-
A total of 509 participants were identified; 125 were excluded, leaving a final dataset of 384 participants who were enrolled. The baseline characteristics are shown in Table 1. The median age was 52 (IQR, 44–58) years for VS patients, which was lower than that for the controls (54; IQR, 47.5–60 years). VS patients had higher body weight (72.11 ± 12.29), BMI (24.82 ± 2.81), SBP (121.97 ± 11.62), DBP (49.36 ± 8.60), PP (49.36 ± 8.60), MAP (89.06 ± 9.04), and FG (5.03 ± 0.50) than the controls (all P < 0.05). No significant differences were found in sex, height, HR, HDL, LDL, TG, TC, Cr, or UA between VS patients and controls. The four significant variables including age, BMI, SBP, and PG were adjusted by later binary multivariate logistic regression and GMDR analysis.
Characteristics VS patients (n = 151) Controls (n = 233) P-values Age, year 52.01 ± 9.04 54.54 ± 9.71 0.023* Sex, male 88.0 (58.3%) 113.0 (48.5%) 0.061 Weight, kg 72.11 ± 12.29 68.28 ± 10.49 0.008** Height, cm 169.97 ± 7.71 168.57 ± 7.20 0.770 BMI, kg/m2 24.82 ± 2.81 23.92 ± 2.49 0.005** HR, time/min 69.87 ± 10.15 68.08 ± 10.37 0.068 SBP, mmHg 121.97 ± 11.62 116.06 ± 10.12 0.000** DBP, mmHg 72.61 ± 8.93 68.64 ± 8.01 0.000** PP, mmHg 49.36 ± 8.60 47.43 ± 8.13 0.024* MAP, mmHg 89.06 ± 9.04 84.45 ± 7.89 0.000** HDL, mmol/L 1.48 ± 0.31 1.52 ± 0.31 0.292 LDL, mmol/L 2.72 ± 0.41 2.74 ± 0.45 0.424 TG, mmol/L 1.09 ± 0.30 1.06 ± 0.28 0.307 TC, mmol/L 5.10 ± 0.73 5.06 ± 0.81 0.925 FG, mmol/L 5.03 ± 0.50 4.92 ± 0.51 0.036* Cr, μmol/L 89.05 ± 11.81 87.72 ± 10.64 0.162 UA, μmol/L 333.30 ± 58.96 323.29 ± 67.87 0.307 Note. Data were represent as mean ± SD; VS: vascular senescence; BMI: body mass index. HR: heart rate; SBP: systolic blood pressure; DBP: diastolic blood pressure; PP: pulse pressure (SBP-DBP); MAP: mean arterial pressure; HDL: high-density lipoprotein; LDL: low-density lipoprotein; TG: triglycerides; TC: total cholesterol; FG: fasting glucose; Cr: creatinine; UA: uric acid; s: standard deviation; *P < 0.05; **P < 0.01. Table 1. Baseline demographic and biochemical characteristics of the study cohort
-
The 14 selected SNPs from 5 genes in the MMP-2 signaling pathway comprised rs4586, rs13900, and rs1860189 in CCL2; rs14070, rs2287074, rs7201, rs243849, and rs1053605 in MMP-2; rs9493150 in CCN2; rs3810818 and rs8182173 in VASN; rs7503726, rs2277698, rs8179090, and rs8179096 in TIMP2. Fourteen SNP loci were genotyped, and homogeneous genotypic and allelic frequencies were analyzed in VS patients and controls. All genotype distributions were within the Hardy–Weinberg equilibrium except for TIMP2 rs8179096. A positive association was found between VS occurrence and the four polymorphisms: CCL2 rs4586, MP2 rs14070, MMP2 rs7201, and MMP2 rs1053605. A significant association was found between VS and the T/C (P = 0.011) and T/T (P = 0.019) genotypes and with the T allele (P = 0.000) of the CCL2 rs4586 polymorphism. Carriers of the T/C genotype of MMP-2 rs14070 had a 2.17-fold increased risk of developing VS compared with those of the C/C genotype (P = 0.001), and those of the T/T genotype had a 19.375-fold increased risk (P = 0.001). Similarly, carriers of the T allele of MMP-2 rs14070 had a greater risk of developing VS than those of the C allele (P = 0.000). Carriers of the C/A (P = 0.001) and C/C (P = 0.001) genotypes of MMP-2 rs7201 demonstrated a stronger association with VS Individuals carrying the A allele of MMP-2 rs7201 had a greater risk of VS than those carrying the C allele (P = 0.000). Significant associations between VS and the T/T genotype (P = 0.016) as well as the T allele (P = 0.008) of MMP-2 rs1053605 were observed. The multivariate regression analyses before and after bootstrap correction are shown in Table 2, and the χ2 analysis is shown in Table 3.
