| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Experimental Studies |
1 From the Department of Radiology, Zhong-Da Hospital, Southeast University, 87 Dingjiaqiao Rd, Nanjing 210009, China. Received April 9, 2004; revision requested June 22; revision received July 23; accepted August 18. Supported by grants 30170279 and 30300094 of the National Natural Science Foundation of China. Address correspondence to G.J.T. (e-mail: gjteng@vip.sina.com).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: This study was approved by the ethical review committee, and written informed consent was obtained from all mothers. Endothelial cells (ECs) were collected from human umbilical veins and cultured in vitro. After 24–48 hours in culture, ECs were distributed into groups supplemented with the following concentrations of bile in the culture medium: 0%, 5.0%, 10.0%, 15.0%, 20.0%, and 25.0%. The cells were harvested 5 days after supplementation with bile. The morphologic features, von Willebrand factor (vWF) level, tetrazolium salt (MTT) assay value of light absorption, total protein level, and nitric oxide synthase (NOS) activity of the ECs were evaluated.
RESULTS: All explanted cells were identified as ECs by using the vWF test. Compared with ECs in the control group without bile, ECs in culture medium with a bile concentration of 5.0%, 10.0%, or 15.0% showed no marked morphologic changes, whereas ECs in culture medium with a bile concentration of 20.0% or 25.0% were reduced greatly in number and looked markedly immature. The MTT value of light absorption, total protein level, and vWF secretion were significantly decreased (P < .05 for all) in ECs in culture medium with 25.0% bile compared with these parameters in ECs in culture medium without bile, although these parameters did not significantly differ between the ECs in culture medium of 5.0% or 10.0% bile and the ECs in culture medium without bile. Compared with NOS activity in ECs when no bile was present in the culture medium, NOS activity in ECs was significantly decreased at all bile concentrations (P < .05).
CONCLUSION: Low concentrations of bile do not markedly inhibit cell growth; the inhibiting effect of bile on ECs progresses with an increase in bile concentration.
© RSNA, 2005
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
The histopathologic findings of TIPS restenosis in both humans and animals are similar to the histopathologic findings of the intravascular restenosis seen after stent placement, which primarily consist of pseudointimal proliferation (1–8). Pseudointimal proliferation tissue consists mainly of smooth muscle cells and extracellular collagenous matrix (1–8). Bile leakage and its effect on pseudointimal proliferation during TIPS creation have been the focus of study for many investigators (1,3,4). Bile duct injury and bile leakage—at least on a microscopic level—seem inevitable during a TIPS procedure owing to the disintegrated nature of hepatic tissue. Several investigators (1–3) have demonstrated in both humans and pigs that bile leakage during TIPS creation may promote pseudointimal proliferation; this idea is also supported by the improved TIPS patency observed when a covered stent is used (2,9–11). The mechanism by which use of the covered stent discourages restenosis seems to be the exclusion of bile leakage and, thereby, a decrease in pseudointimal proliferation. However, this mechanism is inconsistent with results of an in vitro and in vivo investigation (4), which revealed that bile seems to inhibit smooth muscle cell proliferation instead of promoting it. Thus, the purpose of our study was to evaluate the effect of bile on the growth and proliferation of human umbilical vein endothelium cultured in vitro, with a view toward clarifying the effect of bile leakage during TIPS creation.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Twenty 20-cm-long umbilical cords were obtained within 6 hours after delivery of healthy babies by healthy mothers. The umbilical vein was cannulated and flushed with 1000 mL of phosphate-buffered saline; the cord was then filled with 0.1% Type IV collagenase (Worthington Biochemical, Lakewood, NJ) and clamped at the ends. Following incubation for 10 minutes at 37°C, the cord was unclamped and the fluid was collected by gently massaging the cord. The resulting fluid suspension containing endothelial cells was centrifuged at 1000 rotations per minute for 10 minutes, and the cell pellet was resuspended in medium 1640 containing 0.2 mmol/mL L-glutamine (GIBCO BRL, Carlsbad, Calif), 100 U penicillin, 100 g/mL streptomycin (Nanjing Jinling Pharmaceutical, Nanjing, China), and 20% fetal bovine serum (Worthington Biochemical).
