Expression Of Matrix Metalloproteinase-2 And -9 In Peripheral Blood Of Patients With Diabetic Foot

Objectives: Diabetes foot (DF) chronic unhealing ulcers commonly have such fundamental pathological changes as collagen matrix excessive decomposition leading to not deposit in local ulcer, thus not filling wound, keratinocyte failing to migrate to the surface of wound, and accompanied by ischemia, hypoxia, invasion of inflammatory cells, and microbial infection. Those process deepen and widen diabetic foot ulcers. Metalloproteinases play a key role in degradator of extracellular matrix. Previous reports found MMPs and its tissue inhibitory factors (TIMP) lose the normal dynamic balance, with MMPs increasing and TIMP decreasing. Reasearchs also confirmed that serum abnormal MMPs/TIMP proportion existed in diabetic patients with vascular lesions. But biopsy samples of diabetic foot ulcers is restricted by ethics and consent of patients, which made intensive study difficult. The aim of our study is to observe whether high MMP-2 and -9 serum levels exsit and why those get disorder in patients with diabetic foot ulcers. The reseach results might help us prevent and treat diabetic foot in clinic.Methods:1. From March 2007 to September 2009, total 190 in-patients with diabetes was involved, including 78 patients with diabetic foot ulcers (DF group) and 111 diabetes in hospital but not with diabetic foot as NDF group. In DF group, 78 cases included male 44 and female 34, aged 48⁸1 years old (average age 64.3 years), with Duration of diabetes 3⁸ years (average 14.6 years); areas of wounds were ranged from 0.5×1cm¹3×4cm, foot ulcer course from 2 weeks to 5 months, mean 1.9 months. Patients in NDF group include male 60 and female 51 cases, with duration of diabetes from 5 to 18 years, mean 13.1 years. Meanwhile 150 healthy retired people were selected as CON group at random. There were none with foot or other body ulcers, or chronic ulcer history, nearly three months without acute injury history, and all people except for cardiac insufficiency and severe liver, kidney function failure.(1) To measure the levels of TG, CHO, LDL-C, HDL-C, FINS, FPG and HbA1c in serum of CON, NDF and DF groups, count intravenous whole blood leukocyte numbers and determine the concentration AGEs, MMP-2, MMP-9 and TIMP-1 by ELISA in serum of the three groups.(2) To detect the active enzyme expression quantities of MMP-2 and MMP-9 in serum by gelatin zymography.(3) To assay the mRNA expression quantity of MMP-2 and MMP-9 in intravenous whole blood leukocyte by RT-PCR.(4) 2×2 factorial analysis design: all 78 diabetic foot were graded using university of TEXAS wound classification system, so the pathogeny is divided into infection and ischemia. Infectious factors (A) had two levels, yes (a₁) and no (a₁); similarly, ischemia factors (B) had levels of yes (b1) and no (b₂). Through the factorial analysis, we evaluate the influence of two factors and deduce the main effect and interation effect on the expression of MMPs. Firstly, we treated 2×2 factorial analysis as 4 simple factors, which were stage D (a₁b₂), Stage B (a₁b₂), Stage C (a₁b₂), Stage A (a2b2). Secondly, every patient who meet single factor of Stage A, B, C or D, would enter factorial analysis according to completely random method. Thridly, total 16 patients of four factors were involved into 2×2 factorial analysis.2. From March 1999 to September 2009, a total of 242 diabetic patients with recently diagnosed diabetic foot ulcer were involved. Ulcer area, clinical evidence of infection, ischemia and neuropathy (according to ADA diabetic foot help groups) at presentation were recorded. Each ulcer was graded using both Wagner and university of TEXAS wound classification system and divided into neuropathic, ischemic, neuroischemic, and none of neuropathic and ischemic. Patients follow-up was part of the normal treatment. Unhealed ulcers were followed up for a minimum period of six months. Once a patients ulcers had healed completely or a lower-limb amputation or death was performed, the outcome was noted and the patients was deemed to have completed the study.Results:1. The general clinical data of DF, NDF and CON groups No statistical differences were found among the three groups in age, sex, nation and smoke history. There were 80(72.1%) patients with