4882

10000

5479

4521

10000

Holla et al (2004)

Pirhan et al (2008)

Holla et al (2005)

Study ID

102 (075 140)

103 (067 159)

103 (080 132)

104 (065 165)

119 (073 193)

111 (079 155)

OR (95 CI)

6619

3381

10000

5388

4612

10000

weight

weight

Subtotal (I2 = 00 p = 0695)

Subtotal (I2 = 00 p = 0984)

Subtotal (I2 = 782 p = 0032)

Subtotal (I2 = 848 p = 0010)

Subtotal (I2 = 845 p = 0011)

Subtotal (I2 = 631 p = 0067)

1 5790173

1 1930518

Note weights are from randomeffects analysis

(minus1562 CT) CP CaucasianMMP-9


(minus519 AG) CP CaucasianMMP-1

(minus1607 1G2G) AgP AsianMMP-1

(minus1607 1G2G) CP AsianMMP-1

(minus1607 1G2G) CP CaucasianMMP-1

Gurkan et al (2007)

(a) Mutant allele versus wild allele

Figure 2 Continued

10 Disease Markers

286 (082 1001)

056 (030 107)

107 (042 273)

105 (044 251)

133 (069 257)

693 (203 2363)

279 (056 1398)

066 (023 188)

379 (114 1258)

154 (028 855)

2539

4150

3312

10000

5509

4491

10000

5148

4852

10000

104 (057 189)

121 (042 350)

108 (064 182)

087 (021 357)

175 (038 818)

120 (043 337)

061 (016 233)

008 (000 157)

036 (011 112)

7713

2287

10000

6270

3730

10000

5250

4750

10000

Study ID OR (95 CI)

Study ID OR (95 CI)

Subtotal (I2 = 359 p = 0212)

Subtotal (I2 = 00 p = 0511)

Subtotal (I2 = 00 p = 0806)

Subtotal (I2 = 784 p = 0031)

Subtotal (I2 = 815 p = 0020)

Subtotal (I2 = 630 p = 0067)

1 23600423

1 23200043

Note weights are from random effects analysis







weight

weight


Holla et al (2004)

Ustun et al (2008)


Cao et al (2006)


Cao et al (2005)

Holla et al (2006)

Keles et al (2006)

Holla et al (2004)

Pirhan et al (2008)

Holla et al (2005)

Gurkan et al (2007)

(b) Homozygous rare versus homozygous frequent

Figure 2 Continued

Disease Markers 11

089 (053 148)

040 (018 087)

063 (029 136)

5713

4287

10000

54910182

Study ID OR (95 CI)

Subtotal (I2 = 640 p = 0096)

198 (063 620)079 (048 129)110 (047 254)095 (064 142)

107 (055 207)239 (074 776)131 (074 232)

082 (030 226)104 (032 337)091 (042 196)

104 (062 172)094 (053 169)100 (068 146)

110 (063 191)114 (063 206)112 (074 168)

84470792077

10000

81671833

10000

59634037

10000

55794421

10000

54274573

10000

1 7760129

Study ID OR (95 CI)

Subtotal (I2 = 120 p = 0321)

Subtotal (I2 = 270 p = 0242)

Subtotal (I2 = 00 p = 0763)

Subtotal (I2 = 00 p = 0811)

Subtotal (I2 = 00 p = 0940)








weight

weight

de Souza et al (2003)Holla et al (2004)Ustun et al (2008)

Itagaki et al (2004)Cao et al (2006)


Holla et al (2006)

Keles et al (2006)

Holla et al (2004)Pirhan et al (2008)

Holla et al (2005)Gurkan et al (2007)

(c) Heterozygous versus homozygous frequent

Figure 2 Continued

12 Disease Markers

144 (053 393)

072 (039 132)

098 (046 206)

090 (059 137)

100 (057 177)

129 (044 379)

106 (064 175)

079 (018 342)

154 (032 741)

108 (037 313)

069 (017 275)

020 (001 404)

051 (015 172)

1426

5500

3074

10000

8046

1954

10000

6129

3871

10000

6310

3690

10000

124 (078 197)

290 (121 693)

175 (077 394)

080 (036 180)

363 (138 959)

167 (038 731)

5994

4006

10000

5167

4833

10000

Study ID OR (95 CI)

Study ID OR (95 CI)

Subtotal (I2 = 00 p = 0490)

Subtotal (I2 = 00 p = 0691)

Subtotal (I2 = 00 p = 0541)

Subtotal (I2 = 00 p = 0459)

Subtotal (I2 = 646 p = 0093)

Subtotal (I2 = 818 p = 0019)

1 99200101

1 9590104








weight

weight


Holla et al (2004)

Ustun et al (2008)


Cao et al (2006)


Cao et al (2005)

Holla et al (2006)

Keles et al (2006)

Holla et al (2004)

Pirhan et al (2008)

Holla et al (2005)

Gurkan et al (2007)

(d) Homozygous rare versus heterozygous

Figure 2 Continued

Disease Markers 13

228 (077 669)

071 (045 114)

109 (049 241)

091 (062 132)

074 (029 191)

189 (065 551)

114 (056 232)

104 (065 166)

098 (056 172)

102 (071 145)

107 (063 183)

092 (051 166)

100 (067 149)

756

7248

1996

10000

6504

3496

10000

5777

4223

10000

5363

4637

10000


119 (064 222)

386 (127 1176)

196 (063 609)

085 (052 140)

034 (016 074)

057 (023 138)

5805

4195

10000

5539

4461

10000

Study ID OR (95 CI)

Study ID OR (95 CI)

1 6690149

1 11800851

Subtotal (I2 = 499 p = 0136)

Subtotal (I2 = 390 p = 0200)

Subtotal (I2 = 00 p = 0884)

Subtotal (I2 = 00 p = 0699)

Subtotal (I2 = 694 p = 0071)

Subtotal (I2 = 739 p = 0050)




MMP-1 (minus1607 1G2G) CP Caucasian



weight

weight


Holla et al (2004)

Ustun et al (2008)


Cao et al (2005)

Holla et al (2004)

Cao et al (2006)


Holla et al (2006)

Keles et al (2006)

Pirhan et al (2008)

Holla et al (2005)

Gurkan et al (2007)

(e) Heterozygous + homozygous rare versus homozygous frequent

Figure 2 Continued

14 Disease Markers

175 (068 451)

065 (036 115)

100 (049 204)

089 (060 133)

102 (061 172)

124 (044 348)

106 (067 169)

085 (021 346)

167 (036 767)

116 (042 324)

063 (017 239)

010 (001 198)

039 (012 124)

1279

5789

2931

10000

8101

1899

10000

6208

3792

10000

5485

4515

10000

126 (082 195)

362 (159 825)

201 (072 561)

076 (036 162)

368 (151 902)

164 (035 770)

5578

4422

10000

5123

487

10000

Study ID OR (95 CI)

Study ID OR (95 CI)

Subtotal (I2 = 204 p = 0262)

Subtotal (I2 = 382 p = 0198)

Subtotal (I2 = 00 p = 0736)

Subtotal (I2 = 00 p = 0522)

Subtotal (I2 = 797 p = 0027)

Subtotal (I2 = 857 p = 0008)

1 9020111

1 181000554








weight

weight


Holla et al (2004)

Ustun et al (2008)


Cao et al (2006)


Cao et al (2005)

Holla et al (2006)

Keles et al (2006)

Holla et al (2004)

Pirhan et al (2008)

Holla et al (2005)

Gurkan et al (2007)

(f) Homozygous rare versus heterozygous + homozygous frequent

Figure 2 Continued

Disease Markers 15

106 (045 247)097 (062 150)105 (056 200)100 (072 140)

086 (056 133)062 (029 133)079 (055 116)

110 (053 227)045 (019 105)075 (043 130)

102 (065 159)092 (052 162)098 (069 139)

111 (064 192)110 (061 199)110 (074 165)

090 (054 151)044 (020 095)072 (047 110)

150358312666

10000

72332767

10000

46645336

10000

60823918

10000

53494651

10000

61083892

10000

10189 528

Subtotal (I2 = 00 p = 0969)

Subtotal (I2 = 00 p = 0467)

Subtotal (I2 = 595 p = 0116)

Subtotal (I2 = 00 p = 0778)

Subtotal (I2 = 00 p = 0989)

Subtotal (I2 = 571 p = 0127)

Study ID OR (95 CI)







weight




Holla et al (2006)Keles et al (2006)



(g) Heterozygous versus homozygous frequent + homozygous rare

Figure 2 Forest plot of periodontitis risk associated with MMPs polymorphisms under all comparison models

Holla et al [24] that therewas nodifference in the distributionofMMP-9 +279 RQ SNP between the Caucasian CP patientswith mild to moderate disease and those with severe diseaseConcerning the smoking habit of subjects the results byHollaet al [24] suggested no significant difference in the allele andgenotype frequencies of MMP-9 minus1562 CT polymorphismbetween smoking or nonsmoking CP patients and controlswith or without smoking history in Caucasians Moreoverwhen the smokers were excluded the distribution of this SNPin the nonsmoking Caucasian subjects with AgP was similarto that in the healthy group [8]

34 Other MMPs One SNP minus1171 5A6A (in the promoterregion of MMP-3 gene) has been investigated In the studyby Itagaki et al [22] they failed to support the influence of

this polymorphism on susceptibility to both CP (119901 = 0935)and AgP (119901 = 0057) in Asians Moreover as far as theseverity of CP was concerned the results by Itagaki et al[22] also revealed that in Asian population there were nostatistically significant differences in the distribution of thisvariant among three CP phenotypes (severe moderate andslight) (119901 = 0240 0188 and 0114 resp)

Variation inMMP-8 gene particularly of minus799 CT minus381AG and +17 CG SNPs has been investigated in associationwith periodontitis As for MMP-8 minus799 CT polymorphismanalysis of genotypes in periodontitis and healthy controlgroups in the study by Chou et al [11] showed that theminus799 T allele was associated with increased risk of both AgP(adjusted OR = 199 119901 = 004) and CP (adjusted OR =193 119901 = 0007) in Asians Likewise Emingil et al [30] has

16 Disease Markers

also found analogous results for the association between thisvariant and AgP risk (119901 lt 0005) in Caucasians On the con-trary the results by Holla et al [29] suggested no differencesin the allelic and genotype frequencies of this SNP betweenCaucasian CP patients and controls (119901 gt 005) Besides asfor MMP-8 minus381 AG and +17 CG polymorphisms studiesconducted by Holla et al [29] and Emingil et al [30] revealedthat there was no significant association of these two SNPswith the susceptibility to periodontitis in Caucasians Whenstratified by smoking habit a significant difference in T allelecarriers of MMP-8 minus799 CT polymorphism in both AgP(adjusted OR = 233 119901 lt 005) and CP (adjusted OR =184 119901 lt 005) groups versus control group was foundin nonsmokers subgroup analysis in Asian population [11]while studies by Holla et al [29] and Emingil et al [30]showed no association of all these threeMMP-8 variants withthe risk of CP and AgP in Caucasians when the group ofsubjects was divided according to smoking status

A few articles have reported the relation ofMMP-12 minus357AsnSer as well as MMP-13 minus77 AG and 11A12A SNPs toperiodontitis risk in Caucasian population In the studies byGurkan et al [8 25] they could not succeed in establishingthe relationship of MMP-12 minus357 AsnSer variant with thesusceptibility either to AgP (OR = 129 95 CI = 064ndash261119901 = 047) or to severe CP (OR = 080 95 CI = 031ndash203119901 = 056) Similarly a study conducted by Pirhan et al[28] also failed to reveal any significant influence regardingthe distribution of MMP-13 minus77 AG (OR = 011 95 CI =001ndash159 119901 = 011) and 11A12A (data not shown 119901 gt005) polymorphisms on severe CP risk Furthermore in thenonsmoker subgroup analysis the allelic and genotype fre-quencies ofMMP-12minus357AsnSer variant in the nonsmokingsubjects with AgP or CP was similar to those in the healthygroup according to studies by Gurkan et al [8 25]

35 Publication Bias and Sensitivity Analysis The results ofthese two analyses are shown in Appendices S2 and S3

4 Discussion

MMP-1 minus1607 1G2G located on 11q22-q23 chromosomeis one of the most studied SNPs in periodontitis Evidencefrom previous studies revealed that individuals carrying2G2G genotype appeared to be at greater risk for developingperiodontitis than individuals who had 1G1G and 1G2Ggenotypes [26 32] Although the exact mechanism behindthese findings is not known it has reported that the presenceof 2G allele together with an adjacent adenosine creates a corebinding site (51015840-GGA-31015840) which is the consensus sequence forthe Ets family of transcription factors immediately adjacentto an AP-1 site [33] Moreover carriage of 2G allele is alsoshown to augment transcriptional activity by 37-fold andmaypotentially increase the levels of protein expression [34]Thismechanism provides the molecular bases for a more intensedegradation of periodontal extracellular matrix leading toincreased risk of periodontitis

However in our study we could not only find anysignificant association between MMP-1 minus1607 1G2G poly-morphism and periodontitis risk but also failed to associate

MMP-1 minus519 AG and minus422 AT SNPs with the suscep-tibility to periodontitis Several reasons may contribute toour results First an overview of clinical outcomes revealedthat even with the same genotype the presence of a highvariation in MMP-1 expression among periodontitis indi-viduals could be due to additional influence of specificperiodontopathogens and cytokine stimulation [35] Basedon the results of these studies a stronger signaling becauseof intense and sustained stimulation of host cells by peri-odontopathogens and by the inflammatory mediators (suchas IL-1b and TNF-a) characteristically induced by themmay overcome the genetic predisposition and high levels ofMMP-1 are transcribed irrespective of these SNPs [36]

Also it is believed that the combination of severalsignificant gene variants in certain individuals synergisticallyelevate the susceptibility to disease [37] Since role of TIMPsinMMPs function cannot be denied it can also be postulatedthat mutation of the position 2 (Thr in TIMP-1) greatlyaffects the affinity of TIMPs for MMPs and substitution toglycine essentially inactivates TIMP-1 for MMPs inhibition[38] thus potentiating MMPs activity Besides results ofthe linkage disequilibrium and the haplotype frequencies ofMMP-1 and MMP-3 variants both of which are located in11q223 chromosome near to each other indicated that therisk 2G allele in MMP-1 was more frequently linked to thenonrisk 6A allele in MMP-3 suggesting that the risk andnonrisk linkage combination might lead to the functionalcompensation of MMP function to put it in another wayprotective function of host homeostasis [22]

Furthermore previous studies have hypothesized thatcovariates like severity of the disease and smoking maycontribute towards regulation of MMP-1 expression in dis-eased periodontium [39] So we also performed subgroupanalyses according to severity of CP and smoking habit ofsubjects Similarly lack of association between MMP-1 genevariants in terms of CP severity as well as smoking status andperiodontitis risk was observed in the present meta-analysisand systematic review All these above results may lead to theconclusion that an increase in mRNA transcription causedby these MMP-1 promoter SNPs may not necessarily lead toan increased effect of MMP-1 on the extracellular matrix ofperiodontal tissues and many other factors such as bacterialmetabolites cytokines and other gene variants are supposedto be involved in the regulation of MMP-1 expression andfunctionality

The MMP-2 minus735 CT polymorphism is a synonymousmutation resulting in the same amino acid (threonine)at codon 460 regardless of the allele present It has beenshown that variation of this SNP at synonymous sites couldlead to allele-specific structural differences in mRNA thatcould affect mRNA structure dependent mechanisms [40]which could have functional consequences of increasedMMP-2 expression In oral cancer previous studies haveverified that patients withMMP-2 minus735 CC genotype presentincreased risk for developing oral squamous cell carcinomawhen compared to those with CT or TT genotype [41]These findings were consistent with other studies that havelinked this genotype with an increased risk of developmentof lung cancer [42] gastric cardia adenocarcinoma [43]

Disease Markers 17

and abdominal aortic aneurysm [44] suggesting that thispolymorphism is identified as a promising candidate forneoplasms

On the contrary several studies failed to show associationbetween this variant and the susceptibility to periodontitis [823 25] Likewise our results also found no association of thisSNP with the risk of both CP and AgP so did MMP-2 minus1575GA and minus1306 CT as well as minus790 TG SNPs A possibleexplanation would be that the rare allele of these variantscould disrupt a Sp-1 binding site within the promoter regionof MMP-2 gene thus leading to lower MMP-2 promoteractivity [45] which might also contribute towards negativeassociation of these MMP-2 polymorphisms with periodon-titis risk Besides when stratified by the severity of CP andsmoking a similar distribution of all these MMP-2 variantswas observed between periodontitis patients and controlsSo we can make a conclusion that genetically determinedmechanisms may not be important in tuning the effect ofMMP-2 on periodontal tissues

MMP-9minus1562 CT SNP located on 20q112-q131 chromo-some has been under investigation for its association with anincreased risk for the development of cancer and emphysemaas well as many other diseases [46] Based on the evidence ofprevious studies the suggested mechanism behind a positiveassociation of this polymorphism with disease risk mightbe that the MMP-9 expression is primarily controlled at thetranscriptional level where the promoter of MMP-9 generesponds to stimuli of various cytokines and growth factors[47] Furthermore the T allele of this variant can abolish abinding site for a transcription repressor and thus changethe promoter activity ofMMP-9 leading to increased MMP-9 expression Besides an exchange ofC-to-T at positionminus1562can also alter the binding of a nuclear protein to this regionresulting in increased transcriptional activity inmacrophages[48]

However the present study failed to find any associationof both MMP-9 minus1562 CT and +279 RQ SNPs with peri-odontitis risk A possible explanation for this discrepancymay be that not only the variant but several binding sites andalso their length-dependent interaction with nuclear proteinsmay influence the transcriptional activity of the gene dueto its close localization to the transcriptional start site [49]In addition recent evidence indicates that in periodontitischanges in MMPTIMP balance occur as a result of physio-logical ageing and that gender might be a significant fac-tor modifying this balance [50] Although multiple geneticfactors including SNPs are involved in pro- and anti-inflammatory situations effect of other factors like oxidant-antioxidant imbalance and tissue remodeling cannot bedenied and should be simultaneously considered to under-stand the entire picture of periodontitis risk

The minus1171 5A6A variant a well-characterized inser-tiondeletion polymorphism in the promoter region ofMMP-3 gene is considered to be functionally involved in the processof periodontitisThe 5A allele of this polymorphism has beenshown to result in higher MMP-3 expression and enzymeactivity thereby increasing extracellular matrix breakdownbecause of disruption of a binding site for a nuclear factorkappa B which acts as a transcriptional repressor [51]

Moreover some studies reported positive association of thisSNP with periodontitis and concluded that individuals withthe 5A5A genotype were 2-3 times more likely to developperiodontitis [15 19] Conversely several other studiesshowed a nonsignificant trend for association of this variantwith periodontitis suggesting a likely attempt of the hostenvironment to contain and perhaps specifically outbalancethe increased MMP-3 levels to minimize tissue damage[7 22]

Recently several studies have investigated MMP-8 minus799CT minus381 AG and +17 CG variants in different periodontaldiseases However a significant correlationwith periodontitisrisk was only found inMMP-8 minus799 CT polymorphism andit has been reported that T allele carriers have more MMP-8 production in the periodontal environment with bacterialchallenge compared to non-T allele carriers [30] The exactmechanism behind this association is still unknown but Tallele of this variant has been proved to have about 18-fold higher promoter activity than the C allele [52] BesidesMMP-8 activity has also been found to bemodified in variousorgans and body fluids in smokers [53 54] and tobacco-induced degranulation events in neutrophils and increase inproinflammatory mediator burden can influence the expres-sion level of MMP-8 in smokersrsquo periodontal environment[55] However none of the previous studies have succeedin associating these MMP-8 variants with smoking andperiodontitis risk indicating smoking status may not exertan effect on the association of these SNPs with periodontitissusceptibility

MMP-12 minus357 AsnSer as well as MMP-13 minus77 AGand 11A12A SNPs located on 11q222-q223 chromosomeshas been evaluated with the periodontitis risk in a limitednumber of studies And it is suggested that all these poly-morphisms do not appear to have a significant influence onthe susceptibility to periodontitis and are also not associatedwith the clinical severity of periodontitis as well as outcome ofperiodontal therapy and gingival crevicular fluidMMP-12-13levels [28] Moreover recent studies have also revealed thatMMP-2MMP-3MMP-7MMP-8MMP-11 orMMP-12 sin-gle gene knockoutmice failed to show any apparent disorderssuggesting that a single SNP of MMP might not contributeenough in the susceptibility or progression of a disease Alikely explanation for this behavior would be the sharing ofcommon extracellularmatrix substrates by someMMPmem-bers which might even compensate these functions for eachother [56] Furthermore lack of association between thesevariants and periodontitis may also suggest that an increasein theMMP-12 orMMP-13 transcriptionsmay not necessarilylead to an increase in the destructive effect of these enzymeson the periodontal tissues

When compared with previous similar meta-analysis andsystematic reviews [57 58] the present study has severalstrengths First almost all of these prior studies pooled ORsby using the data of trials investigating the mixed populationhowever a meta-analysis of mixed ethnicities is meaninglessfor a genetic association study owing to high populationheterogeneity As a result we excluded the trials if they did notprovide the detailed information for each ethnicity of amixedpopulation moreover in order to get more reliable results all

18 Disease Markers

meta-analyses and subgroup analyses in our study were per-formed according to the racial descent Besides some of theprevious meta-analyses even included studies in which geno-type distributions of control subjects were varied fromHWEhowever the allele-frequency comparison test is valid onlyif HWE conditions prevail Therefore in the current studywe also took into consideration this factor that might biasthe results suggesting that evidence from our meta-analysisshould be considered to be convincing Nevertheless thisstudy still has several potential limitations One potential lim-itation is that our restriction on searching studies publishedin indexed journals and also studies published only in Englishcould introduce an inherent bias for this analysis Moreoverlack of information for the adjustments ofmajor confoundersincluding age gender and environmental factorsmight causeconfounding bias so a more precise analysis would havebeen performed if all individual raw data had been availableFinally there were only two ethnicity groups (Caucasian andAsian) included in the present study Thus it is doubtfulwhether the obtained conclusions were generalizable to otherpopulations Further studies on this topic in different ethnic-ities are expected to be conducted to strengthen our results

In conclusion the present meta-analysis and systematicreview suggested that although studies of the associationbetween MMP-8 minus799 CT variant and the susceptibilityto periodontitis have not yielded consistent results MMP-1 (minus1607 1G2G minus519 AG and minus422 AT) MMP-2 (minus1575GA minus1306 CT minus790 TG and minus735 CT) MMP-3 (minus11715A6A) MMP-8 (minus381 AG and +17 CG) MMP-9 (minus1562CT and +279 RQ) and MMP-12 (minus357 AsnSer) as wellas MMP-13 (minus77 AG and 11A12A) SNPs are not relatedto periodontitis risk However further well-designed studieswith larger sample size and more ethnic groups are requiredto validate the negative association identified in our studyBesides we expect that in the future analyses using poly-morphisms will not only identify individual variations withindisease comparisons but also help in identification of humanresponse to various therapies Consequently even though sig-nificant insights have been gained into the role of MMPs andtheir function a lot of work needs to be done before the rolesofMMPs in development of periodontitis are fully elucidated

Disclosure

The authors Ying Zhu and Pradeep Singh should be regardedas first joint authors

Competing Interests


Authorsrsquo Contributions

The authors Wenyang Li Ying Zhu and Pradeep Singh con-tributed equally to this study

References

[1] F O Costa A N Guimaraes L O M Cota et al ldquoImpact ofdifferent periodontitis case definitions on periodontal researchrdquoJournal of oral science vol 51 no 2 pp 199ndash206 2009

[2] A Endo T Watanabe N Ogata et al ldquoComparative genomeanalysis and identification of competitive and cooperativeinteractions in a polymicrobial diseaserdquo ISME Journal vol 9no 3 pp 629ndash642 2015

[3] U M Irfan D V Dawson and N F Bissada ldquoEpidemiology ofperiodontal disease a review and clinical perspectivesrdquo Journalof the International Academy of Periodontology vol 3 no 1 pp14ndash21 2001

[4] R P Verma and C Hansch ldquoMatrix metalloproteinases(MMPs) chemical-biological functions and (Q)SARsrdquo Bioor-ganic and Medicinal Chemistry vol 15 no 6 pp 2223ndash22682007

[5] J L Lauer-Fields D Juska and G B Fields ldquoMatrix metallo-proteinases and collagen catabolismrdquo Biopolymers vol 66 no1 pp 19ndash32 2002

[6] B S Sekhon ldquoMatrix metalloproteinasesmdashan overviewrdquoResearch and Reports in Biology vol 1 pp 1ndash20 2010

[7] A Letra R M Silva R J Rylands et al ldquoMMP3 and TIMP1variants contribute to chronic periodontitis and may be impli-cated in disease progressionrdquo Journal of Clinical Periodontologyvol 39 no 8 pp 707ndash716 2012

[8] A Gurkan G Emingil B H Saygan et al ldquoMatrixmetalloproteinase-2 -9 and -12 gene polymorphisms ingeneralized aggressive periodontitisrdquo Journal of Periodontologyvol 78 no 12 pp 2338ndash2347 2007

[9] V Fontana P S Silva R F Gerlach and J E Tanus-SantosldquoCirculating matrix metalloproteinases and their inhibitors inhypertensionrdquo Clinica Chimica Acta vol 413 no 7-8 pp 656ndash662 2012

[10] G Li Y Yue Y Tian et al ldquoAssociation of matrix metallopro-teinase (MMP)-1 3 9 interleukin (IL)-2 8 and cyclooxygenase(COX)-2 gene polymorphisms with chronic periodontitis in aChinese populationrdquo Cytokine vol 60 no 2 pp 552ndash560 2012

[11] Y-H Chou Y-P Ho Y-C Lin et al ldquoMMP-8 -799 CgtT geneticpolymorphism is associated with the susceptibility to chronicand aggressive periodontitis in Taiwaneserdquo Journal of ClinicalPeriodontology vol 38 no 12 pp 1078ndash1084 2011

[12] G C Keles S Gunes A P Sumer et al ldquoAssociation ofmatrix metalloproteinase-9 promoter gene polymorphism withchronic periodontitisrdquo Journal of Periodontology vol 77 no 9pp 1510ndash1514 2006

[13] Z Cao C Li and G Zhu ldquoMMP-1 promoter gene polymor-phism and susceptibility to chronic periodontitis in a Chinesepopulationrdquo Tissue Antigens vol 68 no 1 pp 38ndash43 2006

[14] L I Holla M Jurajda A Fassmann N Dvorakova VZnojil and J Vacha ldquoGenetic variations in the matrixmetalloproteinase-1 promoter and risk of susceptibility andorseverity of chronic periodontitis in the Czech populationrdquoJournal of Clinical Periodontology vol 31 no 8 pp 685ndash6902004

[15] C M Astolfi A L Shinohara R A da Silva M C L G SantosS R P Line and A P de Souza ldquoGenetic polymorphismsin the MMP-1 and MMP-3 gene may contribute to chronicperiodontitis in a Brazilian populationrdquo Journal of ClinicalPeriodontology vol 33 no 10 pp 699ndash703 2006

[16] D Chen QWang Z-WMa et al ldquoMMP-2 MMP-9andTIMP-2gene polymorphisms in Chinese patients with generalized

Disease Markers 19

aggressive periodontitisrdquo Journal of Clinical Periodontology vol34 no 5 pp 384ndash389 2007

[17] S M Luczyszyn C M De Souza A P R Braosi et al ldquoAnalysisof the association of an MMP1 promoter polymorphism andtranscript levels with chronic periodontitis and end-stage renaldisease in a Brazilian populationrdquo Archives of Oral Biology vol57 no 7 pp 954ndash963 2012

[18] C E Repeke A P F Trombone S B Ferreira Jr et al ldquoStrongand persistent microbial and inflammatory stimuli overcomethe genetic predisposition to higher matrix metalloproteinase-1 (MMP-1) expression a mechanistic explanation for the lackof association of MMP1-1607 single-nucleotide polymorphismgenotypes with MMP-1 expression in chronic periodontitislesionsrdquo Journal of Clinical Periodontology vol 36 no 9 pp726ndash738 2009

