Annals of Surgical Oncology

, Volume 18, Issue 8, pp 2348–2356

Role of MMP14 Gene Polymorphisms in Susceptibility and Pathological Development to Hepatocellular Carcinoma

Authors

  • Tzy-Yen Chen
    • School of MedicineChung Shan Medical University
    • Department of Internal MedicineChung Shan Medical University Hospital
  • Yi-Ching Li
    • Department of Pharmacology, School of MedicineChung Shan Medical University
    • Department of Medical ResearchChung Shan Medical University Hospital
  • Yu-Fan Liu
    • Department of Biomedical SciencesChung Shan Medical University
  • Chiung-Man Tsai
    • Chest Hospital, Department of Health, Executive Yuan
  • Yi-Hsien Hsieh
    • Institute of Biochemistry and BiotechnologyChung Shan Medical University
  • Chiao-Wen Lin
    • Institute of Biochemistry and BiotechnologyChung Shan Medical University
  • Shun-Fa Yang
    • Department of Medical ResearchChung Shan Medical University Hospital
    • Institute of MedicineChung Shan Medical University
    • Graduate Institute of Applied Science of LivingTainan University of Technology
Translational Research and Biomarkers

DOI: 10.1245/s10434-011-1574-x

Cite this article as:
Chen, T., Li, Y., Liu, Y. et al. Ann Surg Oncol (2011) 18: 2348. doi:10.1245/s10434-011-1574-x

Abstract

Background

Early detection of hepatocellular carcinoma (HCC) is seldom available because of the lack of reliable markers. Matrix metalloproteinase (MMP) 14 is a cell surface proteinase that displays a broad spectrum of activity against extracellular matrix components and promotes the invasion/metastasis of cells. MMP14 is overexpressed in HCC, and the level is correlated with poor overall survival. The purpose of this study was to examine whether the MMP14 gene polymorphisms are associated with the susceptibility and clinicopathological development of HCC.

Methods

A total of 135 patients with HCC and 496 healthy control subjects were recruited. Six single nucleotide polymorphisms (SNPs) of MMP14 genes were analyzed by polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP) genotyping and haplotype-base analysis.

Results

A significant (p < 0.05) lower risk for HCC was shown in the individuals with MMP14 +6767 G/A and +7096 C/C genotypes compared with those with corresponding wild-type homozygotes; high frequency for anti-hepatitis C virus and cirrhosis positive were shown in the HCC patients with MMP14 +7096 TC/CC genotype after adjusting for other confounding factors. The distribution frequency of −165 T: +221 T: +6727 C: +6767 G: +7096 T: +8153 G haplotype and diplotype was significantly higher in the HCC patients than healthy control subjects.

Conclusions

The +6767 and +7096 polymorphic genotypes and haplotype −165 T: +221 T: +6727 C: +6767 G: +7096 T: +8153 G of MMP14 gene might contribute to the prediction of susceptibility and pathological development to HCC.

Matrix metalloproteinases (MMPs) are a group of proteolytic enzymes that play a vital role in the proteolysis of structure and signaling components of extracellular matrix (ECM) and in the influence on differentiation, migration, invasion, and proliferation of cells.1,2 Among these MMPs, MMP14 (also called membrane type 1 MMP, MT1-MMP) usually forms a trimolecular complex with tissue inhibitor of metalloproteinase-2 (TIMP-2) and proMMP-2 on cell surface; proMMP-2 can be activated and released at a low concentration of TIMP-2 to mediate the ECM degradation.1,3,4 MMP14 is able to be a pericellular collagenase against ECM components directly.5 In a variety of cell types, MMP14 is employed to act on their surrounding environment and to promote tissue remodeling, invasion, and metastasis by catalyzing pericellular collagenolysis.69 By this way, it provides advantageous circumstance for cells proliferating in a constrained three-dimensional collagen or fibrin microenvironment. MMP14 is overexpressed in malignant ovarian tumors and HCC, and the level of expression is correlated with poor overall survival.913 Cell behavior, such as invasion, migration, and growth, and angiogenesis formation, can be altered by MMP14 proteolysis. MMP14 is thereby considered a key protein in physiological and pathological processes from normal cell development to cancer cell growth.14,15 Therefore, it has been suggested that the function and activity of MMP14 might involve to the modulation of susceptibility or clinicopathological features of a cancer.

