Regulation, Function, and Expression of Matrix Metalloproteinase-9 in Colon, Lung, and Breast Cancers In Silico and Experimental Methods From Iraqi Metastatic Patients

The in silico expression of MMP9 at gene and protein levels in lung, colon, and breast cancers was evaluated by different computational techniques. Firstly, an extensive literature review was performed, followed by the identification of functional partners of the MMP9 gene using the Search Tool for the Retrieval of Interacting Genes/Proteins (STRING) database and its expression at gene and protein levels was analyzed using databases such as Gene Expression Profiling Interactive Analysis (GEPIA) and The Human Protein Atlas (HPA), respectively. This was performed to identify the genes involved in the progression of lung, colon, and breast cancers, either by the MMP9 gene inducing the function of other genes or by the MMP9 gene getting induced by other genes during tumor progression.

A comprehensive review of the literature was performed to gain insights into the genes playing a crucial role in the development and progression of genes. Several different search terms such as “MMP-9 expression,” “MMP-9 lung cancer,” “MMP-9 colon cancer,” “MMP-9 breast cancer,” “role of MMP-9 in cancers,” “the role of MMP-9 in lung, colon, and breast cancers,” and “genes inducing MMP-9 in lung, colon, and breast cancers” were used on “Google Scholar” to identify the genes significantly involved in lung, colon, and breast cancers.

The functional partners of MMP9 gene were identified using the STRING database, which identifies both physical interactions and functional associations while aiming to integrate all known and predicted associations between proteins^[18,30]. This was performed to identify the known and predicted functional partners of MMP9 at default parameters such as network type (full STRING network), required score (median confidence [0.400]), and size cutoff (no more than 10 interactions), resulting in the interacting genes that showed the highest interactions with MMP9 gene.

The gene expression analysis for MMP9 gene was performed using GEPIA, a web-based tool based on TCGA and GTEx data providing interactive and customizable functions such as differential expression analysis, profiling plotting, correlation analysis, patient survival analysis, similar gene detection, and dimensionality reduction analysis^[19]. It was utilized to identify the expression levels of MMP9 gene in normal tissue compared to tumor tissues, resulting in various expression values of the MMP9 gene in a range of cancers provided in GEPIA, including lung, colon, and breast cancers.

MMP9 protein expression analysis was performed using HPA https://www.proteinatlas.org, an open resource providing the spatial expression levels of transcripts and proteins across cells, tissues, and organs^[20]. The HPA analysis identified various aspects of genome-wide analysis of MMP-9 protein, including protein–protein interactions (PPIs), concentration in cancers, expression clustering, and prognostic probability in provided cancer types in HPA, including lung, colon, and breast cancers.

Five milliliters of blood samples were collected from late-stage colon, lung, and breast cancer patients after the clinical evaluation of each participant in Anbar Cancer Center. Blood was incubated in a water bath for 15 min until it clotted. Then, centrifugation of the samples was performed for 15 min to obtain clear serum to be used in the enzyme-linked immunosorbent assay (ELISA) technique. The serum of patients was stored at −80 °C for 2 months before use for evaluation.

The MMP9 kit comprising all the reagents as well as the samples was equilibrated to room temperature for optimal temperature conditions. The reagents were prepared according to the manufacturer’s protocol, which commenced with the addition of 100 µL of standard or sample in each well, followed by incubation for 90 min at 37 °C. After removal of the standard or sample from each well, 100 µL of biotinylated detection Ab/Ag was added, which was followed by three wash cycles; in addition, 100 µL of horseradish peroxidase (HRP) conjugate was added and incubated for 30 min. Thenceforth, to five washes, a substrate reagent of 90 µL was added, whereas a stop solution of 50 µL was introduced, and the optical density (OD) value was determined at 450 nm.

The statistical analysis involved expressing MMP9 as mean ± standard deviation (SD). Analysis of variance (ANOVA) was used to assess significant differences among groups. Furthermore, Duncan test was utilized to identify variations in the means of MMP9 indicators within the study groups. The sensitivity and specificity of MMP9 were determined using the receiver operating characteristic (ROC) curve. A significance level of P ≤ 0.05 was applied to identify statistically significant differences. In addition, the data was processed using the statistical software program Statistical Package for the Social Sciences (SPSS) v. 23.0 and GraphPad Prism v. 6.

The study data showed that the participants were predominantly males (33.3%), with most belonging to the age groups of 45–54 years (30.0%) and 55–64 years (33.3%). A significantly large percentage of participants were found to have been diagnosed with lung and breast cancer, making up 33.3% and 38.9% of the total, respectively. Most of these instances were detected in the advanced stage (91.25%), suggesting significant disease progression (91.25%). The statistical analysis showed significant differences among participants (P < 0.05), which corresponded with anthropometric and clinical parameters detailed in Table 1.

The highest levels of MMP9 expression were observed in patients with colon cancer (0.62 ± 0.25), followed by lung cancer (0.46 ± 0.22) and then breast cancer (0.35 ± 0.16), compared to controls (0.30 ± 0.15), with statistically significant differences (p < 0.05). These findings are presented in Figure 1A and B. Moreover, there were no significant differences (p > 0.05) in the levels of MMP9 concerning cancer stages and response to chemotherapy, as detailed in Tables 2 and 3.

