Statistical comparisons were performed using SPSS software (SPSS Inc.). Correlation between MMPs and TIMPs expression in RT-PCR analysis was performed using the paired sample t test. Densitometric analyses of the zymograms, Western blots and RT-PCR were
analyzed by on-way ANOVA with a LSD post-hoc test. Data are means ± standard deviation (SD), and statistical significance was considered if p < 0.05.
III. Results:
3-1. Induction MMP-9 in the atrial with AF
Sustained AF model in pigs has been established successfully in our laboratory[Lin et al., 2003; Hsu, 2003]. The electrophysiological and pathological findings in the porcine atria with AF showed the characteristics resembling those in human. Therefore, we have recognized a suitable animal model of sustained AF in adult pigs for studying human AF.
AF is characterized by structural remodeling on cytoskeleton and ECM proteins[Lai et al., 2004; Lin et al., 2005]; however, the underlying molecular mechanisms involved ECM remodeling are largely unknown. To get a better insight into the ECM remodeling at molecular level, we analyzed the mRNA and activity changes in ECM metabolic enzymes, MMPs and TIMPs, in the fibrillating atria. Moreover, we attempt to test hypothesis whether the abnormal expressed MMPs are associated with the disease.
3-1-1. Histological findings in the atria with AF
Histological studies were conducted to identify the potential pathological substrate underlying the abnormalities in sustained AF. Atrial tissues from the AF subjects revealed that disordered or denatured cardiomyocytes were associated with extensive myocytolysis, thinning bundles and fragmentation of myofibrils, and large space within myofibrils[Hsu, 2003]. Masson’s trichrome staining showed that myocardial fibrosis markedly accumulated in interstitial cardiomyocytes in the AF atria(Figure 1A and 1B). The extent of fibrosis (%F) was measured and the result showed that %F of the atrial appendage in the AF group (21.6 ± 3.3%) was significantly greaten than that in the SR group (5.2 ± 1.7%, p < 0.01) (Figure 1C).
3-1-2. Collagen type I protein in the atria with AF
According the pathological results of Masson’s trichrome staining, ECM was significantly accumulated interstitial space in the atria with AF. Therefore, we were interesting to measure the collagen level in the tissue, because collagen was a major
component of ECM. The collagen type I in the AF (n = 14) and SR (n = 12) subjects was determined to Western blotting assay used the-actin as internal control(Figure 2A). The content of collagen typeⅠ(collagen/-actin) in the AF group (1.79 ± 0.14, ranging from 1.51 to 2.03) was significantly increased as compared with that in the SR group (1.17 ± 0.26, ranging from 0.66 to 1.51) ( p < 0.01) (Figure 2B).
3-1-3. Enzyme activities of MMP-2 and MMP-9 in the fibrillating atria
The atria tissue isolated from hearts with AF (n = 14) and SR (n = 12) were used to detect the enzyme activity of two gelatinases, MMP-2 (gelatinase A; 72 kDa) and MMP-9 (gelatinase B; 92 kDa) by zymographic analysis(Figure 3A). As shown inFigure 3A, clear proteolytic bands on the gelatin zymographicy gels shows increased MMP-9 in the AF
subjects as compared with in the SR. The relative MMP-9 activity in the AF group was 1806 ± 753, ranging from 786 to 2809; the relative MMP-9 activity in the SR group was 254
± 150, ranging from 100 to 627(Figure 3B). The result showed that MMP-9 activity in the AF was higher than that in the SR by 7.1 folds (p < 0.01). In the MMP-2, relative activity of this enzyme in the AF group was 923 ± 356, ranging from 496 to 1586); the relative MMP-2 activity in the SR group was 932 ± 402, ranging from 258 to 1498(Figure 3C). The enzyme activity of MMP-2 in the subjects with of with or without AF was insiginificantly different.