Epidemiological studies have provided evidences that genetic factor is one of important variants for mediating an individual’s susceptibility to cancer. Single nucleotide polymorphism (SNP) is the most common type of DNA sequence variation occurring when a single nucleotide in the shared sequence of a gene differs between members of a species or paired chromosomes in an individual, which is thought to be associated with the occurrence and development of certain disease.16 To predict the risk and the prognosis of cancer, genotyping of related SNPs might provide a simple and valuable method. Hepatocellular carcinoma (HCC) is one of the most common malignancies worldwide and the second leading cause of cancer-related death in Taiwan.17 Carcinogenesis of HCC is a multistep and complex process, and it is known that multiple risk factors, including chronic hepatitis B virus (HBV) or hepatitis C virus (HCV) infection, cirrhosis, carcinogen exposure, and a variety of SNP, contribute to hepatocarcinogenesis.1720 Cumulative studies have implicated the associations between HCC cancer risk and SNPs in selected candidate genes. For example, insulin-like growth factor (IGF)-2, IGF-2R, and plasminogen activator inhibitor (PAI)-1 are suggested to be predictive factors related to high risk for HCC.17,21 Therefore, the identification of a SNP or combinative interaction of several SNPs in certain genes related to HCC might help to elucidate the complex process of hepatocarcinogenesis and improve the scientific basis for preventive interventions.

Although the development of HCC may take 20 to 50 years, early detection of this cancer is seldom available because the lack of reliable markers.22 It has been known that the aberrations in some genes might be responsible for certain clinical features of HCC.23 The level of MMP14 has been shown a correlation with tumor recurrence and with poor survival in neck node-positive cases.24 Expression of a gene is able to be affected by a SNP located within the promoter or other regulatory regions of the gene. Thus, study on individual or a series of SNPs in a possible aberrant gene might provide information for the occurrence and development of cancer. In the present study, a case-control association study was performed for six SNPs, which located in the promoter or exon region of MMP14 gene (Fig. 1) to analyze the role of MMP14 gene polymorphisms in susceptibility and pathological development to HCC.
https://static-content.springer.com/image/art%3A10.1245%2Fs10434-011-1574-x/MediaObjects/10434_2011_1574_Fig1_HTML.gif
Fig. 1

Exon and intron structure of MMP14 gene in Homo sapiens and the features of SNP of MMP14 gene, which were used to analyze in this study. Exons are shown by the filled box and are numbered 1 to 10 from the 5’- to 3’-end of the gene; introns are shown by the thin line; the untranslated portions of the gene are indicated by the unfilled box; and the SNP of MMP14 gene are indicated by the black arrow and labeled A to F. The start codon (ATG) and stop codon (TGA) are shown in exons 1 and 10, respectively

Materials and Methods

Subjects and Specimen Collection

This study was a hospital-based case-control study. A total of 135 (92 men and 43 women; mean age = 62.24 ± 11.08 years) patients were recruited as a case group between 2007 and 2010 at Chung Shan Medical University Hospital in Taichung, Taiwan. Meanwhile, 496 race- and ethnic group-matched healthy individuals (411 men and 85 women; mean age = 53.03 ± 14.46 years) who visited the same hospitals were randomly selected from the same geographic area to be a control group. The patients were diagnosed with HCC according to the characteristic criteria of the national guidelines for HCC, such as liver injury diagnosed by histology or cytology irrespective of α-fetoprotein (AFP) titer where imaging data showed one of following three cases: (1) one or more liver masses ≥2 cm in diameter via both computed tomography and magnetic resonance imaging; (2) one imaging data with early enhancement and a high level of AFP ≥ 400 ng/mL; (3) one imaging data with early arterial phase contrast enhancement plus early venous phase contrast washout regardless of AFP level.17 The whole blood specimens, collected from healthy controls and HCC patients, were placed in tubes containing ethylenediaminetetraacetic acid (EDTA) and immediately centrifuged and stored at −80°C. Associated clinicopathological characteristics, such as hepatitis B surface antigen (HBsAg), antibody to HCV (anti-HCV), liver cirrhosis history, Child-Pugh grade, AFP, aspartate aminotransferase (AST), alanine aminotransferase (ALT), and stage of HCC, were verified by chart review. Before the conduction of this study, approval from the Institutional Review Board of Chung Shan Medical University Hospital was obtained, and informed, written consent was obtained from each individual.

