1. Context
Gastric cancer (GC) is one of the most common malignancies (4th in men and 7th in women) with over one million new cases in 2018. In addition, GC includes about 10% of the new diagnoses of cancer cases and comprises 12% of the overall cancer-related deaths in the world (1). Based on reported data, GC is the 5th common carcinoma in men and women among developed countries and has the second rank for men individual in numerous countries of Asia and South America (2). However, due to the recognition of certain risk factors for susceptibility to GC, the worldwide morbidity and mortality rate has declined over the past few decades, but there is still a high prevalence of gastric cancer in some population in developing countries (3, 4). GC is a multi-factorial disease that its carcinogenesis mechanisms are still unknown. Based on previous epidemiological studies, the interplay of both genetic and environmental factors (such as age, sex, body mass index, infectious agents, diet, and lifestyle) play a major role in the gastric carcinogenesis that can be different among ethnic groups (5-7). Different genes are involved in variety of processes such as inflammatory response, DNA repair, cell proliferation, carcinogen detoxification, and antioxidant protection to development and susceptibility of gastric cancer and increase invasive progression and metastasis (8, 9).
Matrix metalloproteinases (MMPs), the main family of zinc-dependent enzymes, play an important role in the digestion of the extracellular matrix (ECM) (10). Furthermore, different studies have shown that MMPs affect cancer development processes such as apoptosis, cell proliferation, and immune system (11).
So far 24 different types of MMPs have been reported and 23 of them had been found in human (12). However, MMPs present in healthy individuals, but they are up-regulated in almost all types of cancer (13, 14). The high expression of different MMP genes has been correlated to metastasis, invasion, and survival of many human cancers (14). These MMPs are gelatinases, stromelysins, membrane-type MMPs, and other MMPs. Cells synthesize MMPs as an inactive zymogen and the N-terminal cleavage of pro-peptide leads to make them activated. Based on the reported of previous studies, 4 members of this family are associated-MMPs with GC, MMP-1, -2, -7, and-9. The roles of MMP-2 (gelatinase A with 72 kDa) and MMP-9 (gelatinase B with 92 kDa) are more important than others in tumor invasion and metastasis, this importance is because of their substrates (11). MMP-2 and MMP-9 digest types IV and V of collagen, fibronectin and gelatin. The basement membrane is the first barrier for a metastatic epithelial tumor that consists of type IV collagen (15).
Among main MMPs genes related to cancer development, more attention has been focused on the Matrix metalloproteinase 2 which is over-expressed in the different human tumors. Expression level and activity of MMP-2 are often associated with invasion, cell migration, and development of tumor cells (16). Known evidence indicates that MMPs, especially MMP-2, play a critical function in the degradation of ECM that are mediated by tumor cells. MMP-2 gene with a total length of 27 kbs is located on chromosome 16q21 and contains 12 introns and 13 exons (17). Over-expression of MMP-2 gene was reported to the GC tissues compared to the matched normal tissues that are correlated to the invasion and metastasis in GC cells (12).
2. Objectives
According to the evidence available on this issue, the present study aimed to conduct a review article of all original studies that have investigated the role of MMP-2 in gastric cancer among different population.
3. Methods
3.1. Search Strategy
We searched on electronic databases such as PubMed, Scopus, Science Direct, and Cochrane Library Database without any language restriction for relevant publications that were published until April 2019. In addition, following keywords and abbreviation terms were used in the search strategy to get relevant results: [“gastric cancer”, “gastric carcinoma”, “GC”, “stomach carcinomas” or “stomach cancer”] and [“matrix metalloproteinase 2”, “MMP-2”, “MMP2”, “72kDa Type IV Collagenase”, “matrix metalloproteinase2” or “MMP2metalloproteinase”].
3.2. Selection Criteria
The criteria for the selection of relevant studies were: (1) the article type must be original research that focused on the association of MMP2 expression in GC patients, (2) all studies must have used human specimens, (3) full-text article must be in English language, and (4) the study must give suitable data about the genetic, expression levels or activity of the MMP2.
3.3. Data Extraction
The required information from the included study was extracted using a standardized form. We documented the most relevant information such as the first author’s name, publication year, publication Journal, geographical location, sample characteristics (size, sex, age, and type), subject or aim, used method, and main results that state the association between expression levels of MMP-2 and gastric cancer.