On the basis of different concentrations of bile administered to the culture medium, five experimental groups of endothelial cells in culture media with bile concentrations of 5.0%, 10.0%, 15.0%, 20.0%, and 25.0% and a control group of endothelial cells in culture medium without bile were created and subjected to morphologic observation, total protein assays, von Willebrand factor (vWF) measurement, and nitric oxide synthase (NOS) activity estimation; endothelial cells in an additional five bile concentrations of 2.5%, 7.5%, 12.5%, 17.5%, 20.0%, and 22.5% were used for tetrazolium salt (MTT) assays. Endothelial cell cultures were established by plating cells at a density of 1 x 105 in each well of a six-well plate (Denmark NUNC, Roskilde, Denmark). Twenty cultures were created for MTT assays, 10 were created for NOS studies, and 15 each were created for total protein assays and vWF measurement. Twenty-four to 48 hours after the endothelial cells were plated, the culture medium was replaced with an appropriate experimental growth medium containing different concentrations of bile as the experimental protocol. The medium was changed every 48 hours until the cells were harvested.
Preparation of bile.—Bile was obtained from a T tube after a cholecystectomy had been performed in a 77-year-old man with chronic cholecystitis secondary to chololithiasis. The collection and use of bile were approved by the ethical review committee of our institution, and written informed consent was obtained from the patient. The collection of bile was performed in aseptic conditions 4 days after cholecystectomy while the patient had a normal body temperature. The bile was frozen immediately after collection and was kept frozen until use.
Medium preparation.—On the basis of results (not shown) from a pilot study, we had determined that bile at a concentration of 25.0% or higher is lethal to endothelial cells. For the experimental groups, endothelial cells were cultured in medium 1640 that was supplemented with 0.2 mmol/mL L-glutamine, 100 U penicillin, 100 g/mL streptomycin, and 20% fetal bovine serum and contained bile at different concentrations. The same culture conditions and culture medium—but no bile—were used for the control group cells.
Cell Assays
Morphologic observation.—The cultured cells were evaluated every day by using an inverted microscope (Olympus CK2; Olympus Opto-Technology, Nagano, Japan). The evaluation was performed by a pathologist with 30 years of experience who was blinded as to which specimens contained bile and which did not. Morphologic evaluation of endothelial cells included assessment of the following: their size and shape, the visibility of the nucleolus and organelles, and the clarity of cellular rims.
MTT assays.—After endothelial cells were seeded in a 96-well plate and after 24 hours in culture, various concentrations of bile (2.5%, 5.0%, 7.5%, 10.0%, 12.5%, 15.0%, 17.5%, 20.0%, 22.5%, and 25.0%) were added to the culture medium. Twenty cultures were created for each bile concentration level and for the control group without bile. After continuous culture for 5 days, 20 µL of MTT liquor (5 mg of MTT per milliliter of phosphate-buffered saline, Fluka, Buchs, Switzerland) was added to every well and the cells were incubated for another 4 hours. After the upper layer of the culture medium was removed, 20 µL of dimethylsulfoxide (Shanghai Bioengineering, Shanghai, China) was added to the medium and oscillated for 10 minutes. The 490-nm wavelength of light absorption was used for measuring the light absorption value of every well with a spectrophotometer (Spectra MAX 250; Molecular Devices, Sunnyvale, Calif).
Total protein assays.—After endothelial cells were seeded in a six-well plate and after 24 hours in culture, various concentrations of bile (5.0%, 10.0%, 15.0%, 20.0%, and 25.0%) were added to the culture medium. Fifteen cultures were created for each bile concentration level and for the control group without bile. After continuous culture for 5 days, the endothelial cells were harvested for protein assays. Total protein in the cultured endothelial cells was measured by using a bicinchoninic acid spectrophotometer assay (BCA Protein Assay Reagent; Pierce, Rockford, Ill). Sample results were compared with a standard curve generated from purified bovine serum albumin (Pierce).
Measurement of vWF.—After endothelial cells were seeded in a six-well plate and after 24 hours in culture, various concentrations of bile (5.0%, 10.0%, 15.0%, 20.0%, and 25.0%) were added to the culture medium. Fifteen cultures were created for each bile concentration level and for the control group without bile. The cells were harvested after 5 days in culture. vWF was measured so that we could verify the presence and function of the endothelial cells according to their secretion of vWF. A vWF kit (Thrombosis Institute, Suzhou University, Suzhou, China) and a standard technique (12) were used.