SNPs VS patients (n = 151) Controls (n = 233) OR (95% CI) P-valuea P-valueb CCL2 rs4586, n (%) CC 64 (42.4) 84 (36.1) TC 52 (34.4) 125 (53.6) 0.546 (0.345, 0.864) 0.011* 0.019* TT 35 (23.2) 24 (10.3) 1.914 (1.037, 3.533) 0.019* 0.029* χ2 (P-value) 5.18 (0.07) MMP-2 rs14070, n (%) CC 54 (35.8) 155 (66.5) TC 70 (46.4) 74 (31.8) 2.715 (1.731, 4.259) 0.001** 0.001** TT 27 (17.9) 4 (1.7) 19.375 (6.483, 57.903) 0.001** 0.001** χ2 (P-value) 2.11 (0.35) MMP-2 rs7201, n (%) AA 58 (38.4) 153 (65.7) CA 71 (47.0) 75 (32.2) 2.497 (1.603, 3.891) 0.001** 0.001** CC 22 (14.6) 5 (2.1) 11.607 (4.198, 32.091) 0.001** 0.001** χ2 (P-value) 1.46 (0.48) MMP-2 rs1053605, n (%) CC 117 (77.5) 199 (85.4) TC 27 (17.9) 32 (13.7) 1.435 (0.819, 2.514) 0.216 0.286 TT 7 (4.6) 2 (0.9) 5.953 (1.216, 29.133) 0.016* 0.017* χ2 (P-value) 0.31 (0.85) Note. aMultivariate logistic regression analysis adjusted for age, systolic blood pressure, BMI and fasting glucose levels; bAfter bootstrap 1,000 replications; SNP: single nucleotide polymorphism; VS: vascular senescence; OR: odds ratio; CI: confidence interval; CCL2: chemokine (C‐C motif) ligand 2; MMP-2: matrix metalloproteinase 2; BMI: body mass index; *P < 0.05; **P < 0.01. Table 2. Genotype and odds ratio of SNPs associated with vascular senescence
SNPs VS patients (n = 151) Controls (n = 233) χ2 P-value CCL2 rs4586, n (%) C 180 (59.6) 293 (62.9) 37.278 < 0.001** T 122 (40.4) 173 (37.1) MMP-2 rs14070, n (%) C 178 (58.9) 357 (76.6) 51.395 < 0.001** T 124 (41.1) 55 (11.8) MMP-2 rs7201, n (%) A 187 (61.9) 381 (81.8) 37.447 < 0.001** C 115 (38.1) 85 (18.2) MMP-2 rs1053605, n (%) C 261 (86.4) 430 (92.3) 6.954 0.008** T 41 (13.6) 36 (7.7) Note. SNP: single nucleotide polymorphism; VS: vascular senescence; CCL2: chemokine (C‐C motif) ligand 2; MMP-2: matrix metalloproteinase 2; **P < 0.01. Table 3. Allele frequencies and χ2 analyses
-
All possible combinations of polymorphisms were analyzed using MDR, and the optimal model had a balanced accuracy of 0.653 and a consistency of cross-validation of 10/10 (Table 4). The GMDR model yielded similar results (Table 5). Entropy measurements were used to construct an interaction map of the CCL2 rs4586, MMP-2 rs14070, and MMP-2 rs1053605 polymorphisms, and CCL2 rs4586 and MMP-2 rs14070 showed strong interactions, with information gain values of 3.01% (Figure 2).
