Discussion

51  

V. Discussion

As we know, the history of MMPs is more than 50 years from the beginning of first MMP member (MMP-1) purified from the tails of tadpoles by Gross and Lapiere in 1962 (Gross and Lapiere, 1962). Many researchers designed different experiments, cloned and sequenced MMP genes, in order to figure out the structures and try to understand the

mechanisms of the big MMPs family contribute to normal physiology and disease pathology.

It is generally accepted that optimal cardiac structure and function are maintained by a tightly regulated myocardial microenvironment. ECM assembles the structural framework and provides a dynamic microenvironment in which molecular signals, including growths factor and cytokines, converge to dictate the cell’s behavior and conduct a well-coordinated heart function (Stefanidakis and Koivunen, 2006). Linask et al. (2005) documented MMP activity regulates the coordination of early heart organogenesis by affecting ventral closure of the heart and gut tubes, asymmetric cell proliferation in the dorsal mesocardium to drive looping direction, and ECM degradation within the dorsal mesocardium allowing looping to proceed toward completion. But nowadays, there is still little known about the production, secretion, and clearance of these important proteinases throughout normal growth and development in human heart.

Contreras-Ramos et al. (2008) presented that the interventricular septum has both mesenchymal and muscular components, and the mesenchymal element originates mainly from fusion of the conotruncal and atrioventricular endocardial cushions. There are at least two proposed mechanisms of the initiate development of the muscular septum. The first process is that the muscular septum forms from coalescence of the part of the ventricular wall is interposed between the enlarging free walls of the developing right and left ventricles;

therefore, as the ventricular cavities become deeper the septum grows passively inward (Goor et al., 1970). The second process is an alternative hypothesis suggesting that the muscular

52  

septum originates from a cluster of cells and the so-called primitive inter-ventricular septum, which expands actively towards the cushions of the atrioventricular canal (Srivastava and Baldwin, 2008). If some faults happened during these processes, the development of ventricular septum will be imperfection.

5-1. Gene Polymorphisms of MMP-2/-9 in VSD patients

Now, a recent PubMed search using the term ”MMP [ti]” lists more than two thousands articles within the last 5 years. To realize MMPs polymorphism, we also search the National Center for Biotechnology Information (NCBI)-SNP database, and found over 700 SNPs of MMP-2 and 300 SNPs of MMP-9 registed. The relatively important and correlative literatures were organized and shown in Appendix 1 for MMP-2 and Appendix 2 for MMP-9.

Gene polymorphism also is considered as an essential cause of VSD formation, especially the Nkx 2.5, TBX5 and GATA4 genes. SNPS of them are frequently discussed.

TBX5 is a vital gene during embryonic differentiation that affects cardiac and upper limbs development (Basson et al., 1997). A TBX5 polymorphism is also associated with ventricular septal defect in the Chinese Han population (Liu et al., 2009). There are many researches point out that NKx2.5 and GATA4 genes mutations related to the occurrences of ASD (atrial septal defect) and VSD (Matsuoka, 2007; Rajagopal et al., 2007; Zhang et al., 2009). The expanding literatures also indicate the other genes like VEGF and NFATc1 also contribute to the process of VSD formation (Gu et al., 2011; Xie et al., 2007; Zhang et al., 2009). These genes’ expressions in the heart and their interactions are necessary for proper cardiac septation (Fig. 5-1) (Srivastava and Baldwin, 2008). To the best of our knowledge, this is the first research studying on the association among MMPs SNPs and MMPs activities to the mechanism of VSD formation and closure.

53  

As we shown above, MMPs play important roles in many physiological functions and MMP polymorphisms have proven to be relative to many diseases. Among many ways to regulate the gene expression of different MMPs, a major mechanism is the sequence change in the promoter region, that is important for the transcription and causes effect in protein level and cell physiology (Vincenti and Brinckerhoff, 2007).