3-1-4. MMP-2 and MMP-9 mRNA levels in AF
According the results of Zymography, the markedly MMP-9 activity was observed in the fibrillating atria. Further, we were interesting to measure the MMP-2 and MMP-9 mRNA levels in the atria, because MMP family was a major protease in the ECM metabolism. The transcript (i.e., mRNA) of MMP-2 and MMP-9 in the atrial tissues with AF and SR were determined by the method of semiquantitative RT-PCR(Figure 4A). In this analysis, the expression of GAPDH mRNA was used as internal control and the relative MMP mRNA was represented as MMP/GAPDH mRNA detected. The relative mRNA level of MMP-9 in the AF group was 0.43 ± 0.18 (ranging from 0.16 to 0.49) and in the SR group was 0.0.10 ± 0.03 (ranging from 0.05 to 0.13)(Figure 4B). The results showed that the relative mRNA level of MMP-2 was 0.90 ± 0.22 (ranging from 0.53 to 1.13) in the AF group and 0.82 ± 0.23 (ranging from 0.50 to 1.11) in the SR group(Figure 4C). The statistical result shows that MMP-9 mRNA level in the AF was higher than that in the SR by 4.3 folds ( p < 0.01); but mRNA level of MMP-2 in the SR was not significantly different.
3-1-5. TIMP-1 and TIMP-2 mRNA levels in AF
Several studies have demonstrated the pathological effects of both MMPs and TIMPs in cardiovascular diseases involved vascular remodeling and cardiac remodeling in congestive heart failure or myocardial infarction[Hijova, 2005]. Therefore, the relative abundances of TIMP-1 and TIMP-2 in the atrial tissues were also determined by semiquantitative RT-PCR analysis(Figure 5A). In this analysis, the expression of GAPDH mRNA was used as internal control and the relative TIMP mRNA was represented as TIMP/GAPDH mRNA detected. The relative mRNA level of TIMP-1 in the AF group was 0.75 ± 0.33 (ranging from 0.35 to 1.22) and in the SR group was 0.88 ± 0.41 (ranging from 0.35 to 1.70)(Figure
5B). The results showed that the relative mRNA level of TIMP-2 was 1.07 ± 0.45 (ranging from 0.52 to 1.63) in the AF group and 0.84 ± 0.26 (ranging from 0.52 to 1.28) in the SR group(Figure 5C). There were insignificantly different between the two groups on TIMP-1 and TIMP-2.
3-2. MMP-2 expression in cardiac fibroblasts treated with Ang II and Ang-(1-7)
Alterations in the normally circulating Ang II and/or Ang-(1-7) concentrations and the ratio of Ang II and Ang-(1-7) might reflected cardiac remodeling in response to physiologic stresses[Schupp et al., 2005]. We were interesting to understand the effects on the MMPs and TIMPs expression in cardiovascular system under abnormal conditions of Ang II and Ang-(1-7). In the following experiments, a cardiac fibroblast cell line H9c2 was used to treat with serious concentrations (10-9, 10-7and 10-5M) of Ang II and Ang-(1-7), respectively.
The mRNA levels of collagen type I, MMP-2, MMP-9, TIMP-1 and TIMP-2, and enzyme activity of MMP-2 as well as MMP-9 in the angiotensin peptides treated cells were detected and compared.
3-2-1. Collagen typeⅠmRNA in the H9c2 treated with Ang II and Ang-(1-7)
The relative collagen type I expression in control, 10-9, 10-7and 10-5M of Ang II treated H9c2 cells was 0.54 ± 0.16, 0.52 ± 0.12, 0.56 ± 0.08, and 0.73 ± 0.10, respectively. In the same condition, the cells were treated with control, 10-9, 10-7and 10-5M of Ang(1-7) which was 0.54 ± 0.16, 0.59 ± 0.22, 0.93 ± 0.10 and 0.63 ± 0.14, respectively(Figure 7A). In this analysis, the expression of GAPDH mRNA was used as internal control and the relative collagen type I mRNA was represented as collagen type I/GAPDH mRNA detected. The
significant data showed 0.92 ± 0.10 in 10-7M of Ang-(1-7); but, the other concentrations were insignificantly different, whether Ang II or Ang-(1-7) treatments(Figure 7B).