Genomic DNA Extraction

Genomic DNA was extracted by QIAamp DNA blood mini kits (Qiagen, Valencia, CA) according to the instructions of manufacturer. DNA was dissolved in TE buffer [10 mM Tris (pH 7.8), 1 mM EDTA] and then quantified by a measurement of OD260. Final preparation was stored at −20°C and used as templates for polymerase chain reaction (PCR).

Polymerase Chain Reaction Restriction Fragment Length Polymorphism

The MMP14 −165 (rs100349), MMP14 +7096 (rs2236307), and MMP14 +8153 (rs3751489) gene polymorphisms were determined by polymerase chain reaction restriction fragment length polymorphism (PCR-RFLP) assay. The forward and reverse primers used for analysis of MMP14 gene polymorphisms were designed as 5′-CACAAAAAGGCAACTTAGAGGTGTT-3′ and 5′-GTGGTTGTTTTAGCCTGAATCCAAT-3′ (for MMP14 -165, 215 bps); 5′-GTAGTCTACACCCACGCCTG-3′ and 5′-GACAAACATCTCCCCTCGGA-3′ (for MMP14 +7096, 442 bps); 5′-GGAAGGATGGCAAATTCGTC-3′ and 5′-CATTTCTAGGCAACAGCAGA-3′ (for MMP14 +8153, 447 bps). The PCR was performed in a 10-μL volume containing 100-ng DNA template, 1.0 μL of 10× PCR buffer (Invitrogen, Carlsbad, CA), 0.25 U of Taq DNA polymerase (Invitrogen, Carslbad, CA), 0.2-mM dNTPs (Promega, Madison, WI), and 200 nM of each primer (MDBio Inc, Taipei, Taiwan). The PCR cycling conditions were 5 min at 94°C followed by 35 cycles of 1 min at 94°C, 1 min at 64°C, and 2 min at 72°C, with a final step at 72°C for 20 min. Ten-microliter aliquot of PCR product was subjected to digestion at 37°C for 4 h in a 15-μL reaction buffer containing 1.5 μL of 10× buffer and 5 U of BsrI, HphI, and NcoI (New England Biolabs, Beverly, MA) for MMP14 −165, +7096, and +8153, respectively. Digested products were separated on a 2% agarose gel and then stained with ethidium bromide. For MMP14 −165, G allele yielded 168- and 47-bp products, whereas T alleles yielded a 215-bp product. For MMP14 +7096, T allele yielded 302- and 140-bp products, and C alleles yielded a 442-bp product. For MMP14 +8153, G allele yielded 447-bp products, and A alleles yielded 290- and 157-bp products. Furthermore, the genotypes determined by PCR-RFLP were confirmed by DNA sequencing analysis. For each assay, a negative control (without DNA template) was added to monitor PCR contamination. After genotyping each genetic polymorphism, 20–25% of the samples in each genotype group were randomly selected for repeated assays to validate the results.