Model Training-balanced accuracy Testing-balanced accuracy Cross-validation consistency MMP-2 rs14070 0.6538 0.6538 10/10 MMP-2 rs14070, MMP-2 rs1053605 0.6730 0.6565 9/10 CCL2 rs4586, MMP-2 rs14070, MMP-2 rs1053605 0.7254 0.7101 10/10 Note. MDR: multifactor dimensionality reduction; CCL2: chemokine (C‐C motif) ligand 2; MMP-2: matrix metalloproteinase 2. Table 4. MDR analysis
Model Training-balanced accuracy Testing-balanced accuracy Sign test
(P-value)Cross-validation consistency MMP-2 rs14070 0.6538 0.6537 10 (0.010) 10/10 MMP-2 rs14070, MMP-2 rs1053605 0.6730 0.6566 10 (0.010) 9/10 CCL2 rs4586, MMP-2 rs14070, MMP-2 rs1053605 0.7254 0.7089 10 (0.010) 10/10 Note. GMDR: generalized multifactor dimensionality regression; SBP: systolic blood pressure; BMI: body mass index; CCL2: chemokine (C‐C motif) ligand 2; MMP-2: matrix metalloproteinase 2. Table 5. GMDR analysis adjusted forage, SBP, BMI, and fasting blood glucose
Figure 2. Interaction map for the risk of vascular senescence. Values inside nodes indicate information gain (IG) of each single nucleotide polymorphism, while values between nodes indicate IG of pairwise combinations. Orange: information gain. Green or blue: information redundancy. CCL2: chemokine (C‐C motif) ligand 2; MMP-2: matrix metalloproteinase 2.
Interactions among polymorphisms in high-risk genotypes were evaluated for CCL2 rs4586, MMP-2 rs14070, and MMP-2 rs1053605, with the strongest relationships observed for [CC+TT+CC] and [TC+TT+CC]. Low-risk genotypes were characterized by [CC+CC+CC] and [TC+CC+CC]. Likewise, for the two-locus model of MMP-2 rs14070 and MMP-2 rs1053605, the high- and low-risk genotypes were [TT+CC] and [CC+CC], respectively (Figure 3).
Figure 3. Distribution of high- and low-risk genotypes in the best three (A) and two-locus (B) models. Left bars represent the percentage of VS participants. Right bars represent the percentage of controls. Cells were at high risk if the percentage of cases/controls ≥ 1.0. VS: vascular senescence; CCL2: chemokine (C‐C motif) ligand 2; MMP-2: matrix metalloproteinase 2.
Association between Gene Polymorphisms and SNP-SNP Interactions of the Matrix Metalloproteinase 2 Signaling Pathway and the Risk of Vascular Senescence
doi: 10.3967/bes2024.016
- Received Date: 2023-10-09
- Accepted Date: 2024-01-19
-
Key words:
- Vascular senescence /
- Pulse wave velocity (PWV) /
- Single nucleotide polymorphism (SNP) /
- Matrix metalloproteinase 2 (MMP-2) /
- Extracellular matrix (ECM) /
- Structural degradation /
- Multifactor dimensionality reduction (MDR)
Abstract:
Citation: | LIAO Zhen Yu, YANG Shuo, HU Song, LIU Jia, MAO Yong Jun, SUN Shu Qin. Association between Gene Polymorphisms and SNP-SNP Interactions of the Matrix Metalloproteinase 2 Signaling Pathway and the Risk of Vascular Senescence[J]. Biomedical and Environmental Sciences, 2024, 37(2): 146-156. doi: 10.3967/bes2024.016 |