4. Results and Discussion
This systematic review study indicated the role of MMP-2 expression in gastric cancer susceptibility. Primarily, the highly sensitive search strategy recognized 152 articles. We reviewed all articles according to the inclusion and exclusion criteria (Figure 1). Finally, 32 original and relevant studies which had evaluated the association between gastric cancer and MMP-2 were selected. The articles were published between 1996 and 2019 (Table 1).
Table 1.Description of the Main Characteristics, the Properties of Samples and Studies, and the Main Results of All Eligible Articles
| Reference | Publication Year | Population Nationality | Sample | Study | Main Results | |||||
|---|---|---|---|---|---|---|---|---|---|---|
| Size | Type | Subject | Methods | |||||||
| Total | GC, M/F | Control, M/F | Age, y | |||||||
| (18) | 2018 | China | 160 | 160 (103/57) | 160 (103/57) | 112 > 60, 48 ≤ 60 | Tissue | MMP-2 expression | IHC and qRT-PCR | CDH17 increased the level of MMP-2 via the canonical NF-κB pathway. |
| (19) | 2018 | China | 160 | 160 (107/53) | 160 (107/53) | 95 > 60, 65 ≤ 60 | Tissue | MMP-2 expression | Gelatin zymographic | AGEs induced the migration and invasion of GC cells by increasing transcription of RAGE, Sp1, and MMP2. |
| (20) | 2017 | China | 94 | 47 (29/18) | 47 (30/17) | 58.36 (40 - 70), 59.85 (40 - 75) | Blood | MMP-2 expression | ELISA | The level of MMP-2 was declined after 3 cycles of treatment by adding glutamine. |
| (21) | 2017 | China | 204 | 204 (140/64) | - | - | Tissue | MMP-2 expression | IHC | Over-expression of MMP-2 with MMP-9 associated with metastasis and poor prognosis in early GC. |
| (22) | 2015 | Poland | 60 | 60 (44/16) | - | 63 (37 - 84) | Tissue | MMP-2 Expression | IHC | Low expression of MMP-2 in tumor and stromal compartments were correlated to poor prognosis. |
| (17) | 2015 | China | 504 | 254 (162/92) | 250 | 53.35 (26 - 75), 53.35 ± 3 | Blood and Tissue | Polymorphism, MMP-2 expression | Genotyping and IHC | MMP-2-1306C/T was statistically associated with the MMP-2 expression. |
| (23) | 2015 | Germany | 128 | 28, 85/43 | - | 78 (70 - 78) | Tissue | MMP-2 expression | IHC, RT-PCR | MMP-2 and MMP-7 at the invasive front of GC were not correlated with mTOR expression. |
| (24) | 2013 | China | 439 | 81 | 358 | - | Tissue | MMP-2 expression | IHC and Real Time PCR | JWA and MMP-2 can play role as promising prognostic biomarkers in resectable GC. |
| (25) | 2014 | China | 105 | 105 | - | 58 (32 - 84) | Tissue | MMP-2 expression | IHC, RT-PCR | Activation of p38 induced metastasis in GC with up-regulation of AP-1/c-fos, MMP-2, and MMP-9. |
| (26) | 2013 | Poland | 108 | 40 (30/10) | 68 (14/54) | GC: 28 - 82, Control: 21 - 64 | Tissue and Blood | MMP-2 expression | IHC and ELISA | The positive reaction of MMP-2 was higher in GC compared to normal tissue. |
| (27) | 2013 | China | 40 | 40 (32/8) | - | 57.9 (18 - 77) | Tissue | MMP-2 expression | IHC | Combined detections for the activity of telomerase and MMP-2 can identify patients at high risk in disease recurrence. |
| (28) | 2012 | Poland | 24 | 24 (18/6) | - | 63.7 (43 - 81) | Tissue | MMP-2 expression | Spectrophotometric | The highly significant association between VEGF and MMP-2 in GC was reported. |
| (29) | 2011 | Poland | 191 | 100 (73/27) | 91 (60/31) | GC: 27 - 83, Control: 21 - 65 | Blood | MMP-2 expression in serum | ELISA | MMP-2 and TIMP-2 expression at serum level was positively lower in GC patients than in control subjects. |
| (30) | 2010 | Taiwan | 189 | 189 (110/79) | - | 62 | Tissue | MMP-2 expression | IHC | The Claudin-4 expression was positively associated with MMP-2 expression. |
