流体中速度场检测方案(粒子图像测速)

INADVERTENT VARIATIONS IN FLUID FLOW ACROSS APARALLEL PLATE FLOWCHAMBER RESULTS IN NON-UNIFORM GENE EXPRESSION Jennifer A. McCann(1)， Sean D. Peterson (2)， Thomas J. Webster (1)， Michael W. Plesniak (2)，and Karen M. Haberstroh (1) (1) Department of Biomedical EngineeringPurdue University (2) School of Mechanical EngineeringPurdue University Gene expression studies are important for understanding generegulation， deducing cell signal cascades， etc. More broadly， thesestudies are often used to understand disease development， as specificgenes are commonly up regulated in clinical pathologies. Thesestudies can be paired with the application of a mechanical stimulus aswell； for example， endothelial cells can be exposed to a defined fluidflow (physiological or pathological) in order to understand the possiblemechanisms for atherosclerosis development. Cyclooxygenase (COX)-2 and endothelial cell nitric oxidesynthase (ecNOS) are genes with known responses to physiologicalfluid shear stress， and1 are important in the development ofatherosclerosis. Specifically， COX-2 is involved in the production ofprostacyclin (PGI)， an inhibitor of platelet aggregation and smoothmuscle cell growth and migration [1]. ecNOS catalyzes the reaction ofarginine to citrulline along with nitric oxide (NO)， which acts as avasodilator and mimics PGI2 in other ways， as a product [2]. Such flow-mediated cell responses are commonly studied withthe parallel plate chamber， first developed by Frangos et al. [3]. Inthese studies， flow is assumed to be laminar and uniform over thelength of the plate. Therefore， all cells in the chamber experience thesame conditions regardless of their plate location. Following theseflow experiments， cells are collected from the whole chamber； thusgene studies represent an average cell response over the entire plate.More recently， many research groups have modified the parallel plateflowc chamberr.. Wewereinterested in determiningwhetherexperimental flow irregularities caused by slight machining variationsfrom one modified flow chamber to the next would affect sensitive cellresponses (such as gene expression) over the chamber area. Therefore，for the first time， this study combined u-PIV flow measurements andgene expression studies to elucidate flow irregularities followingexposure of endothelial cells to fluid flow in a modified parallel plateflow chamber. MATERIALS AND METHODS u-PlV Studies Velocity fields within the flow chamber were obtained usingmicro-Particle Image Velocimetry (u-PIV)..The u-PIV systemconsists of a pulsed Nd：YAG laser， a fast interline transfer CCDcamera， and an epifluorescent microscope which captures twoconsecutive images of a fluorescent particle seeded flow illuminatedby the laser.A 0.00038% concentration of 1 um diameter seedparticles and a 20X microscope lens were used to image a 300x300 umarea within the flow cell.Water (20°℃) was pumped through thechamber by a gear pump at a flow rate of 37.3 ml/min， whichcorresponds to a shear stress of 10 dynes/cm. Five image pairs weretaken at each measurement volume within the chamber and temporallyaveraged to obtain the velocity. DaVis 6.0 software (developed byLaVision) was used to capture and process the image pairs and toaverage the vector fields. Additionally， the channel height wasmeasured at each PIV measurement location. Cell Culture Rat aortic endothelial cells (RAEC) (VEC Technologies) weremaintained in MCDB-131 Complete Medium (VEC Technologies).Cells were grown under standard cell culture conditions， that is， ahumidified， 37℃， 5% CO2， 95% air environment. All cells were usedatpopulationnnnumbersfive tthroughttwelvewithoutfurthercharacterization. Exposure of RAEC to Laminar Flow Prior to fluid flow exposure， endothelial cells were seeded(35，242 cells/cm’) onto etched glass cover slides previously coatedwith fibronectin (10 ug/mL； Sigma). Once confluent (approximately1.5 days)， endothelial cells were exposed to laminar fluid flow in amodified version of the parallel plate flow chamber [3] for 6 hours.The shear stress generated in this system was 10.4 dyne/cm. mRNA Expression by RAEC Exposed to Fluid Flow Following exposure to fluid flow， sterile， silastic strips were usedto divide the glass slides (experimental and control) into six equalregions (top left， top center， top right， bottom left， bottom center， andbottom right). mRNA from endothelial cells in each regi wasdigested with Triazol (Gibco)， extracted with chloroform， andprecipitated with isopropanol. Reverse transcriptase-polymerase chainreaction was performed using sample RNA and the Ambio RetroscriptKit with primers (20 pmol/pL； Gibco) specific to glyceraldehyde 3-phosphate dehydrogenase (GAPDH)， cyclooxygenase-2 (COX-2)， orendothelial cell nitric oxide synthase (ecNOS).Electrophoresis ofcDNA samples was performed in a gel casting system (Biorad) with1X TAE buffer for 1.5 hours at 200V. The amount ofcDNA in eachlane wasdeterminedusingaphosphoimager ((Biorad) anddensitometry. RESULTS AND CONCLUSIONS u-PIV studies revealed that flow characteristics within a modifiedparallel plate chamber were location and geometry dependent.Specifically， the region nearest the inlet experienced the overallhighest average velocity and the two regions in the center of the plateexperienced the overall lowest average velocities. Furthermore， thechannel height was found to be the deepest in the regions with thelowest velocities. mRNA expression corresponded with the velocityprofiles， providing evidence that cellular responses are not tolerant tothese small geometry changes. Based on results of this study， it is clear that different regions offluid flow do exist across this modified version of the parallel plateflow chamber； similar variations in other modified flow chamberscould affect cell studies where average cell responses are not reported，such as representative cell staining/imaging. We would thereforerecommend that custom-built， modified， parallel plate flow chambersare analyzed for flow characteristics across the plate； such data wouldprovide a more exact correlation between shear stress and cellresponses. .ACKNOWLEDGEMENTS The authors would like to thank the Showalter Foundation(Award #1220012825) and the Purdue Special Initiatives Fellowshipfor funding this project， Dr. Fred Pavalko (Department of Cellular andIntegrative Physiology； Indiana UniversityPurdueUniversityIndianapolis) for supplying the flow apparatus， and Karen Hile(Department of Pediatric Urology； Indiana University11PurdueUniversity Indianapolis) for technical support. REFERENCES 1. Topper， J.N.， J. Cai， D. Falb， and M.A. Gimbrone， 1996.“Identification of vascular endothelial genes differentiallyresponsive too fluid mechanical stimuli： Cyclooxygenase-2，manganese superoxide dismutase， and endothelial cell nitricoxide synthase are selectively up-regulated by steady laminarflow，” Proceedings of the National Academy of Sciences， 93， pp.10417-10422. 2. Ignarro， L.J.， G.M. Buga， L.H. Wei， P.M. Bauer， G. Wu， and P.del Soldato， 2001，“Role of arginine-nitric oxide pathway in theregulation of vascular smooth muscle cell proliferation，”Proceedings of the National Academy of Sciences， 98， pp. 4202-4208. 3. Frangos， J.A.， L.V. McIntire， and S.G. Eskin， 1988，“ShearStressInduced Stimulation of MammalianCellMetabolism，”Biotechnology and Bioengineering， 32， pp. 1053-1060. Summer Bioengineering Conference， June Sonesta Beach Resort in Key Biscayne， FloridaStarting page #：