[19] W T Y Loo M Wang L J Jin M N B Cheung and G R LildquoAssociation of matrix metalloproteinase (MMP-1 MMP-3 andMMP-9) and cyclooxygenase-2 gene polymorphisms and theirproteins with chronic periodontitisrdquo Archives of Oral Biologyvol 56 no 10 pp 1081ndash1090 2011

[20] A P de Souza P C Trevilatto R M Scarel-Caminaga R B deBrito Jr S P Barros and S R P Line ldquoAnalysis of the MMP-9 (C-1562 T) and TIMP-2 (G-418C) gene promoter polymor-phisms in patients with chronic periodontitisrdquo Journal ofClinical Periodontology vol 32 no 2 pp 207ndash211 2005

[21] D M Isaza-Guzman C Arias-Osorio M C Martınez-Pabonand S I Tobon-Arroyave ldquoSalivary levels of matrix metal-loproteinase (MMP)-9 and tissue inhibitor of matrix metal-loproteinase (TIMP)-1 a pilot study about the relationshipwith periodontal status and MMP-9-1562CT gene promoterpolymorphismrdquo Archives of Oral Biology vol 56 no 4 pp 401ndash411 2011

[22] M Itagaki T Kubota H Tai et al ldquoMatrix metalloproteinase-1and -3 gene promoter polymorphisms in Japanese patients withperiodontitisrdquo Journal of Clinical Periodontology vol 31 no 9pp 764ndash769 2004

[23] L I Holla A Fassmann A Vasku et al ldquoGenetic variations inthe human gelatinase A (matrix metalloproteinase-2) promoterare not associated with susceptibility to and severity of chronicperiodontitisrdquo Journal of Periodontology vol 76 no 7 pp 1056ndash1060 2005

[24] L I Holla A Fassmann J Muzik J Vanek and A VaskuldquoFunctional polymorphisms in the matrix metalloproteinase-9gene in relation to severity of chronic periodontitisrdquo Journal ofPeriodontology vol 77 no 11 pp 1850ndash1855 2006

[25] A Gurkan G Emingil B H Saygan et al ldquoGene polymor-phisms of matrix metalloproteinase-2 -9 and -12 in periodontalhealth and severe chronic periodontitisrdquo Archives of OralBiology vol 53 no 4 pp 337ndash345 2008

[26] D Pirhan G Atilla G Emingil T Sorsa T Tervahartialaand A Berdeli ldquoEffect of MMP-1 promoter polymorphismson GCF MMP-1 levels and outcome of periodontal therapy inpatients with severe chronic periodontitisrdquo Journal of ClinicalPeriodontology vol 35 no 10 pp 862ndash870 2008

[27] K Ustun N O Alptekin S S Hakki and E E HakkildquoInvestigation ofmatrixmetalloproteinase-1mdash1607 1G2Gpoly-morphism in a Turkish population with periodontitisrdquo Journalof Clinical Periodontology vol 35 no 12 pp 1013ndash1019 2008

[28] D Pirhan G Atilla G Emingil T Tervahartiala T Sorsa andA Berdeli ldquoMMP-13 promoter polymorphisms in patients with

chronic periodontitis effects on GCF MMP-13 levels and out-come of periodontal therapyrdquo Journal of Clinical Periodontologyvol 36 no 6 pp 474ndash481 2009

[29] L I Holla B Hrdlickova J Vokurka and A Fassmann ldquoMatrixmetalloproteinase 8 (MMP8) gene polymorphisms in chronicperiodontitisrdquo Archives of Oral Biology vol 57 no 2 pp 188ndash196 2012

[30] G Emingil B Han A Gurkan et al ldquoMatrix metalloproteinase(MMP)-8 and tissue inhibitor of MMP-1 (TIMP-1) gene poly-morphisms in generalized aggressive periodontitis gingivalcrevicular fluid MMP-8 and TIMP-1 levels and outcome ofperiodontal therapyrdquo Journal of Periodontology vol 85 no 8pp 1070ndash1080 2014

[31] A P de Souza P C Trevilatto R M Scarel-Caminaga R BBrito Jr and S R P Line ldquoMMP-1 promoter polymorphismassociation with chronic periodontitis severity in a Brazilianpopulationrdquo Journal of Clinical Periodontology vol 30 no 2 pp154ndash158 2003

[32] Z Cao C Li L Jin and E F Corbet ldquoAssociation of matrixmetalloproteinase-1 promoter polymorphism with generalizedaggressive periodontitis in a Chinese populationrdquo Journal ofPeriodontal Research vol 40 no 6 pp 427ndash431 2005

[33] J L Rutter T I Mitchell G Buttice et al ldquoA single nucleotidepolymorphism in the matrix metalloproteinase-1 promotercreates an Ets binding site and augments transcriptionrdquo CancerResearch vol 58 no 23 pp 5321ndash5325 1998


[35] A Kasamatsu K Uzawa K Shimada et al ldquoElevation ofgalectin-9 as an inflammatory response in the periodontal liga-ment cells exposed to Porphylomonas gingivalis lipopolysaccha-ride in vitro and in vivordquo International Journal of Biochemistryand Cell Biology vol 37 no 2 pp 397ndash408 2005

[36] C Rossa Jr LMin P Bronson andK L Kirkwood ldquoTranscrip-tional activation of MMP-13 by periodontal pathogenic LPSrequires p38 MAP kinaserdquo Journal of Endotoxin Research vol13 no 2 pp 85ndash93 2007

[37] S A Dowsett L Archila T Foroud D Koller G J Eckert andM J Kowolik ldquoThe effect of shared genetic and environmentalfactors on periodontal disease parameters in untreated adultsiblings in Guatemalardquo Journal of Periodontology vol 73 no 10pp 1160ndash1168 2002

[38] Q Meng V Malinovskii W Huang et al ldquoResidue 2 of TIMP-1 is a major determinant of affinity and specificity for matrixmetalloproteinases but effects of substitutions do not correlatewith those of the corresponding P1rsquo residue of substraterdquo Journalof Biological Chemistry vol 274 no 15 pp 10184ndash10189 1999

[39] Y-C Chang S-F Yang C-C Lai J-Y Liu and Y-SHsieh ldquoRegulation of matrix metalloproteinase production bycytokines pharmacological agents and periodontal pathogensin human periodontal ligament fibroblast culturesrdquo Journal ofPeriodontal Research vol 37 no 3 pp 196ndash203 2002

[40] L X Shen J P Basilion and V P Stanton Jr ldquoSingle-nucleotidepolymorphisms can cause different structural folds of mRNArdquoProceedings of the National Academy of Sciences of the UnitedStates of America vol 96 no 14 pp 7871ndash7876 1999

[41] A C Pereira E Dias do Carmo M A Dias da Silva and L EBlumer Rosa ldquoMatrix metalloproteinase gene polymorphismsand oral cancerrdquo Journal of Clinical and Experimental Dentistryvol 4 no 5 pp e297ndashe301 2012

20 Disease Markers

[42] J Rollin S Regina P Vourcrsquoh et al ldquoInfluence of MMP-2and MMP-9 promoter polymorphisms on gene expression andclinical outcome of non-small cell lung cancerrdquo Lung Cancervol 56 no 2 pp 273ndash280 2007

[43] Y Li D-L Sun Y-N Duan et al ldquoAssociation of functionalpolymorphisms in MMPs genes with gastric cardia adeno-carcinoma and esophageal squamous cell carcinoma in highincidence region of North Chinardquo Molecular Biology Reportsvol 37 no 1 pp 197ndash205 2010

[44] C Saracini P Bolli E Sticchi et al ldquoPolymorphisms of genesinvolved in extracellular matrix remodeling and abdominalaortic aneurysmrdquo Journal of Vascular Surgery vol 55 no 1 pp171ndash179e2 2012

[45] C Yu Y Zhou X Miao P Xiong W Tan and D Lin ldquoFunc-tional haplotypes in the promoter of matrix metalloproteinase-2 predict risk of the occurrence and metastasis of esophagealcancerrdquo Cancer Research vol 64 no 20 pp 7622ndash7628 2004

[46] M R Luizon and V de Almeida Belo ldquoMatrix metallopro-teinase (MMP)-2 and MMP-9 polymorphisms and haplotypesas disease biomarkersrdquo Biomarkers vol 17 no 3 pp 286ndash2882012

[47] S B Kondapaka R Fridman and K B Reddy ldquoEpi-dermal growth factor and amphiregulin up-regulate matrixmetalloproteinase-9 (MMP-9) in human breast cancer cellsrdquoInternational Journal of Cancer vol 70 no 6 pp 722ndash726 1997

[48] B Zhang S Ye S-M Herrmann et al ldquoFunctional polymor-phism in the regulatory region of gelatinase B gene in relationto severity of coronary atherosclerosisrdquo Circulation vol 99 no14 pp 1788ndash1794 1999

[49] T-S Huang C-C Lee A-C Chang et al ldquoShortening ofmicrosatellite deoxy(CA) repeats involved in GL331-induceddown-regulation of matrix metalloproteinase-9 gene expres-sionrdquo Biochemical and Biophysical Research Communicationsvol 300 no 4 pp 901ndash907 2003

[50] K Komosinska-Vassev P Olczyk K Winsz-Szczotka KKuznik-Trocha K Klimek and K Olczyk ldquoAge- and gender-dependent changes in connective tissue remodeling physiolog-ical differences in circulating MMP-3 MMP-10 TIMP-1 andTIMP-2 levelrdquo Gerontology vol 57 no 1 pp 44ndash52 2010

[51] R C Borghaei P L Rawlings Jr M Javadi and J WoloshinldquoNF-120581B binds to a polymorphic repressor element in theMMP-3 promoterrdquo Biochemical and Biophysical Research Communica-tions vol 316 no 1 pp 182ndash188 2004

[52] J Decock J-R Long R C Laxton et al ldquoAssociation ofmatrix metalloproteinase-8 gene variation with breast cancerprognosisrdquo Cancer Research vol 67 no 21 pp 10214ndash102212007

[53] A M Heikkinen T Sorsa J Pitkaniemi et al ldquoSmoking affectsdiagnostic salivary periodontal disease biomarker levels inadolescentsrdquo Journal of Periodontology vol 81 no 9 pp 1299ndash1307 2010

[54] H Ilumets P Rytila I Demedts et al ldquoMatrix metallopro-teinases -8 -9 and -12 in smokers and patients with Stage 0COPDrdquo International Journal of Chronic Obstructive PulmonaryDisease vol 2 no 3 pp 369ndash379 2007

[55] K-Z Liu A Hynes A Man A Alsagheer D L Singerand D A Scott ldquoIncreased local matrix metalloproteinase-8expression in the periodontal connective tissues of smokerswith periodontal diseaserdquo Biochimica et Biophysica ActamdashMolecular Basis of Disease vol 1762 no 8 pp 775ndash780 2006

[56] Z Zhou S S Apte R Soininen et al ldquoImpaired endochondralossification and angiogenesis in mice deficient in membrane-type matrix metalloproteinase Irdquo Proceedings of the NationalAcademy of Sciences of the United States of America vol 97 no8 pp 4052ndash4057 2000

[57] D Li Q Cai L Ma et al ldquoAssociation between MMP-1 g-1607dupG polymorphism and periodontitis susceptibility ameta-analysisrdquo PLoS ONE vol 8 no 3 Article ID e59513 2013

[58] Y Pan D Li Q Cai et al ldquoMMP-9 -1562CgtT contributes toperiodontitis susceptibilityrdquo Journal of Clinical Periodontologyvol 40 no 2 pp 125ndash130 2013


Disease Markers





Editorial Board




Contents


















2 Disease Markers





References


























1 Introduction




2 Disease Markers










Disease Markers 3








3 Results




4 Disease Markers

Seru

m ac

tive M

MP-

9 (n

gm

L)


500

1000

1500

2000

2500p lt 001

p lt 0001

(a)

OND patients

Seru

m ac

tive M

MP-

2 (n

gm

L)

HC MS patients0

100

200

300

500

600

700

(b)

CSF

activ

e MM

P-9

(ng

mL)


025

050

075

100

200

300

400p lt 001

(c)

CSF

activ

e MM

P-2

(ng

mL)


10

20

30

40

(d)






Disease Markers 5

Seru

m IL

-17

(pg

mL)

MS patients0

200

400

600

800

1000

1500

2000

2500

OND patients

(a)

Seru

m IL

-18

(pg

mL)

0

5000

10000

15000

20000


(b)

Seru

m IL

-23

(pg

mL)

0

2000

4000

6000

800050000

100000

150000


(c)

Seru

m M

CP-3

(pg

mL)

0

50

100

150

200


(d)

CSF

IL-1

7 (p

gm

L)

0

10

20

30

40

50


(e)

CSF

IL-1

8 (p

gm

L)

0

2000

4000

6000

8000


(f)

Figure 2 Continued

6 Disease Markers

CSF

IL-2

3 (p

gm

L)

0

50

100

150

200

2000

4000


(g)

CSF

MCP

-3 (p

gm

L)

0

5

10

15

20

25


(h)








Disease Markers 7




OND () MS ()Serum






4 Discussion






8 Disease Markers






Competing Interests


Acknowledgments


References












Disease Markers 9









































1 Introduction








2 Disease Markers













Disease Markers 3











4 Disease Markers












Disease Markers 5



9 Conclusion




Competing Interests


References













6 Disease Markers































Disease Markers 7







































1 Introduction





2 Disease Markers












3 Results


Disease Markers 3

(a) (b) (c)

(d) (e) (f)






4 Disease Markers













Disease Markers 5



GenderMale 23 (793)









4 Discussion




6 Disease Markers






Competing Interests


Acknowledgments


References







Disease Markers 7



































1 Introduction




2 Disease Markers









Disease Markers 3






3 Results



4 Discussion


4 Disease Markers

Seru

m ac

tive M

MP-2

leve

ls (n

gm

L)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

0

5

10

15

20

(a)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

Seru

m ac

tive M

MP-9

leve

ls (n

gm

L)

0

5

10

15

(b)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

Seru

m T

IMP-

2le

vels

(ng

mL)

0

50

100

150

200

250

(c)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

Seru

m T

IMP-

1le

vels

(ng

mL)

0

200

400

600

800

(d)




Disease Markers 5

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

0

200

400

600

800Se

rum

activ

e MM

P-2

TIM

P-2

(times103)

(a)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

0

10

20

30

40

Seru

m ac

tive M

MP-9

TIM

P-1

(times103)

(b)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

00

05

10

15

20

Seru

m ac

tive M

MP-9

act

ive M

MP-2

1 2 3

4 5

6 7

(c)




6 Disease Markers


Competing Interests


Acknowledgments


References




























Disease Markers 7






















1 Introduction





2 Disease Markers










Disease Markers 3





3 Results





4 Disease Markers

Table1MainCh

aracteris

ticso

fincludedstu

dies

Authoryear

Cou

ntry

Ethn

icity

Samples

ize

(casecontrol)

Type

ofperio

dontitis

Matchingcriteria

Genotype

metho

dGene(po

lymorph

ism)amp

HWEin

controls

deSouzae

tal2003

[31]

Brazil

Caucasian

5037

CP(m

oderateo

rsevere)

Smoker

ratio

sPC

R-RF

LPMMP-1(minus1607

1G2G)0

87

Hollaetal2004

[14]

CzechRe

public

Caucasian

133196

CP(m

ildto

mod

erate

tosevere)

Agegender

PCR-RF

LPMMP-1(minus1607

1G2G)0

52MMP-1(minus519AG

)011

MMP-1(minus422AT)0

47

Itagakietal2004

[22]

Japan

Asian

205142

CP(m

ildor

mod

erate

orsevere)

Agegendersm

oker

ratio

sTaqM

anMMP-1(minus1607

1G2G)0

48

MMP-3(minus117

15A6A)0

87

3714

2AgP

Hollaetal2005

[23]

CzechRe

public

Caucasian

149127

CP(m

ildto

mod

erate

tosevere)

Agesmoker

ratio

sPC

R-RF

LP

MMP-2(minus1575

GA

)040

MMP-2(minus1306

CT)

019

MMP-2(minus790TG)0

67

MMP-2(minus735CT)

042

Caoetal2005

[32]

China

Asian

4052

AgP

mdashPC

R-RF

LPMMP-1(minus1607

1G2G)0

78

Caoetal2006

[13]

China

Asian

6050

CP(m

oderateo

rsevere)

mdashPC

R-RF

LPMMP-1(minus1607

1G2G)0

99

Kelese

tal2006

[12]

Turkey

Caucasian

7070

CP(severe)

Agegender

PCR-RF

LPMMP-9(minus1562

CT)

082

Hollaetal2006

[24]

CzechRe

public

Caucasian

169135

CP(m

oderateo

rsevere)

Agegendersm

oker

ratio

sPC

R-RF

LPMMP-9(minus1562

CT)

059

MMP-9(+279RQ)0

25

Chen

etal2007

[16]

China

Asian

7912

8AgP

Agegender

DHPL

CPC

R-RF

LPMMP-2(minus1306

CT)

100

MMP-9(minus1562

CT)

063

Gurkanetal2007

[8]

Turkey

Caucasian

9215

7AgP

Gender

PCR-RF

LPMMP-2(minus735CT)

045

MMP-9(minus1562

CT)

035

MMP-12

(357

AsnSer)0

06

Gurkanetal2008

[25]

Turkey

Caucasian

8710

7CP

(severe)

mdashPC

R-RF

LPMMP-2(minus735CT)

043

MMP-12

(357

AsnSer)0

47

Pirhan

etal2008

[26]

Turkey

Caucasian

10298

CP(severe)

mdashPC

R-RF

LPMMP-1(minus519AG

)079

Ustu

netal2008

[27]

Turkey

Caucasian

12654

CP(m

oderateo

rsevere)

Age

PCR-RF

LPMMP-1(minus1607

1G2G)0

75

Pirhan

etal2009

[28]

Turkey

Caucasian

10298

CP(severe)

mdashPC

R-RF

LPMMP-13

(minus77

AG

)089

MMP-13

(11A

12A)0

92

Chou

etal2011[11]

China

Asian

3611

06CP

(mod

erateto

severe)

Gendersm

oker

ratio

sPC

R-RF

LPMMP-8(minus799CT)

022

9610

6AgP

Hollaetal2012

[29]

CzechRe

public

Caucasian

3412

78CP

(mild

tomod

erate

tosevere)

Agegendersm

oker

ratio

sPC

R-RF

LPMMP-8(+17

CG)0

14MMP-8(minus799CT)

063

Emingiletal2014

[30]

Turkey

Caucasian

100167

AgP

Gender

PCR-RF

LPMMP-8(+17

CG)0

29

MMP-8(minus799CT)

009

MMP-8(minus381A

G)0

87

CPchron

icperio

dontitisAgP

aggressiveperio

dontitisHWE

Hardy-W

einb

ergequilib

riumM

ildchronicperio

dontitispatie

ntsw

ithteethexhibitin

glt3m

mattachmentlossmod

eratechronicperio

dontitis

patientsw

ithteethexhibitin

gge3m

mandlt7m


chronicp

eriodo

ntitispatie

ntsw

ithteethexhibitin

gge7m


ofHWEwas

considered

significant

Disease Markers 5



Selection






Exposure




6 Disease Markers


Scre

enin

gIn

clude

dEl

igib

ility








Iden

tifica

tion


HWE n = 3







(i) Reviews n = 5








Disease Markers 7


ofthep

olym

orph

ismsinMMPs

gene

onperio

dontitisrisk

MMP-1

minus1607

1G2G

Stud

ies

(casescon

trols)

2Gversus

1GOR(95

CI)

1198682(

)119901ℎ119901119888

2G2Gversus

1G1G

OR(95

CI)

1198682(

)119901ℎ119901119888

1G2Gversus

1G1G

OR(95

CI)

1198682(

)119901ℎ119901119888

2G2Gversus

1G2G

OR(95

CI)

1198682(

)119901ℎ119901119888

1G2G+2G

2Gversus

1G1G

OR(95

CI)

1198682(

)119901ℎ119901119888

2G2Gversus

1G1G

+1G

2GOR(95

CI)

1198682(

)119901ℎ119901119888

1G2Gversus

1G1G

+2G

2G

OR(95

CI)

1198682(

)119901ℎ119901119888

Type

ofdisease

CP Caucasian

3(309287)

102(067ndash15

6)10

5(044ndash

251)

095

(064ndash

142)

090

(059ndash

137)

091

(062ndash13

2)089

(060ndash

133)

100(072ndash14

0)631

006710

0063000671000

120032

110

0000049010

00499

013610

00382019

810

0000096910

00

Asian

2(30219

2)17

5(077ndash395)

279

(056ndash

1398)

131(074ndash232

)17

5(077ndash394)

196(063ndash609)

201

(072ndash561)

079

(055ndash116)

84500111000

81500201000

27002421000

64600931000

69400711000

797

002710

0000046

710

00Ag

P

Asian

2(77194)

134(048ndash373)

154(028ndash855)

091

(042ndash19

6)16

7(038ndash731)

114(056ndash

232)

164(035

ndash770)

075

(043ndash13

0)84800101000

78400311000

00076310

0081800191000

39002001000

857

000810

00595011610

00Severe

CPCa

ucasian

Mild

tomod

erate

2(6691)

099

(063ndash15

5)099

(039

ndash250)

116(052

ndash260)

085

(039

ndash184)

110(051ndash238)

089

(043ndash18

5)117(062ndash221)

00061310

0000061310

0000066710

0000087410

0000061310

0000075110

0000087610

00

Severe

2(11091)

153(072ndash324)

244

(047ndash1259)

152(070ndash

330)

131(068ndash251)

168(081ndash350)

147(080ndash

273)

098

(056ndash

173)

657

008810

0063400981000

15802761000

00036910

00511

015310

00423018

810

0000084510

00Asian

Mild

tomod

erate

2(16819

2)12

6(093ndash17

2)16

2(084ndash

312)

140(072ndash269)

119(075ndash18

7)15

1(082ndash280)

127(083ndash19

5)097

(063ndash14

8)601

0113098

7627

010110

00438018

210

0000053410

00560013

210

0021002611000

00078410

00

Severe

2(97192)

199(092ndash426)

293

(071ndash1203)

116(053

ndash256)

219

(126ndash

379)

178(086ndash

367)

255

(095ndash685)

058

(035

ndash098)

73500520546

67000820952

00046

810

00489

01620035

381

02040854

697

0069044

1000517028

7Sm

okinghabitinCP

Caucasian

Non

smok

ing

3(200213)

092

(055ndash15

5)090

(033

ndash243)

087

(054ndash

140)

085

(052

ndash141)

096

(045ndash205)

082

(052

ndash130)

098

(066ndash

146)

661

005310

00631

006710

0034902151000

00043810

00569

009810

0040

10118

1000

00057510

00

Smok

ing

2(10974)

110(071ndash17

2)114(043ndash302)

112(054ndash

234)

115(050ndash

264

)112(055ndash229)

119(053

ndash265)

098

(053

ndash184)

19002671000

329

022210

0000097610

00522014

810

0000068210

00540014

010

0000031910

00MMP-1

minus519AG

Stud

ies

(casescon

trols)

Gversus

AOR(95

CI)

1198682(

)119901ℎ119901119888

GGversus

AA

OR(95

CI)

1198682(

)119901ℎ119901119888

AGversus

AA

OR(95

CI)

1198682(

)119901ℎ119901119888

GGversus

AGOR(95

CI)

1198682(

)119901ℎ119901119888

AG+GGversus

AA

OR(95

CI)

1198682(

)119901ℎ119901119888

GGversus

AA+AG

OR(95

CI)

1198682(

)119901ℎ119901119888

AGversus

AA+GG

OR(95

CI)

1198682(

)119901ℎ119901119888

Type

ofdisease

CP Caucasian

2(235293)

103(080ndash

132)

108(064ndash

182)

100(068ndash14

6)10

6(064ndash

175)

102(071ndash14

5)10

6(067ndash16

9)098

(069ndash

139)

00

09841000

00

08061000

00

0811

1000

00

06911000

00

08841000

00

07361000

00077810

00MMP-2

minus735CT

Stud

ies

(casescon

trols)

Tversus

COR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

CTversus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CTOR(95

CI)

1198682(

)119901ℎ119901119888

CT+TT

versus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CC+CT

OR(95

CI)

1198682(

)119901ℎ119901119888

CTversus

CC+TT

OR(95

CI)

1198682(

)119901ℎ119901119888

Type

ofdisease

CP Caucasian

2(236234)

111(079ndash155)

119(043ndash337)

112(074

ndash168)

108(037

ndash313

)10

0(067ndash14

9)116(041ndash324)

110(074

ndash165)

00069510

0000051110

0000094010

0000054110

0000069910

0000052210

0000098910

00Sm

okinghabitinCP

Caucasian

Non

smok

ing

2(13319

8)113(074

ndash172)

115(032

ndash409)

117(070ndash

194)

099

(027ndash366)

116(071ndash18

8)110(031ndash389)

115(069ndash

190)

00033710

00452017

710

0000078410

0032402241000

00053310

00424018

810

0000091710

00

8 Disease Markers

Table3Con

tinued

MMP-9

minus1562

CT

Stud

ies

(casescon

trols)

Tversus

COR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

CTversus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CTOR(95

CI)

1198682(

)119901ℎ119901119888

CT+TT

versus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CC+CT

OR(95

CI)

1198682(

)119901ℎ119901119888

CTversus

CC+TT

OR(95

CI)

1198682(

)119901ℎ119901119888

Type

ofdisease

CP Caucasian

2(239205)

056

(024ndash

135)

036

(011ndash112

)063

(029ndash

136)

051

(015

ndash172)

057

(023ndash13

8)039

(012

ndash124)

072

(047ndash110)

78200321000

35902120553

64000961000

00045910

00739

005010

0020402620777

571

0127092

4Severe

CPCa

ucasian

Severe

2(163205)

063

(020ndash

197)

044

(013

ndash149)

071

(024ndash

209)

053

(014

ndash195)

065

(019

ndash215

)046

(014

ndash157)

075

(028ndash201)

861

000710

00544013

910

0079300281000

00042810

0084200121000

410019

310

0076000411000

MMP-1matrix


MP-2matrix


MP-9matrix


Pchronicp

eriodo

ntitisAgP

aggressiveg

eneralized

perio

dontitis



orrected

byBo

nferroni

correctio

nOR

odds

ratio

CIconfi

denceinterval


0the

rand

omeffectsmod

elisusedC

onversely

the

fixed

effectsmod

elisused

119901119888lt001

isconsidered

statisticallysig

nificant

Disease Markers 9


Holla et al (2004)

Ustun et al (2008)


Cao et al (2006)


Cao et al (2005)

Holla et al (2006)

Keles et al (2006)

Study ID

165 (090 303)

075 (055 103)

103 (065 161)

102 (067 156)

119 (087 163)

274 (157 481)

175 (077 395)

080 (048 135)

229 (124 420)

134 (048 373)

084 (055 130)

035 (017 069)

056 (024 135)

OR (95 CI)

2551

4121

3327

10000

5398

4602

10000

5118

4882

10000

5479

4521

10000

Holla et al (2004)

Pirhan et al (2008)

Holla et al (2005)

Study ID

102 (075 140)

103 (067 159)

103 (080 132)

104 (065 165)

119 (073 193)

111 (079 155)

OR (95 CI)

6619

3381

10000

5388

4612

10000

weight

weight

Subtotal (I2 = 00 p = 0695)

Subtotal (I2 = 00 p = 0984)

Subtotal (I2 = 782 p = 0032)

Subtotal (I2 = 848 p = 0010)

Subtotal (I2 = 845 p = 0011)

Subtotal (I2 = 631 p = 0067)

1 5790173

1 1930518








Gurkan et al (2007)


Figure 2 Continued

10 Disease Markers

286 (082 1001)

056 (030 107)

107 (042 273)

105 (044 251)

133 (069 257)

693 (203 2363)

279 (056 1398)

