【摘要】
AIM: To examine the expression of MMP9 and TIMP3 mRNA during choroidal neovascularization (CNV) in a murine model and to investigate the role of them in the development of CNV.
METHODS: CNV was induced in C57BL/6J mice by intensive diode laser (810nm) photocoagulation (120mW, 75μm, 0.1s) of the fundus whereafter eyes were enucleated at 1, 3days, 1, 2, and 4 weeks. The MMP9 and TIMP3 mRNA expression were analyzed using in situ hybridization and image analysis system.
RESULTS: Both expression of MMP9 and TIMP3 mRNA had dynamic changes. For MMP9, the expression was 1, 2, 4 wk>3d>1d (P<0.05), whereas TIMP3 mRNA, 3d, 1, 2, 4 wk>1d (P<0.05).
CONCLUSION: The imbalance between the changes of MMP9 and TIMP3 may accelerate the degrading of extracelluar matrix, and then be involved in the pathogenesis of CNV.
KEYWORDS: matrix metalloproteinase9;tissue inhibitor of metalloproteinase3;choroidal neovascularization;mice
【关键词】 matrix metalloproteinase9 tissue inhibitor of metalloproteinase3 choroidal neovascularization mice
INTRODUCTION
Choroidal neovascularization (CNV) is one of the serious complications of some macular diseases such as agerelated macular degeneration and hypermyopic macular degeneration. The development of CNV can cause some pathological changes, such as exudation, hemorrhage and cicatrization, etc., resulting in severe visual disorder. The mechanism of its formation is to be elucidated.
Angiogenesis is a multistep event in which new capillaries are formed from preexisting vessels. In the initial stages of angiogenesis, microvascular cells locally degrade the underlying basement membrane and the surrounding extracelluar matrix (ECM), and then vascular endothelial cells proliferate and migrate into the tissue to be vascularized.
Matrix metalloproteinases (MMPs) and theirs tissue inhibitors of metalloproteinase (TIMPs) are the important factors that affect the metabolism of ECM. In this study, we examined the expression of MMP9 and TIMP3 mRNA in a murine model, and to investigate the role of them in the development of CNV.
MATERIALS AND METHODS
Experimental Animal Twelve healthy adult male C57BL/6J mice (supplied by Shanghai Experimental Animals Center, Chinese Academy of Sciences) weighing 2530g were used in this study.
Experimental CNV Induction After totally dilation of the pupils with 10g/L tropicamide and anesthesia with intraperitoneal injection of 0.3mL 2.5g/L sodium pentobarbital, diode laser (810nm) was delivered into the 20 eyes (10 mice) through slit lamp and contact lens. The laser settings were 120mW intensity, 75μm diameter, and 0.1s duration. Ten burns were delivered around the optic disk. Production of bubbles could be seen at the time of laser treatment, sometimes associated with light sounds. Occasionally hemorrhage could be seen at the sites of photocoagulation. Fundus fluorescein angiography was performed at 1 day, 3 days, 1 week, 2 weeks and 4 weeks after photocoagulation to confirm the development of CNV. The other two mice without any treatment were selected as normal control.
Tissue Processing At 1 day, 3 days, 1 week, 2 weeks and 4 weeks after photocoagulation, the animals were killed and the eyeballs were enucleated, four eyeballs at each time point were used. The eyes were washed with PBS (0.1% DEPC) and embedded in optimum cutting temperature(OCT) compound. After stored in the ultra colder (80℃), serial sections of 8μm thick were taken using cryomicrotome. Sections including satisfying photocoagulation scars were used and fixed in 40g/L paraformaldehyde, and then stored at 20℃ for preparing of in situ hybridization.
In Situ Hybridization
Reagents MMP9, TIMP3 in situ hybridization kits (Boster Biotechnology, Inc., Wuhan, China).
Table 1 Expression of MMP9 vs TIMP3 mRNA during the development of CNV（略）
Oligonnucleotide probe sequence MMP9: 5GCTAT CCAGC TCACC GGTCT CGGGC AGGGA3, 5GGGAA GACGC ACAGC TCTCC CGCCG AGTTG3, 5TAGGT CACGT AGCCC ACTTG GTCCA CCTGG3; TIMP3: 5AGGCT CCAGC TGCCC AGGAG CACGA TGAGC3, 5CCTGT CAGCA GGTAC TGGTA CTTGT TGACC3, 5AAGCA AGGCA GGTAG TAGCA GGACT TGATC3.
Procedure Frozen sections were prepared at room temperature for 60 minutes, and digested with pepsin diluted by 30mL/L citromalic acid/DEPC solution at 37℃ for 5 minutes in order to expose the mRNA. After that, the sections were balanced with 2×SSC/DEPC at room temperature for 15 minutes. The sections were prehybridized at 38℃ for 4 hours and hybridized at 38℃ for 20 hours. After hybridization, the reaction were blocked with blocking reagent at 37℃ for 30 minutes, then the slides were incubated with biotinylated antidigoxigenin (DIG) antibody at room temperature for 2 hours. The slides were incubated with streptavidinbiotinalkaline phosphatase complex (SABCAP) at 37℃ for 30 minutes, which followed by incubated in coloring reagent containing nitroblue tetrazolium salt (NBT) and 5bromo4chloro3indolylphosphate (BCIP) at room temperature. The coloring reaction was controlled under a light microscope for 12 hours. These slides were counterstained with nuclear fast red staining and studied by light microscopy. As a negative control, hybridization was performed with prehybridization reagent instead of hybridization reagents. Cells with cytoplasm stained as royal blue were positive ones.