Estimation of NOS activity.—After endothelial cells were seeded in a six-well plate and after 24 hours in culture, various concentrations of bile (5.0%, 10.0%, 15.0%, 20.0%, and 25.0%) were added to the culture medium. Ten cultures were created for each bile concentration level and for the control group without bile. The cells were harvested after another 5 days in culture, and NOS activity was then estimated by using a standard technique (13).
Statistical Analysis
Kruskal-Wallis tests were used for analysis of all the data, including the MTT value of light absorption, total protein content, vWF level, and NOS activity, and Bonferroni-Dunn t tests were employed for comparison between groups. A P value of .05 or less was considered to indicate a significant difference. All statistical analyses were performed by using SAS version 8.02 (SAS Institute, Cary, NC).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
|
|
|
2 = 29.913, P < .01 [Kruskal-Wallis test]) among the bile concentration groups and decreased as bile concentration increased. There was a significant difference between the value at 25.0% bile and the value at 0% bile (P < .05 [Bonferroni-Dunn t test]) (Table 1).
|
2 = 18.857, P = .002 [Kruskal-Wallis test]) and significantly lower in the group with 25.0% bile than in the group without bile (P < .05 [Bonferroni-Dunn t test]) (Table 2).
|
2 = 27.213, P < .001 [Kruskal-Wallis test]). The vWF levels in medium containing 20.0% or 25.0% bile were significantly decreased compared with the vWF level in medium without bile (P < .05 [Bonferroni-Dunn t test]) (Table 2).
NOS activity.—Endothelial cell NOS activity values varied significantly among the bile concentration groups (2 = 11.667, P < .009 [Kruskal-Wallis test]) and were significantly lower for endothelial cells in medium containing 5.0%, 10.0%, 15.0%, or 20.0% bile than for endothelial cells in medium without bile (P < .05 [Bonferroni-Dunn t test]) (Table 2).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
-month-old TIPS. These findings are similar to those of Saxon et al (3), who found that large biliary fistulas formed in seven of nine TIPS in a porcine model and in seven of eight human TIPS specimens in which parenchymal tract stenosis and occlusion were documented. Other experiments revealed the occurrence of bile leakage in TIPS porcine models, and there was a much higher incidence of stenosis or occlusion of the shunt in cases with bile leakage than in cases without bile leakage (3,4). Therefore, it seems a reasonable speculation that biliary tract injury and bile leakage during TIPS creation may be one of the factors involved in TIPS restenosis. In our previous investigation involving a porcine model (4), we were unable to demonstrate significant differences in histologic quantity of pseudointimal proliferation between the group with bile leakage and the group without bile leakage. However, there were fewer cellular components in the group with bile leakage than in the group without bile leakage. Few endothelial cells covered the TIPS stents with bile leakage, while a good layer of endothelial cells covered the TIPS stents without bile leakage. Additionally, the results of our previous in vitro investigation (4) indicated that bile inhibits smooth muscle cell proliferation. Therefore, we suggest that the mechanism by which bile leakage causes TIPS restenosis might be associated with the promotion of thrombosis instead of smooth muscle cell proliferation. It is known that a defect in endothelial cell linings is associated with coagulation and thrombosis (16), and thus it seems logical that bile leakage may promote the thrombosis of TIPS stents by inhibiting endothelialization of the stent.
The results of this in vitro experiment indicate that bile inhibits endothelial cell growth in a concentration-dependent manner: Low concentrations of bile inhibit endothelial cell growth slightly, while the effect of bile on endothelial cell growth increases with an increase in bile concentration, until a level of bile (
40%) that is lethal to the endothelial cells is reached. This morphologic result is in agreement with our finding that bile inhibits the NOS activity and vWF secretion of endothelial cells.
Injury of biliary tracts and bile leakage seem inevitable during TIPS creation. However, bile is supposed to be diluted immediately owing to high blood flow in the TIPS, and the bile concentration should be low. Bile leakage is verified mainly by the recognition of bile staining in areas of pseudointimal proliferation in the TIPS (1–5). In a case reported by LaBerge et al (14), a biliary tract was transected, but bile staining in the areas of proliferation in the TIPS and endothelial cell layer involvement were not mentioned. Therefore, it is possible that bile leakage may be invisible if the bile leak is too small to be seen.