Therefore, two MMPs polymorphisms in promoter region were selected, MMP-2

-735C>T and MMP-9 -1562 C>T. The MMP-2 promoter polymorphism, located at nucleotide -735, destroys a Sp1-binding element and the T allele is associated with significantly

diminished promoter activity (Yu et al., 2004). The polymorphism of promoter region of MMP-9 results in losing a nuclear repressor protein binding site and decreasing MMP-9 expression as the T allele is present, and thus increasing the enzyme expression compared to the C allele (Zhang et al., 1999).

As we know, non-synonymous SNPs in the coding region of a gene produce an amino acid change, which could also affect the protein tertiary structure and as a consequence, the biophysical and biochemical activity. Hence, we also examined SNPs in MMP-9/rs17576 (Q279R) and rs2250889 (R574P). SNP of MMP-9 rs17576 (Q279R) modifies an amino acid residue within the highly conserved gelatinase-specific fibronectin type II domain (FN2) which is the catalytic domain of the MMP-9 enzyme encoding the sequence required for binding of the enzyme to its substrate. An amino acid exchange in this region of the gene might affect the binding capacities of the protein to its substrate and have functional change (Allan et al., 1995).

SNP rs2250889 locates in the C-terminal hemopexin-like domain and represents a transition of C to G at nucleotide 1740 in exon 10 of the MMP-9 and is associated with the substitution of proline for arginine at 574 residue. This is a substantial change, because proline is a cyclic nonpolar amino acid, whereas arginine is basic and positively charged. The

54  

hemopexin domain of MMP-9 is important for association with TIMP-1 and TIMP-3. In addition, the hemopexin domain of MMP-9 had a high affinity-binding site for gelatin.

Although the specific functional consequences of the substitution of proline for arginine in MMP-9 have not been studied yet, the location of SNP rs2250889 in the C-terminal hemopexin-like domain might lead to an important protein structural modification. This structural modification could either decrease the affinity of MMP-9 for its specific inhibitors, resulting in a higher level of protein activity, or enhance the affinity binding for gelatin which might increase its gelatinase activity (Rodriguez-Pla et al., 2008).

According to the basis mentioned above, we selected these SNPs to try to underlie the relationship of MMP and VSD. The results showed that genotype distribution and allele frequencies of MMP-2 -735C>T between the VSD and the control group had no significant differences. This result suggests that C>T polymorphism located at nucleotide -735 may not influence the incidence of VSD in Taiwan individuals (Table 4-3).

The results of polymorphism in the promoter of MMP-9 -1562 C>T showed no

correlation of this genotype to VSD. However, there was no VSD patient had TT genotype of MMP-9 -1562 C>T in our study. We speculate that the lack of the TT genotype of this

MMP-9 polymorphism can be explained by racial differences, which is supported by other disease studies conducted in Asian populations. There are other reports where only few or even none of the participants were homozygous (TT) for the MMP-9 -1562 C>T

polymorphism (Buss et al., 2009; Chen et al., 2010).

In the aspects of two non-synonymous MMP-9 substitutions, our data demonstrated that there is an association of the MMP-9 Q279R polymorphism and risk of VSD. There are other papers conclude the relationship of MMP-9 Q279R and many clinical diseases, such as melanoma, lumbar-disc herniation, type 2 diabetes, and pelvic organ prolapse (Ahluwalia et al., 2009; Chen et al., 2010; Cotignola et al., 2007). The importance of the R279Q

55  

polymorphism is based on its ability to change the MMP-9 protein structure and its substrate binding affinity, which is related to many diseases mentioned above. The number of subjects with the GA genotype was significantly higher in the VSD group than in the control group.

Although there were few subjects with the AA genotype in the VSD group, there was an obvious difference in the genotype distribution when compared to the control group. We found the interesting fact, and the correlation of MMP-9 Q279R non-synonymous substitution and VSD is worth to work out.

However, we didn’t find significant differences in both the genotype and allele

distribution between the VSD and control groups in Taiwanese population of MMP-9 R574P substitution, which Blankenberg et al. (2003) had presented that the G allele of MMP-9 R574P polymorphism was overrepresented in patients with histologically confirmed giant cell arteritis.