3-2-2. MMP-2 activity in H9c2 cells treated with Ang Ⅱ and Ang -(1-7)
The condition medium isolated from Ang II and Ang-(1-7)-treated in H9c2 cells were used to detect the enzyme activity of gelatinase, MMP-2 (gelatinase A; 72 kDa) by
zymographic analysis(Figure 8A, C). The relative MMP-2 activity in control, 10-9, 10-7and 10-5M of Ang II were 3257 ± 184, 1943 ± 512, 1603 ± 334, and 1737 ± 357, respectively.
In the same condition, the cells were treated with control, 10-9, 10-7and 10-5M of Ang-(1-7) which was 3257 ± 184, 1219 ± 261, 1477 ± 73 and 2140 ± 247, respectively. The enzyme activities were significantly decreased in all concentrations of Ang II(Figure 8B)and Ang-(1-7)(Figure 8D); in Ang-(1-7)-treated, the enzyme activities had dose-dependent, but not in Ang II-treated.
3-2-3. MMPs and TIMPs mRNA levels in the H9c2 cells treated with Ang Ⅱ and Ang-(1-7)
The relative MMP-2 expression in control, 10-9, 10-7and 10-5M of Ang II treated H9c2 cells was 0.72 ± 0.29, 0.71 ± 0.07, 0.67 ± 0.34, and 0.69 ± 0.36, respectively; in MMP-9 expression, the values were 0.40 ± 0.20, 0.34 ± 0.22, 0.25 ± 0.14, and 0.32 ± 0.11,
respectively; in TIMP-1 expression, the values were 0.81 ± 0.23, 0.88 ± 0.15, 0.96 ± 0.17, and 0.94 ± 0.20, respectively; in TIMP2 expression, the values were 0.80 ± 0.35, 0.80 ± 0.20, 0.92
± 0.15, and 0.94 ± 0.12, respectively. In the same condition, the MMP-2 activity in cells treated with control, 10-9, 10-7and 10-5M of Ang(1-7) which was 0.72 ± 0.29, 0.80 ± 0.16, 0.85 ± 0.38, and 0.69 ± 0.18, respectively; in MMP-9 expression, the values were 0.40 ± 0.20,
0.38 ± 0.49, 0.70 ± 0.40 and 0.39 ± 0.19, respectively; in TIMP-1 expression, the values were 0.81 ± 0.23, 0.93 ± 0.14, 0.80 ± 0.10 and 0.77 ± 0.03, respectively; in the TIMP-2 expression, the values were, 0.80 ± 0.35, 0.60 ± 0.22, 0.93 ± 0.10 and 0.63 ± 0.14, respectively. In this analysis, the expression of GAPDH mRNA was used as internal control and the relative MMP-2, MMP-9, TIMP-1 and TIMP-2 mRNA were represented as MMP-2/GAPDH, MMP-9/GAPDH(Figure 9), TIMP-1/GAPDH and TIMP-2/GAPDH(Figure 10)mRNA detected. The relative MMP-9 expression was significant by 10-7M of Ang-(1-7)-treated (Figure 9C). There were no significantly different between the three groups on MMP-2 (Figure 9B), TIMP-1(Figure 10B)and TIMP-2(Figure 10C).
3-3. MMP-2 expression in the different tissues
The MMPs and TIMPs had until been distributed the quantity in any tissues of rat.
The balance of MMPs and TIMPs was demonstrated relative of cardiac diseases. We were interesting the study, we collected different source tissues in rat and analyzed the content of MMP-2, MMP-9, TIMP-1 and TIMP-2 mRNA, enzyme activity of MMP-2 and MMP-9.
The activity of MMP-2 and MMP-9 in the different tissues, including heart, liver, spleen, lung, kidney, brain, muscle, intestine and testis, were detected by gelatin zymography(Figure 13A). The actual activity of MMP-2 was estimated via MMP-2 standard curve(Figure 12).
The activity MMP-2 in the heart, lung, muscle and testis was 10475 ± 7454, 14444 ± 8918, 13782 ± 10288 and 6993 ± 4483, respectively(Figure 13B). We noted the conspicuous gelatinolytic activity in the heart, lung, muscle and testis as compared with others. However, the activity of MMP-9 in the several tissues of rat was undetectable.