Figure 1:

Expression and protein analysis of MMP9. (A) Means of the study groups when compared using the Duncan test; ^a,b Significant differences (p < 0.05). (B) Comparative mean levels of MMP9 among study groups. (C) Sensitivity and specificity graph of colon cancer. (D) Sensitivity and specificity graph of lung cancer. (E) Sensitivity and specificity graph of breast cancer. (F) Differential gene expression of MMP9 in LUAD, COAD, and BRCA tumors. (G) Protein concentrations of MMP9 in each cancer. (H) PPI analysis through STRING. (I) PPI analysis through protein atlas.

Furthermore, MMP9 exhibited its highest sensitivity (80%) and specificity (90%) at a cutoff value of 0.39, demonstrating significant differences (p < 0.05) in screening patients with colon cancer. Conversely, for screening patients with lung cancer, MMP9 showed lower sensitivity (60%) but maintained high specificity (90%) at a cutoff value of 0.37. Notably, for the diagnosis of breast cancer, MMP9 proved to be less efficient, with a sensitivity of 49% and specificity of 50% at a cutoff value of 0.30, as shown in Table 4 and Figure 1C–E.

This study utilized GEPIA to examine the regulation of MMP9 in lung adenocarcinoma (LUAD), colorectal adenocarcinoma (COAD), and breast invasive carcinoma tumors. MMP9 was significantly upregulated in tumor tissues compared to normal tissues, with substantial upregulation in LUAD (30.52 in tumor vs. 10 in normal), COAD (18.86 in tumor vs. 0.69 in normal), and BRCA (29.67 in tumor vs. 2.34 in normal) tumors. The differential expression of MMP9 in LUAD, COAD, and BRCA tumors is shown in Figure 1F.

MMP9 protein concentrations were observed through clustering-based analysis on the tissue-based expression dataset, which revealed higher concentrations of MMP9 in both breast and lung cancers. However, in colon cancer, MMP9 concentrations exhibited a more varied pattern, with some samples showing high concentrations while others displayed lower concentrations. The varying concentrations of MMP9 in each cancer are depicted in Figure 1G.

An extensive literature review was performed to identify, filter, and target the genes and pathways that are implicated with MMP9 gene dysregulation in breast, lung, and colon cancers. A comprehensive analysis identified a total of 34 genes and 21 pathways for the aforementioned cancer types. Specifically, in lung cancer, notable genes, including CRKL, URGCP, SDF-1α, SEMA4b, ATDC, PTTG1, FAK, TIMP2, MMP3, TCF2 (HNF1β), and KISS1, were implicated alongside crucial pathways, namely the plasmin-dependent pathway, NF-κB, CXCR4/ERK/NF-κB, EGFR signaling pathway, and JNK pathways. Furthermore, genes such as TSP1, Clusterin (CLU), RasGRF2, piwi-like protein 2 (Piwil2), CTHRC1, VEGF, TIMP2, MMP2, and uPA, alongside pathways such as Src/PI 3-kinase, NF-κB pathway, MAPK/ERK, PI3K/Akt, JNK-activated AP-1, ROS-dependent ERK1/2, p38-MAPK-activated, and NOTCH1 signaling pathways were implicated in colon cancer^[21,22,23].

Similarly, genes such as Heregulin-β1 (NRG-1), p53, CDC42, CD44, EGF, Ets-1, TGFβ, EGFR, TNF-β, MMP2, TIMP1, TIMP2, syndecan-2, and syndecan-4, along with pathways such as ERK, MAPK, PI3K, PKC, p38 kinase, JAK3/ERK, PI3K/AKT, and NF-κB pathways were implicated in breast cancer^[24,25,26]. Notably, TIMP2 gene and NF-κB pathway were found to be implicated across all three cancers. MMP2 gene, along with the MAPK and PI3K pathways, were identified in both colon and breast cancers^[27,28,29]. The genes and pathways implicated with MMP9 in aforementioned cancers are listed in Table 5.

Identification of the MMP9 functional partners was conducted through the STRING database. Genes such as TIMP1, TIMP2, LCN2, CD44, THBS1, TIMP3, CTSG, DMP1, ELN, and MMP1 exhibited the highest interactions with MMP9, with scores ranging from 0.999 to 0.974, respectively. The detailed list of genes and their corresponding scores is provided in Table 6. A visual representation of interactions between MMP9 and its functional partners can be observed in Figure 1H.

Moreover, direct interactions of MMP9 with VCAN, MEP1A, and MEP1B were identified. A high-confidence interaction was observed with VCAN, with an MI score of 0.6, while medium-confidence interactions were noted with MEP1A and MEP1B, each with an MI score of 0.56. The interaction network is illustrated in Figure 1I.

Furthermore, MMP9 tissue RNA expression clustering was performed to identify the nearest neighboring genes correlated with the annotated functions. The analysis revealed that MMP9 is part of the “lymphoid tissue & bone marrow – innate immune response” cluster with confidence 1. The top 15 neighbor genes NKG7, CLEC12A, CTSW, NFE2, GNLY, PIK3R5, RETN, PADI4, IL18RAP, FMNL1, FOLR3, SLFN14, ADGRG3, HK3, and GYPE showed correlation scores ranging from 0.984 to 0.963 in clusters 6, 81, and 77. The top neighboring genes, their description, along with their respective correlation scores and cluster numbers are listed in Table 7.