5-2. Plasma activities of MMP-2/-9 in VSD patients and their relation with spontaneous closure rate of VSD

Plasma MMP-2 activities of VSD patients were also examined by zymography,

categorized by their different genotypes (CC, CT, and TT) of MMP-2 -735 polymorphism and conducted in statistic. As mentioned earlier, the MMP-2 C>T polymorphism located at

nucleotide -735 should significantly diminish promoter activity. However, the statistical results in this study showed that there was no significant correlation between MMP-2 -735 genotypes and their activity. The result also showed that the T allele of MMP-2 -735

polymorphism has the highest MMP-2 activity. Our results were different from the results of Yu C et al. (2004) that they described C allele could enhance MMP-2 protein transcription and corresponded to higher MMP-2 activity.

56  

The polymorphism at position -1562 was expected to change the promoter activity of MMP-9 (Van den Steen et al., 2002; Zhang et al., 1999); however, our results showed no relationship between the genotypes of this polymorphism and MMP-9 promoter activity. Our results are in agreement with those of Demacq et al. (2006) who obtained plasma samples from healthy subjects.

Surprisingly, despite the different genotypes of the Q279R or R574P polymorphism, there were no significant differences in MMP-9 activity in genotypes of MMP-9 R279Q and R574P polymorphisms, indicating that these SNPs may have effects on protein structure and

substrate binding efficiency, but have no notable effect on enzyme activity.

VSD and VSD/Ao were used as the independent variables and tried to disclosure the relationship between MMP-2/-9 and different VSD severities. There is a notable finding that MMP-9 but not MMP-2 activities show the increasing tendency with different VSD defect levels. The VSD patients were further categorized into three levels, Trivial (VSD/Ao ratio  0.2), Small (0.2 < VSD/Ao  0.3 and Median (VSD/Ao > 0.3) groups, according to the examinations of 2-D echocardiography. The patients with large VSD were excluded since surgery is mandatory for such patients with poor medical control. There is a notable finding that circulating MMP-2 and MMP-9 activities, but not circulating TIMP-3 concentrations bear a positive association with the levels of VSD severity. This is the first study to specifically identify the relationships between circulating MMP-2 as well as MMP-9 and severity of VSD in children patients. In the literature search, the elevated level of amniotic fluid MMP-9 has been reported in ASD cases (Abdallah et al., 2012). However, no investigation addressing the changes of circulating MMPs in the children patients either with ASD or VSD has ever been documented.

The rate of spontaneous closure of VSD has been reported to be between 11% and 70.8%

in various researches (Ahunbay et al., 1999; Kidd et al., 1993; Mehta and Chidambaram,

57  

1992). In our study, spontaneous closure rate was detected in 17 % of perimembranous defect.

This variation may be due to study population, criteria for diagnosis, and methods of investigation, follow-up period, and the percentage of different types of VSD.

In several studies, it was proposed that ventricular septal aneurysm is an important mechanism of closure and shows a more favorable prognosis in perimembranous defects.

(Freedom et al., 1974; Ramaciotti et al., 1995). We had investigated on the circulating MMP-2 and MMP-9 activities in the VSD patients who have spontaneous closure and non-closure in this study. Although MMP-2 activity has no significant difference in the patients with different levels of VSD severity, MMP-9 activity showed significant difference related to the VSD closure. These data showed that spontaneous closure group has a higher level of circulating MMP-9 activity. Therefore, we proposed that in vivo proteases might play a role related to the spontaneous closure of VSD. It seemed to infringe the instincts; but with the expanding and growing data, some authors also suggest that MMPs may act as good candidates in helping physicians resolve some pathologic conditions (Hettiaratchi et al., 2007;

Vincenti and Brinckerhoff, 2007).

We expect great difficulties in striving to realize the interaction between MMP family and mechanism of VSD because of their intricate and complicated roles in physiology and pathogenesis. There remains an urgent need to study further on this matter.