066 (023 188)

379 (114 1258)

154 (028 855)

2539

4150

3312

10000

5509

4491

10000

5148

4852

10000

104 (057 189)

121 (042 350)

108 (064 182)

087 (021 357)

175 (038 818)

120 (043 337)

061 (016 233)

008 (000 157)

036 (011 112)

7713

2287

10000

6270

3730

10000

5250

4750

10000

Study ID OR (95 CI)

Study ID OR (95 CI)

Subtotal (I2 = 359 p = 0212)

Subtotal (I2 = 00 p = 0511)

Subtotal (I2 = 00 p = 0806)

Subtotal (I2 = 784 p = 0031)

Subtotal (I2 = 815 p = 0020)

Subtotal (I2 = 630 p = 0067)

1 23600423

1 23200043








weight

weight


Holla et al (2004)

Ustun et al (2008)


Cao et al (2006)


Cao et al (2005)

Holla et al (2006)

Keles et al (2006)

Holla et al (2004)

Pirhan et al (2008)

Holla et al (2005)

Gurkan et al (2007)


Figure 2 Continued

Disease Markers 11

089 (053 148)

040 (018 087)

063 (029 136)

5713

4287

10000

54910182

Study ID OR (95 CI)

Subtotal (I2 = 640 p = 0096)

198 (063 620)079 (048 129)110 (047 254)095 (064 142)

107 (055 207)239 (074 776)131 (074 232)

082 (030 226)104 (032 337)091 (042 196)

104 (062 172)094 (053 169)100 (068 146)

110 (063 191)114 (063 206)112 (074 168)

84470792077

10000

81671833

10000

59634037

10000

55794421

10000

54274573

10000

1 7760129

Study ID OR (95 CI)

Subtotal (I2 = 120 p = 0321)

Subtotal (I2 = 270 p = 0242)

Subtotal (I2 = 00 p = 0763)

Subtotal (I2 = 00 p = 0811)

Subtotal (I2 = 00 p = 0940)








weight

weight




Holla et al (2006)

Keles et al (2006)




Figure 2 Continued

12 Disease Markers

144 (053 393)

072 (039 132)

098 (046 206)

090 (059 137)

100 (057 177)

129 (044 379)

106 (064 175)

079 (018 342)

154 (032 741)

108 (037 313)

069 (017 275)

020 (001 404)

051 (015 172)

1426

5500

3074

10000

8046

1954

10000

6129

3871

10000

6310

3690

10000

124 (078 197)

290 (121 693)

175 (077 394)

080 (036 180)

363 (138 959)

167 (038 731)

5994

4006

10000

5167

4833

10000

Study ID OR (95 CI)

Study ID OR (95 CI)

Subtotal (I2 = 00 p = 0490)

Subtotal (I2 = 00 p = 0691)

Subtotal (I2 = 00 p = 0541)

Subtotal (I2 = 00 p = 0459)

Subtotal (I2 = 646 p = 0093)

Subtotal (I2 = 818 p = 0019)

1 99200101

1 9590104








weight

weight


Holla et al (2004)

Ustun et al (2008)


Cao et al (2006)


Cao et al (2005)

Holla et al (2006)

Keles et al (2006)

Holla et al (2004)

Pirhan et al (2008)

Holla et al (2005)

Gurkan et al (2007)


Figure 2 Continued

Disease Markers 13

228 (077 669)

071 (045 114)

109 (049 241)

091 (062 132)

074 (029 191)

189 (065 551)

114 (056 232)

104 (065 166)

098 (056 172)

102 (071 145)

107 (063 183)

092 (051 166)

100 (067 149)

756

7248

1996

10000

6504

3496

10000

5777

4223

10000

5363

4637

10000


119 (064 222)

386 (127 1176)

196 (063 609)

085 (052 140)

034 (016 074)

057 (023 138)

5805

4195

10000

5539

4461

10000

Study ID OR (95 CI)

Study ID OR (95 CI)

1 6690149

1 11800851

Subtotal (I2 = 499 p = 0136)

Subtotal (I2 = 390 p = 0200)

Subtotal (I2 = 00 p = 0884)

Subtotal (I2 = 00 p = 0699)

Subtotal (I2 = 694 p = 0071)

Subtotal (I2 = 739 p = 0050)







weight

weight


Holla et al (2004)

Ustun et al (2008)


Cao et al (2005)

Holla et al (2004)

Cao et al (2006)


Holla et al (2006)

Keles et al (2006)

Pirhan et al (2008)

Holla et al (2005)

Gurkan et al (2007)


Figure 2 Continued

14 Disease Markers

175 (068 451)

065 (036 115)

100 (049 204)

089 (060 133)

102 (061 172)

124 (044 348)

106 (067 169)

085 (021 346)

167 (036 767)

116 (042 324)

063 (017 239)

010 (001 198)

039 (012 124)

1279

5789

2931

10000

8101

1899

10000

6208

3792

10000

5485

4515

10000

126 (082 195)

362 (159 825)

201 (072 561)

076 (036 162)

368 (151 902)

164 (035 770)

5578

4422

10000

5123

487

10000

Study ID OR (95 CI)

Study ID OR (95 CI)

Subtotal (I2 = 204 p = 0262)

Subtotal (I2 = 382 p = 0198)

Subtotal (I2 = 00 p = 0736)

Subtotal (I2 = 00 p = 0522)

Subtotal (I2 = 797 p = 0027)

Subtotal (I2 = 857 p = 0008)

1 9020111

1 181000554








weight

weight


Holla et al (2004)

Ustun et al (2008)


Cao et al (2006)


Cao et al (2005)

Holla et al (2006)

Keles et al (2006)

Holla et al (2004)

Pirhan et al (2008)

Holla et al (2005)

Gurkan et al (2007)


Figure 2 Continued

Disease Markers 15

106 (045 247)097 (062 150)105 (056 200)100 (072 140)

086 (056 133)062 (029 133)079 (055 116)

110 (053 227)045 (019 105)075 (043 130)

102 (065 159)092 (052 162)098 (069 139)

111 (064 192)110 (061 199)110 (074 165)

090 (054 151)044 (020 095)072 (047 110)

150358312666

10000

72332767

10000

46645336

10000

60823918

10000

53494651

10000

61083892

10000

10189 528

Subtotal (I2 = 00 p = 0969)

Subtotal (I2 = 00 p = 0467)

Subtotal (I2 = 595 p = 0116)

Subtotal (I2 = 00 p = 0778)

Subtotal (I2 = 00 p = 0989)

Subtotal (I2 = 571 p = 0127)

Study ID OR (95 CI)







weight













16 Disease Markers




4 Discussion







Disease Markers 17










18 Disease Markers



Disclosure


Competing Interests




References

















Disease Markers 19




























20 Disease Markers


















Disease Markers





Editorial Board




Contents


















2 Disease Markers





References


























1 Introduction




2 Disease Markers










Disease Markers 3








3 Results




4 Disease Markers

Seru

m ac

tive M

MP-

9 (n

gm

L)


500

1000

1500

2000

2500p lt 001

p lt 0001

(a)

OND patients

Seru

m ac

tive M

MP-

2 (n

gm

L)

HC MS patients0

100

200

300

500

600

700

(b)

CSF

activ

e MM

P-9

(ng

mL)


025

050

075

100

200

300

400p lt 001

(c)

CSF

activ

e MM

P-2

(ng

mL)


10

20

30

40

(d)






Disease Markers 5

Seru

m IL

-17

(pg

mL)

MS patients0

200

400

600

800

1000

1500

2000

2500

OND patients

(a)

Seru

m IL

-18

(pg

mL)

0

5000

10000

15000

20000


(b)

Seru

m IL

-23

(pg

mL)

0

2000

4000

6000

800050000

100000

150000


(c)

Seru

m M

CP-3

(pg

mL)

0

50

100

150

200


(d)

CSF

IL-1

7 (p

gm

L)

0

10

20

30

40

50


(e)

CSF

IL-1

8 (p

gm

L)

0

2000

4000

6000

8000


(f)

Figure 2 Continued

6 Disease Markers

CSF

IL-2

3 (p

gm

L)

0

50

100

150

200

2000

4000


(g)

CSF

MCP

-3 (p

gm

L)

0

5

10

15

20

25


(h)








Disease Markers 7




OND () MS ()Serum






4 Discussion






8 Disease Markers






Competing Interests


Acknowledgments


References












Disease Markers 9









































1 Introduction








2 Disease Markers













Disease Markers 3











4 Disease Markers












Disease Markers 5



9 Conclusion




Competing Interests


References













6 Disease Markers































Disease Markers 7







































1 Introduction





2 Disease Markers












3 Results


Disease Markers 3

(a) (b) (c)

(d) (e) (f)






4 Disease Markers













Disease Markers 5



GenderMale 23 (793)









4 Discussion




6 Disease Markers






Competing Interests


Acknowledgments


References







Disease Markers 7



































1 Introduction




2 Disease Markers









Disease Markers 3






3 Results



4 Discussion


4 Disease Markers

Seru

m ac

tive M

MP-2

leve

ls (n

gm

L)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

0

5

10

15

20

(a)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

Seru

m ac

tive M

MP-9

leve

ls (n

gm

L)

0

5

10

15

(b)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

Seru

m T

IMP-

2le

vels

(ng

mL)

0

50

100

150

200

250

(c)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

Seru

m T

IMP-

1le

vels

(ng

mL)

0

200

400

600

800

(d)




Disease Markers 5

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

0

200

400

600

800Se

rum

activ

e MM

P-2

TIM

P-2

(times103)

(a)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

0

10

20

30

40

Seru

m ac

tive M

MP-9

TIM

P-1

(times103)

(b)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

00

05

10

15

20

Seru

m ac

tive M

MP-9

act

ive M

MP-2

1 2 3

4 5

6 7

(c)




6 Disease Markers


Competing Interests


Acknowledgments


References




























Disease Markers 7






















1 Introduction





2 Disease Markers










Disease Markers 3





3 Results





4 Disease Markers

Table1MainCh

aracteris

ticso

fincludedstu

dies

Authoryear

Cou

ntry

Ethn

icity

Samples

ize

(casecontrol)

Type

ofperio

dontitis

Matchingcriteria

Genotype

metho

dGene(po

lymorph

ism)amp

HWEin

controls

deSouzae

tal2003

[31]

Brazil

Caucasian

5037

CP(m

oderateo

rsevere)

Smoker

ratio

sPC

R-RF

LPMMP-1(minus1607

1G2G)0

87

Hollaetal2004

[14]

CzechRe

public

Caucasian

133196

CP(m

ildto

mod

erate

tosevere)

Agegender

PCR-RF

LPMMP-1(minus1607

1G2G)0

52MMP-1(minus519AG

)011

MMP-1(minus422AT)0

47

Itagakietal2004

[22]

Japan

Asian

205142

CP(m

ildor

mod

erate

orsevere)

Agegendersm

oker

ratio

sTaqM

anMMP-1(minus1607

1G2G)0

48

MMP-3(minus117

15A6A)0

87

3714

2AgP

Hollaetal2005

[23]

CzechRe

public

Caucasian

149127

CP(m

ildto

mod

erate

tosevere)

Agesmoker

ratio

sPC

R-RF

LP

MMP-2(minus1575

GA

)040

MMP-2(minus1306

CT)

019

MMP-2(minus790TG)0

67

MMP-2(minus735CT)

042

Caoetal2005

[32]

China

Asian

4052

AgP

mdashPC

R-RF

LPMMP-1(minus1607

1G2G)0

78

Caoetal2006

[13]

China

Asian

6050

CP(m

oderateo

rsevere)

mdashPC

R-RF

LPMMP-1(minus1607

1G2G)0

99

Kelese

tal2006

[12]

Turkey

Caucasian

7070

CP(severe)

Agegender

PCR-RF

LPMMP-9(minus1562

CT)

082

Hollaetal2006

[24]

CzechRe

public

Caucasian

169135

CP(m

oderateo

rsevere)

Agegendersm

oker

ratio

sPC

R-RF

LPMMP-9(minus1562

CT)

059

MMP-9(+279RQ)0

25

Chen

etal2007

[16]

China

Asian

7912

8AgP

Agegender

DHPL

CPC

R-RF

LPMMP-2(minus1306

CT)

100

MMP-9(minus1562

CT)

063

Gurkanetal2007

[8]

Turkey

Caucasian

9215

7AgP

Gender

PCR-RF

LPMMP-2(minus735CT)

045

MMP-9(minus1562

CT)

035

MMP-12

(357

AsnSer)0

06

Gurkanetal2008

[25]

Turkey

Caucasian

8710

7CP

(severe)

mdashPC

R-RF

LPMMP-2(minus735CT)

043

MMP-12

(357

AsnSer)0

47

Pirhan

etal2008

[26]

Turkey

Caucasian

10298

CP(severe)

mdashPC

R-RF

LPMMP-1(minus519AG

)079

Ustu

netal2008

[27]

Turkey

Caucasian

12654

CP(m

oderateo

rsevere)

Age

PCR-RF

LPMMP-1(minus1607

1G2G)0

75

Pirhan

etal2009

[28]

Turkey

Caucasian

10298

CP(severe)

mdashPC

R-RF

LPMMP-13

(minus77

AG

)089

MMP-13

(11A

12A)0

92

Chou

etal2011[11]

China

Asian

3611

06CP

(mod

erateto

severe)

Gendersm

oker

ratio

sPC

R-RF

LPMMP-8(minus799CT)

022

9610

6AgP

Hollaetal2012

[29]

CzechRe

public

Caucasian

3412

78CP

(mild

tomod

erate

tosevere)

Agegendersm

oker

ratio

sPC

R-RF

LPMMP-8(+17

CG)0

14MMP-8(minus799CT)

063

Emingiletal2014

[30]

Turkey

Caucasian

100167

AgP

Gender

PCR-RF

LPMMP-8(+17

CG)0

29

MMP-8(minus799CT)

009

MMP-8(minus381A

G)0

87

CPchron

icperio

dontitisAgP

aggressiveperio

dontitisHWE

Hardy-W

einb

ergequilib

riumM

ildchronicperio

dontitispatie

ntsw

ithteethexhibitin

glt3m

mattachmentlossmod

eratechronicperio

dontitis

patientsw

ithteethexhibitin

gge3m

mandlt7m


chronicp

eriodo

ntitispatie

ntsw

ithteethexhibitin

gge7m


ofHWEwas

considered

significant

Disease Markers 5



Selection






Exposure




6 Disease Markers


Scre

enin

gIn

clude

dEl

igib

ility








Iden

tifica

tion


HWE n = 3







(i) Reviews n = 5








Disease Markers 7


ofthep

olym

orph

ismsinMMPs

gene

onperio

dontitisrisk

MMP-1

minus1607

1G2G

Stud

ies

(casescon

trols)

2Gversus

1GOR(95

CI)

1198682(

)119901ℎ119901119888

2G2Gversus

1G1G

OR(95

CI)

1198682(

)119901ℎ119901119888

1G2Gversus

1G1G

OR(95

CI)

1198682(

)119901ℎ119901119888

2G2Gversus

1G2G

OR(95

CI)

1198682(

)119901ℎ119901119888

1G2G+2G

2Gversus

1G1G

OR(95

CI)

1198682(

)119901ℎ119901119888

2G2Gversus

1G1G

+1G

2GOR(95

CI)

1198682(

)119901ℎ119901119888

1G2Gversus

1G1G

+2G

2G

OR(95

CI)

1198682(

)119901ℎ119901119888

Type

ofdisease

CP Caucasian

3(309287)

102(067ndash15

6)10

5(044ndash

251)

095

(064ndash

142)

090

(059ndash

137)

091

(062ndash13

2)089

(060ndash

133)

100(072ndash14

0)631

006710

0063000671000

120032

110

0000049010

00499

013610

00382019

810

0000096910

00

Asian

2(30219

2)17

5(077ndash395)

279

(056ndash

1398)

131(074ndash232

)17

5(077ndash394)

196(063ndash609)

201

(072ndash561)

079

(055ndash116)

84500111000

81500201000

27002421000

64600931000

69400711000

797

002710

0000046

710

00Ag

P

Asian

2(77194)

134(048ndash373)

154(028ndash855)

091

(042ndash19

6)16

7(038ndash731)

114(056ndash

232)

164(035

ndash770)

075

(043ndash13

0)84800101000

78400311000

00076310

0081800191000

39002001000

857

000810

00595011610

00Severe

CPCa

ucasian

Mild

tomod

erate

2(6691)

099

(063ndash15

5)099

(039

ndash250)

116(052

ndash260)

085

(039

ndash184)

110(051ndash238)

089

(043ndash18

5)117(062ndash221)

00061310

0000061310

0000066710

0000087410

0000061310

0000075110

0000087610

00

Severe

2(11091)

153(072ndash324)

244

(047ndash1259)

152(070ndash

330)

131(068ndash251)

168(081ndash350)

147(080ndash

273)

098

(056ndash

173)

657

008810

0063400981000

15802761000

00036910

00511

015310

00423018

810

0000084510

00Asian

Mild

tomod

erate

2(16819

2)12

6(093ndash17

2)16

2(084ndash

312)

140(072ndash269)

119(075ndash18

7)15

1(082ndash280)

127(083ndash19

5)097

(063ndash14

8)601

0113098

7627

010110

00438018

210

0000053410

00560013

210

0021002611000

00078410

00

Severe

2(97192)

199(092ndash426)

293

(071ndash1203)

116(053

ndash256)

219

(126ndash

379)

178(086ndash

367)

255

(095ndash685)

058

(035

ndash098)

73500520546

67000820952

00046

810

00489

01620035

381

02040854

697

0069044

1000517028

7Sm

okinghabitinCP

Caucasian

Non

smok

ing

3(200213)

092

(055ndash15

5)090

(033

ndash243)

087

(054ndash

140)

085

(052

ndash141)

096

(045ndash205)

082

(052

ndash130)

098

(066ndash

146)

661

005310

00631

006710

0034902151000

00043810

00569

009810

0040

10118

1000

00057510

00

Smok

ing

2(10974)

110(071ndash17

2)114(043ndash302)

112(054ndash

234)

115(050ndash

264

)112(055ndash229)

119(053

ndash265)

098

(053

ndash184)

19002671000

329

022210

0000097610

00522014

810

0000068210

00540014

010

0000031910

00MMP-1

minus519AG

Stud

ies

(casescon

trols)

Gversus

AOR(95

CI)

1198682(

)119901ℎ119901119888

GGversus

AA

OR(95

CI)

1198682(

)119901ℎ119901119888

AGversus

AA

OR(95

CI)

1198682(

)119901ℎ119901119888

GGversus

AGOR(95

CI)

1198682(

)119901ℎ119901119888

AG+GGversus

AA

OR(95

CI)

1198682(

)119901ℎ119901119888

GGversus

AA+AG

OR(95

CI)

1198682(

)119901ℎ119901119888

AGversus

AA+GG

OR(95

CI)

1198682(

)119901ℎ119901119888

Type

ofdisease

CP Caucasian

2(235293)

103(080ndash

132)

108(064ndash

182)

100(068ndash14

6)10

6(064ndash

175)

102(071ndash14

5)10

6(067ndash16

9)098

(069ndash

139)

00

09841000

00

08061000

00

0811

1000

00

06911000

00

08841000

00

07361000

00077810

00MMP-2

minus735CT

Stud

ies

(casescon

trols)

Tversus

COR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

CTversus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CTOR(95

CI)

1198682(

)119901ℎ119901119888

CT+TT

versus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CC+CT

OR(95

CI)

1198682(

)119901ℎ119901119888

CTversus

CC+TT

OR(95

CI)

1198682(

)119901ℎ119901119888

Type

ofdisease

CP Caucasian

2(236234)

111(079ndash155)

119(043ndash337)

112(074

ndash168)

108(037

ndash313

)10

0(067ndash14

9)116(041ndash324)

110(074

ndash165)

00069510

0000051110

0000094010

0000054110

0000069910

0000052210

0000098910

00Sm

okinghabitinCP

Caucasian

Non

smok

ing

2(13319

8)113(074

ndash172)

115(032

ndash409)

117(070ndash

194)

099

(027ndash366)

116(071ndash18

8)110(031ndash389)

115(069ndash

190)

00033710

00452017

710

0000078410

0032402241000

00053310

00424018

810

0000091710

00

8 Disease Markers

Table3Con

tinued

MMP-9

minus1562

CT

Stud

ies

(casescon

trols)

Tversus

COR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

CTversus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CTOR(95

CI)

1198682(

)119901ℎ119901119888

CT+TT

versus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CC+CT

OR(95

CI)

1198682(

)119901ℎ119901119888

CTversus

CC+TT

OR(95

CI)

1198682(

)119901ℎ119901119888

Type

ofdisease

CP Caucasian

2(239205)

056

(024ndash

135)

036

(011ndash112

)063

(029ndash

136)

051

(015

ndash172)

057

(023ndash13

8)039

(012

ndash124)

072

(047ndash110)

78200321000

35902120553

64000961000

00045910

00739

005010

0020402620777

571

0127092

4Severe

CPCa

ucasian

Severe

2(163205)

063

(020ndash

197)

044

(013

ndash149)

071

(024ndash

209)

053

(014

ndash195)

065

(019

ndash215

)046

(014

ndash157)

075

(028ndash201)

861

000710

00544013

910

0079300281000

00042810

0084200121000

410019

310

0076000411000

MMP-1matrix


MP-2matrix


MP-9matrix


Pchronicp

eriodo

ntitisAgP

aggressiveg

eneralized

perio

dontitis



orrected

byBo

nferroni

correctio

nOR

odds

ratio

CIconfi

denceinterval


0the

rand

omeffectsmod

elisusedC

onversely

the

fixed

effectsmod

elisused

119901119888lt001

isconsidered

statisticallysig

nificant

Disease Markers 9


Holla et al (2004)

Ustun et al (2008)


Cao et al (2006)


Cao et al (2005)

Holla et al (2006)

Keles et al (2006)

Study ID

165 (090 303)

075 (055 103)

103 (065 161)

102 (067 156)

119 (087 163)

274 (157 481)

175 (077 395)

080 (048 135)

229 (124 420)

134 (048 373)

084 (055 130)

035 (017 069)

056 (024 135)

OR (95 CI)

2551

4121

3327

10000

5398

4602

10000

5118

4882

10000

5479

4521

10000

Holla et al (2004)

Pirhan et al (2008)

Holla et al (2005)

Study ID

102 (075 140)

103 (067 159)

103 (080 132)

104 (065 165)

119 (073 193)

111 (079 155)

OR (95 CI)

6619

3381

10000

5388

4612

10000

weight

weight

Subtotal (I2 = 00 p = 0695)

Subtotal (I2 = 00 p = 0984)

Subtotal (I2 = 782 p = 0032)

Subtotal (I2 = 848 p = 0010)

Subtotal (I2 = 845 p = 0011)

Subtotal (I2 = 631 p = 0067)

1 5790173

1 1930518








Gurkan et al (2007)


Figure 2 Continued

10 Disease Markers

286 (082 1001)

056 (030 107)

107 (042 273)

105 (044 251)

133 (069 257)

693 (203 2363)

279 (056 1398)

066 (023 188)

379 (114 1258)

154 (028 855)

2539

4150

3312

10000

5509

4491

10000

5148

4852

10000

104 (057 189)

121 (042 350)

108 (064 182)

087 (021 357)

175 (038 818)

120 (043 337)

061 (016 233)

008 (000 157)

036 (011 112)

7713

2287

10000

6270

3730

10000

5250

4750

10000

Study ID OR (95 CI)

Study ID OR (95 CI)

Subtotal (I2 = 359 p = 0212)

Subtotal (I2 = 00 p = 0511)

Subtotal (I2 = 00 p = 0806)

Subtotal (I2 = 784 p = 0031)

Subtotal (I2 = 815 p = 0020)

Subtotal (I2 = 630 p = 0067)

1 23600423

1 23200043








weight

weight


Holla et al (2004)

Ustun et al (2008)


Cao et al (2006)


Cao et al (2005)

Holla et al (2006)

Keles et al (2006)

Holla et al (2004)

Pirhan et al (2008)

Holla et al (2005)

Gurkan et al (2007)


Figure 2 Continued

Disease Markers 11

089 (053 148)

040 (018 087)

063 (029 136)

5713

4287

10000

54910182

Study ID OR (95 CI)

Subtotal (I2 = 640 p = 0096)

198 (063 620)079 (048 129)110 (047 254)095 (064 142)

107 (055 207)239 (074 776)131 (074 232)

082 (030 226)104 (032 337)091 (042 196)

104 (062 172)094 (053 169)100 (068 146)

110 (063 191)114 (063 206)112 (074 168)

84470792077

10000

81671833

10000

59634037

10000

55794421

10000

54274573

10000

1 7760129

Study ID OR (95 CI)

Subtotal (I2 = 120 p = 0321)

Subtotal (I2 = 270 p = 0242)

Subtotal (I2 = 00 p = 0763)

Subtotal (I2 = 00 p = 0811)

Subtotal (I2 = 00 p = 0940)








weight

weight




Holla et al (2006)

Keles et al (2006)




Figure 2 Continued

12 Disease Markers

144 (053 393)

072 (039 132)

098 (046 206)

090 (059 137)

100 (057 177)

129 (044 379)

106 (064 175)

079 (018 342)

154 (032 741)

108 (037 313)

069 (017 275)

020 (001 404)

051 (015 172)

1426

5500

3074

10000

8046

1954

10000

6129

3871

10000

6310

3690

10000

124 (078 197)

290 (121 693)

175 (077 394)

080 (036 180)

363 (138 959)

167 (038 731)

5994

4006

10000

5167

4833

10000

Study ID OR (95 CI)

Study ID OR (95 CI)

Subtotal (I2 = 00 p = 0490)

Subtotal (I2 = 00 p = 0691)

Subtotal (I2 = 00 p = 0541)

Subtotal (I2 = 00 p = 0459)

Subtotal (I2 = 646 p = 0093)

Subtotal (I2 = 818 p = 0019)

1 99200101

1 9590104








weight

weight


Holla et al (2004)

Ustun et al (2008)


Cao et al (2006)


Cao et al (2005)

Holla et al (2006)

Keles et al (2006)

Holla et al (2004)

Pirhan et al (2008)

Holla et al (2005)

Gurkan et al (2007)


Figure 2 Continued

Disease Markers 13

228 (077 669)

071 (045 114)

109 (049 241)

091 (062 132)

074 (029 191)

189 (065 551)

114 (056 232)

104 (065 166)

098 (056 172)

102 (071 145)

107 (063 183)

092 (051 166)

100 (067 149)

756

7248

1996

10000

6504

3496

10000

5777

4223

10000

5363

4637

10000


119 (064 222)

386 (127 1176)

196 (063 609)

085 (052 140)

034 (016 074)

057 (023 138)

5805

4195

10000

5539

4461

10000

Study ID OR (95 CI)

Study ID OR (95 CI)

1 6690149

1 11800851

Subtotal (I2 = 499 p = 0136)

Subtotal (I2 = 390 p = 0200)

Subtotal (I2 = 00 p = 0884)

Subtotal (I2 = 00 p = 0699)

Subtotal (I2 = 694 p = 0071)

Subtotal (I2 = 739 p = 0050)







weight

weight


Holla et al (2004)

Ustun et al (2008)


Cao et al (2005)

Holla et al (2004)

Cao et al (2006)


Holla et al (2006)

Keles et al (2006)

Pirhan et al (2008)

Holla et al (2005)

Gurkan et al (2007)


Figure 2 Continued

14 Disease Markers

175 (068 451)

065 (036 115)

100 (049 204)

089 (060 133)

102 (061 172)

124 (044 348)

106 (067 169)

085 (021 346)

167 (036 767)

116 (042 324)

063 (017 239)

010 (001 198)

039 (012 124)

1279

5789

2931

10000

8101

1899

10000

6208

3792

10000

5485

4515

10000

126 (082 195)

362 (159 825)

201 (072 561)

076 (036 162)

368 (151 902)

164 (035 770)

5578

4422

10000

5123

487

10000

Study ID OR (95 CI)

