MMP1-1607 polymorphism increases the risk for periapical lesion development through the upregulation MMP-1 expression in association with pro-inflammatory milieu elements

ABSTRACT Increased matrix metalloproteinases (MMPs) activity is a hallmark of periapical granulomas. However, the factors underlying the MMPs expression modulation in healthy and diseased periapical tissues remains to be determined. Objective In this study, we evaluated the association between the MMP1-1607 polymorphism (rs1799750) and pro-inflammatory milieu elements with MMP-1 mRNA levels in vivo. Material and Methods MMP1-1607 SNP and the mRNA levels of MMP-1, TNF-a, IFN-g, IL-17A, IL-21, IL-10, IL-4, IL-9, and FOXp3 were determined via RealTimePCR in DNA/RNA samples from patients presenting periapical granulomas (N=111, for both genotyping and expression analysis) and control subjects (N=214 for genotyping and N=26 for expression analysis). The Shapiro-Wilk, Fisher, Pearson, Chi-square ordinal least squares regression tests were used for data analysis (p<0.05 was considered statistically significant). Results The MMP1-1607 1G/2G and 1G/2G+2G/2G genotypes were significantly more prevalent in the patients than in controls, comprising a risk factor for periapical lesions development. MMP-1 mRNA levels were higher in periapical lesions than in healthy periodontal ligament samples, as well as higher in active than in inactive lesions. The polymorphic allele 2G carriers presented a significantly higher MMP-1 mRNA expression when compared with the 1G/1G genotype group. The ordered logistic regression demonstrated a significant correlation between the genetic polymorphism and the expression levels of MMP-1. Additionally, the pro- and anti-inflammatory cytokines IL-17A, IFN-g, TNF-a, IL-21, IL-10, IL-9, and IL-4 were significant as complementary explanatory variables of MMP-1 expression. Conclusion The MMP1-1607 SNP was identified as a risk factor for periapical lesions development, possibly due to its association with increased MMP-1 mRNA levels in periapical lesions. The MMP-1 expression is also under the control of the inflammatory milieu elements, being the cytokines TNF-a, IL-21, IL-17A, and IFN-g associated with increased MMP-1 levels in periapical lesions, while IL-10, IL-9, or IL-4 presented an inverse association.


INTRODUCTION
Periapical lesions are characterized by the destruction of periapical tissues as a consequence of the local host response, which is triggered by bacterial infection of pulpal and endodontic environment 18, 30 . Among multiple elements involved ytokines are critical determinants of lesions outcome acting as major modulators of the main pathways responsible for tissue destruction 13,18 . While the RANK-RANKL-OPG system, the main responsible for osteoclastogenesis control, is supposed to be one determinant of periapical lesions activity, the balance between MMPs (matrix metalloproteinases, a family of zinc-and calcium-dependent proteases) and TIMPs (the endogenous tissue inhibitors of metalloproteinases) is supposed to determine the outcome of the protease-mediated catabolic response that contributes to the degradation of soft and mineralized tissues surrounding the root apex 18, 19,26,34, . In physiological circumstances, MMPs play a role in degradation and remodeling of both extracellular and bone matrix proteins. However, the increase in the expression and/or activation of MMPs without a parallel increase of TIMPs to counteract catabolic proteolysis mediate numerous pathological processes, including the periapical and periodontal diseases 6,23,31 . Indeed, the high matrix proteinase activity is a hallmark of periapical granulomas, and is supposed to derive from a combined action of multiple MMPs such as MMP-1, MMP-2, MMP-8, MMP-9, and MMP-13 1,28 .
Within the MMPs found to be overexpressed in contribute to lesion development by mediating since MMP-1 is the main proteolytic enzyme that tissue matrix most abundant components. Indeed, while MMP-1 levels are relatively low in healthy periapical tissues, its levels are upregulated in human and experimental periapical lesions 12, 20 .
It is important to mention that a high individual variation in the levels of MMP-1 was described in periapical lesions 29 in MMP-1 levels may directly influence the of the factors that determine the control of MMP-1 expression in periapical environment may comprise important information to predict and prevent lesions development as well to develop therapeutic strategies to limit lesions progression. Variations in the promoter region of MMP1 gene account for heritable differences in MMPthe susceptibility to certain pathologies 4,29,32 . A -1607 position of MMP1 gene promoter (MMP1-1607, rs:1799750) is particularly associated with increased MMP-1 mRNA transcriptional activity 29, 32 . While previous studies suggest that genetic variations in MMP-2 and MMP-3 genes account for increased risk to periapical lesion formation 26 , the possible association of MMP-1 genetic variants was not investigated.
background, it is also important to consider that different cytokines may also account for the variance of MMP-1 levels in periapical tissues 2 . Indeed, strong and persistent microbial and genetic control of MMP-1 expression in periodontitis context 29 . Accordingly , are described to directly increase the levels of MMPto counteract such effect and to downregulate MMPs expression 8,14 . Interestingly, a marked dichotomy clusters is associated with periapical activity/ inactivity status 3 , suggesting that such mediators could in fact determine lesions activity via the regulation of MMPs expression 3 .
Therefore, in the present study we investigated the possible association of MMP1-1607 SNP with periapical lesions development risk, and simultaneously analyzed the impact of such polymorphism and host response mediators (TNF, IL-21, IL17, IFN-g, IL-10, FOXp3, IL-9, and IL-4) in the modulation of MMP-1 mRNA transcription in human chronic periapical granulomas.