| (15) | 2009 | Poland | 53 | 34 (24/10) | 19 (11/8) | 60.4 | Serum | MMP-2 expression in serum | zymographic | MMP-2 expression at serum level in GC individuals was not significantly higher than that in healthy group. |
| (31) | 2008 | China | 44 | 44 | - | - | Tissue | MMP-2 expression | IHC | Metastatic GC presents higher MMP-2 immunoreactivity than primary GC. |
| (32) | 2008 | Germany | 116 | 116 (69/47) | - | 64 (33 - 85) | Tissue | MMP-2 expression | IHC | MMP-2 plays a critical role in the invasion of GC |
| (33) | 2007 | Taiwan | 523 | 240 (143/97) | 283 (170/113) | 59.95 (22 - 88) | Blood | polymorphism | PCR-direct sequencing, PCR-RFLP | 1306 C/T polymorphism of MMP2 is correlated with invasion and progression of GC |
| (34) | 2006 | Finland | 329 | 329 (171/158) | - | 66 | Tissue | MMP-2 expression | IHC | Epithelial MMP-2 expression in GC correlated with aggressive forms OF COX-2 |
| (35) | 2006 | Japan | 229 | 229 (166/63) | - | 65.5 (38 - 88) | Tissue | MMP-2 expression | Tissue microarray and Immunostaining | MMP-2, MMP-9 and VEGF play role in the angiogenesis and progression of GC. |
| (36) | 2006 | China | 30 | 30 (17/13) | - | 56.8 (26 - 82) | Lymph node | MMP-2 expression | IHC, RT-PCR | Expression of MMP-2 has a positive association with tumor invasion, tumor differentiation, and lymph node metastasis of GC. |
| (37) | 2006 | Netherlands | 81 | 81 (60/21) | - | 65.9 (35 - 91) | Tissue | MMP-2 expression | gelatin zymography and ELISA | MMP-2 is positively correlated to the prognosis of GC than other MMPs or TIMPs. |
| (38) | 2005 | China | 97 | 65 (50/15) | 32 | 60.08 (31 - 81) | Tissue | MMP-2 expression | IHC | MMP-2 gene plays the main role in metastasis, invasion, and prognosis of GC. |
| (39) | 2005 | China | 77 | 67 (43/24) | 10 | 62.2 (23 - 78) | Tissue | MMP-2 expression | RT-PCR analysis of MMP-2mRNA | Expression of MMP-2 in tumor tissues didn’t correlate with depth of invasion. MMP-2 may play a critical function in the development, invasion, and metastasis of GC. |
| (13) | 2003 | China | 1145 | 356 (308/48) | 789 (672/117) | GC:58.4 (41 - 72), Control: 57.6 (45 - 76) | Blood | MMP-2 Polymorphism | PCR based DHPLC, DNA sequencing | 1306 C/T polymorphism in MMP2 is correlated with development risk but not metastasis of GC. |
| (40) | 2001 | Germany | 114 | 114 (66/48) | - | 64 (33 - 85) | Tissue | MMP-2 expression | IHC | MMP-2 expression is positively associated with lymph node metastasis and tumor progression in GC |
| (41) | 1999 | China | 25 | 20 (13/7) | 5benign ulcer patients | 61.8 (38 - 74) | Tissue | MMP-2 expression | RT-PCR | MMP-2 can play the main role in GC invasion and metastatic progression. |
| (42) | 1998 | Germany | 203 | 203 (108/95) | - | 63.8 (22 - 87) | Tissue | MMP-2 expression | IHC | Immunohistochemical detection of MMP-2 was associated with the prognosis of GC. |
| (43) | 1998 | Italy | 25 | 25 (11/14) | - | - | Tissue | MMP-2 expression | RT-PCR, zymography | MMP-2 suggested as a new molecular-level prognostic factor. |
| (44) | 1997 | Japan | 68 | 68 (42/26) | - | 65.6 | Tissue | MMP-2 expression | RT-PCR, IHC, zymography | Activation of MMP-2 can be clinically relevant with GC susceptibility. |
| (45) | 1996 | Netherlands | 50 | 50 (38/12) | - | 66.3 | Tissue | MMP-2 expression | quantitative zymography | The mean level of MMP-2 was positively increased in carcinomas than in tumor-free adjacent mucosa of the stomach. |
| (46) | 1996 | Japan | 46 | 46 | 46 | - | Tissue | MMP-2 expression | IHC, immunoassays, zymography | Activation of pro-MMP2 can be the main step for spreading of GC cells. |
Abbreviations: AGEs, advanced glycation end products; ELISA, enzyme-linked immunosorbent assay; GC, gastric carcinoma; IHC, immunohistochemistry; mTOR, mammalian target of rapamycin; qRT-PCR, Quantitative real-time polymerase chain reaction; VEGF, vascular endothelial growth factor.