Study ID OR (95 CI)

Subtotal (I2 = 204 p = 0262)

Subtotal (I2 = 382 p = 0198)

Subtotal (I2 = 00 p = 0736)

Subtotal (I2 = 00 p = 0522)

Subtotal (I2 = 797 p = 0027)

Subtotal (I2 = 857 p = 0008)

1 9020111

1 181000554








weight

weight


Holla et al (2004)

Ustun et al (2008)


Cao et al (2006)


Cao et al (2005)

Holla et al (2006)

Keles et al (2006)

Holla et al (2004)

Pirhan et al (2008)

Holla et al (2005)

Gurkan et al (2007)


Figure 2 Continued

Disease Markers 15

106 (045 247)097 (062 150)105 (056 200)100 (072 140)

086 (056 133)062 (029 133)079 (055 116)

110 (053 227)045 (019 105)075 (043 130)

102 (065 159)092 (052 162)098 (069 139)

111 (064 192)110 (061 199)110 (074 165)

090 (054 151)044 (020 095)072 (047 110)

150358312666

10000

72332767

10000

46645336

10000

60823918

10000

53494651

10000

61083892

10000

10189 528

Subtotal (I2 = 00 p = 0969)

Subtotal (I2 = 00 p = 0467)

Subtotal (I2 = 595 p = 0116)

Subtotal (I2 = 00 p = 0778)

Subtotal (I2 = 00 p = 0989)

Subtotal (I2 = 571 p = 0127)

Study ID OR (95 CI)







weight













16 Disease Markers




4 Discussion







Disease Markers 17










18 Disease Markers



Disclosure


Competing Interests




References

















Disease Markers 19




























20 Disease Markers




















Editorial Board




Contents


















2 Disease Markers





References


























1 Introduction




2 Disease Markers










Disease Markers 3








3 Results




4 Disease Markers

Seru

m ac

tive M

MP-

9 (n

gm

L)


500

1000

1500

2000

2500p lt 001

p lt 0001

(a)

OND patients

Seru

m ac

tive M

MP-

2 (n

gm

L)

HC MS patients0

100

200

300

500

600

700

(b)

CSF

activ

e MM

P-9

(ng

mL)


025

050

075

100

200

300

400p lt 001

(c)

CSF

activ

e MM

P-2

(ng

mL)


10

20

30

40

(d)






Disease Markers 5

Seru

m IL

-17

(pg

mL)

MS patients0

200

400

600

800

1000

1500

2000

2500

OND patients

(a)

Seru

m IL

-18

(pg

mL)

0

5000

10000

15000

20000


(b)

Seru

m IL

-23

(pg

mL)

0

2000

4000

6000

800050000

100000

150000


(c)

Seru

m M

CP-3

(pg

mL)

0

50

100

150

200


(d)

CSF

IL-1

7 (p

gm

L)

0

10

20

30

40

50


(e)

CSF

IL-1

8 (p

gm

L)

0

2000

4000

6000

8000


(f)

Figure 2 Continued

6 Disease Markers

CSF

IL-2

3 (p

gm

L)

0

50

100

150

200

2000

4000


(g)

CSF

MCP

-3 (p

gm

L)

0

5

10

15

20

25


(h)








Disease Markers 7




OND () MS ()Serum






4 Discussion






8 Disease Markers






Competing Interests


Acknowledgments


References












Disease Markers 9









































1 Introduction








2 Disease Markers













Disease Markers 3











4 Disease Markers












Disease Markers 5



9 Conclusion




Competing Interests


References













6 Disease Markers































Disease Markers 7







































1 Introduction





2 Disease Markers












3 Results


Disease Markers 3

(a) (b) (c)

(d) (e) (f)






4 Disease Markers













Disease Markers 5



GenderMale 23 (793)









4 Discussion




6 Disease Markers






Competing Interests


Acknowledgments


References







Disease Markers 7



































1 Introduction




2 Disease Markers









Disease Markers 3






3 Results



4 Discussion


4 Disease Markers

Seru

m ac

tive M

MP-2

leve

ls (n

gm

L)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

0

5

10

15

20

(a)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

Seru

m ac

tive M

MP-9

leve

ls (n

gm

L)

0

5

10

15

(b)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

Seru

m T

IMP-

2le

vels

(ng

mL)

0

50

100

150

200

250

(c)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

Seru

m T

IMP-

1le

vels

(ng

mL)

0

200

400

600

800

(d)




Disease Markers 5

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

0

200

400

600

800Se

rum

activ

e MM

P-2

TIM

P-2

(times103)

(a)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

0

10

20

30

40

Seru

m ac

tive M

MP-9

TIM

P-1

(times103)

(b)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

00

05

10

15

20

Seru

m ac

tive M

MP-9

act

ive M

MP-2

1 2 3

4 5

6 7

(c)




6 Disease Markers


Competing Interests


Acknowledgments


References




























Disease Markers 7






















1 Introduction





2 Disease Markers










Disease Markers 3





3 Results





4 Disease Markers

Table1MainCh

aracteris

ticso

fincludedstu

dies

Authoryear

Cou

ntry

Ethn

icity

Samples

ize

(casecontrol)

Type

ofperio

dontitis

Matchingcriteria

Genotype

metho

dGene(po

lymorph

ism)amp

HWEin

controls

deSouzae

tal2003

[31]

Brazil

Caucasian

5037

CP(m

oderateo

rsevere)

Smoker

ratio

sPC

R-RF

LPMMP-1(minus1607

1G2G)0

87

Hollaetal2004

[14]

CzechRe

public

Caucasian

133196

CP(m

ildto

mod

erate

tosevere)

Agegender

PCR-RF

LPMMP-1(minus1607

1G2G)0

52MMP-1(minus519AG

)011

MMP-1(minus422AT)0

47

Itagakietal2004

[22]

Japan

Asian

205142

CP(m

ildor

mod

erate

orsevere)

Agegendersm

oker

ratio

sTaqM

anMMP-1(minus1607

1G2G)0

48

MMP-3(minus117

15A6A)0

87

3714

2AgP

Hollaetal2005

[23]

CzechRe

public

Caucasian

149127

CP(m

ildto

mod

erate

tosevere)

Agesmoker

ratio

sPC

R-RF

LP

MMP-2(minus1575

GA

)040

MMP-2(minus1306

CT)

019

MMP-2(minus790TG)0

67

MMP-2(minus735CT)

042

Caoetal2005

[32]

China

Asian

4052

AgP

mdashPC

R-RF

LPMMP-1(minus1607

1G2G)0

78

Caoetal2006

[13]

China

Asian

6050

CP(m

oderateo

rsevere)

mdashPC

R-RF

LPMMP-1(minus1607

1G2G)0

99

Kelese

tal2006

[12]

Turkey

Caucasian

7070

CP(severe)

Agegender

PCR-RF

LPMMP-9(minus1562

CT)

082

Hollaetal2006

[24]

CzechRe

public

Caucasian

169135

CP(m

oderateo

rsevere)

Agegendersm

oker

ratio

sPC

R-RF

LPMMP-9(minus1562

CT)

059

MMP-9(+279RQ)0

25

Chen

etal2007

[16]

China

Asian

7912

8AgP

Agegender

DHPL

CPC

R-RF

LPMMP-2(minus1306

CT)

100

MMP-9(minus1562

CT)

063

Gurkanetal2007

[8]

Turkey

Caucasian

9215

7AgP

Gender

PCR-RF

LPMMP-2(minus735CT)

045

MMP-9(minus1562

CT)

035

MMP-12

(357

AsnSer)0

06

Gurkanetal2008

[25]

Turkey

Caucasian

8710

7CP

(severe)

mdashPC

R-RF

LPMMP-2(minus735CT)

043

MMP-12

(357

AsnSer)0

47

Pirhan

etal2008

[26]

Turkey

Caucasian

10298

CP(severe)

mdashPC

R-RF

LPMMP-1(minus519AG

)079

Ustu

netal2008

[27]

Turkey

Caucasian

12654

CP(m

oderateo

rsevere)

Age

PCR-RF

LPMMP-1(minus1607

1G2G)0

75

Pirhan

etal2009

[28]

Turkey

Caucasian

10298

CP(severe)

mdashPC

R-RF

LPMMP-13

(minus77

AG

)089

MMP-13

(11A

12A)0

92

Chou

etal2011[11]

China

Asian

3611

06CP

(mod

erateto

severe)

Gendersm

oker

ratio

sPC

R-RF

LPMMP-8(minus799CT)

022

9610

6AgP

Hollaetal2012

[29]

CzechRe

public

Caucasian

3412

78CP

(mild

tomod

erate

tosevere)

Agegendersm

oker

ratio

sPC

R-RF

LPMMP-8(+17

CG)0

14MMP-8(minus799CT)

063

Emingiletal2014

[30]

Turkey

Caucasian

100167

AgP

Gender

PCR-RF

LPMMP-8(+17

CG)0

29

MMP-8(minus799CT)

009

MMP-8(minus381A

G)0

87

CPchron

icperio

dontitisAgP

aggressiveperio

dontitisHWE

Hardy-W

einb

ergequilib

riumM

ildchronicperio

dontitispatie

ntsw

ithteethexhibitin

glt3m

mattachmentlossmod

eratechronicperio

dontitis

patientsw

ithteethexhibitin

gge3m

mandlt7m


chronicp

eriodo

ntitispatie

ntsw

ithteethexhibitin

gge7m


ofHWEwas

considered

significant

Disease Markers 5



Selection






Exposure




6 Disease Markers


Scre

enin

gIn

clude

dEl

igib

ility








Iden

tifica

tion


HWE n = 3







(i) Reviews n = 5








Disease Markers 7


ofthep

olym

orph

ismsinMMPs

gene

onperio

dontitisrisk

MMP-1

minus1607

1G2G

Stud

ies

(casescon

trols)

2Gversus

1GOR(95

CI)

1198682(

)119901ℎ119901119888

2G2Gversus

1G1G

OR(95

CI)

1198682(

)119901ℎ119901119888

1G2Gversus

1G1G

OR(95

CI)

1198682(

)119901ℎ119901119888

2G2Gversus

1G2G

OR(95

CI)

1198682(

)119901ℎ119901119888

1G2G+2G

2Gversus

1G1G

OR(95

CI)

1198682(

)119901ℎ119901119888

2G2Gversus

1G1G

+1G

2GOR(95

CI)

1198682(

)119901ℎ119901119888

1G2Gversus

1G1G

+2G

2G

OR(95

CI)

1198682(

)119901ℎ119901119888

Type

ofdisease

CP Caucasian

3(309287)

102(067ndash15

6)10

5(044ndash

251)

095

(064ndash

142)

090

(059ndash

137)

091

(062ndash13

2)089

(060ndash

133)

100(072ndash14

0)631

006710

0063000671000

120032

110

0000049010

00499

013610

00382019

810

0000096910

00

Asian

2(30219

2)17

5(077ndash395)

279

(056ndash

1398)

131(074ndash232

)17

5(077ndash394)

196(063ndash609)

201

(072ndash561)

079

(055ndash116)

84500111000

81500201000

27002421000

64600931000

69400711000

797

002710

0000046

710

00Ag

P

Asian

2(77194)

134(048ndash373)

154(028ndash855)

091

(042ndash19

6)16

7(038ndash731)

114(056ndash

232)

164(035

ndash770)

075

(043ndash13

0)84800101000

78400311000

00076310

0081800191000

39002001000

857

000810

00595011610

00Severe

CPCa

ucasian

Mild

tomod

erate

2(6691)

099

(063ndash15

5)099

(039

ndash250)

116(052

ndash260)

085

(039

ndash184)

110(051ndash238)

089

(043ndash18

5)117(062ndash221)

00061310

0000061310

0000066710

0000087410

0000061310

0000075110

0000087610

00

Severe

2(11091)

153(072ndash324)

244

(047ndash1259)

152(070ndash

330)

131(068ndash251)

168(081ndash350)

147(080ndash

273)

098

(056ndash

173)

657

008810

0063400981000

15802761000

00036910

00511

015310

00423018

810

0000084510

00Asian

Mild

tomod

erate

2(16819

2)12

6(093ndash17

2)16

2(084ndash

312)

140(072ndash269)

119(075ndash18

7)15

1(082ndash280)

127(083ndash19

5)097

(063ndash14

8)601

0113098

7627

010110

00438018

210

0000053410

00560013

210

0021002611000

00078410

00

Severe

2(97192)

199(092ndash426)

293

(071ndash1203)

116(053

ndash256)

219

(126ndash

379)

178(086ndash

367)

255

(095ndash685)

058

(035

ndash098)

73500520546

67000820952

00046

810

00489

01620035

381

02040854

697

0069044

1000517028

7Sm

okinghabitinCP

Caucasian

Non

smok

ing

3(200213)

092

(055ndash15

5)090

(033

ndash243)

087

(054ndash

140)

085

(052

ndash141)

096

(045ndash205)

082

(052

ndash130)

098

(066ndash

146)

661

005310

00631

006710

0034902151000

00043810

00569

009810

0040

10118

1000

00057510

00

Smok

ing

2(10974)

110(071ndash17

2)114(043ndash302)

112(054ndash

234)

115(050ndash

264

)112(055ndash229)

119(053

ndash265)

098

(053

ndash184)

19002671000

329

022210

0000097610

00522014

810

0000068210

00540014

010

0000031910

00MMP-1

minus519AG

Stud

ies

(casescon

trols)

Gversus

AOR(95

CI)

1198682(

)119901ℎ119901119888

GGversus

AA

OR(95

CI)

1198682(

)119901ℎ119901119888

AGversus

AA

OR(95

CI)

1198682(

)119901ℎ119901119888

GGversus

AGOR(95

CI)

1198682(

)119901ℎ119901119888

AG+GGversus

AA

OR(95

CI)

1198682(

)119901ℎ119901119888

GGversus

AA+AG

OR(95

CI)

1198682(

)119901ℎ119901119888

AGversus

AA+GG

OR(95

CI)

1198682(

)119901ℎ119901119888

Type

ofdisease

CP Caucasian

2(235293)

103(080ndash

132)

108(064ndash

182)

100(068ndash14

6)10

6(064ndash

175)

102(071ndash14

5)10

6(067ndash16

9)098

(069ndash

139)

00

09841000

00

08061000

00

0811

1000

00

06911000

00

08841000

00

07361000

00077810

00MMP-2

minus735CT

Stud

ies

(casescon

trols)

Tversus

COR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

CTversus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CTOR(95

CI)

1198682(

)119901ℎ119901119888

CT+TT

versus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CC+CT

OR(95

CI)

1198682(

)119901ℎ119901119888

CTversus

CC+TT

OR(95

CI)

1198682(

)119901ℎ119901119888

Type

ofdisease

CP Caucasian

2(236234)

111(079ndash155)

119(043ndash337)

112(074

ndash168)

108(037

ndash313

)10

0(067ndash14

9)116(041ndash324)

110(074

ndash165)

00069510

0000051110

0000094010

0000054110

0000069910

0000052210

0000098910

00Sm

okinghabitinCP

Caucasian

Non

smok

ing

2(13319

8)113(074

ndash172)

115(032

ndash409)

117(070ndash

194)

099

(027ndash366)

116(071ndash18

8)110(031ndash389)

115(069ndash

190)

00033710

00452017

710

0000078410

0032402241000

00053310

00424018

810

0000091710

00

8 Disease Markers

Table3Con

tinued

MMP-9

minus1562

CT

Stud

ies

(casescon

trols)

Tversus

COR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

CTversus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CTOR(95

CI)

1198682(

)119901ℎ119901119888

CT+TT

versus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CC+CT

OR(95

CI)

1198682(

)119901ℎ119901119888

CTversus

CC+TT

OR(95

CI)

1198682(

)119901ℎ119901119888

Type

ofdisease

CP Caucasian

2(239205)

056

(024ndash

135)

036

(011ndash112

)063

(029ndash

136)

051

(015

ndash172)

057

(023ndash13

8)039

(012

ndash124)

072

(047ndash110)

78200321000

35902120553

64000961000

00045910

00739

005010

0020402620777

571

0127092

4Severe

CPCa

ucasian

Severe

2(163205)

063

(020ndash

197)

044

(013

ndash149)

071

(024ndash

209)

053

(014

ndash195)

065

(019

ndash215

)046

(014

ndash157)

075

(028ndash201)

861

000710

00544013

910

0079300281000

00042810

0084200121000

410019

310

0076000411000

MMP-1matrix


MP-2matrix


MP-9matrix


Pchronicp

eriodo

ntitisAgP

aggressiveg

eneralized

perio

dontitis



orrected

byBo

nferroni

correctio

nOR

odds

ratio

CIconfi

denceinterval


0the

rand

omeffectsmod

elisusedC

onversely

the

fixed

effectsmod

elisused

119901119888lt001

isconsidered

statisticallysig

nificant

Disease Markers 9


Holla et al (2004)

Ustun et al (2008)


Cao et al (2006)


Cao et al (2005)

Holla et al (2006)

Keles et al (2006)

Study ID

165 (090 303)

075 (055 103)

103 (065 161)

102 (067 156)

119 (087 163)

274 (157 481)

175 (077 395)

080 (048 135)

229 (124 420)

134 (048 373)

084 (055 130)

035 (017 069)

056 (024 135)

OR (95 CI)

2551

4121

3327

10000

5398

4602

10000

5118

4882

10000

5479

4521

10000

Holla et al (2004)

Pirhan et al (2008)

Holla et al (2005)

Study ID

102 (075 140)

103 (067 159)

103 (080 132)

104 (065 165)

119 (073 193)

111 (079 155)

OR (95 CI)

6619

3381

10000

5388

4612

10000

weight

weight

Subtotal (I2 = 00 p = 0695)

Subtotal (I2 = 00 p = 0984)

Subtotal (I2 = 782 p = 0032)

Subtotal (I2 = 848 p = 0010)

Subtotal (I2 = 845 p = 0011)

Subtotal (I2 = 631 p = 0067)

1 5790173

1 1930518








Gurkan et al (2007)


Figure 2 Continued

10 Disease Markers

286 (082 1001)

056 (030 107)

107 (042 273)

105 (044 251)

133 (069 257)

693 (203 2363)

279 (056 1398)

066 (023 188)

379 (114 1258)

154 (028 855)

2539

4150

3312

10000

5509

4491

10000

5148

4852

10000

104 (057 189)

121 (042 350)

108 (064 182)

087 (021 357)

175 (038 818)

120 (043 337)

061 (016 233)

008 (000 157)

036 (011 112)

7713

2287

10000

6270

3730

10000

5250

4750

10000

Study ID OR (95 CI)

Study ID OR (95 CI)

Subtotal (I2 = 359 p = 0212)

Subtotal (I2 = 00 p = 0511)

Subtotal (I2 = 00 p = 0806)

Subtotal (I2 = 784 p = 0031)

Subtotal (I2 = 815 p = 0020)

Subtotal (I2 = 630 p = 0067)

1 23600423

1 23200043








weight

weight


Holla et al (2004)

Ustun et al (2008)


Cao et al (2006)


Cao et al (2005)

Holla et al (2006)

Keles et al (2006)

Holla et al (2004)

Pirhan et al (2008)

Holla et al (2005)

Gurkan et al (2007)


Figure 2 Continued

Disease Markers 11

089 (053 148)

040 (018 087)

063 (029 136)

5713

4287

10000

54910182

Study ID OR (95 CI)

Subtotal (I2 = 640 p = 0096)

198 (063 620)079 (048 129)110 (047 254)095 (064 142)

107 (055 207)239 (074 776)131 (074 232)

082 (030 226)104 (032 337)091 (042 196)

104 (062 172)094 (053 169)100 (068 146)

110 (063 191)114 (063 206)112 (074 168)

84470792077

10000

81671833

10000

59634037

10000

55794421

10000

54274573

10000

1 7760129

Study ID OR (95 CI)

Subtotal (I2 = 120 p = 0321)

Subtotal (I2 = 270 p = 0242)

Subtotal (I2 = 00 p = 0763)

Subtotal (I2 = 00 p = 0811)

Subtotal (I2 = 00 p = 0940)








weight

weight




Holla et al (2006)

Keles et al (2006)




Figure 2 Continued

12 Disease Markers

144 (053 393)

072 (039 132)

098 (046 206)

090 (059 137)

100 (057 177)

129 (044 379)

106 (064 175)

079 (018 342)

154 (032 741)

108 (037 313)

069 (017 275)

020 (001 404)

051 (015 172)

1426

5500

3074

10000

8046

1954

10000

6129

3871

10000

6310

3690

10000

124 (078 197)

290 (121 693)

175 (077 394)

080 (036 180)

363 (138 959)

167 (038 731)

5994

4006

10000

5167

4833

10000

Study ID OR (95 CI)

Study ID OR (95 CI)

Subtotal (I2 = 00 p = 0490)

Subtotal (I2 = 00 p = 0691)

Subtotal (I2 = 00 p = 0541)

Subtotal (I2 = 00 p = 0459)

Subtotal (I2 = 646 p = 0093)

Subtotal (I2 = 818 p = 0019)

1 99200101

1 9590104








weight

weight


Holla et al (2004)

Ustun et al (2008)


Cao et al (2006)


Cao et al (2005)

Holla et al (2006)

Keles et al (2006)

Holla et al (2004)

Pirhan et al (2008)

Holla et al (2005)

Gurkan et al (2007)


Figure 2 Continued

Disease Markers 13

228 (077 669)

071 (045 114)

109 (049 241)

091 (062 132)

074 (029 191)

189 (065 551)

114 (056 232)

104 (065 166)

098 (056 172)

102 (071 145)

107 (063 183)

092 (051 166)

100 (067 149)

756

7248

1996

10000

6504

3496

10000

5777

4223

10000

5363

4637

10000


119 (064 222)

386 (127 1176)

196 (063 609)

085 (052 140)

034 (016 074)

057 (023 138)

5805

4195

10000

5539

4461

10000

Study ID OR (95 CI)

Study ID OR (95 CI)

1 6690149

1 11800851

Subtotal (I2 = 499 p = 0136)

Subtotal (I2 = 390 p = 0200)

Subtotal (I2 = 00 p = 0884)

Subtotal (I2 = 00 p = 0699)

Subtotal (I2 = 694 p = 0071)

Subtotal (I2 = 739 p = 0050)







weight

weight


Holla et al (2004)

Ustun et al (2008)


Cao et al (2005)

Holla et al (2004)

Cao et al (2006)


Holla et al (2006)

Keles et al (2006)

Pirhan et al (2008)

Holla et al (2005)

Gurkan et al (2007)


Figure 2 Continued

14 Disease Markers

175 (068 451)

065 (036 115)

100 (049 204)

089 (060 133)

102 (061 172)

124 (044 348)

106 (067 169)

085 (021 346)

167 (036 767)

116 (042 324)

063 (017 239)

010 (001 198)

039 (012 124)

1279

5789

2931

10000

8101

1899

10000

6208

3792

10000

5485

4515

10000

126 (082 195)

362 (159 825)

201 (072 561)

076 (036 162)

368 (151 902)

164 (035 770)

5578

4422

10000

5123

487

10000

Study ID OR (95 CI)

Study ID OR (95 CI)

Subtotal (I2 = 204 p = 0262)

Subtotal (I2 = 382 p = 0198)

Subtotal (I2 = 00 p = 0736)

Subtotal (I2 = 00 p = 0522)

Subtotal (I2 = 797 p = 0027)

Subtotal (I2 = 857 p = 0008)

1 9020111

1 181000554








weight

weight


Holla et al (2004)

Ustun et al (2008)


Cao et al (2006)


Cao et al (2005)

Holla et al (2006)

Keles et al (2006)

Holla et al (2004)

Pirhan et al (2008)

Holla et al (2005)

Gurkan et al (2007)


Figure 2 Continued

Disease Markers 15

106 (045 247)097 (062 150)105 (056 200)100 (072 140)

086 (056 133)062 (029 133)079 (055 116)

110 (053 227)045 (019 105)075 (043 130)

102 (065 159)092 (052 162)098 (069 139)

111 (064 192)110 (061 199)110 (074 165)

090 (054 151)044 (020 095)072 (047 110)

150358312666

10000

72332767

10000

46645336

10000

60823918

10000

53494651

10000

61083892

10000

10189 528

Subtotal (I2 = 00 p = 0969)

Subtotal (I2 = 00 p = 0467)

Subtotal (I2 = 595 p = 0116)

Subtotal (I2 = 00 p = 0778)

Subtotal (I2 = 00 p = 0989)

Subtotal (I2 = 571 p = 0127)

Study ID OR (95 CI)







weight













16 Disease Markers




4 Discussion







Disease Markers 17










18 Disease Markers



Disclosure


Competing Interests




References

















Disease Markers 19




























20 Disease Markers


















Editorial Board




Contents


















2 Disease Markers





References


























1 Introduction




2 Disease Markers










Disease Markers 3








3 Results




4 Disease Markers

Seru

m ac

tive M

MP-

9 (n

gm

L)


500

1000

1500

2000

2500p lt 001

p lt 0001

(a)

OND patients

Seru

m ac

tive M

MP-

2 (n

gm

L)

HC MS patients0

100

200

300

500

600

700

(b)

CSF

activ

e MM

P-9

(ng

mL)


025

050

075

100

200

300

400p lt 001

(c)

CSF

activ

e MM

P-2

(ng

mL)


10

20

30

40

(d)






Disease Markers 5

Seru

m IL

-17

(pg

mL)

MS patients0

200

400

600

800

1000

1500

2000

2500

OND patients

(a)

Seru

m IL

-18

(pg

mL)

0

5000

10000

15000

20000


(b)

Seru

m IL

-23

(pg

mL)

0

2000

4000

6000

800050000

100000

150000


(c)

Seru

m M

CP-3

(pg

mL)

0

50

100

150

200


(d)

CSF

IL-1

7 (p

gm

L)

0

10

20

30

40

50


(e)

CSF

IL-1

8 (p

gm

L)

0

2000

4000

6000

8000


(f)

Figure 2 Continued

6 Disease Markers

CSF

IL-2

3 (p

gm

L)

0

50

100

150

200

2000

4000


(g)

CSF

MCP

-3 (p

gm

L)

0

5

10

15

20

25


(h)








Disease Markers 7




OND () MS ()Serum






4 Discussion






8 Disease Markers






Competing Interests


Acknowledgments


References












Disease Markers 9









































1 Introduction








2 Disease Markers













Disease Markers 3











4 Disease Markers












Disease Markers 5



9 Conclusion




Competing Interests


References













6 Disease Markers































Disease Markers 7







































1 Introduction





2 Disease Markers












3 Results


Disease Markers 3

(a) (b) (c)

(d) (e) (f)






4 Disease Markers













Disease Markers 5



GenderMale 23 (793)









4 Discussion




6 Disease Markers






Competing Interests


Acknowledgments


References







Disease Markers 7



































1 Introduction




2 Disease Markers









Disease Markers 3






3 Results



4 Discussion


4 Disease Markers

Seru

m ac

tive M

MP-2

leve

ls (n

gm

L)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

0

5

10

15

20

(a)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

Seru

m ac

tive M

MP-9

leve

ls (n

gm

L)

0

5

10

15

(b)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

Seru

m T

IMP-

2le

vels

(ng

mL)

0

50

100

150

200

250

(c)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

Seru

m T

IMP-

1le

vels

(ng

mL)

0

200

400

600

800

(d)




Disease Markers 5

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

0

200

400

600

800Se

rum

activ

e MM

P-2

TIM

P-2

(times103)

(a)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

0

10

20

30

40

Seru

m ac

tive M

MP-9

TIM

P-1

(times103)

(b)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

00

05

10

15

20

Seru

m ac

tive M

MP-9

act

ive M

MP-2

1 2 3

4 5

6 7

(c)




6 Disease Markers


Competing Interests


Acknowledgments


References




























Disease Markers 7






















1 Introduction





2 Disease Markers










Disease Markers 3





3 Results





4 Disease Markers

Table1MainCh

aracteris

ticso

fincludedstu

dies

Authoryear

Cou

ntry

Ethn

icity

Samples

ize

(casecontrol)

