Forty patients with coronary artery disease were included in this study. Each patient had at leasta 90% narrowing in more than one major coronary artery (except acute myocardial infarction). All patients gave their informed consent. The patients were recruited from those undergoing diagnostic coronary artery catheterization. The indications for catheterization for all patients were the presence of stable or unstable angina pectoris, or a suspected significant myocardial ischemia. Only patients with at least ≥1 coronary stenosis of ≥ 90% (by visual analysis of the angiogram) and had no previous medication history were included in this study. The exclusion criteria were the presence of anemia, myocardial infarction, chronic inflammatory illness and neoplastic disease.

For each patient, the following data was collected and recorded via interview: name, registration number, gender, age, previous revascularizations, a familial history of cardiovascular diseases, and a medical history of diabetes mellitus, hypertension, hypercholesterolemia, and smoking history. Two physicians, experienced with coronary angiography, recorded the cardiovascular anatomy and the presence of collateral circulation after consulting with each other. They used a modified TIMI system to measure the degree of collateral circulation formation. Grade 0 referred to the absence of collateral circulation, grade 1 referred to an invisible blood stream in a recipient epicardial vessel, and grade 2 referred to a partial or complete filling of the recipient epicardial vessels by collaterals (Figure 1).

The mononuclear cell layer was isolated from peripheral blood using a mixture of polysaccharide and radiopaque contrast medium. During the coronary angiography, 40 mL of blood was collected from the femoral venous catheter, placed for catheterization before the angiography procedure began. The blood was immediately placed in a 50-mL polypropylene heparinized tube and kept on ice until it was used for monocyte isolation. After layering 20 mL of heparinized blood into a 15 mL tube containing Ficoll-Paque, it was centrifuged at 1,800 rpm at 18~20℃ for 30 minutes. From the upper layer in each tube, 8 mL of serum was collected and stored at 4℃ for later use. The middle phase, including the monocytes, was isolated and transferred to a 15 mL polypropylene centrifuge tube. The isolated monocyte layer was rinsed twice with a sterilephosphatebufferedsaline, then transferred to a centrifuge tube and centrifuged at 800 rpm at 18~20℃ for 10 minutes. After that, the supernatant was removed. The cell pellets were resuspended with 2 mL of Dulbecco's modifi edeagle's medium and a mixture of 2% fetal bovine serum and antibiotics. These included GIBCOBRL, high glucose powder with L-glutamine, 25 mM HEPES buffer with sodium chloride 4,750 mg/L replaces 6,400 mg/L with pyridoxine hydrochloride without sodium pyruvate without sodium bicarbonate. After sorting the cells by size and positioning them into tissue culture dishes (Corning), they were cultured in an incubator (Forma) at 95% room air and 5% CO₂ at 37℃ for 1 hour, in order to induce monocyte adhesion. After confirming the monocyte adhesion, the culture medium was removed. Cells were replaced by a culture medium containing an 8 mL serum collected from the patients and cultured in a Triple Gas Incubator (Jouan. One dish contained 21% O₂ and 5% CO₂ and the other dish contained 1% O₂, 5% CO₂ and 94% N₂. After culturing the cells under normal and hypoxic conditions for 17 hours, the RNA was extracted.

The cells, obtained under normal and hypoxic conditions, were treated with 1 mL RNAzol™B (Biotec Laboratories Inc, Houston, Texas) in each sample and they were exposed to ice for 5 minutes. After that, the cells were centrifuged at 14,000 rpm for 15 minutes. The supernatant was transferred to a new tube and treated with isopropanol of an equivalent concentration in an ice pan for 10 minutes. This was followed by centrifugation at 12,000 rpm for 15 minutes. After centrifugation, the supernatant was removed, and the pellet was rinsed with 70% ethanol and then dried in air. DEPC distilled water was applied to the pellet and it was heated to dissolve at 60℃ for 30 minutes. The RNA was quantified using a spectrophotometer at a wavelength A²⁶⁰.

In 20 µL reaction bowel, the following substances were mixed: 1 µg RNA, 5 mM MgCl₂, 1 × RNA PCR buffer, RNase-free distilled water, 1 mM dNTP Mixture, 1 units/mL RNase inhibitor, 0.25 units/L Reverse Transcriptase, and 2.5 µL Random 9mer (TaKaRa, Tokyo, Japan). After that, the above substances were allowed to react at 30℃ for 10 minutes, at 42℃ for 30 minutes, at 99℃ for 5 minutes and at 5℃ for 5 minutes to prepare the cDNA. The product was then stored at 4℃ before it was used. In a mixture of 4 µL cDNA, 1 × PCR buffer (50 mM KCL, 10 mL Tris-Cl [pH = 8.0]), 200 µM dNTPs and 1 units of Taq DNA polymerase (TaKaRa, Tokyo, Japan), 20 pmol of primer was added to prepare 20 µL in total, followed by PCR using a DNA thermal cycler. The PCR procedure consisted of the following three cycles: denaturation at 94℃ for 5 minutes, annealing at 60 - 66℃ for 1 minute and extension at 72℃ for 10 minutes.