Figure 1.
Flow chart for the selection procedure of the included studies
The total of 32 articles were included to be reviewed. The most of the population studied in these papers were from China (14 articles: 47%) and other studies were from Poland (5 articles), Germany (4 articles), Japan (3 articles), Netherlands (2 articles), Taiwan (2 articles), Finland (1 article), and Italy (1 article). Since the gastric cancer rate in the countries of East Asia and then in Eastern and Central Europe is higher than in other countries or parts of the world (47), therefore most of the studies have been reported from these countries.
Among 32 included articles, the total number of studied samples were 5661 individuals, among these cases, 3709 individuals were patients with gastric cancer and 2318 individuals were considered as controls. All available data such as mean age, sex status (male/female ratio), and sample types are presented in Table 1. Based on the obtained data from reported articles, study specimens were tissue (N: 24), blood (N: 5), tissue and blood (N: 2) and, lymph node (N:1).
These samples were used to study MMP-2 expression (25 cases), MMP-2 expression and immunoreactivity (2 cases), MMP-2 expression and polymorphism (1 case), serum level analysis of MMP-2 (2 cases), and polymorphism (2 cases).
In each research, several methods have been used to reach the mentioned aims, and the IHC method has been used more than others. Accordingly, the most used method in the reviewed articles was the IHC method (19 cases) and then RT-PCR (9 cases ), Gelatin Zymography (6 cases), ELISA (4 cases), PCR-RFLP (1 case), PCR-direct sequencing (1 case), RNA/DNA calculator with spectrophotometric (1 case), Tissue Microarray (TMA) (1 case), Immunostaining (1case), Sandwich enzyme immunoassay (1 case), PCR Based DHPLC analysis and DNA sequencing (1 case).
The analysis of the present study showed that the expression of MMP-2 in the cancer subjects was mostly higher than control individuals. However, according to the study by Mroczko et al. 2011, the serum level of MMP-2 gene was positively lower in GC patients than healthy individuals (29). Consistent with results of Mroczko et al. (29) 2011 and Lukaszewicz-Zajac et al. (26) 2013 that serum level of MMP-2 was significantly lower in GC patients. In addition, according to Emara et al. 2009, MMP-2 level was not significantly higher in GC patients than controls individuals (15).
Based on the findings of the present study, more than 50% of reported studies showed that abnormal changes in MMP-2 expression play a critical role in tumor metastasis in GC. In addition, 8 studies examined the role of MMP-2 in the tumor invasion and resulted that MMP-2 expression significantly correlated with tumor invasion (Table 1). Further, Bornschein et al. showed that MMP-2 had a higher expression level in the invasive front compared to the center section of the tumor (23). These findings support the hypothesis that over-expression of MMP-2 gene may play a crucial role in degrading type IV Collagen, gelatin, and laminin in basal membrane and other components of extracellular matrices, which is a vital step toward the development, invasion, and tumor metastasis. Additionally, the results of the other study by Zheng et al. showed that IL-1β could activate p38 and increase gastric adenocarcinoma (GA) cell migration and invasion (35). IL-1β- induced GA cell migration and invasion occur via activation of the 38 signaling pathway which leads to AP-1 activation and up-regulation of MMP-2.
Zheng et al. (35) and Partyka et al. (28) investigated the correlation between vascular endothelial growth factor (VEGF) and MMP-2 in cancerous tissue of metastatic patients. They found that MMP-2 and VEGF were positively associated with the tumor size, depth of invasion, lymphatic, and venous metastasis. Regarding their results, MMP-2 plays an important role in the “angiogenic switch” and tumor cells can synthesize and secrete high levels of MMP-2 paracrine and/or autocrine to stimulate angiogenesis and increase VEGF release (35).
In the other study, Chen et al., examined how JWA, a multifunctional microtubule-binding protein regulates GC angiogenesis via MMP-2 and the role of JWA and MMP-2 in the progression and prognosis of GC. JWA inhibits GC angiogenesis via Sp1-mediated MMP-2 expression. Sp1 was the transcription factor of MMP-2; it has been reported that Sp1 up-regulates the MMP-2 gene in cancers and promotes angiogenesis in GC (24).