Type

ofperio

dontitis

Matchingcriteria

Genotype

metho

dGene(po

lymorph

ism)amp

HWEin

controls

deSouzae

tal2003

[31]

Brazil

Caucasian

5037

CP(m

oderateo

rsevere)

Smoker

ratio

sPC

R-RF

LPMMP-1(minus1607

1G2G)0

87

Hollaetal2004

[14]

CzechRe

public

Caucasian

133196

CP(m

ildto

mod

erate

tosevere)

Agegender

PCR-RF

LPMMP-1(minus1607

1G2G)0

52MMP-1(minus519AG

)011

MMP-1(minus422AT)0

47

Itagakietal2004

[22]

Japan

Asian

205142

CP(m

ildor

mod

erate

orsevere)

Agegendersm

oker

ratio

sTaqM

anMMP-1(minus1607

1G2G)0

48

MMP-3(minus117

15A6A)0

87

3714

2AgP

Hollaetal2005

[23]

CzechRe

public

Caucasian

149127

CP(m

ildto

mod

erate

tosevere)

Agesmoker

ratio

sPC

R-RF

LP

MMP-2(minus1575

GA

)040

MMP-2(minus1306

CT)

019

MMP-2(minus790TG)0

67

MMP-2(minus735CT)

042

Caoetal2005

[32]

China

Asian

4052

AgP

mdashPC

R-RF

LPMMP-1(minus1607

1G2G)0

78

Caoetal2006

[13]

China

Asian

6050

CP(m

oderateo

rsevere)

mdashPC

R-RF

LPMMP-1(minus1607

1G2G)0

99

Kelese

tal2006

[12]

Turkey

Caucasian

7070

CP(severe)

Agegender

PCR-RF

LPMMP-9(minus1562

CT)

082

Hollaetal2006

[24]

CzechRe

public

Caucasian

169135

CP(m

oderateo

rsevere)

Agegendersm

oker

ratio

sPC

R-RF

LPMMP-9(minus1562

CT)

059

MMP-9(+279RQ)0

25

Chen

etal2007

[16]

China

Asian

7912

8AgP

Agegender

DHPL

CPC

R-RF

LPMMP-2(minus1306

CT)

100

MMP-9(minus1562

CT)

063

Gurkanetal2007

[8]

Turkey

Caucasian

9215

7AgP

Gender

PCR-RF

LPMMP-2(minus735CT)

045

MMP-9(minus1562

CT)

035

MMP-12

(357

AsnSer)0

06

Gurkanetal2008

[25]

Turkey

Caucasian

8710

7CP

(severe)

mdashPC

R-RF

LPMMP-2(minus735CT)

043

MMP-12

(357

AsnSer)0

47

Pirhan

etal2008

[26]

Turkey

Caucasian

10298

CP(severe)

mdashPC

R-RF

LPMMP-1(minus519AG

)079

Ustu

netal2008

[27]

Turkey

Caucasian

12654

CP(m

oderateo

rsevere)

Age

PCR-RF

LPMMP-1(minus1607

1G2G)0

75

Pirhan

etal2009

[28]

Turkey

Caucasian

10298

CP(severe)

mdashPC

R-RF

LPMMP-13

(minus77

AG

)089

MMP-13

(11A

12A)0

92

Chou

etal2011[11]

China

Asian

3611

06CP

(mod

erateto

severe)

Gendersm

oker

ratio

sPC

R-RF

LPMMP-8(minus799CT)

022

9610

6AgP

Hollaetal2012

[29]

CzechRe

public

Caucasian

3412

78CP

(mild

tomod

erate

tosevere)

Agegendersm

oker

ratio

sPC

R-RF

LPMMP-8(+17

CG)0

14MMP-8(minus799CT)

063

Emingiletal2014

[30]

Turkey

Caucasian

100167

AgP

Gender

PCR-RF

LPMMP-8(+17

CG)0

29

MMP-8(minus799CT)

009

MMP-8(minus381A

G)0

87

CPchron

icperio

dontitisAgP

aggressiveperio

dontitisHWE

Hardy-W

einb

ergequilib

riumM

ildchronicperio

dontitispatie

ntsw

ithteethexhibitin

glt3m

mattachmentlossmod

eratechronicperio

dontitis

patientsw

ithteethexhibitin

gge3m

mandlt7m


chronicp

eriodo

ntitispatie

ntsw

ithteethexhibitin

gge7m


ofHWEwas

considered

significant

Disease Markers 5



Selection






Exposure




6 Disease Markers


Scre

enin

gIn

clude

dEl

igib

ility








Iden

tifica

tion


HWE n = 3







(i) Reviews n = 5








Disease Markers 7


ofthep

olym

orph

ismsinMMPs

gene

onperio

dontitisrisk

MMP-1

minus1607

1G2G

Stud

ies

(casescon

trols)

2Gversus

1GOR(95

CI)

1198682(

)119901ℎ119901119888

2G2Gversus

1G1G

OR(95

CI)

1198682(

)119901ℎ119901119888

1G2Gversus

1G1G

OR(95

CI)

1198682(

)119901ℎ119901119888

2G2Gversus

1G2G

OR(95

CI)

1198682(

)119901ℎ119901119888

1G2G+2G

2Gversus

1G1G

OR(95

CI)

1198682(

)119901ℎ119901119888

2G2Gversus

1G1G

+1G

2GOR(95

CI)

1198682(

)119901ℎ119901119888

1G2Gversus

1G1G

+2G

2G

OR(95

CI)

1198682(

)119901ℎ119901119888

Type

ofdisease

CP Caucasian

3(309287)

102(067ndash15

6)10

5(044ndash

251)

095

(064ndash

142)

090

(059ndash

137)

091

(062ndash13

2)089

(060ndash

133)

100(072ndash14

0)631

006710

0063000671000

120032

110

0000049010

00499

013610

00382019

810

0000096910

00

Asian

2(30219

2)17

5(077ndash395)

279

(056ndash

1398)

131(074ndash232

)17

5(077ndash394)

196(063ndash609)

201

(072ndash561)

079

(055ndash116)

84500111000

81500201000

27002421000

64600931000

69400711000

797

002710

0000046

710

00Ag

P

Asian

2(77194)

134(048ndash373)

154(028ndash855)

091

(042ndash19

6)16

7(038ndash731)

114(056ndash

232)

164(035

ndash770)

075

(043ndash13

0)84800101000

78400311000

00076310

0081800191000

39002001000

857

000810

00595011610

00Severe

CPCa

ucasian

Mild

tomod

erate

2(6691)

099

(063ndash15

5)099

(039

ndash250)

116(052

ndash260)

085

(039

ndash184)

110(051ndash238)

089

(043ndash18

5)117(062ndash221)

00061310

0000061310

0000066710

0000087410

0000061310

0000075110

0000087610

00

Severe

2(11091)

153(072ndash324)

244

(047ndash1259)

152(070ndash

330)

131(068ndash251)

168(081ndash350)

147(080ndash

273)

098

(056ndash

173)

657

008810

0063400981000

15802761000

00036910

00511

015310

00423018

810

0000084510

00Asian

Mild

tomod

erate

2(16819

2)12

6(093ndash17

2)16

2(084ndash

312)

140(072ndash269)

119(075ndash18

7)15

1(082ndash280)

127(083ndash19

5)097

(063ndash14

8)601

0113098

7627

010110

00438018

210

0000053410

00560013

210

0021002611000

00078410

00

Severe

2(97192)

199(092ndash426)

293

(071ndash1203)

116(053

ndash256)

219

(126ndash

379)

178(086ndash

367)

255

(095ndash685)

058

(035

ndash098)

73500520546

67000820952

00046

810

00489

01620035

381

02040854

697

0069044

1000517028

7Sm

okinghabitinCP

Caucasian

Non

smok

ing

3(200213)

092

(055ndash15

5)090

(033

ndash243)

087

(054ndash

140)

085

(052

ndash141)

096

(045ndash205)

082

(052

ndash130)

098

(066ndash

146)

661

005310

00631

006710

0034902151000

00043810

00569

009810

0040

10118

1000

00057510

00

Smok

ing

2(10974)

110(071ndash17

2)114(043ndash302)

112(054ndash

234)

115(050ndash

264

)112(055ndash229)

119(053

ndash265)

098

(053

ndash184)

19002671000

329

022210

0000097610

00522014

810

0000068210

00540014

010

0000031910

00MMP-1

minus519AG

Stud

ies

(casescon

trols)

Gversus

AOR(95

CI)

1198682(

)119901ℎ119901119888

GGversus

AA

OR(95

CI)

1198682(

)119901ℎ119901119888

AGversus

AA

OR(95

CI)

1198682(

)119901ℎ119901119888

GGversus

AGOR(95

CI)

1198682(

)119901ℎ119901119888

AG+GGversus

AA

OR(95

CI)

1198682(

)119901ℎ119901119888

GGversus

AA+AG

OR(95

CI)

1198682(

)119901ℎ119901119888

AGversus

AA+GG

OR(95

CI)

1198682(

)119901ℎ119901119888

Type

ofdisease

CP Caucasian

2(235293)

103(080ndash

132)

108(064ndash

182)

100(068ndash14

6)10

6(064ndash

175)

102(071ndash14

5)10

6(067ndash16

9)098

(069ndash

139)

00

09841000

00

08061000

00

0811

1000

00

06911000

00

08841000

00

07361000

00077810

00MMP-2

minus735CT

Stud

ies

(casescon

trols)

Tversus

COR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

CTversus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CTOR(95

CI)

1198682(

)119901ℎ119901119888

CT+TT

versus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CC+CT

OR(95

CI)

1198682(

)119901ℎ119901119888

CTversus

CC+TT

OR(95

CI)

1198682(

)119901ℎ119901119888

Type

ofdisease

CP Caucasian

2(236234)

111(079ndash155)

119(043ndash337)

112(074

ndash168)

108(037

ndash313

)10

0(067ndash14

9)116(041ndash324)

110(074

ndash165)

00069510

0000051110

0000094010

0000054110

0000069910

0000052210

0000098910

00Sm

okinghabitinCP

Caucasian

Non

smok

ing

2(13319

8)113(074

ndash172)

115(032

ndash409)

117(070ndash

194)

099

(027ndash366)

116(071ndash18

8)110(031ndash389)

115(069ndash

190)

00033710

00452017

710

0000078410

0032402241000

00053310

00424018

810

0000091710

00

8 Disease Markers

Table3Con

tinued

MMP-9

minus1562

CT

Stud

ies

(casescon

trols)

Tversus

COR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

CTversus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CTOR(95

CI)

1198682(

)119901ℎ119901119888

CT+TT

versus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CC+CT

OR(95

CI)

1198682(

)119901ℎ119901119888

CTversus

CC+TT

OR(95

CI)

1198682(

)119901ℎ119901119888

Type

ofdisease

CP Caucasian

2(239205)

056

(024ndash

135)

036

(011ndash112

)063

(029ndash

136)

051

(015

ndash172)

057

(023ndash13

8)039

(012

ndash124)

072

(047ndash110)

78200321000

35902120553

64000961000

00045910

00739

005010

0020402620777

571

0127092

4Severe

CPCa

ucasian

Severe

2(163205)

063

(020ndash

197)

044

(013

ndash149)

071

(024ndash

209)

053

(014

ndash195)

065

(019

ndash215

)046

(014

ndash157)

075

(028ndash201)

861

000710

00544013

910

0079300281000

00042810

0084200121000

410019

310

0076000411000

MMP-1matrix


MP-2matrix


MP-9matrix


Pchronicp

eriodo

ntitisAgP

aggressiveg

eneralized

perio

dontitis



orrected

byBo

nferroni

correctio

nOR

odds

ratio

CIconfi

denceinterval


0the

rand

omeffectsmod

elisusedC

onversely

the

fixed

effectsmod

elisused

119901119888lt001

isconsidered

statisticallysig

nificant

Disease Markers 9


Holla et al (2004)

Ustun et al (2008)


Cao et al (2006)


Cao et al (2005)

Holla et al (2006)

Keles et al (2006)

Study ID

165 (090 303)

075 (055 103)

103 (065 161)

102 (067 156)

119 (087 163)

274 (157 481)

175 (077 395)

080 (048 135)

229 (124 420)

134 (048 373)

084 (055 130)

035 (017 069)

056 (024 135)

OR (95 CI)

2551

4121

3327

10000

5398

4602

10000

5118

4882

10000

5479

4521

10000

Holla et al (2004)

Pirhan et al (2008)

Holla et al (2005)

Study ID

102 (075 140)

103 (067 159)

103 (080 132)

104 (065 165)

119 (073 193)

111 (079 155)

OR (95 CI)

6619

3381

10000

5388

4612

10000

weight

weight

Subtotal (I2 = 00 p = 0695)

Subtotal (I2 = 00 p = 0984)

Subtotal (I2 = 782 p = 0032)

Subtotal (I2 = 848 p = 0010)

Subtotal (I2 = 845 p = 0011)

Subtotal (I2 = 631 p = 0067)

1 5790173

1 1930518








Gurkan et al (2007)


Figure 2 Continued

10 Disease Markers

286 (082 1001)

056 (030 107)

107 (042 273)

105 (044 251)

133 (069 257)

693 (203 2363)

279 (056 1398)

066 (023 188)

379 (114 1258)

154 (028 855)

2539

4150

3312

10000

5509

4491

10000

5148

4852

10000

104 (057 189)

121 (042 350)

108 (064 182)

087 (021 357)

175 (038 818)

120 (043 337)

061 (016 233)

008 (000 157)

036 (011 112)

7713

2287

10000

6270

3730

10000

5250

4750

10000

Study ID OR (95 CI)

Study ID OR (95 CI)

Subtotal (I2 = 359 p = 0212)

Subtotal (I2 = 00 p = 0511)

Subtotal (I2 = 00 p = 0806)

Subtotal (I2 = 784 p = 0031)

Subtotal (I2 = 815 p = 0020)

Subtotal (I2 = 630 p = 0067)

1 23600423

1 23200043








weight

weight


Holla et al (2004)

Ustun et al (2008)


Cao et al (2006)


Cao et al (2005)

Holla et al (2006)

Keles et al (2006)

Holla et al (2004)

Pirhan et al (2008)

Holla et al (2005)

Gurkan et al (2007)


Figure 2 Continued

Disease Markers 11

089 (053 148)

040 (018 087)

063 (029 136)

5713

4287

10000

54910182

Study ID OR (95 CI)

Subtotal (I2 = 640 p = 0096)

198 (063 620)079 (048 129)110 (047 254)095 (064 142)

107 (055 207)239 (074 776)131 (074 232)

082 (030 226)104 (032 337)091 (042 196)

104 (062 172)094 (053 169)100 (068 146)

110 (063 191)114 (063 206)112 (074 168)

84470792077

10000

81671833

10000

59634037

10000

55794421

10000

54274573

10000

1 7760129

Study ID OR (95 CI)

Subtotal (I2 = 120 p = 0321)

Subtotal (I2 = 270 p = 0242)

Subtotal (I2 = 00 p = 0763)

Subtotal (I2 = 00 p = 0811)

Subtotal (I2 = 00 p = 0940)








weight

weight




Holla et al (2006)

Keles et al (2006)




Figure 2 Continued

12 Disease Markers

144 (053 393)

072 (039 132)

098 (046 206)

090 (059 137)

100 (057 177)

129 (044 379)

106 (064 175)

079 (018 342)

154 (032 741)

108 (037 313)

069 (017 275)

020 (001 404)

051 (015 172)

1426

5500

3074

10000

8046

1954

10000

6129

3871

10000

6310

3690

10000

124 (078 197)

290 (121 693)

175 (077 394)

080 (036 180)

363 (138 959)

167 (038 731)

5994

4006

10000

5167

4833

10000

Study ID OR (95 CI)

Study ID OR (95 CI)

Subtotal (I2 = 00 p = 0490)

Subtotal (I2 = 00 p = 0691)

Subtotal (I2 = 00 p = 0541)

Subtotal (I2 = 00 p = 0459)

Subtotal (I2 = 646 p = 0093)

Subtotal (I2 = 818 p = 0019)

1 99200101

1 9590104








weight

weight


Holla et al (2004)

Ustun et al (2008)


Cao et al (2006)


Cao et al (2005)

Holla et al (2006)

Keles et al (2006)

Holla et al (2004)

Pirhan et al (2008)

Holla et al (2005)

Gurkan et al (2007)


Figure 2 Continued

Disease Markers 13

228 (077 669)

071 (045 114)

109 (049 241)

091 (062 132)

074 (029 191)

189 (065 551)

114 (056 232)

104 (065 166)

098 (056 172)

102 (071 145)

107 (063 183)

092 (051 166)

100 (067 149)

756

7248

1996

10000

6504

3496

10000

5777

4223

10000

5363

4637

10000


119 (064 222)

386 (127 1176)

196 (063 609)

085 (052 140)

034 (016 074)

057 (023 138)

5805

4195

10000

5539

4461

10000

Study ID OR (95 CI)

Study ID OR (95 CI)

1 6690149

1 11800851

Subtotal (I2 = 499 p = 0136)

Subtotal (I2 = 390 p = 0200)

Subtotal (I2 = 00 p = 0884)

Subtotal (I2 = 00 p = 0699)

Subtotal (I2 = 694 p = 0071)

Subtotal (I2 = 739 p = 0050)







weight

weight


Holla et al (2004)

Ustun et al (2008)


Cao et al (2005)

Holla et al (2004)

Cao et al (2006)


Holla et al (2006)

Keles et al (2006)

Pirhan et al (2008)

Holla et al (2005)

Gurkan et al (2007)


Figure 2 Continued

14 Disease Markers

175 (068 451)

065 (036 115)

100 (049 204)

089 (060 133)

102 (061 172)

124 (044 348)

106 (067 169)

085 (021 346)

167 (036 767)

116 (042 324)

063 (017 239)

010 (001 198)

039 (012 124)

1279

5789

2931

10000

8101

1899

10000

6208

3792

10000

5485

4515

10000

126 (082 195)

362 (159 825)

201 (072 561)

076 (036 162)

368 (151 902)

164 (035 770)

5578

4422

10000

5123

487

10000

Study ID OR (95 CI)

Study ID OR (95 CI)

Subtotal (I2 = 204 p = 0262)

Subtotal (I2 = 382 p = 0198)

Subtotal (I2 = 00 p = 0736)

Subtotal (I2 = 00 p = 0522)

Subtotal (I2 = 797 p = 0027)

Subtotal (I2 = 857 p = 0008)

1 9020111

1 181000554








weight

weight


Holla et al (2004)

Ustun et al (2008)


Cao et al (2006)


Cao et al (2005)

Holla et al (2006)

Keles et al (2006)

Holla et al (2004)

Pirhan et al (2008)

Holla et al (2005)

Gurkan et al (2007)


Figure 2 Continued

Disease Markers 15

106 (045 247)097 (062 150)105 (056 200)100 (072 140)

086 (056 133)062 (029 133)079 (055 116)

110 (053 227)045 (019 105)075 (043 130)

102 (065 159)092 (052 162)098 (069 139)

111 (064 192)110 (061 199)110 (074 165)

090 (054 151)044 (020 095)072 (047 110)

150358312666

10000

72332767

10000

46645336

10000

60823918

10000

53494651

10000

61083892

10000

10189 528

Subtotal (I2 = 00 p = 0969)

Subtotal (I2 = 00 p = 0467)

Subtotal (I2 = 595 p = 0116)

Subtotal (I2 = 00 p = 0778)

Subtotal (I2 = 00 p = 0989)

Subtotal (I2 = 571 p = 0127)

Study ID OR (95 CI)







weight













16 Disease Markers




4 Discussion







Disease Markers 17










18 Disease Markers



Disclosure


Competing Interests




References

















Disease Markers 19




























20 Disease Markers


















Contents


















2 Disease Markers





References


























1 Introduction




2 Disease Markers










Disease Markers 3








3 Results




4 Disease Markers

Seru

m ac

tive M

MP-

9 (n

gm

L)


500

1000

1500

2000

2500p lt 001

p lt 0001

(a)

OND patients

Seru

m ac

tive M

MP-

2 (n

gm

L)

HC MS patients0

100

200

300

500

600

700

(b)

CSF

activ

e MM

P-9

(ng

mL)


025

050

075

100

200

300

400p lt 001

(c)

CSF

activ

e MM

P-2

(ng

mL)


10

20

30

40

(d)






Disease Markers 5

Seru

m IL

-17

(pg

mL)

MS patients0

200

400

600

800

1000

1500

2000

2500

OND patients

(a)

Seru

m IL

-18

(pg

mL)

0

5000

10000

15000

20000


(b)

Seru

m IL

-23

(pg

mL)

0

2000

4000

6000

800050000

100000

150000


(c)

Seru

m M

CP-3

(pg

mL)

0

50

100

150

200


(d)

CSF

IL-1

7 (p

gm

L)

0

10

20

30

40

50


(e)

CSF

IL-1

8 (p

gm

L)

0

2000

4000

6000

8000


(f)

Figure 2 Continued

6 Disease Markers

CSF

IL-2

3 (p

gm

L)

0

50

100

150

200

2000

4000


(g)

CSF

MCP

-3 (p

gm

L)

0

5

10

15

20

25


(h)








Disease Markers 7




OND () MS ()Serum






4 Discussion






8 Disease Markers






Competing Interests


Acknowledgments


References












Disease Markers 9









































1 Introduction








2 Disease Markers













Disease Markers 3











4 Disease Markers












Disease Markers 5



9 Conclusion




Competing Interests


References













6 Disease Markers































Disease Markers 7







































1 Introduction





2 Disease Markers












3 Results


Disease Markers 3

(a) (b) (c)

(d) (e) (f)






4 Disease Markers













Disease Markers 5



GenderMale 23 (793)









4 Discussion




6 Disease Markers






Competing Interests


Acknowledgments


References







Disease Markers 7



































1 Introduction




2 Disease Markers









Disease Markers 3






3 Results



4 Discussion


4 Disease Markers

Seru

m ac

tive M

MP-2

leve

ls (n

gm

L)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

0

5

10

15

20

(a)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

Seru

m ac

tive M

MP-9

leve

ls (n

gm

L)

0

5

10

15

(b)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

Seru

m T

IMP-

2le

vels

(ng

mL)

0

50

100

150

200

250

(c)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

Seru

m T

IMP-

1le

vels

(ng

mL)

0

200

400

600

800

(d)




Disease Markers 5

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

0

200

400

600

800Se

rum

activ

e MM

P-2

TIM

P-2

(times103)

(a)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

0

10

20

30

40

Seru

m ac

tive M

MP-9

TIM

P-1

(times103)

(b)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

00

05

10

15

20

Seru

m ac

tive M

MP-9

act

ive M

MP-2

1 2 3

4 5

6 7

(c)




6 Disease Markers


Competing Interests


Acknowledgments


References




























Disease Markers 7






















1 Introduction





2 Disease Markers










Disease Markers 3





3 Results





4 Disease Markers

Table1MainCh

aracteris

ticso

fincludedstu

dies

Authoryear

Cou

ntry

Ethn

icity

Samples

ize

(casecontrol)

Type

ofperio

dontitis

Matchingcriteria

Genotype

metho

dGene(po

lymorph

ism)amp

HWEin

controls

deSouzae

tal2003

[31]

Brazil

Caucasian

5037

CP(m

oderateo

rsevere)

Smoker

ratio

sPC

R-RF

LPMMP-1(minus1607

1G2G)0

87

Hollaetal2004

[14]

CzechRe

public

Caucasian

133196

CP(m

ildto

mod

erate

tosevere)

Agegender

PCR-RF

LPMMP-1(minus1607

1G2G)0

52MMP-1(minus519AG

)011

MMP-1(minus422AT)0

47

Itagakietal2004

[22]

Japan

Asian

205142

CP(m

ildor

mod

erate

orsevere)

Agegendersm

oker

ratio

sTaqM

anMMP-1(minus1607

1G2G)0

48

MMP-3(minus117

15A6A)0

87

3714

2AgP

Hollaetal2005

[23]

CzechRe

public

Caucasian

149127

CP(m

ildto

mod

erate

tosevere)

Agesmoker

ratio

sPC

R-RF

LP

MMP-2(minus1575

GA

)040

MMP-2(minus1306

CT)

019

MMP-2(minus790TG)0

67

MMP-2(minus735CT)

042

Caoetal2005

[32]

China

Asian

4052

AgP

mdashPC

R-RF

LPMMP-1(minus1607

1G2G)0

78

Caoetal2006

[13]

China

Asian

6050

CP(m

oderateo

rsevere)

mdashPC

R-RF

LPMMP-1(minus1607

1G2G)0

99

Kelese

tal2006

[12]

Turkey

Caucasian

7070

CP(severe)

Agegender

PCR-RF

LPMMP-9(minus1562

CT)

082

Hollaetal2006

[24]

CzechRe

public

Caucasian

169135

CP(m

oderateo

rsevere)

Agegendersm

oker

ratio

sPC

R-RF

LPMMP-9(minus1562

CT)

059

MMP-9(+279RQ)0

25

Chen

etal2007

[16]

China

Asian

7912

8AgP

Agegender

DHPL

CPC

R-RF

LPMMP-2(minus1306

CT)

100

MMP-9(minus1562

CT)

063

Gurkanetal2007

[8]

Turkey

Caucasian

9215

7AgP

Gender

PCR-RF

LPMMP-2(minus735CT)

045

MMP-9(minus1562

CT)

035

MMP-12

(357

AsnSer)0

06

Gurkanetal2008

[25]

Turkey

Caucasian

8710

7CP

(severe)

mdashPC

R-RF

LPMMP-2(minus735CT)

043

MMP-12

(357

AsnSer)0

47

Pirhan

etal2008

[26]

Turkey

Caucasian

10298

CP(severe)

mdashPC

R-RF

LPMMP-1(minus519AG

)079

Ustu

netal2008

[27]

Turkey

Caucasian

12654

CP(m

oderateo

rsevere)

Age

PCR-RF

LPMMP-1(minus1607

1G2G)0

75

Pirhan

etal2009

[28]

Turkey

Caucasian

10298

CP(severe)

mdashPC

R-RF

LPMMP-13

(minus77

AG

)089

MMP-13

(11A

12A)0

92

Chou

etal2011[11]

China

Asian

3611

06CP

(mod

erateto

severe)

Gendersm

oker

ratio

sPC

R-RF

LPMMP-8(minus799CT)

022

9610

6AgP

Hollaetal2012

[29]

CzechRe

public

Caucasian

3412

78CP

(mild

tomod

erate

tosevere)

Agegendersm

oker

ratio

sPC

R-RF

LPMMP-8(+17

CG)0

14MMP-8(minus799CT)

063

Emingiletal2014

[30]

Turkey

Caucasian

100167

AgP

Gender

PCR-RF

LPMMP-8(+17

CG)0

29

MMP-8(minus799CT)

009

MMP-8(minus381A

G)0

87

CPchron

icperio

dontitisAgP

aggressiveperio

dontitisHWE

Hardy-W

einb

ergequilib

riumM

ildchronicperio

dontitispatie

ntsw

ithteethexhibitin

glt3m

mattachmentlossmod

eratechronicperio

dontitis

patientsw

ithteethexhibitin

gge3m

mandlt7m


chronicp

eriodo

ntitispatie

ntsw

ithteethexhibitin

gge7m


ofHWEwas

considered

significant

Disease Markers 5



Selection






Exposure




6 Disease Markers


Scre

enin

gIn

clude

dEl

igib

ility








Iden

tifica

tion


HWE n = 3







(i) Reviews n = 5








Disease Markers 7


ofthep

olym

orph

ismsinMMPs

gene

onperio

dontitisrisk

MMP-1

minus1607

1G2G

Stud

ies

(casescon

trols)

2Gversus

1GOR(95

CI)

1198682(

)119901ℎ119901119888

2G2Gversus

1G1G

OR(95

CI)

1198682(

)119901ℎ119901119888

1G2Gversus

1G1G

OR(95

CI)

1198682(

)119901ℎ119901119888

2G2Gversus

1G2G

OR(95

CI)