Subjects and samples
This study had institutional review board approval of Bauru School of Dentistry, University of São Paulo. Patients and control subjects were selected as previously described 3 . Patients presenting periapical lesions were referred to endodontic surgery after conventional root canal treatment failure; periapical lesions diagnosis was performed as previously described 25 based on histopathological and radiographic analysis, being periapical lesions characterized radiographically as rarefaction lesions with the disappearance of the periodontal ligament space and discontinuity of the presence of periradicular radiolucency that did not resolve, persisting as before acceptable endodontic treatment (i.e., having all canals instrumented and obturated, with no voids in the obturation mass, the apical terminus of the obturation at 1/1.5 mm from the radiographic apex), or that increased in size with evidences of continuous bone resorption) 2 . elements Periapical granulomas (N=111) were collected from patients (N=111, aged 19-59 years; 51 females and 59 males) during periapical surgery and divided in two roughly similar fragments and stored in both formalin (for routine histological examination performed after hematoxylin-eosin staining) and RNAlater solution (Ambion -Thermo Waltham, Massachusetts, USA) (for molecular analysis). Test samples were macrophages, and without the presence of an epithelial lining. Periapical cysts, where cavities squamous epithelium, and partially epithelized lesions (epithelized granulomas) were excluded from the study. Periapical lesions were also categorized into putative active and inactive, based in the molecular profile of RANKL/OPG mRNA expression, as previously described 24 . Patients' epithelial buccal cells were sampled from inner cheek buccal mucosa scrapping after a mouthwash with 3% glucose for genotyping purposes.
The control group (N=214) (patients aged 18-52 years; 103 females and 111 males) was comprised of subjects without a clinical history of pulpal or periapical pathology and were free from periapical lesions. Epithelial buccal cells of control subjects were sampled from inner cheek buccal mucosa scrapping after a mouthwash with 3% glucose for genotyping purposes. A subgroup of control subjects (N=26) (patients aged 18-35 years; 13 females and 11 males) was comprised of subjects scheduled for premolar extraction due to orthodontic purposes; in addition to epithelial buccal cells sampling, during the surgery periapical tissue samples were isolated, stored in RNA later solution and used as control specimens.
Patients and controls with medical conditions metabolism or other assisted drug therapy (i.e. therapy) during the last six months before the study were excluded. Patients and controls with preexisting conditions, such as periodontal disease and pregnant or lactating women, were also excluded.

DNA extraction and analysis of MMP1-1607 SNP (rs1799750)
DNA was extracted from epithelial buccal cells with sequential phenol/chloroform solution as previously described 11 . Extracted DNA was used for genotyping. DNA integrity was checked and the allelic discrimination of MMP1-1607 SNP Inc.; Waltham, Massachusetts, USA) chemistry as previously described 21 . For reaction quality control, a sample of known genotype was included in the plate and a no DNA template sample was included as negative control. Only genotypes with an automatic call rate >95% were considered, error rate was <3%. Samples that failed to provide a genotype were repeated in additional reactions; genotyping was performed blinded to group status, as previously described 7 .