diabetic macroangiopathy and 68(61.3%) with diabetic mircroangiopathy in NDF, and 59(75.6%) and 45(57.7%) cases respectively in DF group. These morbidity rates of macroangiopathy and mircroangiopathy are similar with diabetic groups(χ2大血管=0.5466,P=0.4597;χ2微血管=0.247,P=0.6223). Blood pressure values of NDF and DF patients both exceed CON peple, but there were not different between the DF and NDF cases.2. The biochemistry index of DF, NDF and CON groupsCompared with CON group, many biochemistry value including FPG, HbA1C, TG, CHO, AGEs, WBC, HOMA-IR, MMP-2 and MMP-9 increaed in all diabetes(P<0.05). There were no statistical difference in serum level of FPG, HbA1C, AGEs and TG between NDF and DF groups(P>0.05). The count of intravenous whole blood leukocyte were (4.35±0.47)×109/L,(6.17±0.84)×109/L and (7.62±1.13)×109/L,in CON, NDF and DF groups, respectively, and the differences were significant (P<0.05). Although serum AGEs levels in NDF and DF groups(93±10.5mg/L vs 106.5±22.9mg/L) were higher than that in CON group(35.1±7.3mg/L), there were not different between the two diabetes groups. MMP-2 and MMP-9 levels in diabetes patients of both NDF [46.1(18.2)μg/L), 136.1(107.4,275.1)μg/L] and DF groups [81.7(48.8)μg/L, 624.4(485.3,994.6)μg/L] were significantly higher than in healthy subjects [34.4(16.4)μg/L, 86.7(60.7)μg/L], as a marked difference between NDF and DF groups (P<0.05). Meanwhile, significant TIMP-1 decrease was also observed in the NDF group and DF group compared with it in the CON group [334.5(166.3)μg/L, 252.1(131.6)μg/L, vs 378.8(190.0)μg/L] (P<0.05), but there were no notable difference between NDF and DF groups (P>0.05). The MMP-2 to TIMP-1 radio was higher in the diabetes with DF than the others [0.3(0.1) vs 0.1(0.1), 0.1(0.1)], with no difference between NDF and CON groups (P>0.05).3. The spearman rank correlation analysis and regression analysis on MMP-2 with other indexSerum HbA1C, TG, AGEs, course and SBP were positively correlated with MMP-2, with the related coefficient value 0.387, 0.169, 0.221, 0.214, and 0.135 respectively. And TIMP-1 was highly inversely correlated with MMP-2(r=-0.852,P<0.05). No correlation was found between MMP-2 and WBC numbers(r=0.014, P>0.05). The linear multivariate stepwise regression analysis showed that, optimum regression equation was LnMMP-2 = 6.264-0.591×LnTIMP-1 + 0.392×Ln (course) + 0.281×HbA1C + 0.152×age.4. DF risk factors logistic regression analysisDF as dependent variable y, several indexes as independent variables(duration of diabetes, MMP-2/TIMP-1 ratio, MMP-2, HbA1C, AGEs, FPG, WBC count, blood lipid, blood pressure, and insulin resistance index), logistic regression analysis indicated that the role of MMP-2 was as an independent predictor of DF, and diabetic macroangiopathy, microangiopathy and hypertriglyceridemia were also risk factors of diabetic foot.5. The spearman rank correlation analysis and regression analysis on MMP-9 with other indexThe relevance of AGEs and MMP-9 was not strong(r=0.281, P<0.05). Optimum regression equation of the linear multivariate stepwise regression analysis was, LnMMP-9=3.475+0.612×AGEs+0.435×WBC+0.281×Ln(course)+0.041×age6. With high serum MMP-9 levels for exposure factor, we calculated the occurrence of the relative risk of DF (OR value) in diabetes with higher MMP-9 levels. As MMP-9 were skewness distribution, we used percentile method to calculate healthy cases MMP-9 upper boundary value (95 percentile) for 169.6μg/L. Thus serum MMP-9 levels higher than 169.6μg/L, was defined as high blood MMP-9. Adopting case-control study material of 4 case, OR value was 20.682, with confidence interval (8.955, 47.767) (χ2=66.534). The result suggested that the risk of diabetic foot increased 20 times to diabetes with high MMP-9 than normal.7. The active enzyme expression quantities of MMP-2 and MMP-9 in serum of three groups by gelatin zymography were not different. The activity of MMP-2 in CON, NDF and DF cases was 0.4±0.0, 7.0±3.3, 8.2±4.5 (A·mm²Â·g^-1·L^-1), respectively, and there was no significant difference between DF and NDF group. However, active MMP-9 increased in patients with diabetic foot than patients without DF(40.1±8.3 vs 17.0±6.8 A·mm²Â·g^-1·L^-1), all diabetes had higher active MMP-9 than healthy(1.5±0.3 A·mm²Â·g^-1·L^-1) ( P<0.05) .8. Of 242 patients with firstly diagnosed diabetic foot ulcers, lower-limb amputations were performed for 19.8%(48cases) of ulcers, whereas 60.7%(147cases) healed within 6 months and 14% (34cases) were not healed at study termination. 