Ten studies surveyed the MMP-2 expression and its effect on prognosis in GC patients. Among these, Donizy et al. (22), Lukaszewicz-Zajac et al. (26), Kubben et al. (37), Ji et al. (39), and Caenazzo et al (43) suggested the high expression of MMP-2 as an independent and molecular prognostic factor for gastric cancer. However, the other authors had a different idea, they declared that MMP-2 alone was not enough and suggested other molecules to accompany MMP-2 as a prognostic factor. According to the findings of other research by Yao et al. (21) and Ji et al. (39) MMP-9 was as a helpful factor along with MMP2. Also, Allgayer et al. suggested that consideration of interrelated tumor-associated proteases like uPA receptor in combination with MMP-2 may improve its prognostic power (42). The results of Wang et al. showed the evaluation of both telomerase activity (TA) and MMP2 protein can more effectively detect patients who are susceptible to disease recurrence and prognosis (27). As mentioned above, Chen et al. findings pointed out that JWA and MMP-2 may serve as prognostic biomarkers in GC (24).
Survey of MMP-2-1306 C/T polymorphism which reported by Miao et al. (13), Zhang et al. (17), and Wu et al. (33) proved that this SNP correlated with GC susceptibility, lymphatic or venous invasion, and progression of gastric cancer but not associated with the tumor diameter, the depth of tissue infiltration, lymphatic metastasis, survival rate, age, sex, H. pylori infection, Lauren’s classification, tumor status, depth of invasion or lymph node metastasis, and metastasis of GCA (Gastric Cardia Adenocarcinoma). Miao et al. indicated that MMP-2-1306 C/T polymorphism is associated with the risk of gastric cancer development (13), but this result is not consistent with Wu et al. (33). In the study carried out by Zhang et al. showed that susceptibility of GC for patients with CC+CT genotype in MMP-2-1306 C/T SNP was 1.803 times more than the individuals who have TT genotype (17), and Miao et al. proved it in their study too (13). However, Wu et al. reported that the C/T allele frequencies of MMP-2-1306 in GC patients did not differ from those of controls. In addition in the other study conducted by Wu et al. the cases with MMP-2 1306 C/C genotype were significantly more susceptible for lymphatic and venous invasion than cases with C/T or T/T genotype, but they did not differ in the survival rate (33).
Eventually, Wang et al. examined the effect of glutamine-enriched nutritional support on intestinal mucosal barrier function, MMP-2, MMP-9, and immune function in patients with advanced gastric cancer during the perioperative chemotherapy. The results showed that after three cycles of treatment by adding glutamine, MMP-2 level was positively decreased (20).
Jiang et.al surveyed the effect of CDH17 on MMP-2 expression by NF-κB pathway. Its results showed a positive relation; decreased or increased levels of MMP-2 was significantly regulated by CDH17 knockdown or overexpression, respectively. This effect was mediated by the NF-κB pathway in GC cells. In the present study, GC tissues had a considerably higher level of CDH17 mRNA than the matched para-carcinoma tissues in the same patient. CDH17 expression was associated with clinical lymph node metastasis. CDH17 induce these activities through the NF-κB/MMP-2 pathway (18).
Based on Deng et al. RAGE/ERK/Sp1/MMP2 pathway induced by glucose-derived AGEs (Advanced glycation end products) may result in GC progression and stimulating the invasion and metastasis of it. They found the accumulation of glucose-derived AGEs in cancer tissues and blood of GC patients and it’s resulting in over-expression of RAGE, Sp1, and MMP2. This study revealed the following sequence: glucose-derived AGEs binding to RAGE, activating MEK1/2/ERK pathway, overexpression of Sp1, up-regulating MMP2 expression, and GC cells invasion, respectively (19).
5. Conclusions
In summary, our review indicated that increased MMP-2 expression had been seen in gastric cancer. MMP-2 over-expression may play a crucial role in degrading ECM as well as stimulate angiogenesis and increase VEGF releasing. Therefore, MMP2 overexpression can play an important role in tumor metastasis, tumor size, invasion, and lymph node invasion in the GC. Furthermore, high-level expression of MMP-2 may be strongly associated with poor prognosis in GC patients. Thus, the detection of MMP-2 expression may serve as an independent prognostic factor for GC. In addition, MMP-2 polymorphism such as -1306C/T can be correlated with GC susceptibility. However, further studies with larger sample sizes and more comprehensive and meta-analysis data are still required for the achievement to conclusive results.