RNA extraction and RealTime-PCR
In brief, total RNA was extracted from samples by using the RNeasy kit (Qiagen Inc, Valencia, California, USA) according to the manufacturers' instructions. The integrity of RNA samples was Bioanalyzer (Agilent Technologies, Santa Clara, California, USA) according to the manufacturers' instructions. After RNA extraction, complementary a reverse transcription reaction using QuantiTectRT kit (Qiagen Inc, Valencia, California, USA). All cytokines/Th markers (TNF-IL-17A, IL-21, IL-10, IL-4, IL-9, FOXp3) mRNA levels were measured by means of RealTimePCR using TaqMan Waltham, Massachusetts, USA) in a Viia7 instrument (Thermo Waltham, Massachusetts, USA) using inventoried optimized primers/probes sets (Invitrogen, Carlsbad, CA), with basic reaction conditions (40 cycles) 95°C (10'), 94°C (1'), 56°C (1'), and 72°C (2'). The analysis of RANKL and OPG mRNA levels were also determined in all the lesions (also by RealTimePCR using TaqMan chemistry), in order to categorize each sample in putative active and inactive lesions based on the RANKL/OPG ratio as previously described 24 . The results are depicted as the relative level of gene expression; calculated expression in each sample using the 2 Ct method.

Data analysis
The Shapiro-Wilk test was performed to test the distribution of all test groups prior to comparative analysis; p>0.05 was considered indicative of normal distribution. Chi-squared test with 1 degree of freedom was performed to test the Hardy-Weinberg equilibrium in the allele frequencies of the study population. The differences in the demographic data for the study population and genotype and allele distribution among groups were tested by Fisher's exact test and the equality of proportions test. Correlations between the MMP1-1607 genotype and the mRNA expression of pro , IL-17A, IL-21, and IL-10) biomarkers was assayed by the Pearson's product-moment correlation coefficient. The interrelation between the MMP1-1607 genotype and the MMP-1 mRNA levels was assayed by ordered logistic regression.

MMP1 -1607 SNP frequency analysis
The subject sample included in this study was similarly composed by male and female subjects ( Table 1). The frequency of MMP1-1607 SNP genotypes and alleles in control group was similar to that previously reported for the Brazilian population 24 and the Chi-square test for the distribution of genotypes in patients and controls rendered a p-value of 0.469 and 0.243, respectively, which was compatible with the Hardy-Weinberg equilibrium. When the patients presenting periapical lesions were compared with the controls, it was observed that the frequency of MMP1-1607 1G/2G was higher in the patients group (p=0.0110, OR=2.143, Cl=1.209 to 3.799), as well the combined mutant genotypes 1G/2G+2G/2G was more frequent in the patients (p=0.0326, OR=1.764, Cl=1.047 to 2.971) group than in controls (Table 1). When the frequency of the alleles was compared between the patients and controls groups, no differences were found. The frequency of the MMP1-1607 SNP genotypes and alleles was also compared within patient's presenting active and inactive periapical found in the frequencies (Table 2).

Association between MMP1-1607 with MMP-1 mRNA expression
Our data initially showed a weak MMP-1 expression in healthy periapical tissues sampled expression was evidenced in the periapical lesions harvested from patients ( Figure 1). In addition, our data demonstrate that MMP-1 mRNA levels lesions (Figure 1). When analyzing the possible association between MMP-1 mRNA levels and the MMP1-1607 SNP genotypes in control group, we found that polymorphic allele 2G (1G/2G+2G/2G MMP-1 mRNA expression when compared with the 1G/1G genotype group (Figure 1). Similar results were observed within the periapical lesions, where the carriers of polymorphic allele 2G (1G/2G+2G/2G MMP-1 mRNA expression than 1G/1G genotype carriers (Figure 1). When the periapical lesions were evaluated after stratification by lesion activity/inactivity status, it was observed that the expression of MMP-1 was higher in 1G/2G+2G/2G versus 1G/1G genotypes in both subgroups ( Figure  1). Additionally, the ordered logistic regression 1607 SNP over the mRNA MMP-1 expression levels, demonstrated by a likelihood ratio (LR) chi-squared of 6.09 (probability of > chi-squared=0.0136).

Regression analysis
Since MMP1-1607 and multiple cytokines were found to be associated with MMP-1 mRNA levels in periapical lesions, we subsequently performed a OLS regression analysis to model the relative contribution of the MMP1-1607 SNP and each biomarker to the overall MMP-1 regulation    (Table 3 and Figure 3).