1198682(

)119901ℎ119901119888

1G2G+2G

2Gversus

1G1G

OR(95

CI)

1198682(

)119901ℎ119901119888

2G2Gversus

1G1G

+1G

2GOR(95

CI)

1198682(

)119901ℎ119901119888

1G2Gversus

1G1G

+2G

2G

OR(95

CI)

1198682(

)119901ℎ119901119888

Type

ofdisease

CP Caucasian

3(309287)

102(067ndash15

6)10

5(044ndash

251)

095

(064ndash

142)

090

(059ndash

137)

091

(062ndash13

2)089

(060ndash

133)

100(072ndash14

0)631

006710

0063000671000

120032

110

0000049010

00499

013610

00382019

810

0000096910

00

Asian

2(30219

2)17

5(077ndash395)

279

(056ndash

1398)

131(074ndash232

)17

5(077ndash394)

196(063ndash609)

201

(072ndash561)

079

(055ndash116)

84500111000

81500201000

27002421000

64600931000

69400711000

797

002710

0000046

710

00Ag

P

Asian

2(77194)

134(048ndash373)

154(028ndash855)

091

(042ndash19

6)16

7(038ndash731)

114(056ndash

232)

164(035

ndash770)

075

(043ndash13

0)84800101000

78400311000

00076310

0081800191000

39002001000

857

000810

00595011610

00Severe

CPCa

ucasian

Mild

tomod

erate

2(6691)

099

(063ndash15

5)099

(039

ndash250)

116(052

ndash260)

085

(039

ndash184)

110(051ndash238)

089

(043ndash18

5)117(062ndash221)

00061310

0000061310

0000066710

0000087410

0000061310

0000075110

0000087610

00

Severe

2(11091)

153(072ndash324)

244

(047ndash1259)

152(070ndash

330)

131(068ndash251)

168(081ndash350)

147(080ndash

273)

098

(056ndash

173)

657

008810

0063400981000

15802761000

00036910

00511

015310

00423018

810

0000084510

00Asian

Mild

tomod

erate

2(16819

2)12

6(093ndash17

2)16

2(084ndash

312)

140(072ndash269)

119(075ndash18

7)15

1(082ndash280)

127(083ndash19

5)097

(063ndash14

8)601

0113098

7627

010110

00438018

210

0000053410

00560013

210

0021002611000

00078410

00

Severe

2(97192)

199(092ndash426)

293

(071ndash1203)

116(053

ndash256)

219

(126ndash

379)

178(086ndash

367)

255

(095ndash685)

058

(035

ndash098)

73500520546

67000820952

00046

810

00489

01620035

381

02040854

697

0069044

1000517028

7Sm

okinghabitinCP

Caucasian

Non

smok

ing

3(200213)

092

(055ndash15

5)090

(033

ndash243)

087

(054ndash

140)

085

(052

ndash141)

096

(045ndash205)

082

(052

ndash130)

098

(066ndash

146)

661

005310

00631

006710

0034902151000

00043810

00569

009810

0040

10118

1000

00057510

00

Smok

ing

2(10974)

110(071ndash17

2)114(043ndash302)

112(054ndash

234)

115(050ndash

264

)112(055ndash229)

119(053

ndash265)

098

(053

ndash184)

19002671000

329

022210

0000097610

00522014

810

0000068210

00540014

010

0000031910

00MMP-1

minus519AG

Stud

ies

(casescon

trols)

Gversus

AOR(95

CI)

1198682(

)119901ℎ119901119888

GGversus

AA

OR(95

CI)

1198682(

)119901ℎ119901119888

AGversus

AA

OR(95

CI)

1198682(

)119901ℎ119901119888

GGversus

AGOR(95

CI)

1198682(

)119901ℎ119901119888

AG+GGversus

AA

OR(95

CI)

1198682(

)119901ℎ119901119888

GGversus

AA+AG

OR(95

CI)

1198682(

)119901ℎ119901119888

AGversus

AA+GG

OR(95

CI)

1198682(

)119901ℎ119901119888

Type

ofdisease

CP Caucasian

2(235293)

103(080ndash

132)

108(064ndash

182)

100(068ndash14

6)10

6(064ndash

175)

102(071ndash14

5)10

6(067ndash16

9)098

(069ndash

139)

00

09841000

00

08061000

00

0811

1000

00

06911000

00

08841000

00

07361000

00077810

00MMP-2

minus735CT

Stud

ies

(casescon

trols)

Tversus

COR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

CTversus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CTOR(95

CI)

1198682(

)119901ℎ119901119888

CT+TT

versus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CC+CT

OR(95

CI)

1198682(

)119901ℎ119901119888

CTversus

CC+TT

OR(95

CI)

1198682(

)119901ℎ119901119888

Type

ofdisease

CP Caucasian

2(236234)

111(079ndash155)

119(043ndash337)

112(074

ndash168)

108(037

ndash313

)10

0(067ndash14

9)116(041ndash324)

110(074

ndash165)

00069510

0000051110

0000094010

0000054110

0000069910

0000052210

0000098910

00Sm

okinghabitinCP

Caucasian

Non

smok

ing

2(13319

8)113(074

ndash172)

115(032

ndash409)

117(070ndash

194)

099

(027ndash366)

116(071ndash18

8)110(031ndash389)

115(069ndash

190)

00033710

00452017

710

0000078410

0032402241000

00053310

00424018

810

0000091710

00

8 Disease Markers

Table3Con

tinued

MMP-9

minus1562

CT

Stud

ies

(casescon

trols)

Tversus

COR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

CTversus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CTOR(95

CI)

1198682(

)119901ℎ119901119888

CT+TT

versus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CC+CT

OR(95

CI)

1198682(

)119901ℎ119901119888

CTversus

CC+TT

OR(95

CI)

1198682(

)119901ℎ119901119888

Type

ofdisease

CP Caucasian

2(239205)

056

(024ndash

135)

036

(011ndash112

)063

(029ndash

136)

051

(015

ndash172)

057

(023ndash13

8)039

(012

ndash124)

072

(047ndash110)

78200321000

35902120553

64000961000

00045910

00739

005010

0020402620777

571

0127092

4Severe

CPCa

ucasian

Severe

2(163205)

063

(020ndash

197)

044

(013

ndash149)

071

(024ndash

209)

053

(014

ndash195)

065

(019

ndash215

)046

(014

ndash157)

075

(028ndash201)

861

000710

00544013

910

0079300281000

00042810

0084200121000

410019

310

0076000411000

MMP-1matrix


MP-2matrix


MP-9matrix


Pchronicp

eriodo

ntitisAgP

aggressiveg

eneralized

perio

dontitis



orrected

byBo

nferroni

correctio

nOR

odds

ratio

CIconfi

denceinterval


0the

rand

omeffectsmod

elisusedC

onversely

the

fixed

effectsmod

elisused

119901119888lt001

isconsidered

statisticallysig

nificant

Disease Markers 9


Holla et al (2004)

Ustun et al (2008)


Cao et al (2006)


Cao et al (2005)

Holla et al (2006)

Keles et al (2006)

Study ID

165 (090 303)

075 (055 103)

103 (065 161)

102 (067 156)

119 (087 163)

274 (157 481)

175 (077 395)

080 (048 135)

229 (124 420)

134 (048 373)

084 (055 130)

035 (017 069)

056 (024 135)

OR (95 CI)

2551

4121

3327

10000

5398

4602

10000

5118

4882

10000

5479

4521

10000

Holla et al (2004)

Pirhan et al (2008)

Holla et al (2005)

Study ID

102 (075 140)

103 (067 159)

103 (080 132)

104 (065 165)

119 (073 193)

111 (079 155)

OR (95 CI)

6619

3381

10000

5388

4612

10000

weight

weight

Subtotal (I2 = 00 p = 0695)

Subtotal (I2 = 00 p = 0984)

Subtotal (I2 = 782 p = 0032)

Subtotal (I2 = 848 p = 0010)

Subtotal (I2 = 845 p = 0011)

Subtotal (I2 = 631 p = 0067)

1 5790173

1 1930518








Gurkan et al (2007)


Figure 2 Continued

10 Disease Markers

286 (082 1001)

056 (030 107)

107 (042 273)

105 (044 251)

133 (069 257)

693 (203 2363)

279 (056 1398)

066 (023 188)

379 (114 1258)

154 (028 855)

2539

4150

3312

10000

5509

4491

10000

5148

4852

10000

104 (057 189)

121 (042 350)

108 (064 182)

087 (021 357)

175 (038 818)

120 (043 337)

061 (016 233)

008 (000 157)

036 (011 112)

7713

2287

10000

6270

3730

10000

5250

4750

10000

Study ID OR (95 CI)

Study ID OR (95 CI)

Subtotal (I2 = 359 p = 0212)

Subtotal (I2 = 00 p = 0511)

Subtotal (I2 = 00 p = 0806)

Subtotal (I2 = 784 p = 0031)

Subtotal (I2 = 815 p = 0020)

Subtotal (I2 = 630 p = 0067)

1 23600423

1 23200043








weight

weight


Holla et al (2004)

Ustun et al (2008)


Cao et al (2006)


Cao et al (2005)

Holla et al (2006)

Keles et al (2006)

Holla et al (2004)

Pirhan et al (2008)

Holla et al (2005)

Gurkan et al (2007)


Figure 2 Continued

Disease Markers 11

089 (053 148)

040 (018 087)

063 (029 136)

5713

4287

10000

54910182

Study ID OR (95 CI)

Subtotal (I2 = 640 p = 0096)

198 (063 620)079 (048 129)110 (047 254)095 (064 142)

107 (055 207)239 (074 776)131 (074 232)

082 (030 226)104 (032 337)091 (042 196)

104 (062 172)094 (053 169)100 (068 146)

110 (063 191)114 (063 206)112 (074 168)

84470792077

10000

81671833

10000

59634037

10000

55794421

10000

54274573

10000

1 7760129

Study ID OR (95 CI)

Subtotal (I2 = 120 p = 0321)

Subtotal (I2 = 270 p = 0242)

Subtotal (I2 = 00 p = 0763)

Subtotal (I2 = 00 p = 0811)

Subtotal (I2 = 00 p = 0940)








weight

weight




Holla et al (2006)

Keles et al (2006)




Figure 2 Continued

12 Disease Markers

144 (053 393)

072 (039 132)

098 (046 206)

090 (059 137)

100 (057 177)

129 (044 379)

106 (064 175)

079 (018 342)

154 (032 741)

108 (037 313)

069 (017 275)

020 (001 404)

051 (015 172)

1426

5500

3074

10000

8046

1954

10000

6129

3871

10000

6310

3690

10000

124 (078 197)

290 (121 693)

175 (077 394)

080 (036 180)

363 (138 959)

167 (038 731)

5994

4006

10000

5167

4833

10000

Study ID OR (95 CI)

Study ID OR (95 CI)

Subtotal (I2 = 00 p = 0490)

Subtotal (I2 = 00 p = 0691)

Subtotal (I2 = 00 p = 0541)

Subtotal (I2 = 00 p = 0459)

Subtotal (I2 = 646 p = 0093)

Subtotal (I2 = 818 p = 0019)

1 99200101

1 9590104








weight

weight


Holla et al (2004)

Ustun et al (2008)


Cao et al (2006)


Cao et al (2005)

Holla et al (2006)

Keles et al (2006)

Holla et al (2004)

Pirhan et al (2008)

Holla et al (2005)

Gurkan et al (2007)


Figure 2 Continued

Disease Markers 13

228 (077 669)

071 (045 114)

109 (049 241)

091 (062 132)

074 (029 191)

189 (065 551)

114 (056 232)

104 (065 166)

098 (056 172)

102 (071 145)

107 (063 183)

092 (051 166)

100 (067 149)

756

7248

1996

10000

6504

3496

10000

5777

4223

10000

5363

4637

10000


119 (064 222)

386 (127 1176)

196 (063 609)

085 (052 140)

034 (016 074)

057 (023 138)

5805

4195

10000

5539

4461

10000

Study ID OR (95 CI)

Study ID OR (95 CI)

1 6690149

1 11800851

Subtotal (I2 = 499 p = 0136)

Subtotal (I2 = 390 p = 0200)

Subtotal (I2 = 00 p = 0884)

Subtotal (I2 = 00 p = 0699)

Subtotal (I2 = 694 p = 0071)

Subtotal (I2 = 739 p = 0050)







weight

weight


Holla et al (2004)

Ustun et al (2008)


Cao et al (2005)

Holla et al (2004)

Cao et al (2006)


Holla et al (2006)

Keles et al (2006)

Pirhan et al (2008)

Holla et al (2005)

Gurkan et al (2007)


Figure 2 Continued

14 Disease Markers

175 (068 451)

065 (036 115)

100 (049 204)

089 (060 133)

102 (061 172)

124 (044 348)

106 (067 169)

085 (021 346)

167 (036 767)

116 (042 324)

063 (017 239)

010 (001 198)

039 (012 124)

1279

5789

2931

10000

8101

1899

10000

6208

3792

10000

5485

4515

10000

126 (082 195)

362 (159 825)

201 (072 561)

076 (036 162)

368 (151 902)

164 (035 770)

5578

4422

10000

5123

487

10000

Study ID OR (95 CI)

Study ID OR (95 CI)

Subtotal (I2 = 204 p = 0262)

Subtotal (I2 = 382 p = 0198)

Subtotal (I2 = 00 p = 0736)

Subtotal (I2 = 00 p = 0522)

Subtotal (I2 = 797 p = 0027)

Subtotal (I2 = 857 p = 0008)

1 9020111

1 181000554








weight

weight


Holla et al (2004)

Ustun et al (2008)


Cao et al (2006)


Cao et al (2005)

Holla et al (2006)

Keles et al (2006)

Holla et al (2004)

Pirhan et al (2008)

Holla et al (2005)

Gurkan et al (2007)


Figure 2 Continued

Disease Markers 15

106 (045 247)097 (062 150)105 (056 200)100 (072 140)

086 (056 133)062 (029 133)079 (055 116)

110 (053 227)045 (019 105)075 (043 130)

102 (065 159)092 (052 162)098 (069 139)

111 (064 192)110 (061 199)110 (074 165)

090 (054 151)044 (020 095)072 (047 110)

150358312666

10000

72332767

10000

46645336

10000

60823918

10000

53494651

10000

61083892

10000

10189 528

Subtotal (I2 = 00 p = 0969)

Subtotal (I2 = 00 p = 0467)

Subtotal (I2 = 595 p = 0116)

Subtotal (I2 = 00 p = 0778)

Subtotal (I2 = 00 p = 0989)

Subtotal (I2 = 571 p = 0127)

Study ID OR (95 CI)







weight













16 Disease Markers




4 Discussion







Disease Markers 17










18 Disease Markers



Disclosure


Competing Interests




References

















Disease Markers 19




























20 Disease Markers





























2 Disease Markers





References


























1 Introduction




2 Disease Markers










Disease Markers 3








3 Results




4 Disease Markers

Seru

m ac

tive M

MP-

9 (n

gm

L)


500

1000

1500

2000

2500p lt 001

p lt 0001

(a)

OND patients

Seru

m ac

tive M

MP-

2 (n

gm

L)

HC MS patients0

100

200

300

500

600

700

(b)

CSF

activ

e MM

P-9

(ng

mL)


025

050

075

100

200

300

400p lt 001

(c)

CSF

activ

e MM

P-2

(ng

mL)


10

20

30

40

(d)






Disease Markers 5

Seru

m IL

-17

(pg

mL)

MS patients0

200

400

600

800

1000

1500

2000

2500

OND patients

(a)

Seru

m IL

-18

(pg

mL)

0

5000

10000

15000

20000


(b)

Seru

m IL

-23

(pg

mL)

0

2000

4000

6000

800050000

100000

150000


(c)

Seru

m M

CP-3

(pg

mL)

0

50

100

150

200


(d)

CSF

IL-1

7 (p

gm

L)

0

10

20

30

40

50


(e)

CSF

IL-1

8 (p

gm

L)

0

2000

4000

6000

8000


(f)

Figure 2 Continued

6 Disease Markers

CSF

IL-2

3 (p

gm

L)

0

50

100

150

200

2000

4000


(g)

CSF

MCP

-3 (p

gm

L)

0

5

10

15

20

25


(h)








Disease Markers 7




OND () MS ()Serum






4 Discussion






8 Disease Markers






Competing Interests


Acknowledgments


References












Disease Markers 9









































1 Introduction








2 Disease Markers













Disease Markers 3











4 Disease Markers












Disease Markers 5



9 Conclusion




Competing Interests


References













6 Disease Markers































Disease Markers 7







































1 Introduction





2 Disease Markers












3 Results


Disease Markers 3

(a) (b) (c)

(d) (e) (f)






4 Disease Markers













Disease Markers 5



GenderMale 23 (793)









4 Discussion




6 Disease Markers






Competing Interests


Acknowledgments


References







Disease Markers 7



































1 Introduction




2 Disease Markers









Disease Markers 3






3 Results



4 Discussion


4 Disease Markers

Seru

m ac

tive M

MP-2

leve

ls (n

gm

L)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

0

5

10

15

20

(a)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

Seru

m ac

tive M

MP-9

leve

ls (n

gm

L)

0

5

10

15

(b)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

Seru

m T

IMP-

2le

vels

(ng

mL)

0

50

100

150

200

250

(c)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

Seru

m T

IMP-

1le

vels

(ng

mL)

0

200

400

600

800

(d)




Disease Markers 5

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

0

200

400

600

800Se

rum

activ

e MM

P-2

TIM

P-2

(times103)

(a)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

0

10

20

30

40

Seru

m ac

tive M

MP-9

TIM

P-1

(times103)

(b)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

00

05

10

15

20

Seru

m ac

tive M

MP-9

act

ive M

MP-2

1 2 3

4 5

6 7

(c)




6 Disease Markers


Competing Interests


Acknowledgments


References




























Disease Markers 7






















1 Introduction





2 Disease Markers










Disease Markers 3





3 Results





4 Disease Markers

Table1MainCh

aracteris

ticso

fincludedstu

dies

Authoryear

Cou

ntry

Ethn

icity

Samples

ize

(casecontrol)

Type

ofperio

dontitis

Matchingcriteria

Genotype

metho

dGene(po

lymorph

ism)amp

HWEin

controls

deSouzae

tal2003

[31]

Brazil

Caucasian

5037

CP(m

oderateo

rsevere)

Smoker

ratio

sPC

R-RF

LPMMP-1(minus1607

1G2G)0

87

Hollaetal2004

[14]

CzechRe

public

Caucasian

133196

CP(m

ildto

mod

erate

tosevere)

Agegender

PCR-RF

LPMMP-1(minus1607

1G2G)0

52MMP-1(minus519AG

)011

MMP-1(minus422AT)0

47

Itagakietal2004

[22]

Japan

Asian

205142

CP(m

ildor

mod

erate

orsevere)

Agegendersm

oker

ratio

sTaqM

anMMP-1(minus1607

1G2G)0

48

MMP-3(minus117

15A6A)0

87

3714

2AgP

Hollaetal2005

[23]

CzechRe

public

Caucasian

149127

CP(m

ildto

mod

erate

tosevere)

Agesmoker

ratio

sPC

R-RF

LP

MMP-2(minus1575

GA

)040

MMP-2(minus1306

CT)

019

MMP-2(minus790TG)0

67

MMP-2(minus735CT)

042

Caoetal2005

[32]

China

Asian

4052

AgP

mdashPC

R-RF

LPMMP-1(minus1607

1G2G)0

78

Caoetal2006

[13]

China

Asian

6050

CP(m

oderateo

rsevere)

mdashPC

R-RF

LPMMP-1(minus1607

1G2G)0

99

Kelese

tal2006

[12]

Turkey

Caucasian

7070

CP(severe)

Agegender

PCR-RF

LPMMP-9(minus1562

CT)

082

Hollaetal2006

[24]

CzechRe

public

Caucasian

169135

CP(m

oderateo

rsevere)

Agegendersm

oker

ratio

sPC

R-RF

LPMMP-9(minus1562

CT)

059

MMP-9(+279RQ)0

25

Chen

etal2007

[16]

China

Asian

7912

8AgP

Agegender

DHPL

CPC

R-RF

LPMMP-2(minus1306

CT)

100

MMP-9(minus1562

CT)

063

Gurkanetal2007

[8]

Turkey

Caucasian

9215

7AgP

Gender

PCR-RF

LPMMP-2(minus735CT)

045

MMP-9(minus1562

CT)

035

MMP-12

(357

AsnSer)0

06

Gurkanetal2008

[25]

Turkey

Caucasian

8710

7CP

(severe)

mdashPC

R-RF

LPMMP-2(minus735CT)

043

MMP-12

(357

AsnSer)0

47

Pirhan

etal2008

[26]

Turkey

Caucasian

10298

CP(severe)

mdashPC

R-RF

LPMMP-1(minus519AG

)079

Ustu

netal2008

[27]

Turkey

Caucasian

12654

CP(m

oderateo

rsevere)

Age

PCR-RF

LPMMP-1(minus1607

1G2G)0

75

Pirhan

etal2009

[28]

Turkey

Caucasian

10298

CP(severe)

mdashPC

R-RF

LPMMP-13

(minus77

AG

)089

MMP-13

(11A

12A)0

92

Chou

etal2011[11]

China

Asian

3611

06CP

(mod

erateto

severe)

Gendersm

oker

ratio

sPC

R-RF

LPMMP-8(minus799CT)

022

9610

6AgP

Hollaetal2012

[29]

CzechRe

public

Caucasian

3412

78CP

(mild

tomod

erate

tosevere)

Agegendersm

oker

ratio

sPC

R-RF

LPMMP-8(+17

CG)0

14MMP-8(minus799CT)

063

Emingiletal2014

[30]

Turkey

Caucasian

100167

AgP

Gender

PCR-RF

LPMMP-8(+17

CG)0

29

MMP-8(minus799CT)

009

MMP-8(minus381A

G)0

87

CPchron

icperio

dontitisAgP

aggressiveperio

dontitisHWE

Hardy-W

einb

ergequilib

riumM

ildchronicperio

dontitispatie

ntsw

ithteethexhibitin

glt3m

mattachmentlossmod

eratechronicperio

dontitis

patientsw

ithteethexhibitin

gge3m

mandlt7m


chronicp

eriodo

ntitispatie

ntsw

ithteethexhibitin

gge7m


ofHWEwas

considered

significant

Disease Markers 5



Selection






Exposure




6 Disease Markers


Scre

enin

gIn

clude

dEl

igib

ility








Iden

tifica

tion


HWE n = 3







(i) Reviews n = 5








Disease Markers 7


ofthep

olym

orph

ismsinMMPs

gene

onperio

dontitisrisk

MMP-1

minus1607

1G2G

Stud

ies

(casescon

trols)

2Gversus

1GOR(95

CI)

1198682(

)119901ℎ119901119888

2G2Gversus

1G1G

OR(95

CI)

1198682(

)119901ℎ119901119888

1G2Gversus

1G1G

OR(95

CI)

1198682(

)119901ℎ119901119888

2G2Gversus

1G2G

OR(95

CI)

1198682(

)119901ℎ119901119888

1G2G+2G

2Gversus

1G1G

OR(95

CI)

1198682(

)119901ℎ119901119888

2G2Gversus

1G1G

+1G

2GOR(95

CI)

1198682(

)119901ℎ119901119888

1G2Gversus

1G1G

+2G

2G

OR(95

CI)

1198682(

)119901ℎ119901119888

Type

ofdisease

CP Caucasian

3(309287)

102(067ndash15

6)10

5(044ndash

251)

095

(064ndash

142)

090

(059ndash

137)

091

(062ndash13

2)089

(060ndash

133)

100(072ndash14

0)631

006710

0063000671000

120032

110

0000049010

00499

013610

00382019

810

0000096910

00

Asian

2(30219

2)17

5(077ndash395)

279

(056ndash

1398)

131(074ndash232

)17

5(077ndash394)

196(063ndash609)

201

(072ndash561)

079

(055ndash116)

84500111000

81500201000

27002421000

64600931000

69400711000

797

002710

0000046

710

00Ag

P

Asian

2(77194)

134(048ndash373)

154(028ndash855)

091

(042ndash19

6)16

7(038ndash731)

114(056ndash

232)

164(035

ndash770)

075

(043ndash13

0)84800101000

78400311000

00076310

0081800191000

39002001000

857

000810

00595011610

00Severe

CPCa

ucasian

Mild

tomod

erate

2(6691)

099

(063ndash15

5)099

(039

ndash250)

116(052

ndash260)

085

(039

ndash184)

110(051ndash238)

089

(043ndash18

5)117(062ndash221)

00061310

0000061310

0000066710

0000087410

0000061310

0000075110

0000087610

00

Severe

2(11091)

153(072ndash324)

244

(047ndash1259)

152(070ndash

330)

131(068ndash251)

168(081ndash350)

147(080ndash

273)

098

(056ndash

173)

657

008810

0063400981000

15802761000

00036910

00511

015310

00423018

810

0000084510

00Asian

Mild

tomod

erate

2(16819

2)12

6(093ndash17

2)16

2(084ndash

312)

140(072ndash269)

119(075ndash18

7)15

1(082ndash280)

127(083ndash19

5)097

(063ndash14

8)601

0113098

7627

010110

00438018

210

0000053410

00560013

210

0021002611000

00078410

00

Severe

2(97192)

199(092ndash426)

293

(071ndash1203)

116(053

ndash256)

219

(126ndash

379)

178(086ndash

367)

255

(095ndash685)

058

(035

ndash098)

73500520546

67000820952

00046

810

00489

01620035

381

02040854

697

0069044

1000517028

7Sm

okinghabitinCP

Caucasian

Non

smok

ing

3(200213)

092

(055ndash15

5)090

(033

ndash243)

087

(054ndash

140)

085

(052

ndash141)

096

(045ndash205)

082

(052

ndash130)

098

(066ndash

146)

661

005310

00631

006710

0034902151000

00043810

00569

009810

0040

10118

1000

00057510

00

Smok

ing

2(10974)

110(071ndash17

2)114(043ndash302)

112(054ndash

234)

115(050ndash

264

)112(055ndash229)

119(053

ndash265)

098

(053

ndash184)

19002671000

329

022210

0000097610

00522014

810

0000068210

00540014

010

0000031910

00MMP-1

minus519AG

Stud

ies

(casescon

trols)

Gversus

AOR(95

CI)

1198682(

)119901ℎ119901119888

GGversus

AA

OR(95

CI)

1198682(

)119901ℎ119901119888

AGversus

AA

OR(95

CI)

1198682(

)119901ℎ119901119888

GGversus

AGOR(95

CI)

1198682(

)119901ℎ119901119888

AG+GGversus

AA

OR(95

CI)

1198682(

)119901ℎ119901119888

GGversus

AA+AG

OR(95

CI)

1198682(

)119901ℎ119901119888

AGversus

AA+GG

OR(95

CI)

1198682(

)119901ℎ119901119888

Type

ofdisease

CP Caucasian

2(235293)

103(080ndash

132)

108(064ndash

182)

100(068ndash14

6)10

6(064ndash

175)

102(071ndash14

5)10

6(067ndash16

9)098

(069ndash

139)

00

09841000

00

08061000

00

0811

1000

00

06911000

00

08841000

00

07361000

00077810

00MMP-2

minus735CT

Stud

ies

(casescon

trols)

Tversus

COR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

CTversus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CTOR(95

CI)

1198682(

)119901ℎ119901119888

CT+TT

versus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CC+CT

OR(95

CI)

1198682(

)119901ℎ119901119888

CTversus

CC+TT

OR(95

CI)

1198682(

)119901ℎ119901119888

Type

ofdisease

CP Caucasian

2(236234)

111(079ndash155)

119(043ndash337)

112(074

ndash168)

108(037

ndash313

)10

0(067ndash14

9)116(041ndash324)

110(074

ndash165)

00069510

0000051110

0000094010

0000054110

0000069910

0000052210

0000098910

00Sm

okinghabitinCP

Caucasian

Non

smok

ing

2(13319

8)113(074

ndash172)

115(032

ndash409)

117(070ndash

194)

099

(027ndash366)