Real-Time PCR

The allelic discrimination of the MMP14 +221 (rs1042703), MMP14 +6727 (rs2236302), and MMP14 +6767 (rs1042704) gene polymorphisms were assessed with the ABI StepOne™ Real-Time PCR System (Applied Biosystems) and analyzed using SDS v3.0 software (Applied Biosystems), using the TaqMan assay (assay IDs: C_3036225_20 for MMP14 +221, C_15954018_20 for MMP14 +6727, and C_11436237_20 for MMP14 +6767). The final volume for each reaction was 5 μL, containing 2.5 μL TaqMan Genotyping Master Mix, 0.125 μL TaqMan probes mix, and 10 ng genomic DNA. The real time PCR reaction included an initial denaturation step at 95°C for 10 min, followed by 40 cycles, each consisting of 95°C for 15 s and 60°C for 1 min.

Statistical Analysis

Hardy–Weinberg equilibrium was assessed using a goodness-of-fit χ2 test for biallelic markers. Mann–Whitney U test and Fisher’s exact test were used to compare the differences of demographic characteristics distributions between healthy control subjects and patients with HCC. The adjusted odds ratios (AORs) with their 95% confidence intervals (CIs) of the association between genotype frequencies and HCC risk as well as clinical pathological characteristics were estimated by multiple logistic regression models after controlling for other covariates. The haplotype-based analysis was using the Phase program. A p value < 0.05 was considered significant. The data were analyzed on SAS statistical software (Version 9.1, 2005; SAS Institute Inc., Cary, NC).

Results

The statistical analysis of demographic characteristics showed that 207 of 496 healthy control subjects (41.7%) and 47 of 135 HCC patients (34.8%) had the experience of alcohol consumption; 205 of 496 healthy control subjects (41.3%) and 55 of 135 HCC patients (40.7%) had the experience of smoking. The distributions of age and gender in our recruited individuals are described in Materials and Methods-subjects and specimen collection section. A significant difference of the distributions of age (p < 0.0001) and gender (p < 0.0001) but not alcohol (p = 0.087) and tobacco consumptions (p = 0.491) between healthy control subjects and HCC patients was shown. Hence, the adjusted odds ratios with 95% confidence intervals were estimated by multiple logistic regression models after controlling for age and gender in each comparison.

In our recruited control group, all analyzed gene markers were in Hardy–Weinberg equilibrium (p > 0.05). The data in Table 1 show that the alleles with the highest distribution frequency for −165, +221, +6727, +6767, +7096, and +8153 loci of MMP14 genes were G/T, T/T, C/C, G/G, T/C, and G/G, respectively, in both HCC patients and healthy control subjects. According to the AORs with their 95% CIs for HCC of MMP14 gene polymorphisms, +6767 G/A and +7096 C/C exhibited a significant (p < 0.05) lower risk of 0.253-fold (95% CI, 0.066–0.964) and 0.299-fold (95% CI, 0.138–0.647), respectively, to have HCC compared with their corresponding wild-type homozygotes after adjusting for other confounding factors. To explore the impact of polymorphic genotype of MMP14 on pathological development of HCC, we further classified the HCC patients and healthy control subjects into two subgroups each: one subgroup with at least one polymorphic allele and the other subgroup with homozygous wild-type alleles. The data of statistical analysis showed that no significant difference in any allele frequency distribution between HCC patients and healthy control subjects, and none of genotypic frequencies of these SNPs but +7096, showed a significant association with clinical pathological variables in HCC patients. Compared with the wild-type genotype, patients with at least one polymorphic C allele of MMP14 +7096 showed a low ratio of 0.290-fold (95% CI, 0.094–0.899) for HBsAg-positive but a high ratio of 2.844-fold (95% CI, 1.022–7.909) and 3.278-fold (95% CI, 1.028–10.45) for anti-HCV-positive and liver cirrhosis, respectively (Table 2). Whereas the levels of HCC clinical pathological markers, such as AFP, AST, and ALT, had no significant differences between wild-type and polymorphic genotypes of each MMP14 SNP in HCC patients (Table 3).
Table 1

Distribution frequency of MMP-14 genotypes in 496 healthy control subjects and 135 patients with HCC

Variable

Controls (N = 496)

n (%)

Patients (N = 135)

n (%)

OR (95% CI)

AOR (95% CI)

−165

 GG

155 (31.3%)

42 (31.1%)

1.00

1.00

 GT

262 (52.8%)