DISCUSSION
The increased expression of proteolytic enzymes, such as MMP-1, is supposed to contribute to the onset and progression of periapical lesions by promoting the degradation of soft and mineralized tissues surrounding the tooth apex. In this context, the control of MMP-1 expression in periapical environment may comprise important information to predict lesions development or to develop therapeutic strategies to limit lesion's progression.
Our results initially demonstrate that MMP-1 mRNA levels were increased in periapical lesions when compared with healthy periapical tissues, in accordance with previous reports 1,31 . In addition, higher MMP-1 expression than inactive lesions, reinforcing the hypothesis that MMP-1 in fact contribute to lesions maintenance or progression. Accordingly, previous studies demonstrate that active lesions present a reduced expression of healing markers when compared with inactive 15,23 . It is important to highlight that the overall MMP-1 expression in the periapical lesions, despite its activity status, was evidently variable. Therefore, our next step was to investigate if a genetic variation in the MMP-1 gene promoter, namely MMP1-1607 SNP, previously described as a regulator of MMP-1 expression levels 29 , could account for the MMP-1 mRNA modulation in periapical lesions as well as for an increased risk for lesions development.
Regarding the genetic association point of view, our results demonstrate a higher frequency of the 1G/2G genotype, as well as of the combined 1G/2G+2G/2G polymorphic genotypes, in the subjects presenting periapical lesions than in the control group. Therefore, such genotypes add a develop periapical lesions. Interestingly, a previous study demonstrate the association of MMP-2 and MMP-3 variants with periapical lesion formation in individuals with untreated deep carious lesions 26 , suggesting that individuals genetically prone to increased production of proteolytic enzymes may experience an increased risk for periapical lesions. It is important to consider that the increased risk observed in this study may be underestimated, since the control population does not comprise individuals strictly exposed to the major risk factors for periapical lesions such as untreated deep carious lesions 9,26 . While the selection of different control populations may in fact impact the power and odds of genetic studies 16 a given risk factor, such as MMP1-1607 SNP, in a case-control setting that theoretically minimizes the changes of association, reinforces that this factor in fact comprises a risk element 9,15,26 . However, no differences were observed in the frequency of the MMP1-1607 genotypes and alleles in the patients presenting active or inactive lesions; however, the limited power of such fragmented analysis due to the dilution of the experimental sample within two After the identification of positive genetic association of MMP1-1607 with periapical lesion's risk, we next evaluated from the functional viewpoint if the MMP-1 genetic variants were in fact associated with MMP-1 transcripts levels in the lesions. In fact, our results demonstrate that the MMP1-1607 2G allele was associated with increased MMP-1 mRNA expression in both healthy periapical tissues and in the periapical lesions. Accordingly, such polymorphism was previously described to affect gene transcription in vitro, since the 2G allele, together with an adjacent adenosine, creates a core binding site (5´-GGA-3´) for transcription factors immediately adjacent to an AP-1 site, causing a 22 .
In addition, the ordered logistic regression analysis demonstrate that the MMP1-1607 SNP is tissues (LR chi-squared of 6.09, probability of > chi-squared=0.0136). Unfortunately, the ordered nature of the genetic data precludes the computation of an R-squared capable of predicting the relative effect of the polymorphism on the MMP-1 expression variation, nevertheless the sequential iterative computation of OLS regressions including and excluding the polymorphism as explanatory variable in various models (including incrementally all other explanatory variables) points to a contribution ranging between 2 and 5% (data not shown). This seemingly weak effect must be interpreted carefully, since OLS models are a the context of biological processes. More important than the computed predictive value is the fact that the genetic polymorphism demonstrated a majority of the regressions computed.