The mRNA of MMP-2 was measured in the different tissues of rat with heart, liver, spleen, lung, kidney, brain, muscle, intestine and testis, were detected by the method of
semiquantitative RT-PCR. In this analysis, the expression of 18s RNA mRNA was used as internal control and the relative MMP-2 mRNA was respresented as MMP-2/18s RNA mRNA detected(Figure 14A). The values were that a markedly high expression in heart (5.15 ± 1.06), lung (5.05 ± 0.95) and muscle (8.02 ± 0.79) was detected as compared with those in the others(Figure 14B).
3-4. Ratios of MMP2/TIMP-2 and MMP-9/TIMP-2
MMPs and TIMPs played a crucial role in physiological and pathological matrix turnover[Moche et al., 2005]. Inhibition of MMPs or targeted deletion of MMP gene attenuated cardiac remodeling. Correlation between MMP-9 and TIMP-1 (MMP-9/TIMP-1) and MMP-2 and TIMP-2 (MMP-2/TIMP-2) can be estimated and applied as a physiological marker for diagnosis of disease processing.
The ratio of MMP-9/TIMP-1 showed positive correlation with mRNA levels in the AF group (0.32 ± 0.23, ranging from 0.10 to 0.75) compared with the SR group (0.11 ± 0.05, ranging from 0.05 to 0.19). The result showed that ratio of MMP-9/TIMP-1 in the AF was higher than that in the SR by 2.9 folds (p < 0.05). In contrast, the ratio of MMP-2/TIMP-2 was insignificantly different(Figure 6).
Similarly, the ratio of MMP-9/TIMP-1 in 10-7M Ang-(1-7)-treated showed positive correlation with mRNA levels control (0.70 ± 0.40). The results showed that ratio of MMP-9/TIMP-1 treated 10-7M of Ang-(1-7) in H9c2 cells was higher than that in the control by 1.8 folds (p < 0.05). In contrast, the ratio of other concentration with Ang II or Ang-(1-7)-treated was insignificantly different(Figure 11).
IV. Discussion
The major finding of this study was that the process of fibrillation follow AF might regulate MMP-9 mRNA through cardiac fibroblast activation. In constant, cardiac
fibroblasts were associated with, high levels of active MMPs. This raises the possibility that an in vivo interaction between the fibrillar collagen network and cardiac fibroblasts works as a “motor” to generate active MMPs, which may, in turn, be involved in myocardical ECM turnover. A point of view about cardiac fibroblasts play an important role to regulate the ECM remodeling has been a focus of increasing recent investigation. Accurate
understanding the molecular mechanism of ECM remodeling would help make possible the development of successful therapeutics strategies during the AF.
In this article, we had successful established the sustained AF model by pacing-induced, the myocardial fibrosis markedly accumulated in interstitial cardiomyocytes in the AF by Masson’s trichrome staining(Figure 1). These results were corresponsed to Masson’s trichrome staining. First at all, Western blot demonstrated the increased protein levels of collagen typeⅠinthe AF(Figure 2). The values indicated that MMP-9 activity showed different between SR and AF and the quantitative values demonstrated that the increase of MMP-9 activity was more pronounced in the AF than SR. In fact, the zymographic activity of MMP-9 increased in the AF was reported here for the first time. These AF had benefits to imitate actual human pathological AF symptom compared with other animal models.
Several lines of evidence point to the importance of MMP-2 and MMP-9 in cardiac
remodeling after myocardial infarction[Lu et al., 2000; Spinale, 2002; Spinale et al., 2000], as well as in vascular remodeling after injury and during atherogenesis[Johnson andGalis, 2004; Whatling et al., 2004]. Therefore, we presented a useful detection method that would be applied to follow the process of fibrosis in the AF atria by MMP-9 activity.