70 (51.9%)

0.986 (0.641–1.157)

0.918 (0.548–1.036)

 TT

79 (15.9%)

23 (17%)

1.074 (0.604–1.912)

0.996 (0.477–2.079)

+221

 TT

496 (100%)

134 (99.3%)

1.00

1.00

 TC

0 (0%)

1 (0.7%)

 CC

0 (0%)

0 (0%)

+6727

 CC

405 (81.7%)

108 (80%)

1.00

1.00

 CG

85 (17.1%)

24 (17.8%)

1.059 (0.642–1.746)

1.604 (0.88–2.923)

 GG

6 (1.2%)

3 (2.2%)

1.875 (0.461–7.619)

1.549 (0.28–8.559)

+6767

 GG

468 (94.4%)

132 (97.8%)

1.00

1.00

 GA

28 (5.6%)

3 (2.2%)

0.380 (0.114–1.269)

0.253 (0.066–0.964)*

 AA

0 (0%)

0 (0%)

+7096

 TT

130 (26.2%)

49 (36.3%)

1.00

1.00

 TC

251 (50.6%)

67 (49.6%)

0.708 (0.463–1.083)

0.662 (0.399–1.098)

 CC

115 (23.2%)

19 (14.1%)

0.438 (0.244–0.738)*

0.299 (0.138–0.647)*

+8153

 GG

481 (97%)

130 (96.3%)

1.00

1.00

 GA

15 (3%)

4 (3%)

0.987 (0.322–3.023)

1.365 (0.362–5.145)

 AA

0 (0%)

1 (0.7%)

Odds ratios (ORs) with their 95% confidence intervals (CIs) were estimated by logistic regression models

Adjusted odds ratios (AORs) with their 95% CIs were estimated by multiple logistic regression models after controlling for age and gender

*P values < 0.05 are statistically significant

Table 2

Adjusted odds ratio (AOR) and 95% confidence interval (CI) of clinical status and MMP-14 +7096 genotypic frequencies in 135 patients with HCC

Variable

Genotypic frequencies

TT (N = 49)

n (%)

TC + CC (N = 86)

n (%)

OR (95% CI)

AOR (95% CI)

Clinical stage

 Stage I/II

26 (53.1%)

58 (67.4%)

1.00

1.00

 Stage III/IV

23 (46.9%)

28 (32.6%)

0.546 (0.266–1.121)

0.699 (0.255–1.915)

Tumor size

 ≤T2

27 (55.1%)

60 (69.8%)

1.00

1.00

 >T2

22 (44.9%)

26 (30.2%)

0.532 (0.257–1.100)

0.717 (0.26–1.975)

Lymph node metastasis

 No

47 (95.9%)

81 (94.2%)

1.00

1.00

 Yes

2 (4.1%)

5 (5.8%)

1.451 (0.271–7.774)

0.762 (0.062–9.308)

Distant metastasis

 No

46 (93.9%)

83 (96.5%)

1.00

1.00

 Yes

3 (6.1%)

3 (3.5%)

0.554 (0.107–2.858)

0.560 (0.058–5.416)

Child-Pugh grade

 A

36 (73.5%)

63 (73.3%)

1.00

1.00

 B or C

13 (26.5%)

23 (26.7%)

1.011 (0.457–2.236)

0.777 (0.264–2.286)

HBsAg

 Negative

24 (49.0%)

57 (66.3%)

1.00

1.00

 Positive

25 (51.0%)

29 (33.7%)

0.488 (0.239–1)*

0.29 (0.094–0.899)*

Anti-HCV

 Negative

28 (57.1%)

35 (40.7%)

1.00

1.00

 Positive

21 (42.9%)

51 (59.3%)

1.943 (0.954–3.955)

2.844 (1.022–7.909)*

Liver cirrhosis

 Negative

16 (32.7%)

19 (22.1%)

1.00

1.00

 Positive

33 (67.3%)

67 (77.9%)

1.71 (0.78–3.748)