116(071ndash18

8)110(031ndash389)

115(069ndash

190)

00033710

00452017

710

0000078410

0032402241000

00053310

00424018

810

0000091710

00

8 Disease Markers

Table3Con

tinued

MMP-9

minus1562

CT

Stud

ies

(casescon

trols)

Tversus

COR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

CTversus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CTOR(95

CI)

1198682(

)119901ℎ119901119888

CT+TT

versus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CC+CT

OR(95

CI)

1198682(

)119901ℎ119901119888

CTversus

CC+TT

OR(95

CI)

1198682(

)119901ℎ119901119888

Type

ofdisease

CP Caucasian

2(239205)

056

(024ndash

135)

036

(011ndash112

)063

(029ndash

136)

051

(015

ndash172)

057

(023ndash13

8)039

(012

ndash124)

072

(047ndash110)

78200321000

35902120553

64000961000

00045910

00739

005010

0020402620777

571

0127092

4Severe

CPCa

ucasian

Severe

2(163205)

063

(020ndash

197)

044

(013

ndash149)

071

(024ndash

209)

053

(014

ndash195)

065

(019

ndash215

)046

(014

ndash157)

075

(028ndash201)

861

000710

00544013

910

0079300281000

00042810

0084200121000

410019

310

0076000411000

MMP-1matrix


MP-2matrix


MP-9matrix


Pchronicp

eriodo

ntitisAgP

aggressiveg

eneralized

perio

dontitis



orrected

byBo

nferroni

correctio

nOR

odds

ratio

CIconfi

denceinterval


0the

rand

omeffectsmod

elisusedC

onversely

the

fixed

effectsmod

elisused

119901119888lt001

isconsidered

statisticallysig

nificant

Disease Markers 9


Holla et al (2004)

Ustun et al (2008)


Cao et al (2006)


Cao et al (2005)

Holla et al (2006)

Keles et al (2006)

Study ID

165 (090 303)

075 (055 103)

103 (065 161)

102 (067 156)

119 (087 163)

274 (157 481)

175 (077 395)

080 (048 135)

229 (124 420)

134 (048 373)

084 (055 130)

035 (017 069)

056 (024 135)

OR (95 CI)

2551

4121

3327

10000

5398

4602

10000

5118

4882

10000

5479

4521

10000

Holla et al (2004)

Pirhan et al (2008)

Holla et al (2005)

Study ID

102 (075 140)

103 (067 159)

103 (080 132)

104 (065 165)

119 (073 193)

111 (079 155)

OR (95 CI)

6619

3381

10000

5388

4612

10000

weight

weight

Subtotal (I2 = 00 p = 0695)

Subtotal (I2 = 00 p = 0984)

Subtotal (I2 = 782 p = 0032)

Subtotal (I2 = 848 p = 0010)

Subtotal (I2 = 845 p = 0011)

Subtotal (I2 = 631 p = 0067)

1 5790173

1 1930518








Gurkan et al (2007)


Figure 2 Continued

10 Disease Markers

286 (082 1001)

056 (030 107)

107 (042 273)

105 (044 251)

133 (069 257)

693 (203 2363)

279 (056 1398)

066 (023 188)

379 (114 1258)

154 (028 855)

2539

4150

3312

10000

5509

4491

10000

5148

4852

10000

104 (057 189)

121 (042 350)

108 (064 182)

087 (021 357)

175 (038 818)

120 (043 337)

061 (016 233)

008 (000 157)

036 (011 112)

7713

2287

10000

6270

3730

10000

5250

4750

10000

Study ID OR (95 CI)

Study ID OR (95 CI)

Subtotal (I2 = 359 p = 0212)

Subtotal (I2 = 00 p = 0511)

Subtotal (I2 = 00 p = 0806)

Subtotal (I2 = 784 p = 0031)

Subtotal (I2 = 815 p = 0020)

Subtotal (I2 = 630 p = 0067)

1 23600423

1 23200043








weight

weight


Holla et al (2004)

Ustun et al (2008)


Cao et al (2006)


Cao et al (2005)

Holla et al (2006)

Keles et al (2006)

Holla et al (2004)

Pirhan et al (2008)

Holla et al (2005)

Gurkan et al (2007)


Figure 2 Continued

Disease Markers 11

089 (053 148)

040 (018 087)

063 (029 136)

5713

4287

10000

54910182

Study ID OR (95 CI)

Subtotal (I2 = 640 p = 0096)

198 (063 620)079 (048 129)110 (047 254)095 (064 142)

107 (055 207)239 (074 776)131 (074 232)

082 (030 226)104 (032 337)091 (042 196)

104 (062 172)094 (053 169)100 (068 146)

110 (063 191)114 (063 206)112 (074 168)

84470792077

10000

81671833

10000

59634037

10000

55794421

10000

54274573

10000

1 7760129

Study ID OR (95 CI)

Subtotal (I2 = 120 p = 0321)

Subtotal (I2 = 270 p = 0242)

Subtotal (I2 = 00 p = 0763)

Subtotal (I2 = 00 p = 0811)

Subtotal (I2 = 00 p = 0940)








weight

weight




Holla et al (2006)

Keles et al (2006)




Figure 2 Continued

12 Disease Markers

144 (053 393)

072 (039 132)

098 (046 206)

090 (059 137)

100 (057 177)

129 (044 379)

106 (064 175)

079 (018 342)

154 (032 741)

108 (037 313)

069 (017 275)

020 (001 404)

051 (015 172)

1426

5500

3074

10000

8046

1954

10000

6129

3871

10000

6310

3690

10000

124 (078 197)

290 (121 693)

175 (077 394)

080 (036 180)

363 (138 959)

167 (038 731)

5994

4006

10000

5167

4833

10000

Study ID OR (95 CI)

Study ID OR (95 CI)

Subtotal (I2 = 00 p = 0490)

Subtotal (I2 = 00 p = 0691)

Subtotal (I2 = 00 p = 0541)

Subtotal (I2 = 00 p = 0459)

Subtotal (I2 = 646 p = 0093)

Subtotal (I2 = 818 p = 0019)

1 99200101

1 9590104








weight

weight


Holla et al (2004)

Ustun et al (2008)


Cao et al (2006)


Cao et al (2005)

Holla et al (2006)

Keles et al (2006)

Holla et al (2004)

Pirhan et al (2008)

Holla et al (2005)

Gurkan et al (2007)


Figure 2 Continued

Disease Markers 13

228 (077 669)

071 (045 114)

109 (049 241)

091 (062 132)

074 (029 191)

189 (065 551)

114 (056 232)

104 (065 166)

098 (056 172)

102 (071 145)

107 (063 183)

092 (051 166)

100 (067 149)

756

7248

1996

10000

6504

3496

10000

5777

4223

10000

5363

4637

10000


119 (064 222)

386 (127 1176)

196 (063 609)

085 (052 140)

034 (016 074)

057 (023 138)

5805

4195

10000

5539

4461

10000

Study ID OR (95 CI)

Study ID OR (95 CI)

1 6690149

1 11800851

Subtotal (I2 = 499 p = 0136)

Subtotal (I2 = 390 p = 0200)

Subtotal (I2 = 00 p = 0884)

Subtotal (I2 = 00 p = 0699)

Subtotal (I2 = 694 p = 0071)

Subtotal (I2 = 739 p = 0050)







weight

weight


Holla et al (2004)

Ustun et al (2008)


Cao et al (2005)

Holla et al (2004)

Cao et al (2006)


Holla et al (2006)

Keles et al (2006)

Pirhan et al (2008)

Holla et al (2005)

Gurkan et al (2007)


Figure 2 Continued

14 Disease Markers

175 (068 451)

065 (036 115)

100 (049 204)

089 (060 133)

102 (061 172)

124 (044 348)

106 (067 169)

085 (021 346)

167 (036 767)

116 (042 324)

063 (017 239)

010 (001 198)

039 (012 124)

1279

5789

2931

10000

8101

1899

10000

6208

3792

10000

5485

4515

10000

126 (082 195)

362 (159 825)

201 (072 561)

076 (036 162)

368 (151 902)

164 (035 770)

5578

4422

10000

5123

487

10000

Study ID OR (95 CI)

Study ID OR (95 CI)

Subtotal (I2 = 204 p = 0262)

Subtotal (I2 = 382 p = 0198)

Subtotal (I2 = 00 p = 0736)

Subtotal (I2 = 00 p = 0522)

Subtotal (I2 = 797 p = 0027)

Subtotal (I2 = 857 p = 0008)

1 9020111

1 181000554








weight

weight


Holla et al (2004)

Ustun et al (2008)


Cao et al (2006)


Cao et al (2005)

Holla et al (2006)

Keles et al (2006)

Holla et al (2004)

Pirhan et al (2008)

Holla et al (2005)

Gurkan et al (2007)


Figure 2 Continued

Disease Markers 15

106 (045 247)097 (062 150)105 (056 200)100 (072 140)

086 (056 133)062 (029 133)079 (055 116)

110 (053 227)045 (019 105)075 (043 130)

102 (065 159)092 (052 162)098 (069 139)

111 (064 192)110 (061 199)110 (074 165)

090 (054 151)044 (020 095)072 (047 110)

150358312666

10000

72332767

10000

46645336

10000

60823918

10000

53494651

10000

61083892

10000

10189 528

Subtotal (I2 = 00 p = 0969)

Subtotal (I2 = 00 p = 0467)

Subtotal (I2 = 595 p = 0116)

Subtotal (I2 = 00 p = 0778)

Subtotal (I2 = 00 p = 0989)

Subtotal (I2 = 571 p = 0127)

Study ID OR (95 CI)







weight













16 Disease Markers




4 Discussion







Disease Markers 17










18 Disease Markers



Disclosure


Competing Interests




References

















Disease Markers 19




























20 Disease Markers


















2 Disease Markers





References


























1 Introduction




2 Disease Markers










Disease Markers 3








3 Results




4 Disease Markers

Seru

m ac

tive M

MP-

9 (n

gm

L)


500

1000

1500

2000

2500p lt 001

p lt 0001

(a)

OND patients

Seru

m ac

tive M

MP-

2 (n

gm

L)

HC MS patients0

100

200

300

500

600

700

(b)

CSF

activ

e MM

P-9

(ng

mL)


025

050

075

100

200

300

400p lt 001

(c)

CSF

activ

e MM

P-2

(ng

mL)


10

20

30

40

(d)






Disease Markers 5

Seru

m IL

-17

(pg

mL)

MS patients0

200

400

600

800

1000

1500

2000

2500

OND patients

(a)

Seru

m IL

-18

(pg

mL)

0

5000

10000

15000

20000


(b)

Seru

m IL

-23

(pg

mL)

0

2000

4000

6000

800050000

100000

150000


(c)

Seru

m M

CP-3

(pg

mL)

0

50

100

150

200


(d)

CSF

IL-1

7 (p

gm

L)

0

10

20

30

40

50


(e)

CSF

IL-1

8 (p

gm

L)

0

2000

4000

6000

8000


(f)

Figure 2 Continued

6 Disease Markers

CSF

IL-2

3 (p

gm

L)

0

50

100

150

200

2000

4000


(g)

CSF

MCP

-3 (p

gm

L)

0

5

10

15

20

25


(h)








Disease Markers 7




OND () MS ()Serum






4 Discussion






8 Disease Markers






Competing Interests


Acknowledgments


References












Disease Markers 9









































1 Introduction








2 Disease Markers













Disease Markers 3











4 Disease Markers












Disease Markers 5



9 Conclusion




Competing Interests


References













6 Disease Markers































Disease Markers 7







































1 Introduction





2 Disease Markers












3 Results


Disease Markers 3

(a) (b) (c)

(d) (e) (f)






4 Disease Markers













Disease Markers 5



GenderMale 23 (793)









4 Discussion




6 Disease Markers






Competing Interests


Acknowledgments


References







Disease Markers 7



































1 Introduction




2 Disease Markers









Disease Markers 3






3 Results



4 Discussion


4 Disease Markers

Seru

m ac

tive M

MP-2

leve

ls (n

gm

L)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

0

5

10

15

20

(a)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

Seru

m ac

tive M

MP-9

leve

ls (n

gm

L)

0

5

10

15

(b)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

Seru

m T

IMP-

2le

vels

(ng

mL)

0

50

100

150

200

250

(c)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

Seru

m T

IMP-

1le

vels

(ng

mL)

0

200

400

600

800

(d)




Disease Markers 5

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

0

200

400

600

800Se

rum

activ

e MM

P-2

TIM

P-2

(times103)

(a)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

0

10

20

30

40

Seru

m ac

tive M

MP-9

TIM

P-1

(times103)

(b)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

00

05

10

15

20

Seru

m ac

tive M

MP-9

act

ive M

MP-2

1 2 3

4 5

6 7

(c)




6 Disease Markers


Competing Interests


Acknowledgments


References




























Disease Markers 7






















1 Introduction





2 Disease Markers










Disease Markers 3





3 Results





4 Disease Markers

Table1MainCh

aracteris

ticso

fincludedstu

dies

Authoryear

Cou

ntry

Ethn

icity

Samples

ize

(casecontrol)

Type

ofperio

dontitis

Matchingcriteria

Genotype

metho

dGene(po

lymorph

ism)amp

HWEin

controls

deSouzae

tal2003

[31]

Brazil

Caucasian

5037

CP(m

oderateo

rsevere)

Smoker

ratio

sPC

R-RF

LPMMP-1(minus1607

1G2G)0

87

Hollaetal2004

[14]

CzechRe

public

Caucasian

133196

CP(m

ildto

mod

erate

tosevere)

Agegender

PCR-RF

LPMMP-1(minus1607

1G2G)0

52MMP-1(minus519AG

)011

MMP-1(minus422AT)0

47

Itagakietal2004

[22]

Japan

Asian

205142

CP(m

ildor

mod

erate

orsevere)

Agegendersm

oker

ratio

sTaqM

anMMP-1(minus1607

1G2G)0

48

MMP-3(minus117

15A6A)0

87

3714

2AgP

Hollaetal2005

[23]

CzechRe

public

Caucasian

149127

CP(m

ildto

mod

erate

tosevere)

Agesmoker

ratio

sPC

R-RF

LP

MMP-2(minus1575

GA

)040

MMP-2(minus1306

CT)

019

MMP-2(minus790TG)0

67

MMP-2(minus735CT)

042

Caoetal2005

[32]

China

Asian

4052

AgP

mdashPC

R-RF

LPMMP-1(minus1607

1G2G)0

78

Caoetal2006

[13]

China

Asian

6050

CP(m

oderateo

rsevere)

mdashPC

R-RF

LPMMP-1(minus1607

1G2G)0

99

Kelese

tal2006

[12]

Turkey

Caucasian

7070

CP(severe)

Agegender

PCR-RF

LPMMP-9(minus1562

CT)

082

Hollaetal2006

[24]

CzechRe

public

Caucasian

169135

CP(m

oderateo

rsevere)

Agegendersm

oker

ratio

sPC

R-RF

LPMMP-9(minus1562

CT)

059

MMP-9(+279RQ)0

25

Chen

etal2007

[16]

China

Asian

7912

8AgP

Agegender

DHPL

CPC

R-RF

LPMMP-2(minus1306

CT)

100

MMP-9(minus1562

CT)

063

Gurkanetal2007

[8]

Turkey

Caucasian

9215

7AgP

Gender

PCR-RF

LPMMP-2(minus735CT)

045

MMP-9(minus1562

CT)

035

MMP-12

(357

AsnSer)0

06

Gurkanetal2008

[25]

Turkey

Caucasian

8710

7CP

(severe)

mdashPC

R-RF

LPMMP-2(minus735CT)

043

MMP-12

(357

AsnSer)0

47

Pirhan

etal2008

[26]

Turkey

Caucasian

10298

CP(severe)

mdashPC

R-RF

LPMMP-1(minus519AG

)079

Ustu

netal2008

[27]

Turkey

Caucasian

12654

CP(m

oderateo

rsevere)

Age

PCR-RF

LPMMP-1(minus1607

1G2G)0

75

Pirhan

etal2009

[28]

Turkey

Caucasian

10298

CP(severe)

mdashPC

R-RF

LPMMP-13

(minus77

AG

)089

MMP-13

(11A

12A)0

92

Chou

etal2011[11]

China

Asian

3611

06CP

(mod

erateto

severe)

Gendersm

oker

ratio

sPC

R-RF

LPMMP-8(minus799CT)

022

9610

6AgP

Hollaetal2012

[29]

CzechRe

public

Caucasian

3412

78CP

(mild

tomod

erate

tosevere)

Agegendersm

oker

ratio

sPC

R-RF

LPMMP-8(+17

CG)0

14MMP-8(minus799CT)

063

Emingiletal2014

[30]

Turkey

Caucasian

100167

AgP

Gender

PCR-RF

LPMMP-8(+17

CG)0

29

MMP-8(minus799CT)

009

MMP-8(minus381A

G)0

87

CPchron

icperio

dontitisAgP

aggressiveperio

dontitisHWE

Hardy-W

einb

ergequilib

riumM

ildchronicperio

dontitispatie

ntsw

ithteethexhibitin

glt3m

mattachmentlossmod

eratechronicperio

dontitis

patientsw

ithteethexhibitin

gge3m

mandlt7m


chronicp

eriodo

ntitispatie

ntsw

ithteethexhibitin

gge7m


ofHWEwas

considered

significant

Disease Markers 5



Selection






Exposure




6 Disease Markers


Scre

enin

gIn

clude

dEl

igib

ility








Iden

tifica

tion


HWE n = 3







(i) Reviews n = 5








Disease Markers 7


ofthep

olym

orph

ismsinMMPs

gene

onperio

dontitisrisk

MMP-1

minus1607

1G2G

Stud

ies

(casescon

trols)

2Gversus

1GOR(95

CI)

1198682(

)119901ℎ119901119888

2G2Gversus

1G1G

OR(95

CI)

1198682(

)119901ℎ119901119888

1G2Gversus

1G1G

OR(95

CI)

1198682(

)119901ℎ119901119888

2G2Gversus

1G2G

OR(95

CI)

1198682(

)119901ℎ119901119888

1G2G+2G

2Gversus

1G1G

OR(95

CI)

1198682(

)119901ℎ119901119888

2G2Gversus

1G1G

+1G

2GOR(95

CI)

1198682(

)119901ℎ119901119888

1G2Gversus

1G1G

+2G

2G

OR(95

CI)

1198682(

)119901ℎ119901119888

Type

ofdisease

CP Caucasian

3(309287)

102(067ndash15

6)10

5(044ndash

251)

095

(064ndash

142)

090

(059ndash

137)

091

(062ndash13

2)089

(060ndash

133)

100(072ndash14

0)631

006710

0063000671000

120032

110

0000049010

00499

013610

00382019

810

0000096910

00

Asian

2(30219

2)17

5(077ndash395)

279

(056ndash

1398)

131(074ndash232

)17

5(077ndash394)

196(063ndash609)

201

(072ndash561)

079

(055ndash116)

84500111000

81500201000

27002421000

64600931000

69400711000

797

002710

0000046

710

00Ag

P

Asian

2(77194)

134(048ndash373)

154(028ndash855)

091

(042ndash19

6)16

7(038ndash731)

114(056ndash

232)

164(035

ndash770)

075

(043ndash13

0)84800101000

78400311000

00076310

0081800191000

39002001000

857

000810

00595011610

00Severe

CPCa

ucasian

Mild

tomod

erate

2(6691)

099

(063ndash15

5)099

(039

ndash250)

116(052

ndash260)

085

(039

ndash184)

110(051ndash238)

089

(043ndash18

5)117(062ndash221)

00061310

0000061310

0000066710

0000087410

0000061310

0000075110

0000087610

00

Severe

2(11091)

153(072ndash324)

244

(047ndash1259)

152(070ndash

330)

131(068ndash251)

168(081ndash350)

147(080ndash

273)

098

(056ndash

173)

657

008810

0063400981000

15802761000

00036910

00511

015310

00423018

810

0000084510

00Asian

Mild

tomod

erate

2(16819

2)12

6(093ndash17

2)16

2(084ndash

312)

140(072ndash269)

119(075ndash18

7)15

1(082ndash280)

127(083ndash19

5)097

(063ndash14

8)601

0113098

7627

010110

00438018

210

0000053410

00560013

210

0021002611000

00078410

00

Severe

2(97192)

199(092ndash426)

293

(071ndash1203)

116(053

ndash256)

219

(126ndash

379)

178(086ndash

367)

255

(095ndash685)

058

(035

ndash098)

73500520546

67000820952

00046

810

00489

01620035

381

02040854

697

0069044

1000517028

7Sm

okinghabitinCP

Caucasian

Non

smok

ing

3(200213)

092

(055ndash15

5)090

(033

ndash243)

087

(054ndash

140)

085

(052

ndash141)

096

(045ndash205)

082

(052

ndash130)

098

(066ndash

146)

661

005310

00631

006710

0034902151000

00043810

00569

009810

0040

10118

1000

00057510

00

Smok

ing

2(10974)

110(071ndash17

2)114(043ndash302)

112(054ndash

234)

115(050ndash

264

)112(055ndash229)

119(053

ndash265)

098

(053

ndash184)

19002671000

329

022210

0000097610

00522014

810

0000068210

00540014

010

0000031910

00MMP-1

minus519AG

Stud

ies

(casescon

trols)

Gversus

AOR(95

CI)

1198682(

)119901ℎ119901119888

GGversus

AA

OR(95

CI)

1198682(

)119901ℎ119901119888

AGversus

AA

OR(95

CI)

1198682(

)119901ℎ119901119888

GGversus

AGOR(95

CI)

1198682(

)119901ℎ119901119888

AG+GGversus

AA

OR(95

CI)

1198682(

)119901ℎ119901119888

GGversus

AA+AG

OR(95

CI)

1198682(

)119901ℎ119901119888

AGversus

AA+GG

OR(95

CI)

1198682(

)119901ℎ119901119888

Type

ofdisease

CP Caucasian

2(235293)

103(080ndash

132)

108(064ndash

182)

100(068ndash14

6)10

6(064ndash

175)

102(071ndash14

5)10

6(067ndash16

9)098

(069ndash

139)

00

09841000

00

08061000

00

0811

1000

00

06911000

00

08841000

00

07361000

00077810

00MMP-2

minus735CT

Stud

ies

(casescon

trols)

Tversus

COR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

CTversus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CTOR(95

CI)

1198682(

)119901ℎ119901119888

CT+TT

versus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CC+CT

OR(95

CI)

1198682(

)119901ℎ119901119888

CTversus

CC+TT

OR(95

CI)

1198682(

)119901ℎ119901119888

Type

ofdisease

CP Caucasian

2(236234)

111(079ndash155)

119(043ndash337)

112(074

ndash168)

108(037

ndash313

)10

0(067ndash14

9)116(041ndash324)

110(074

ndash165)

00069510

0000051110

0000094010

0000054110

0000069910

0000052210

0000098910

00Sm

okinghabitinCP

Caucasian

Non

smok

ing

2(13319

8)113(074

ndash172)

115(032

ndash409)

117(070ndash

194)

099

(027ndash366)

116(071ndash18

8)110(031ndash389)

115(069ndash

190)

00033710

00452017

710

0000078410

0032402241000

00053310

00424018

810

0000091710

00

8 Disease Markers

Table3Con

tinued

MMP-9

minus1562

CT

Stud

ies

(casescon

trols)

Tversus

COR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

CTversus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CTOR(95

CI)

1198682(

)119901ℎ119901119888

CT+TT

versus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CC+CT

OR(95

CI)

1198682(

)119901ℎ119901119888

CTversus

CC+TT

OR(95

CI)

1198682(

)119901ℎ119901119888

Type

ofdisease

CP Caucasian

2(239205)

056

(024ndash

135)

036

(011ndash112

)063

(029ndash

136)

051

(015

ndash172)

057

(023ndash13

8)039

(012

ndash124)

072

(047ndash110)

78200321000

35902120553

64000961000

00045910

00739

005010

0020402620777

571

0127092

4Severe

CPCa

ucasian

Severe

2(163205)

063

(020ndash

197)

044

(013

ndash149)

071

(024ndash

209)

053

(014

ndash195)

065

(019

ndash215

)046

(014

ndash157)

075

(028ndash201)

861

000710

00544013

910

0079300281000

00042810

0084200121000

410019

310

0076000411000

MMP-1matrix


MP-2matrix


MP-9matrix


Pchronicp

eriodo

ntitisAgP

aggressiveg

eneralized

perio

dontitis



orrected

byBo

nferroni

correctio

nOR

odds

ratio

CIconfi

denceinterval


0the

rand

omeffectsmod

elisusedC

onversely

the

fixed

effectsmod

elisused

119901119888lt001

isconsidered

statisticallysig

nificant

Disease Markers 9


Holla et al (2004)

Ustun et al (2008)


Cao et al (2006)


Cao et al (2005)

Holla et al (2006)

Keles et al (2006)

Study ID

165 (090 303)

075 (055 103)

103 (065 161)

102 (067 156)

119 (087 163)

274 (157 481)

175 (077 395)

080 (048 135)

229 (124 420)

134 (048 373)

084 (055 130)

035 (017 069)

056 (024 135)

OR (95 CI)

2551

4121

3327

10000

5398

4602

10000

5118

4882

10000

5479

4521

10000

Holla et al (2004)

Pirhan et al (2008)

Holla et al (2005)

Study ID

102 (075 140)

103 (067 159)

103 (080 132)

104 (065 165)

119 (073 193)

111 (079 155)

OR (95 CI)

6619

3381

10000

5388

4612

10000

weight

weight

Subtotal (I2 = 00 p = 0695)

Subtotal (I2 = 00 p = 0984)

Subtotal (I2 = 782 p = 0032)

Subtotal (I2 = 848 p = 0010)

Subtotal (I2 = 845 p = 0011)

Subtotal (I2 = 631 p = 0067)

1 5790173

1 1930518








Gurkan et al (2007)


Figure 2 Continued

10 Disease Markers

286 (082 1001)

056 (030 107)

107 (042 273)

105 (044 251)

133 (069 257)

693 (203 2363)

279 (056 1398)

066 (023 188)

379 (114 1258)

154 (028 855)

2539

4150

3312

10000

5509

4491

10000

5148

4852

10000

104 (057 189)

121 (042 350)

108 (064 182)

087 (021 357)

175 (038 818)

120 (043 337)

061 (016 233)

008 (000 157)

036 (011 112)

7713

2287

10000

6270

3730

10000

5250

4750

10000

Study ID OR (95 CI)

Study ID OR (95 CI)

Subtotal (I2 = 359 p = 0212)

Subtotal (I2 = 00 p = 0511)

Subtotal (I2 = 00 p = 0806)

Subtotal (I2 = 784 p = 0031)

Subtotal (I2 = 815 p = 0020)

Subtotal (I2 = 630 p = 0067)

1 23600423

1 23200043








weight

weight


Holla et al (2004)

Ustun et al (2008)


Cao et al (2006)


Cao et al (2005)

Holla et al (2006)

Keles et al (2006)

Holla et al (2004)

Pirhan et al (2008)

Holla et al (2005)

Gurkan et al (2007)


Figure 2 Continued

Disease Markers 11

089 (053 148)

040 (018 087)

063 (029 136)

5713

4287

10000

54910182

Study ID OR (95 CI)

Subtotal (I2 = 640 p = 0096)

198 (063 620)079 (048 129)110 (047 254)095 (064 142)

107 (055 207)239 (074 776)131 (074 232)

082 (030 226)104 (032 337)091 (042 196)

104 (062 172)094 (053 169)100 (068 146)

110 (063 191)114 (063 206)112 (074 168)

84470792077

10000

81671833

10000

59634037

10000

55794421

10000

54274573

10000

1 7760129

Study ID OR (95 CI)

Subtotal (I2 = 120 p = 0321)

Subtotal (I2 = 270 p = 0242)

Subtotal (I2 = 00 p = 0763)

Subtotal (I2 = 00 p = 0811)

Subtotal (I2 = 00 p = 0940)








weight

weight




Holla et al (2006)

Keles et al (2006)




Figure 2 Continued

12 Disease Markers

144 (053 393)