5-3. Correlations between plasma MMP-2/-9 activities, concentrations of TIMP-3 and BNP

For the purpose of exporting the relationships between MMP-2/-9 and TIMP; therefore, the activity of circulating MMP-2/-9 and the TIMP-3 concentrations were measured in VSD patients. Although there were no differences in the activity of circulating MMP-2/ -9 among the genotypes of each polymorphism, TIMP-3 concentrations were higher in subjects with the

58  

CC genotype of MMP-2 -735 C>T, and QQ genotype of the MMP-9 R279Q polymorphism.

We used Pearson’s correlation for further inquiries. Plasma MMP-9 activity had positive correlation with TIMP-3 concentration.

TIMPs regulate the activity of zinc metalloproteases, and that TIMP-3 can form a tight complex with proMMP-2 and proMMP-9 to regulate gelatinase activities in vivo TIMP-3 also has unique characteristics of ECM-binding; it is highly expressed in the heart and involved in maintaining cardiac structure and function (Nagase and Woessner, 1999). In addition, MMP-9 and TIMP-3 are typical inducible ECM metabolic proteins, and the damage caused by

ischemia-reperfusion injury may induce the synthesis of MMP-9 and TIMP-3 (Murphy et al., 1986). In heart diseases, TIMP-3 may contribute to the regulation of myocardial remodeling, it deficiency disrupts matrix homeostasis and causes spontaneous left ventricular dilation, cardiomyocyte hypertrophy and contractile dysfunction (Fedak et al., 2003). Our published paper proposed that the interaction between MMP-2, MMP-9 and TIMP-3 may contribute to atrial ECM remodeling of atrial fibrillation (Yong and Guoping, 2008). In addition, many studies also appear that the association of MMP-2, MMP-9 and TIMP-3 in heart diseases (Givvimani et al., 2010), heart remodeling (Lin et al., 2010; Yeh et al., 2013) and cardiocytes function properly (Manso et al., 2010). Although there is no former report investigating the relationship of circulating MMP-2, MMP-9 activity and TIMP-3 concentration in VSD patients, our report provides first insights into the roles of MMP-9 and its inhibitor in the complex network of ECM in VSD patients.

It is worth mentioning that BNP is secreted mainly in the ventricles in response to increased wall stress and plays important roles in maintaining cardiorenal homeostasis, which is primarily mediated by the second messenger cyclic guanosine monophosphate (Lee et al., 1997). Although the production of BNP and its role as local regulator of ventricular fibrosis are well established, the interaction of BNP in the regulation of the cardiac interstitium

59  

remains unclear (Collier et al., 2001). Regarding the correlations observed in our work, we found not only that there is cross talk between MMP-9 and BNP but also both of them have a similar relationship with the severity of VSD. Interestingly, we can’t find such a trend

between MMP-2 and BNP. Furthermore, a previous study showed that gene expression and zymographic MMP-9 activity were significantly higher in BNP-transgenic mice than in non-transgenic mice (Sopata and Wize, 1979). These investigators suggested that transient MMP-9 expression induced by elevated levels of BNP during the earliest phase after

myocardial infarction is a cardioprotective mechanism. On the basis of these findings and our results, we speculate that MMP-9 and BNP may act together in the pathogenesis of VSD.

Some limitations of our study merit considerations. First, our sample size is fairly modest, which would primarily affect the power of our analysis to detect significant

associations. This is attributed to the difficulty in collecting children's blood samples. Second, patients were recruited from only one hospital system and may not be representative of the general population. However, we used incident cases to lessen the selection bias. Nevertheless, it would be important to confirm these findings in multicenter studies. Third, it is really a difficult task to disentangle the role of each gelatinase within a single study because of its complexity of regulation in biological network.

Last, the echocardiography studies of healthy controls and the patients enrolled were not carried out in this present study, which should be undertaken optimized pharmacological treatment and that may cause an effect on circulating MMP-9 activity and TIMP-3

concentration.

60  

Figure 5-1. Pathways of regulating region specific cardiac morphogenesis. FhF, first heart field; SHF, second heart field (Srivastava and Baldwin, 2008).

61