3.278 (1.028–10.45)*

ORs with their 95% CIs were estimated by logistic regression models

AORs with their 95% CIs were estimated by multiple logistic regression models, after controlling for age and gender

>T2: multiple tumor >5 cm or tumor involving a major branch of the portal or hepatic vein(s)

*P values < 0.05 are statistically significant

Table 3

Association of MMP-14 genotypic frequencies with HCC laboratory status

Characteristic

α-Fetoprotein (ng/mL)

AST (IU/L)

ALT (IU/L)

AST/ALT ratio

−165

 GG

2831.4 ± 1469

261.5 ± 80.2

203 ± 61.9

1.49 ± 0.1

 GT/TT

3661 ± 1781.9

144.2 ± 34.6

115.7 ± 16.7

1.51 ± 0.13

 P value

0.77

0.073

0.073

0.937

+221a

 TT

3187 ± 1298.7

181.6 ± 30.5

140.6 ± 22.6

1.5 ± 0.1

 TC/CC

32335.6

57

449

0.13

 P value

N.D.

N.D.

N.D.

N.D.

+6727

 CC

3165.4 ± 1528.9

171.9 ± 34.2

136.8 ± 25.9

1.53 ± 0.11

 CG/GG

4353.2 ± 2344.9

215.5 ± 66.5

166.9 ± 44.7

1.4 ± 0.16

 P value

0.718

0.567

0.596

0.603

+6767

 GG

3453.2 ± 1336.3

179.4 ± 30.9

141 ± 22.8

1.51 ± 0.1

 GA/AA

1189.8 ± 1186.19

235 ± 158.5

222.3 ± 176.5

1.29 ± 0.27

 P value

0.799

0.788

0.597

0.737

+7096

 TT

3777.4 ± 2693.1

207.5 ± 69.8

170 ± 52.7

1.51 ± 0.16

 TC/CC

3189.6 ± 1375.7

165.4 ± 26.5

127.4 ± 18.9

1.5 ± 0.12

 P value

0.83

0.506

0.366

0.936

+8153

 GG

3512.4 ± 1356.5

183.7 ± 31.4

144 ± 23.3

1.52 ± 0.1

 GA/AA

557.1 ± 480.1

100.8 ± 45.4

113.4 ± 60.1

1.05 ± 0.27

 P value

0.671

0.607

0.799

0.359

Mann–Whitney U test was used between two groups

Data are mean ± standard error of the mean unless otherwise indicated

a+221 only one case in TC/CC group

The haplotype distributions of MMP14 −165, +221, +6727, +6767, +7096, and +8153 were further evaluated, and 12 haplotypes were derived from these 6 SNPs in our recruited individuals. The most common haplotype in control was GTCGCG (37.5%), and it thereby been chosen as reference. Comparing with the reference, three minor haplotypes, TTCGCG, TTCGTG, and TTCATG, significantly (p < 0.05) changed the risk for HCC to 0.281-fold (95% CI, 0.126–0.625), 1.537-fold (95% CI, 1.113–2.122), and 0.148-fold (95% CI, 0.02–1.099), respectively (Table 4). Because the risk for HCC was changed by these haplotypes, we thereby inferred that the diplotypes should decrease or increase the HCC risk more remarkable than their corresponding haplotypes. The speculation is applied to that of TTCGTG only. As the data show in Table 5, TTCGTG/TTCGTG diplotype conferred a 3.079-fold (95% CI, 1.588–5.968), approximately twofold of TTCGTG/others diplotype (odds ratio (OR), 1.537; 95% CI, 1.008–2.284) increased risk of HCC.
Table 4

Distribution frequency of MMP-14 haplotype in healthy control subjects and HCC patients

Variable

Controls (N = 992)

n (%)

Patients (N = 270)

n (%)

OR (95% CI)

P

−165 G/T

+221 T/C

+6727 C/G

+6767 G/A

+7096 T/C

+8153 G/A

G

T

C

G

C

G

372 (37.5%)

90 (33.3%)

Reference

 

G

T

C

G

C

A

6 (0.6%)