It is compulsory to consider that numerous SNPs are in fact supposed to contribute to a given biological effect. Interestingly, in a previous study our group described the association between MMP1-1607 and the MMP-1 mRNA levels in healthy periodontal tissues, while in periodontal lesions the seems to overcome the genetic predisposition to higher MMP-1 29 . In this context, since periapical previously associated with periapical lesions activity 3 could contribute to the modulation of MMP-1 expression levels. Our results demonstrate that the cytokines associated with lesion activity clusters, namely TNF-, IL-21, IL-17A, and IFN-g were positively correlated with MMP-1 levels upon individual linear regression analyzes. Accordingly, all this cytokines were previously described as positive regulators of MMP-1 expression in different models 8 . When the OLS regression analysis was applied to consider the occurrence of the SNP and the level of expression of cytokines simultaneously, IL-17A and of 0.3897 and 0.3261, respectively. Interestingly, the cytokine cluster associated with the highest degree of periapical lesions activity 3 , but were also individually implicated with lesions' activity in the clusters ranked in 2 and 3 positions 3 , reinforcing that their involvement in lesions' progression may in fact involve the upregulation of MMP-1. The OLS regression analysis also demonstrated that 21 were described as components of the top 3 ranked cytokine clusters associated with periapical lesions activity 3 . It is mandatory to mention that the OLS regression computed simultaneously the explanatory effect of the cytokines and the MMP1-1607 SNP over the overall regulation of MMP-1 levels in periapical lesions.
On the other hand, the cytokines prevalent in clusters associated with lesions inactivity, namely IL-10, IL-9, and IL-4 3 , were found to be inversely correlated with MMP-1 levels in periapical lesions upon individual linear regression analyzes. Similarly, the OLS regression analysis demonstrated -0.2552, -0.6067, and -0.4059, respectively). While some studies demonstrate that IL-10 can block MMPs production (especially MT1-MMP, MMP-1, and MMP-9) by tumoral cells 33 , there are no direct evidences of the direct inhibition of MMP-1 by IL-10, IL-9, or IL-4. However, it is possible to hypothesize that the negative correlation is derived properties of such cytokines, especially the most studied IL-10, result in the downregulation of the cytokines responsible for MMP-1 upregulation, such as TNF- 27 . It is also interesting to explore the lack of correlation between FOXp3 and MMP-1 expression in both linear and OLS regression analysis. FOXp3 is the prototypic transcription factor of Tregs (regulatory T cells), a T cell subset with potent immunoregulatory properties, described to attenuate experimental periodontal and periapical lesions progression in vivo 11,17 . In human periapical lesions, the expression of FOXp3 as well as its methylation levels are indicators of lesions inactivity status 2 . In active periapical lesions, increased FOXp3 methylation was associated with reduced FOXp3 mRNA expression, as well as reduced levels of IL-10, a characteristic immunoregulatory product of Tregs 5 . However, since FOXp3 is a regulator of Tregs biology, which in turn are supposed to (as well as via other immunoregulatory molecules such as TGF-b and CTLA-4), the existence of multiple elements in the immunoregulatory cascade between FOXp3 and MMP-1 (i.e., in a hypothetical oversimplified pathway, FOXp3 regulates the development of Tregs, which produce IL-10, which in turn can downregulate the levels of MMP-1 expression) 5 may account for the lack of correlation between such factors. Indeed, while in vivo evidences link Tregs with the modulation of events coordinated by MMPs and TIMPs, such as current evidences points to an indirect regulation of MMPs/TIMPs balance derived from cytokines produced and regulated by Tregs instead of a direct regulation 17 . Additionally, it is noteworthy that computed regression models fail to capture the subtle and intermingled regulatory interdependence between multiple explanatory variables and that the simultaneous inclusion of ordinal and continuous data in the OLS tends to underestimate the relative contribution of the continuous variables.
Taken together, our results demonstrate that elements multiple host factors seem to contribute to the local modulation of MMP-1 levels in periapical lesions. The MMP1-1607 SNP was demonstrated to be functional; being the 2G genotypes associated with as a risk factor for periapical lesions development in the genotypic case-control association analysis. In addition, the cytokines TNF-, IL-21, IL-17A, and MMP-1 levels in periapical lesions, while IL-10, IL-9, or IL-22 presented an inverse association. Still, further cause-and-effect studies are required to dissect the intracellular networks involved in the regulation of MMP-1 levels upon multiple and different genetic backgrounds scenarios. Such studies may support (or discard) the application of MMP1-1607 SNP as biomarker for periapical lesions risk, as well to develop strategies to modulate MMP-1 levels in the clinical practice, and therefore contribute to improve the diagnosis and clinical management of these pathologies.

CONCLUSION
factor for periapical lesions development as well as to be associated with increased MMP-1 mRNA expression in periapical tissues and lesions. The expression of MMP-1 is also under the control of being the cytokines TNF-, IL-21, IL-17A, and IFN-g associated with increased MMP-1 levels in periapical lesions, while IL-10, IL-9, or IL-4 presented a negative correlation with MMP-1 expression.