Image analysis The analysis was performed with KS 400 image analysis software package. Slides of in situ hybridization were selected and the images were entered into computer. We selected corresponding areas at the sites of photocoagulation scars, measured total areas and positive expression areas, and then calculated the corresponding positive expression ratio. The gray scale of background was set as 250, and positive area was 160180. Comparison of positive ratios between different groups was analyzed using a oneway analysis of variance (ANOVA) with SAS 6.12 software package(SAS Institute Inc., Cary, NC 275132414 USA). Difference was considered significant at P<0.05.
RESULTS
Results of FFA At 1 day and 3 days after photocoagulation, obvious fluorescence leakage could not be detected. At 1 week, 2 weeks and 4 weeks after laser treatment, fluorescence leakage could be seen at many sites of photocoagulation scars, along with the time, the areas of fluorescence leakage expanded and its intensity increased, which proved the development of CNV (Figure 1).
Expression of MMP9 and TIMP3 mRNA In normal mouse, only a few expressions of MMP9 and TIMP3 mRNA could be seen in choroids and areas near the retinal pigment epithelium (RPE). At 1 day after laser photocoagulation, local elevation of retina and breakdown of integrity of RPE could be observed at sites of photocoagulation, wherein with a small quantity of signals of MMP9 and TIMP3 mRNA. At 3 days after laser photocoagulation, the pigment proliferated and became confused. The expression of MMP9 and TIMP3 mRNA increased with a different degree. At 1 week after laser photocoagulation, CNV developed and last for at least 4 weeks after photocoagulation. During this period, pigment was still confused, and spindleshaped RPE cells covering the CNV tissues could also be seen. Both expression of MMP9 and TIMP3 mRNA were obviously in the areas of CNV and surrounding tissues (Figure 2).
Figure 1 Photograph of FFA 1 week after photocoagulation At 1 week after laser photocoagulation, fluorescence leakage could be seen at the site of photocoagulation（略）
After image analysis using the KS 400 image analysis software package, the positive ratio were calculated and showed in Table 1 and Figure 3.
As a result, both expression of MMP9 and TIMP3 mRNA had a dynamic changes during the development of CNV. They all increased significantly in the early stage after laser treatment, and then stabilized at a high level. For MMP9 mRNA, the expression was 1, 2, 4wk>3d>1d (P<0.05), whereas TIMP3 mRNA, 3d, 1, 2, 4wk>1d (P<0.05). Namely, the expression of TIMP3 mRNA became stable at 3 days after photocoagulation, whereas MMP9 mRNA increased constantly until 1 week after laser treatment.
DISCUSSION
MMPs are a group of extracellular enzymes that share a number of common characteristics, such as the presence of zinc and a conserved amino acid sequence at their active sites. They are a highly regulated family of at least 14 structurally related enzymes capable of degrading most, if not all, of the components of the ECM.
The MMPs are grouped into collagenases (MMP1, 8, 13), gelatinases (MMP2, 9), stromelysins (MMP3, 7, 10, 11, 12), and membranetype matrix metalloproteinase (MMP14, 15, 16, 17). These enzymes and their inhibitors, TIMPs, Figure 2 Photography of in situ hybridization in CNV tissues are involved in maintaining the dynamic equilibrium of ECM.
A: negative control; B: expression of MMP9 mRNA (1 week after laser photocoagulation); C: expression of TIMP3 mRNA (1 week after laser photocoagulation) original magnification×200
Figure 3 The changes of expression of MMP9 and TIMP3 mRNA during the development of CNV（略）
Several ocular tissues secrete MMPs and TIMPs in culture, including cornea, trabecular meshwork and RPE. MMPs have also been found in aqueous humor, tears and sclera. MMP9, the 92kDa gelatinase B, is present in human Bruchs membrane, vitreous and interphotoreceptor matrix (IPM)[13].
In order to investigate the possible role of MMPs in CNV, Steen et al[4] analyzed the mRNA expression of MMPs in subfoveal fibrovascular membranes from five patients with agerelated macular degeneration. They found that MMP9 expression was distinctly expressed by cells at the margins of the membranes and often in proximity to a thickened Bruchs membranelike layer under the retinal pigment epithelial cells. It may be an attempt by the RPE cells to reform a Bruchs membranelike structure around the CNV lesion in an effort to create a physical barrier that inhibits further CNV growth. The presence of MMP9expressing cells next to this Bruchslike layer possibly heralds a processing collagen IV degradation so that the CNV membrane can grow further into the subretinal space. Das et al[5] also observed that levels of MMPs (MMP2 and MMP9) in retinas were significantly increased in an animal model with ischemiainduced retinal neovascularization, namely, newborn mice exposed to the variable oxygen cycle. Neovascularization was significantly inhibited (72% reduction) with intraperitoneal administration of a synthetic MMP inhibitor, BB94 (1mg/kg).
[1] [2] 下一页 |