2 (0.7%)

1.378 (0.274–6.94)

0.485

G

T

C

G

T

G

100 (10.1%)

30 (11.1%)

1.24 (0.776–1.981)

0.217

G

T

G

G

T

G

94 (9.5%)

29 (10.7%)

1.275 (0.792–2.052)

0.189

T

T

C

G

C

G

103 (10.4%)

7 (2.6%)

0.281 (0.126–0.625)*

<0.001

T

T

C

G

T

G

277 (27.9%)

103 (38.2%)

1.537 (1.113–2.122)*

0.006

T

T

C

G

T

A

9 (1.0%)

1 (0.4%)

0.459 (0.057–3.672)

0.395

T

T

C

A

T

G

28 (2.8%)

1 (0.4%)

0.148 (0.02–1.099)*

0.017

T

T

G

G

T

G

3 (0.3%)

1 (0.4%)

1.378 (0.142–13.401)

0.582

G

T

C

A

C

G

0 (0%)

2 (0.7%)

G

C

C

G

C

G

0 (0%)

1 (0.4%)

T

T

C

G

C

A

0 (0%)

3 (1.1%)

*P values < 0.05 are statistically significant

Table 5

Haplotype pairs frequency of MMP14 in healthy control subjects and patients with HCC

Haplotype/haplotype

Controls (N = 496)

n (%)

Patients (N = 135)

n (%)

OR (95% CI)

P

Others/others

248 (50%)

50 (37%)

Reference

 

TTCGTG/others

219 (44.2%)

67 (49.6%)

1.517 (1.008–2.284)*

0.028

TTCGTG/TTCGTG

29 (5.8%)

18 (13.4%)

3.079 (1.588–5.968)*

0.001

TTCGTG/others and TTCGTG/TTCGTG

248 (50%)

85 (63%)

1.7 (1.15–2.514)*

0.005

*P values < 0.05 are statistically significant

Discussion

The familiar major etiologies for HCC in Taiwan include infection with HBV or HCV, habitual alcohol consumption, disease history of liver cirrhosis, and family history of HCC.2527 In this study, however, lower ratio of individuals with alcohol consumption in HCC patients (34.8%) than in control (41.7%) was observed. It was hence suggested that these risk factors cannot fully explain the entire pathogenesis of HCC. Many gene polymorphisms and somatic mutations have been identified as being associated with the risk of HCC.17,21,28,29 Increasing evidence has revealed that the genomic changes progressively might alter the cellular phenotype to evolve from preneoplastic stage into HCC.23 These findings suggest that genetic components may play a pivotal role in the occurrence of HCC. Multiple gene alterations, such as allelic deletion, insertion, polymorphism, mutation, and methylation change, are marked in HCC, and therefore cause genetic and molecular aberrations.30 Comparing HCC patients with control subjects without HCC, genetic information is particularly valuable to mark a target gene for predicting risk and pathological development of HCC.

The overexpression of MMP14, which promotes intrahepatic metastasis, is present in HCC.11,31,32 The outcome of HCC patients is hence affected along with the bioactivity or availability of MMP14.33,34 Our data (Table 1) demonstrated that MMP14 +6767 G/A and +7096 C/C genotypes have a lower risk for HCC compared with their corresponding wild-type genotypes. The site of +6767 and +7096 is located in the exon 5 (+6639 ~ + 6800) and near 5′-end of exon 6 (+7092 ~ + 7252), respectively, of MMP14 gene. The +6767 G/A cause protein residue N (asparagine) is substituted for D (aspartic acid), which might change the original activity or function of MMP14 protein. Although the +7096 C/C cause a synonymous change in amino acid residue (from glycine into glycine), the splicing efficiency of MMP14 mRNA might be altered because the nucleotide substitution is so close to 5′-end of exon; the translational rate also might be changed via the modification of the mRNA stability or the ribosome binding.35 It has been reported that the overall overexpression of MMP14 is associated with poor HCC prognosis, and MMP14 was found in centromere leading to chromosome instability in Madin Darby canine kidney (MDCK) cells.33,34,36,37 The +6767 G/A and +7096 C/C allele might reduce the bioactivity or expression of MMP14, and the cellular behaviors are thereby aberrant. The ratiocination provides a reasonable explanation for the low risk to have HCC of these two MMP14 polymorphic alleles (Table 1).