The PCR product was treated onto 2% agarose gel. Electrophoresis was performed and the band was stained with ethidium bromide for observation under ultraviolet light. The PCR band was measured by densitometer. The VEGF and HIF-1 (alpha) mRNA levels were measured using a RT-PCR procedure consisting of 30 cycle and an annealing temperature of 60℃/1min for VEGF, 65℃/1min for HIF-1 (alpha). Each primer sequences were recorded in Table 1. The amount of calculatedfold inductions of VEGF and HIF-1 (alpha) mRNA for the hypoxic condition was done by dividing the hypoxic product value by the normoxic product value. The procedure was repeated three times and then averaged.

All data were described as means±SD (SD: standard deviation). The data were tested with Student's t-test and Chi-square tests. The three collateral circulation groups (VEGF and HIF-1 (alpha) mRNA) were compared using analysis of variance (ANOVA) and Tukey's test for multiple comparisons. A p value<0.05 was considered as significant.

The statistical analysis was performed using SPSS-PC10.0 (Statistical package for the Social Sciences, SPSS Inc. Chicago, IL, USA) for MS Windows.

625 patients who had undergone coronary angiography at Hanyang University Hospital were recruited for our study. The exclusion criteria of this study were as follows: (i) stenosis of less than 90% on coronary angiography (541 patients); (ii) acute myocardial infarction (23 patients); and (iii) anemia, chronic inflammation and unknown data (21 patients). Finally, a total of 40 patients were enrolled in this study. The patients were assigned to the collateral circulation and non-collateral circulation groups. The collateral circulation group was 60% male and the non-collateral circulation group was 40%. No significant difference was noticed between the two groups (p=0.135). For risk factors like coronary artery diseases, including age, hypertension, smoking history, total cholesterol and low-density lipoprotein (LDL), there were no significant differences between the two groups. Diabetes mellitus was more prevalent in the collateral circulation group than in the non-collateral circulation group (p=0.038) (Table 2). There were no significant differences between the two groups for the cardiovascular drugs or heparin used prior to the coronary angiography. During an ecchocardiography, the ejection fraction showed no significant difference between the two groups (p=0.103).

Coronary artery collaterals were measured by the physician performing the catheterization, using visual analysis and using the conventional criteria. Regarding the coronary angiography, 40% of the 84 patients with a stenosis of more than 90% (34/84), had non-collateral circulation. Nineteen per cent of them (16/84) were Grade I and 40% of them (34/84) were Grade II. On the other hand, of the final 40 patients, 50% (20/40) were non-collateral circulation, 12% (5/40) were Grade I and 38% (15/40) were Grade II.

The expression of VEGF mRNA and HIF-1 (alpha) mRNA, depending on the collateral circulation development, are presented in Figure 2. For the subgroup analysis, the patient groups were classified into three subgroups on the basis of the modified TIMI system. There were no significant differences in the mean normoxic VEGF/18S and HIF-1 (alpha) mRNA value among the 3 collateral groups. The fold inductions of HIF-1 (alpha) mRNA under the hypoxic condition for 0+ collaterals, 1+ collaterals and 2+ collaterals were 1.20±0.39, 1.28±0.29 and 1.47±0.63, respectively. This indicates there were no significant differences among the three groups (Figure 3). In contrast, the fold inductions of VEGF mRNA under hypoxic condition for 0+ collaterals, 1+ collaterals and 2+ collaterals were 1.65±0.62, 3.77±3.44 and 3.68±3.43, respectively. This indicates that the collateral circulation group had a significantly higher fold inductions than the non-collateral circulation group (Figure 4).

On the other hand, to minimize errors in evaluating the collateral circulation and to more accurately measure the HIF-1 (alpha) mRNA and VEGF mRNA levels (due to the collateral circulation formation), patients were allocated to the collateral circulation group and the non-collateral circulation group. Under the hypoxic condition, the fold inductions of HIF-1 (alpha) mRNA for the non-collaterals and collaterals were 1.20±0.39 and 1.42±0.57, respectively, indicating there was no significant difference between the two groups (p=0.165) (Figure 5). Conversely, under the hypoxic conditions, the fold inductions of VEGF mRNA for the non-collateral circulation group and for the collateral circulation group were 1.65±0.62 and 3.70±3.34, respectively, showing a significant difference between the two groups (p=0.014)(Figure 6, 7).