5.4% (13 cases) of patients died. During six months follow up, 2 cases with ischemic ulcer were performed autologous stem cell transplantation, and 1 performed angioplasty. The 3 cases were cured before termination, with the healing time 7.6, 10 and 4.8 weeks.9. Wagner grade, and TEXAS wound classification system grade and stage all showed positive trends with increased numbers of amputation(Wagner grade,χ²trend=79.6420; TEXAS grade,χ2trend=32.8046; TEXAS stage,χ²trend=39.2448)(P<0.001). 10. The risk of amputation for different stage of TEXAS wound classification system The rate of low limb amputation was 1.4% in stage A, 29.8% in stage B,10.8% in stage C, and 41.2% in stage D. Compared to ulcers without ischemic symptom, the risk of amputation with ischemic wound increased twice [RR=1.954,95%CI (1.044, 3.655)]. And infection in combination with ischemia, the risk of amputation increased 48 folds compared with single infection factor [RR=48.300,95%CI(6.209, 375.703)].11 Healing times were not significantly different for each grade of the Wagner or the TEXAS wound classification system (P >0.05), but there was a significant stepwise increase in healing time with each stage of the TEXAS system by log-rank test(χ²=11.234, df=3, P=0.012).12. Expressions of MMP-2 and MMP-9 mRNA in blood WBC by RT-PCRWe found zero gene expression of both MMP-2 and MMP-9 in healthy leukocyte and weak MMP-2 mRNA expression in diabetic without ulcers. Though gene levels of patients with diabetic foot ulcers were higher than those in NDF group(0.27±0.09 vs 0.53±0.05), the expression of MMP-2 was low. Unlike MMP-2, the mRNA levels of MMP-9 were significantly increased in patiens with ulcers compared to patients without ulcer(7.42±3.39 vs 0.53±0.05)(F=81.485, P<0.0001).13. Analysis results of factorial experimentBoth infection and ischemia increased the mRNA expression of MMP-9 in white cells of cases with diabetic foot. On the effect of infection, gene quantity increased 7.17 averagely, and 2.73 on the effect of ischemia. The research suggested that infection played a key role in expression of MMP-9 mRNA, and ischemia also affected it. And the later might strengthen the expression of MMP-9 in leukocyte of whole blood in patients with diabetic foot(Ftotal=96.57,P<0.0001; Finfection=23.41,P<0.0001; Fischemia=34.63,P<0.0001).14. As the ulcers became deeper, no distinct difference of MMP-9 mRNA was observed among 2, 3, and 4 grade ulcers of TEXAs university wound classification system(P>0.05).Conclusion:Analysis of our results listing above suggests:1. The serum MMP-2 level began to increase before diabetes were suffered foot chronic ulcer and increase obviously after that, accompany to lower TIMP-1 and higher MMP-2/TIMP-1. MMP-2 was highly negatively correlated with TIMP, positively correlated with course and HbA1C. MMP-2 was an indepent factor of diabetic foot.2. The serum MMP-9 level in diabetes without foot ulcer increased 1.5-fold than healthy, and 7-fold in patients with diabetic foot. It was strong postively correlated with numeration of leukocyte. The diabetes with high serum MMP-9 increased suffering foot ulcer 20 folds than ones with normal level.3. Active MMP-2 was weak in serum of patients with diabetic foot, but active enzyme of MMP-9 was higher by gelatin zymography.4. The TEXAS wound classification system was associated with increased risk of amputation and prolonged ulcer time compared with Wagner system. It was suggest that the TEXAS wound classification system be a better predictor of outcomes. It was easy for evaluating grade and stage and fit for clinic using.5. Intravenous whole blood leukocyte in patients with diabetic foot ulcer better expressed gene MMP-9 in mRNA level, but less MMP-2. Both infection and ischemia increased the mRNA expression of MMP-9 in white cells of cases with diabetic foot. The research suggested that infection played a key role in expression of MMP-9 mRNA, and ischemia also affected it. And the later might strengthen the expression of MMP-9 in leukocyte of whole blood in patients with diabetic. The expression of MMP-9 was not associated with the depth of ulcers significantly.