AF is a characteristic feature of heart failure and there is evidence that an ECM imbalance between levels of MMP-9 and TIMP-1 was associated with cardiac remodeling [Hornebeck et al., 2005; Moshal et al., 2005]. Increased cardiac expression of TIMP-1 and TIMP-2 were related to cardiac fibrosis and dysfunction in the chronic pressure-overloaded human heart [Heymans et al., 2005]. In accordance with the documents, we further analyzed the MMPs and TIMPs mRNA levels in the AF. As shown inFigure 3 and Figure 4,the MMP-9 mRNA levels were corresponded with MMP-9 activity. Although the mRNA levels of TIMP-1 and TIMP-2 were down-regulated in the AF, however, the trends of reduction did not reach statistical significance (p > 0.05)(Figure 5). These results seem to correspond to the induction of MMP-9 mRNA levels and zymographic activity. Besider, TIMP-1 would down regulated MMP-9 expression in different tissues. To summarize these data, we supposed that the imbalance between MMP-9 and TIMP-1 mRNA levels in the AF were reached statistical significance in the AF as compare with SR (p < 0.01). This imbalance ratio of MMP-1/TIMP-1 might reveal the tendency toward fibrillation.
In the application of genome research, we would get the whole human genome sequence easily and application to molecular biology. But, we had problems with got the whole pig genome sequence. Therefore, we capitalize on human and rat genome sequence to design MMP-9 and MMP-2 primers for pig. Because, there were almost 70% conserved sequences on MMP-9 and MMP-2 between human and rat. After RT-PCR analysis, the specific product of MMP-9 and MMP-2 were cloned using the TA cloning kit and sequenced. After automated sequencing, comparative analyses were performed with the human MMP-9 and MMP-2 mRNA sequence and demonstrate that the conserve sequence of MMP-9 and MMP-2 from pig compare with human would reach to 70%. Cardiac fibroblasts were the most abundant cell type in the heart and play a major role in synthesizing components of the
cardiac ECM[Eghbali, 1992]. Cardiac fibroblasts had also been shown to play an important
role in the regulation of collagen degradation by MMPs[Cleutjens et al., 1995]. We used the culture systems to mechanically stimulate cardiac fibroblasts allows for the systematic
examination of mechanical regulation of cardiac fibroblast function in the absence of systemic effects. In the present study, the rat cardiac fibroblast H9c2 cells were selected as an in vitro cell line model to investigate what was an important effects in the process of fibrosis in AF.
In fact, the primary cultured of cardiac fibroblasts from AF atria should be selected as an in vitro cell line model; however, the whole pig genome was difficult to get. For the reasons, rat cardiac fibroblast H9c2 cells were selected and challenged with Ang II and Ang-(1-7) for 12 and 24 h. As shown inFigure 7, the critical step of fibrosis was based on the induction of collagen type I mRNA levels. The values indicate that Ang II and Ang-(1-7) induced the collagen type I mRNA levels significant in a dose-dependent manner, especially in the dosage of 10-7M of Ang-(1-7) which had significantly different than control (p < 0.05). It was an interesting finding that Ang-(1-7) would induce collagen type I mRNA levels. However, the regulate mechanism of Ang-(1-7) on collagen type I induction was still unknown. After the H9c2 cells challenged with Ang II and Ang-(1-7), we anticipated that the MMP-2 and
MMP-9 mRNA levels would corresponded with collagen type I induction. However, it was not anticipated that the signal ratios of MMP-2/GAPDH, TIMP-1/GAPDH and
TIMP-2/GAPDH were not increased in dose-dependent manner after Ang II and Ang-(1-7) treatment, respectively(Figure 9 and 10). In contrast, the MMP-9 mRNA levels were induced in the dosage of 10-7M of Ang-(1-7) which had significantly different than control (p
< 0.05). But, the mRNA levels of MMP-9 in H9c2 cells were too weak to be detected by RT-PCR. Then we even tried to amplify the MMP-9 gene expression for 38 cycles by RT-PCR. Here, the difference from MMP-9 mRNA levels and zymographic activity between cardiac fibroblast H9c2 cells and AF model, we though about the species different between rat and pig.