072 (039 132)

098 (046 206)

090 (059 137)

100 (057 177)

129 (044 379)

106 (064 175)

079 (018 342)

154 (032 741)

108 (037 313)

069 (017 275)

020 (001 404)

051 (015 172)

1426

5500

3074

10000

8046

1954

10000

6129

3871

10000

6310

3690

10000

124 (078 197)

290 (121 693)

175 (077 394)

080 (036 180)

363 (138 959)

167 (038 731)

5994

4006

10000

5167

4833

10000

Study ID OR (95 CI)

Study ID OR (95 CI)

Subtotal (I2 = 00 p = 0490)

Subtotal (I2 = 00 p = 0691)

Subtotal (I2 = 00 p = 0541)

Subtotal (I2 = 00 p = 0459)

Subtotal (I2 = 646 p = 0093)

Subtotal (I2 = 818 p = 0019)

1 99200101

1 9590104








weight

weight


Holla et al (2004)

Ustun et al (2008)


Cao et al (2006)


Cao et al (2005)

Holla et al (2006)

Keles et al (2006)

Holla et al (2004)

Pirhan et al (2008)

Holla et al (2005)

Gurkan et al (2007)


Figure 2 Continued

Disease Markers 13

228 (077 669)

071 (045 114)

109 (049 241)

091 (062 132)

074 (029 191)

189 (065 551)

114 (056 232)

104 (065 166)

098 (056 172)

102 (071 145)

107 (063 183)

092 (051 166)

100 (067 149)

756

7248

1996

10000

6504

3496

10000

5777

4223

10000

5363

4637

10000


119 (064 222)

386 (127 1176)

196 (063 609)

085 (052 140)

034 (016 074)

057 (023 138)

5805

4195

10000

5539

4461

10000

Study ID OR (95 CI)

Study ID OR (95 CI)

1 6690149

1 11800851

Subtotal (I2 = 499 p = 0136)

Subtotal (I2 = 390 p = 0200)

Subtotal (I2 = 00 p = 0884)

Subtotal (I2 = 00 p = 0699)

Subtotal (I2 = 694 p = 0071)

Subtotal (I2 = 739 p = 0050)







weight

weight


Holla et al (2004)

Ustun et al (2008)


Cao et al (2005)

Holla et al (2004)

Cao et al (2006)


Holla et al (2006)

Keles et al (2006)

Pirhan et al (2008)

Holla et al (2005)

Gurkan et al (2007)


Figure 2 Continued

14 Disease Markers

175 (068 451)

065 (036 115)

100 (049 204)

089 (060 133)

102 (061 172)

124 (044 348)

106 (067 169)

085 (021 346)

167 (036 767)

116 (042 324)

063 (017 239)

010 (001 198)

039 (012 124)

1279

5789

2931

10000

8101

1899

10000

6208

3792

10000

5485

4515

10000

126 (082 195)

362 (159 825)

201 (072 561)

076 (036 162)

368 (151 902)

164 (035 770)

5578

4422

10000

5123

487

10000

Study ID OR (95 CI)

Study ID OR (95 CI)

Subtotal (I2 = 204 p = 0262)

Subtotal (I2 = 382 p = 0198)

Subtotal (I2 = 00 p = 0736)

Subtotal (I2 = 00 p = 0522)

Subtotal (I2 = 797 p = 0027)

Subtotal (I2 = 857 p = 0008)

1 9020111

1 181000554








weight

weight


Holla et al (2004)

Ustun et al (2008)


Cao et al (2006)


Cao et al (2005)

Holla et al (2006)

Keles et al (2006)

Holla et al (2004)

Pirhan et al (2008)

Holla et al (2005)

Gurkan et al (2007)


Figure 2 Continued

Disease Markers 15

106 (045 247)097 (062 150)105 (056 200)100 (072 140)

086 (056 133)062 (029 133)079 (055 116)

110 (053 227)045 (019 105)075 (043 130)

102 (065 159)092 (052 162)098 (069 139)

111 (064 192)110 (061 199)110 (074 165)

090 (054 151)044 (020 095)072 (047 110)

150358312666

10000

72332767

10000

46645336

10000

60823918

10000

53494651

10000

61083892

10000

10189 528

Subtotal (I2 = 00 p = 0969)

Subtotal (I2 = 00 p = 0467)

Subtotal (I2 = 595 p = 0116)

Subtotal (I2 = 00 p = 0778)

Subtotal (I2 = 00 p = 0989)

Subtotal (I2 = 571 p = 0127)

Study ID OR (95 CI)







weight













16 Disease Markers




4 Discussion







Disease Markers 17










18 Disease Markers



Disclosure


Competing Interests




References

















Disease Markers 19




























20 Disease Markers





























1 Introduction




2 Disease Markers










Disease Markers 3








3 Results




4 Disease Markers

Seru

m ac

tive M

MP-

9 (n

gm

L)


500

1000

1500

2000

2500p lt 001

p lt 0001

(a)

OND patients

Seru

m ac

tive M

MP-

2 (n

gm

L)

HC MS patients0

100

200

300

500

600

700

(b)

CSF

activ

e MM

P-9

(ng

mL)


025

050

075

100

200

300

400p lt 001

(c)

CSF

activ

e MM

P-2

(ng

mL)


10

20

30

40

(d)






Disease Markers 5

Seru

m IL

-17

(pg

mL)

MS patients0

200

400

600

800

1000

1500

2000

2500

OND patients

(a)

Seru

m IL

-18

(pg

mL)

0

5000

10000

15000

20000


(b)

Seru

m IL

-23

(pg

mL)

0

2000

4000

6000

800050000

100000

150000


(c)

Seru

m M

CP-3

(pg

mL)

0

50

100

150

200


(d)

CSF

IL-1

7 (p

gm

L)

0

10

20

30

40

50


(e)

CSF

IL-1

8 (p

gm

L)

0

2000

4000

6000

8000


(f)

Figure 2 Continued

6 Disease Markers

CSF

IL-2

3 (p

gm

L)

0

50

100

150

200

2000

4000


(g)

CSF

MCP

-3 (p

gm

L)

0

5

10

15

20

25


(h)








Disease Markers 7




OND () MS ()Serum






4 Discussion






8 Disease Markers






Competing Interests


Acknowledgments


References












Disease Markers 9









































1 Introduction








2 Disease Markers













Disease Markers 3











4 Disease Markers












Disease Markers 5



9 Conclusion




Competing Interests


References













6 Disease Markers































Disease Markers 7







































1 Introduction





2 Disease Markers












3 Results


Disease Markers 3

(a) (b) (c)

(d) (e) (f)






4 Disease Markers













Disease Markers 5



GenderMale 23 (793)









4 Discussion




6 Disease Markers






Competing Interests


Acknowledgments


References







Disease Markers 7



































1 Introduction




2 Disease Markers









Disease Markers 3






3 Results



4 Discussion


4 Disease Markers

Seru

m ac

tive M

MP-2

leve

ls (n

gm

L)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

0

5

10

15

20

(a)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

Seru

m ac

tive M

MP-9

leve

ls (n

gm

L)

0

5

10

15

(b)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

Seru

m T

IMP-

2le

vels

(ng

mL)

0

50

100

150

200

250

(c)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

Seru

m T

IMP-

1le

vels

(ng

mL)

0

200

400

600

800

(d)




Disease Markers 5

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

0

200

400

600

800Se

rum

activ

e MM

P-2

TIM

P-2

(times103)

(a)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

0

10

20

30

40

Seru

m ac

tive M

MP-9

TIM

P-1

(times103)

(b)

T0(n

=16)

T3(n

=19)

T6(n

=19)

T9(n

=19)

T12(n

=20)

T15(n

=18)

T18(n

=19)

T21(n

=18)

00

05

10

15

20

Seru

m ac

tive M

MP-9

act

ive M

MP-2

1 2 3

4 5

6 7

(c)




6 Disease Markers


Competing Interests


Acknowledgments


References




























Disease Markers 7






















1 Introduction





2 Disease Markers










Disease Markers 3





3 Results





4 Disease Markers

Table1MainCh

aracteris

ticso

fincludedstu

dies

Authoryear

Cou

ntry

Ethn

icity

Samples

ize

(casecontrol)

Type

ofperio

dontitis

Matchingcriteria

Genotype

metho

dGene(po

lymorph

ism)amp

HWEin

controls

deSouzae

tal2003

[31]

Brazil

Caucasian

5037

CP(m

oderateo

rsevere)

Smoker

ratio

sPC

R-RF

LPMMP-1(minus1607

1G2G)0

87

Hollaetal2004

[14]

CzechRe

public

Caucasian

133196

CP(m

ildto

mod

erate

tosevere)

Agegender

PCR-RF

LPMMP-1(minus1607

1G2G)0

52MMP-1(minus519AG

)011

MMP-1(minus422AT)0

47

Itagakietal2004

[22]

Japan

Asian

205142

CP(m

ildor

mod

erate

orsevere)

Agegendersm

oker

ratio

sTaqM

anMMP-1(minus1607

1G2G)0

48

MMP-3(minus117

15A6A)0

87

3714

2AgP

Hollaetal2005

[23]

CzechRe

public

Caucasian

149127

CP(m

ildto

mod

erate

tosevere)

Agesmoker

ratio

sPC

R-RF

LP

MMP-2(minus1575

GA

)040

MMP-2(minus1306

CT)

019

MMP-2(minus790TG)0

67

MMP-2(minus735CT)

042

Caoetal2005

[32]

China

Asian

4052

AgP

mdashPC

R-RF

LPMMP-1(minus1607

1G2G)0

78

Caoetal2006

[13]

China

Asian

6050

CP(m

oderateo

rsevere)

mdashPC

R-RF

LPMMP-1(minus1607

1G2G)0

99

Kelese

tal2006

[12]

Turkey

Caucasian

7070

CP(severe)

Agegender

PCR-RF

LPMMP-9(minus1562

CT)

082

Hollaetal2006

[24]

CzechRe

public

Caucasian

169135

CP(m

oderateo

rsevere)

Agegendersm

oker

ratio

sPC

R-RF

LPMMP-9(minus1562

CT)

059

MMP-9(+279RQ)0

25

Chen

etal2007

[16]

China

Asian

7912

8AgP

Agegender

DHPL

CPC

R-RF

LPMMP-2(minus1306

CT)

100

MMP-9(minus1562

CT)

063

Gurkanetal2007

[8]

Turkey

Caucasian

9215

7AgP

Gender

PCR-RF

LPMMP-2(minus735CT)

045

MMP-9(minus1562

CT)

035

MMP-12

(357

AsnSer)0

06

Gurkanetal2008

[25]

Turkey

Caucasian

8710

7CP

(severe)

mdashPC

R-RF

LPMMP-2(minus735CT)

043

MMP-12

(357

AsnSer)0

47

Pirhan

etal2008

[26]

Turkey

Caucasian

10298

CP(severe)

mdashPC

R-RF

LPMMP-1(minus519AG

)079

Ustu

netal2008

[27]

Turkey

Caucasian

12654

CP(m

oderateo

rsevere)

Age

PCR-RF

LPMMP-1(minus1607

1G2G)0

75

Pirhan

etal2009

[28]

Turkey

Caucasian

10298

CP(severe)

mdashPC

R-RF

LPMMP-13

(minus77

AG

)089

MMP-13

(11A

12A)0

92

Chou

etal2011[11]

China

Asian

3611

06CP

(mod

erateto

severe)

Gendersm

oker

ratio

sPC

R-RF

LPMMP-8(minus799CT)

022

9610

6AgP

Hollaetal2012

[29]

CzechRe

public

Caucasian

3412

78CP

(mild

tomod

erate

tosevere)

Agegendersm

oker

ratio

sPC

R-RF

LPMMP-8(+17

CG)0

14MMP-8(minus799CT)

063

Emingiletal2014

[30]

Turkey

Caucasian

100167

AgP

Gender

PCR-RF

LPMMP-8(+17

CG)0

29

MMP-8(minus799CT)

009

MMP-8(minus381A

G)0

87

CPchron

icperio

dontitisAgP

aggressiveperio

dontitisHWE

Hardy-W

einb

ergequilib

riumM

ildchronicperio

dontitispatie

ntsw

ithteethexhibitin

glt3m

mattachmentlossmod

eratechronicperio

dontitis

patientsw

ithteethexhibitin

gge3m

mandlt7m


chronicp

eriodo

ntitispatie

ntsw

ithteethexhibitin

gge7m


ofHWEwas

considered

significant

Disease Markers 5



Selection






Exposure




6 Disease Markers


Scre

enin

gIn

clude

dEl

igib

ility








Iden

tifica

tion


HWE n = 3







(i) Reviews n = 5








Disease Markers 7


ofthep

olym

orph

ismsinMMPs

gene

onperio

dontitisrisk

MMP-1

minus1607

1G2G

Stud

ies

(casescon

trols)

2Gversus

1GOR(95

CI)

1198682(

)119901ℎ119901119888

2G2Gversus

1G1G

OR(95

CI)

1198682(

)119901ℎ119901119888

1G2Gversus

1G1G

OR(95

CI)

1198682(

)119901ℎ119901119888

2G2Gversus

1G2G

OR(95

CI)

1198682(

)119901ℎ119901119888

1G2G+2G

2Gversus

1G1G

OR(95

CI)

1198682(

)119901ℎ119901119888

2G2Gversus

1G1G

+1G

2GOR(95

CI)

1198682(

)119901ℎ119901119888

1G2Gversus

1G1G

+2G

2G

OR(95

CI)

1198682(

)119901ℎ119901119888

Type

ofdisease

CP Caucasian

3(309287)

102(067ndash15

6)10

5(044ndash

251)

095

(064ndash

142)

090

(059ndash

137)

091

(062ndash13

2)089

(060ndash

133)

100(072ndash14

0)631

006710

0063000671000

120032

110

0000049010

00499

013610

00382019

810

0000096910

00

Asian

2(30219

2)17

5(077ndash395)

279

(056ndash

1398)

131(074ndash232

)17

5(077ndash394)

196(063ndash609)

201

(072ndash561)

079

(055ndash116)

84500111000

81500201000

27002421000

64600931000

69400711000

797

002710

0000046

710

00Ag

P

Asian

2(77194)

134(048ndash373)

154(028ndash855)

091

(042ndash19

6)16

7(038ndash731)

114(056ndash

232)

164(035

ndash770)

075

(043ndash13

0)84800101000

78400311000

00076310

0081800191000

39002001000

857

000810

00595011610

00Severe

CPCa

ucasian

Mild

tomod

erate

2(6691)

099

(063ndash15

5)099

(039

ndash250)

116(052

ndash260)

085

(039

ndash184)

110(051ndash238)

089

(043ndash18

5)117(062ndash221)

00061310

0000061310

0000066710

0000087410

0000061310

0000075110

0000087610

00

Severe

2(11091)

153(072ndash324)

244

(047ndash1259)

152(070ndash

330)

131(068ndash251)

168(081ndash350)

147(080ndash

273)

098

(056ndash

173)

657

008810

0063400981000

15802761000

00036910

00511

015310

00423018

810

0000084510

00Asian

Mild

tomod

erate

2(16819

2)12

6(093ndash17

2)16

2(084ndash

312)

140(072ndash269)

119(075ndash18

7)15

1(082ndash280)

127(083ndash19

5)097

(063ndash14

8)601

0113098

7627

010110

00438018

210

0000053410

00560013

210

0021002611000

00078410

00

Severe

2(97192)

199(092ndash426)

293

(071ndash1203)

116(053

ndash256)

219

(126ndash

379)

178(086ndash

367)

255

(095ndash685)

058

(035

ndash098)

73500520546

67000820952

00046

810

00489

01620035

381

02040854

697

0069044

1000517028

7Sm

okinghabitinCP

Caucasian

Non

smok

ing

3(200213)

092

(055ndash15

5)090

(033

ndash243)

087

(054ndash

140)

085

(052

ndash141)

096

(045ndash205)

082

(052

ndash130)

098

(066ndash

146)

661

005310

00631

006710

0034902151000

00043810

00569

009810

0040

10118

1000

00057510

00

Smok

ing

2(10974)

110(071ndash17

2)114(043ndash302)

112(054ndash

234)

115(050ndash

264

)112(055ndash229)

119(053

ndash265)

098

(053

ndash184)

19002671000

329

022210

0000097610

00522014

810

0000068210

00540014

010

0000031910

00MMP-1

minus519AG

Stud

ies

(casescon

trols)

Gversus

AOR(95

CI)

1198682(

)119901ℎ119901119888

GGversus

AA

OR(95

CI)

1198682(

)119901ℎ119901119888

AGversus

AA

OR(95

CI)

1198682(

)119901ℎ119901119888

GGversus

AGOR(95

CI)

1198682(

)119901ℎ119901119888

AG+GGversus

AA

OR(95

CI)

1198682(

)119901ℎ119901119888

GGversus

AA+AG

OR(95

CI)

1198682(

)119901ℎ119901119888

AGversus

AA+GG

OR(95

CI)

1198682(

)119901ℎ119901119888

Type

ofdisease

CP Caucasian

2(235293)

103(080ndash

132)

108(064ndash

182)

100(068ndash14

6)10

6(064ndash

175)

102(071ndash14

5)10

6(067ndash16

9)098

(069ndash

139)

00

09841000

00

08061000

00

0811

1000

00

06911000

00

08841000

00

07361000

00077810

00MMP-2

minus735CT

Stud

ies

(casescon

trols)

Tversus

COR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

CTversus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CTOR(95

CI)

1198682(

)119901ℎ119901119888

CT+TT

versus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CC+CT

OR(95

CI)

1198682(

)119901ℎ119901119888

CTversus

CC+TT

OR(95

CI)

1198682(

)119901ℎ119901119888

Type

ofdisease

CP Caucasian

2(236234)

111(079ndash155)

119(043ndash337)

112(074

ndash168)

108(037

ndash313

)10

0(067ndash14

9)116(041ndash324)

110(074

ndash165)

00069510

0000051110

0000094010

0000054110

0000069910

0000052210

0000098910

00Sm

okinghabitinCP

Caucasian

Non

smok

ing

2(13319

8)113(074

ndash172)

115(032

ndash409)

117(070ndash

194)

099

(027ndash366)

116(071ndash18

8)110(031ndash389)

115(069ndash

190)

00033710

00452017

710

0000078410

0032402241000

00053310

00424018

810

0000091710

00

8 Disease Markers

Table3Con

tinued

MMP-9

minus1562

CT

Stud

ies

(casescon

trols)

Tversus

COR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

CTversus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CTOR(95

CI)

1198682(

)119901ℎ119901119888

CT+TT

versus

CCOR(95

CI)

1198682(

)119901ℎ119901119888

TTversus

CC+CT

OR(95

CI)

1198682(

)119901ℎ119901119888

CTversus

CC+TT

OR(95

CI)

1198682(

)119901ℎ119901119888

Type

ofdisease

CP Caucasian

2(239205)

056

(024ndash

135)

036

(011ndash112

)063

(029ndash

136)

051

(015

ndash172)

057

(023ndash13

8)039

(012

ndash124)

072

(047ndash110)

78200321000

35902120553

64000961000

00045910

00739

005010

0020402620777

571

0127092

4Severe

CPCa

ucasian

Severe

2(163205)

063

(020ndash

197)

044

(013

ndash149)

071

(024ndash

209)

053

(014

ndash195)

065

(019

ndash215

)046

(014

ndash157)

075

(028ndash201)

861

000710

00544013

910

0079300281000

00042810

0084200121000

410019

310

0076000411000

MMP-1matrix


MP-2matrix


MP-9matrix


Pchronicp

eriodo

ntitisAgP

aggressiveg

eneralized

perio

dontitis



orrected

byBo

nferroni

correctio

nOR

odds

ratio

CIconfi

denceinterval


0the

rand

omeffectsmod

elisusedC

onversely

the

fixed

effectsmod

elisused

119901119888lt001

isconsidered

statisticallysig

nificant

Disease Markers 9


Holla et al (2004)

Ustun et al (2008)


Cao et al (2006)


Cao et al (2005)

Holla et al (2006)

Keles et al (2006)

Study ID

165 (090 303)

075 (055 103)

103 (065 161)

102 (067 156)

119 (087 163)

274 (157 481)

175 (077 395)

080 (048 135)

229 (124 420)

134 (048 373)

084 (055 130)

035 (017 069)

056 (024 135)

OR (95 CI)

2551

4121

3327

10000

5398

4602

10000

5118

4882

10000

5479

4521

10000

Holla et al (2004)

Pirhan et al (2008)

Holla et al (2005)

Study ID

102 (075 140)

103 (067 159)

103 (080 132)

104 (065 165)

119 (073 193)

111 (079 155)

OR (95 CI)

6619

3381

10000

5388

4612

10000

weight

weight

Subtotal (I2 = 00 p = 0695)

Subtotal (I2 = 00 p = 0984)

Subtotal (I2 = 782 p = 0032)

Subtotal (I2 = 848 p = 0010)

Subtotal (I2 = 845 p = 0011)

Subtotal (I2 = 631 p = 0067)

1 5790173

1 1930518








Gurkan et al (2007)


Figure 2 Continued

10 Disease Markers

286 (082 1001)

056 (030 107)

107 (042 273)

105 (044 251)

133 (069 257)

693 (203 2363)

279 (056 1398)

066 (023 188)

379 (114 1258)

154 (028 855)

2539

4150

3312

10000

5509

4491

10000

5148

4852

10000

104 (057 189)

121 (042 350)

108 (064 182)

087 (021 357)

175 (038 818)

120 (043 337)

061 (016 233)

008 (000 157)

036 (011 112)

7713

2287

10000

6270

3730

10000

5250

4750

10000

Study ID OR (95 CI)

Study ID OR (95 CI)

Subtotal (I2 = 359 p = 0212)

Subtotal (I2 = 00 p = 0511)

Subtotal (I2 = 00 p = 0806)

Subtotal (I2 = 784 p = 0031)

Subtotal (I2 = 815 p = 0020)

Subtotal (I2 = 630 p = 0067)

1 23600423

1 23200043








weight

weight


Holla et al (2004)

Ustun et al (2008)


Cao et al (2006)


Cao et al (2005)

Holla et al (2006)

Keles et al (2006)

Holla et al (2004)

Pirhan et al (2008)

Holla et al (2005)

Gurkan et al (2007)


Figure 2 Continued

Disease Markers 11

089 (053 148)

040 (018 087)

063 (029 136)

5713

4287

10000

54910182

Study ID OR (95 CI)

Subtotal (I2 = 640 p = 0096)

198 (063 620)079 (048 129)110 (047 254)095 (064 142)

107 (055 207)239 (074 776)131 (074 232)

082 (030 226)104 (032 337)091 (042 196)

104 (062 172)094 (053 169)100 (068 146)

110 (063 191)114 (063 206)112 (074 168)

84470792077

10000

81671833

10000

59634037

10000

55794421

10000

54274573

10000

1 7760129

Study ID OR (95 CI)

Subtotal (I2 = 120 p = 0321)

Subtotal (I2 = 270 p = 0242)

Subtotal (I2 = 00 p = 0763)

Subtotal (I2 = 00 p = 0811)

Subtotal (I2 = 00 p = 0940)








weight

weight




Holla et al (2006)

Keles et al (2006)




Figure 2 Continued

12 Disease Markers

144 (053 393)

072 (039 132)

098 (046 206)

090 (059 137)

100 (057 177)

129 (044 379)

106 (064 175)

079 (018 342)

154 (032 741)

108 (037 313)

069 (017 275)

020 (001 404)

051 (015 172)

1426

5500

3074

10000

8046

1954

10000

6129

3871

10000

6310

3690

10000

124 (078 197)

290 (121 693)

175 (077 394)

080 (036 180)

363 (138 959)

167 (038 731)

5994

4006

10000

5167

4833

10000

Study ID OR (95 CI)

Study ID OR (95 CI)

Subtotal (I2 = 00 p = 0490)

Subtotal (I2 = 00 p = 0691)

Subtotal (I2 = 00 p = 0541)

Subtotal (I2 = 00 p = 0459)

Subtotal (I2 = 646 p = 0093)

Subtotal (I2 = 818 p = 0019)

1 99200101

1 9590104








weight

weight


Holla et al (2004)

Ustun et al (2008)


Cao et al (2006)


Cao et al (2005)

Holla et al (2006)

Keles et al (2006)

Holla et al (2004)

Pirhan et al (2008)

Holla et al (2005)

Gurkan et al (2007)


Figure 2 Continued

Disease Markers 13

228 (077 669)

071 (045 114)

109 (049 241)

091 (062 132)

074 (029 191)

189 (065 551)

114 (056 232)

104 (065 166)

098 (056 172)

102 (071 145)

107 (063 183)

092 (051 166)

100 (067 149)

756

7248

1996

10000

6504

3496

10000

5777

4223

10000

5363

4637

10000


119 (064 222)

386 (127 1176)

196 (063 609)

085 (052 140)

034 (016 074)

057 (023 138)

5805

4195

10000

5539

4461

10000

Study ID OR (95 CI)

Study ID OR (95 CI)

1 6690149

1 11800851

Subtotal (I2 = 499 p = 0136)

Subtotal (I2 = 390 p = 0200)

Subtotal (I2 = 00 p = 0884)

Subtotal (I2 = 00 p = 0699)

Subtotal (I2 = 694 p = 0071)

Subtotal (I2 = 739 p = 0050)







weight

weight


Holla et al (2004)

Ustun et al (2008)


Cao et al (2005)

Holla et al (2004)

Cao et al (2006)


Holla et al (2006)

Keles et al (2006)

Pirhan et al (2008)

Holla et al (2005)

Gurkan et al (2007)


Figure 2 Continued

14 Disease Markers

175 (068 451)

065 (036 115)

100 (049 204)

089 (060 133)

102 (061 172)

124 (044 348)

106 (067 169)

085 (021 346)

167 (036 767)

116 (042 324)

063 (017 239)

010 (001 198)

039 (012 124)

1279

5789

2931

10000

8101

1899

10000

6208

3792

10000

5485

4515

10000

126 (082 195)

362 (159 825)

201 (072 561)

076 (036 162)

368 (151 902)

164 (035 770)

5578

4422

10000

5123

487

10000

Study ID OR (95 CI)

Study ID OR (95 CI)

Subtotal (I2 = 204 p = 0262)

Subtotal (I2 = 382 p = 0198)

Subtotal (I2 = 00 p = 0736)

Subtotal (I2 = 00 p = 0522)

Subtotal (I2 = 797 p = 0027)

Subtotal (I2 = 857 p = 0008)

1 9020111

1 181000554








weight

weight


Holla et al (2004)

Ustun et al (2008)


Cao et al (2006)


Cao et al (2005)

Holla et al (2006)

Keles et al (2006)

Holla et al (2004)

Pirhan et al (2008)

Holla et al (2005)

Gurkan et al (2007)


Figure 2 Continued

Disease Markers 15

106 (045 247)097 (062 150)105 (056 200)100 (072 140)

086 (056 133)062 (029 133)079 (055 116)

110 (053 227)045 (019 105)075 (043 130)

102 (065 159)092 (052 162)098 (069 139)

111 (064 192)110 (061 199)110 (074 165)

090 (054 151)044 (020 095)072 (047 110)

150358312666

10000

72332767

10000

46645336

10000

60823918

10000

53494651

10000

61083892

10000

10189 528

Subtotal (I2 = 00 p = 0969)

Subtotal (I2 = 00 p = 0467)

Subtotal (I2 = 595 p = 0116)

Subtotal (I2 = 00 p = 0778)

Subtotal (I2 = 00 p = 0989)

Subtotal (I2 = 571 p = 0127)

Study ID OR (95 CI)







weight













16 Disease Markers




4 Discussion







Disease Markers 17










18 Disease Markers



Disclosure


Competing Interests




References

















Disease Markers 19




























20 Disease Markers


















Documents

Matrix Metalloproteinases as a Pleiotropic Biomarker in ...Disease Markers Matrix Metalloproteinases as a Pleiotropic Biomarker in Medicine and Biology Guest Editors: Jacek Kurzepa,