Comparing the levels of clinical pathological markers, such as AFP, AST, and ALT, no significant difference between the wild-type and polymorphic genotypes of each MMP14 SNP in HCC patients was observed (Table 3). Nevertheless, significant (p < 0.05) pathological characteristics of HCC appeared in the HCC patients with at least one polymorphic C allele of MMP14 +7096 genotypes (Table 2). In Asia, hepatitis virus infection is a highly endemic factor for HCC. One study surveyed the demographic, clinical, and virological characteristics of 414 patients from Asia who had HCC; the results showed that HBV infection is the most frequent cause for HCC in China, whereas HCV is more common in Japan.38 Interestingly, the results of low (AOR, 0.29; 95% CI, 0.094–0.899) and high (AOR, 2.844; 95% CI, 1.022–7.909) ratio of HBsAg-positive and anti-HCV-positive, respectively, indicates that the HCC patients of Taiwanese population with MMP14 +7096 TC/CC genotypes might be more sensitive to HCV infection than to HBV. Hepatitis virus infection is associated with the increase of oxidative stress in liver cells and results in DNA changes and instability, thus, increasing the risk of developing cirrhosis and/or HCC.3942

For certain genes, a SNP arising in the coding, promoter, or regulatory region may have functional consequences. Many SNPs are silent, that is to say, with no direct effect on the gene products. However, by virtue of the linkage disequilibrium existing across the human genome, they can still be used as genetic markers to locate adjacent functional variants that contribute to a disease. When each SNP constructing the haplotypes has a true contribution to the susceptibility of disease, even though unapparent, haplotype analysis provide a greater statistical power and sometimes advantageous over an individual SNP analysis for the detection of an association between the alleles and a disease phenotype.43 The −165 G/T polymorphism in the promoter region may affect the transcriptional activity. The +221 C/T, +6767 G/A, and +8153 G/A polymorphisms cause non-synonymous substitutions (Pro8Ser, Asp273Asn, and Arg431His, respectively), which might affect the function of protein; and even the synonymous +6727 C/G (Pro259Pro) and +7096 T/C (Gly285Gly) polymorphisms might be able to change the translational rate or mRNA stability of MMP14 protein. However, the functional effect of each polymorphism constructing the haplotype −165 G/T: +221 T/C: +6727 C/G: +6767 G/A: +7096 T/C: +8153 G/A still cannot be speculated from the above-mentioned information. In this study, the TTCGTG haplotype and diplotype that had a correlation to the high risk of HCC were obtained by the haplotype-based statistical analysis (Tables 4, 5). It is possible that the haplotype TTCGTG of MMP14 is in linkage disequilibrium with other functional polymorphisms that are responsible for the susceptibility to HCC.

In conclusion, MMP14 +6767 G/A and +7096 C/C both exhibited a significant (p < 0.05) lower risk to have HCC compared with their corresponding wild-type homozygotes and the HCC patients with MMP14 +7096 TC/CC genotype showed high frequency for anti-hepatitis C virus and cirrhosis positivity after adjusting for other confounding factors. The distribution frequency of −165 T: +221 T: +6727 C: +6767 G: +7096 T: +8153 G haplotype and diplotype was significantly higher in the HCC patients than the healthy control subjects. It was suggested that the +6767 and +7096 polymorphic genotypes and haplotype −165 T: +221 T: +6727 C: +6767 G: +7096 T: +8153 G of MMP14 gene might contribute to the prediction of susceptibility and pathological development to HCC.

Acknowledgment

This study was supported by a research grant from Chung Shan Medical University Hospital, Taiwan (CSH-2011-C-006).

Conflict of interest

No interest conflicts.

Copyright information

© Society of Surgical Oncology 2011