The effects of MMPs were determined in different tissues of rat. We further studied the MMP-2 zymographic activity in different tissues of rat by gelatin zymography. In this study, MMP-2 zymographic activity was significantly greater in heart, lung and muscle as compared with other tissues. The induction of MMP-2 zymographic activity in heart, lung and muscle than other tissues was a novel observation. A recent study showed that ECM protein
remodeling enhances MMP-2 enzyme stability as well as its activity[Itoh et al., 1998].
Taken together, we thought that certain tissues had high level of MMPs zymographic activity would trend to develop fibrosis after tissue injury or damaged. Several previous rat studies had reported that the high level of MMP-2 was detected in common, but might be predict to increase during cardiac remodeling. Weber et al.[Weber et al., 1992]reported that the severity of cardiac fibrosis might become significantly apparent with the development of remodeling by increasing MMP activity. That was, once the heart was damaged, ECM that connected cardiomyocytes would be degenerated by increased MMP for the adaptation of cardiomyocytes enlargement and fibroblasts invasion. The similar results were observed in lung and muscle, too[Cheng et al., 2005; Hsu et al., 2005].
The MMP-2 mRNA levels were only induced in heart, lung and muscle based on the different mRNA levels between MMP-2 in multiple tissues of rat(Figure 13). The result suggested the development of heart failure characterized by the unusual expression of MMP-2 in the fibrosis. We believe that pathologic cardiac dysfunction can be predicted to occur if the experiment period was prolonged. In fact, cardiac expression of TIMP-1 and TIMP-2 was significantly increased in chronic pressure-overloaded human hearts compared with controls and is related to the degree of interstitial fibrosis[Heymans et al., 2005]. We supposed that the MMPs zymographic activity was as manifest cardiac dysfunction marker.
End-stage of human dilated cardiomyopathy (DCM) was characterized by myocytes loss and fibrosis, and associated with ventricular dilatation and reduced cardiac function. MMPs
and TIMPs have been involved in the myocardial remodeling[Rouet-Benzineb et al., 1999].
In the AF model, the imblance of MMP-9/TIMP-1 was compared with the SR (p < 0.05).
The increased ratio of MMP-9/TIMP-1 associated with the process of fibrosis during fibrillating atria(Figure 6). Recently studies showed that the imbalance between
macrophage-derived MMP-9 and TIMP-1 in bleomycin-induced pulmonary fibrosis in mice [Li et al., 2004]. In fact, the imbalance between MMP-9 and TIMP-1 in AF might be reported here for the first time. Furthermore, we would through the selected inhibitors of MMP-9 RNAi and chemical agent treatment for the correlation between MMP-9 and TIMP-1 in the AF was more clear. In the previous studies, Ang II mediated fibroblast function and collagen production over a 72-h period, while increasing MMP-2 expression and activity [Mookerjee et al., 2005]. But, there were not significant induction of MMPs mRNA and activity in H9c2 cells. Although the ratio of MMP-9/TIMP-1 had significantly different (p <
0.05) on the concentration of 10-7M of Ang-(1-7) treated-cells(Figure 11). But, it was stranged in the induction of MMP-9 mRNA levels by single dosage of Ang-(1-7) treatment.
We speculated that the H9c2 cells were not sensitive to Ang II and Ang-(1-7) treatment.
Our findings in the presentation studies include the induction of MMP-9 mRNA levels, zymographic activity and imbalance of MMP-9/TIMP-1 ratio were predicted what an important role of MMP-9 during the ECM remodeling in AF. They might be, at least partially, the possible molecular mechanisms behind how regulated the ECM remodeling during the increased risk of AF reported in pig. Particularly, because cardiac fibrosis was a more typical end-stage condition, it is more beneficial to alleviate these problems before the
Our findings in the presentation studies include the induction of MMP-9 mRNA levels, zymographic activity and imbalance of MMP-9/TIMP-1 ratio were predicted what an important role of MMP-9 during the ECM remodeling in AF. They might be, at least partially, the possible molecular mechanisms behind how regulated the ECM remodeling during the increased risk of AF reported in pig. Particularly, because cardiac fibrosis was a more typical end-stage condition, it is more beneficial to alleviate these problems before the