Ducoin et al (17) found that more inflammatory cells and less collagen tissue were demonstrated in thicker pseudointimal proliferation tissue than in thinner pseudointimal tissue and that more inflammatory macrophages were found in pseudointimal tissue that showed bile staining. Therefore, it has been suggested that bile may play an important role in an inflammatory reaction that leads to pseudointimal proliferation in TIPS. Ducoin et al (17) agreed with the theory that bile promotes thrombosis owing to a persistent inflammatory reaction. Similar opinions have been voiced by LaBerge and colleagues (14) and Sanyal et al (18). Our results suggest that more extensive bile leakage with visible bile staining may cause a severe inflammatory reaction that leads to marked thrombosis and that less extensive bile leakage without visible bile staining may cause slight inflammation that leads to late restenosis caused by pseudointimal proliferation.
vWF is a cytokine that plays an important role in platelet adhesion and accumulation. It can be used for identifying endothelial cells and estimating endothelial cell function. The results of this in vitro study demonstrated that a certain concentration of bile might inhibit endothelial cell growth and vWF secretion. These results agree with our previous experimental results and speculations in suggesting that bile may inhibit stent endothelialization and then promote thrombosis and restenosis of TIPS.
NOS, as an important message factor, has been widely mentioned in reports of restenosis after percutaneous transluminal angioplasty and stent placement (19). It is well known that NOS plays important roles in vessel healing after injury that include vessel dilation, regulation of smooth muscle cell proliferation, and inhibition of platelet adhesion and thrombosis. TIPS restenosis is pathologically similar to vascular restenosis, but the pseudointimal proliferation in TIPS restenosis seems much faster and more severe than that in vascular restenosis (3,4,6). Bile leakage is one of the major factors that is present in TIPS creation but not vascular stent placement. The results of NOS activity assessment in this in vitro experiment indicate that bile, by inhibiting NOS synthesis through a suppression of endothelial NOS activity and thus contributing to thrombosis of the shunt, may be involved in TIPS restenosis.
There were several limitations to this study: First, because the origination of endothelial cells in TIPS is uncertain, the endothelial cells cultured in this study, which were obtained by using umbilical veins instead of a TIPS restenosis model, may not exactly resemble the endothelial cells in restenotic TIPS. A further study involving the use of restenotic TIPS in a swine model is being pursued in our institution. Second, only a single individual’s bile was used in this study, and results may have varied if different bile compositions had been used. A further study involving multiple samples of bile from different patients is important for clarifying our results.
Another limitation was that it is almost impossible to know the true bile concentration in a TIPS environment. The concentrations of bile studied in this experiment seem much higher than those in a TIPS environment because bile is easily diluted by the blood flow in a TIPS. However, it is realistic to think that high concentrations of bile can occur momentarily in the TIPS environment, especially when the TIPS is dysfunctional. In addition, the results of colorimetric MTT assays may have been interfered with by the color of the bile, especially at higher concentrations.
In conclusion, results of this in vitro study show that low concentrations of bile do not markedly inhibit cell growth; the inhibiting effect of bile on endothelial cells progresses with an increase in bile concentration.
Practical application: The restenosis of TIPS seems to correlate with bile leakage during TIPS creation in both pigs and humans (1–5,14,15). However, in our previous in vitro and in vivo investigations (4,6), we observed that the proliferation of smooth muscle cells, the dominant cellular component of the restenotic tissue in TIPS, was not promoted by bile. The combination of data from the present study and our previous results suggests that bile leakage is an important factor in TIPS restenosis and that bile leakage contributes to restenosis apparently by inhibiting the growth and activity of endothelial cells rather than by promoting the proliferation of smooth muscle cells. Therefore, it is practical to reduce a TIPS restenosis by isolating a bile leak into the shunt by using a stent-graft.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Authors stated no financial relationship to disclose.
Author contributions: Guarantors of integrity of entire study, G.J.T., Q.L.; study concepts and design, G.J.T., Q.L.; literature research, G.J.T., Q.L.; experimental studies, G.J.T., Q.L.; data acquisition, Q.L.; data analysis/interpretation, G.J.T., Q.L.; statistical analysis, G.J.T., Q.L.; manuscript preparation and definition of intellectual content, G.J.T., Q.L.; manuscript editing, revision/review, and final version approval, G.J.T.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
