Track 13. Respiratory Mechanics

5219 We-Th, no. 10 (P64) Flow in a 3D alveolated bend: validation of CFD predictions with experimental results C. van Ertbruggen 1 , N. Buchmann 2, R. Theunissen 2, E Corieri 2, M.L. Riethmuller 2, C. Darquenne 1 . 1Dept. Medicine, University of California at San Diego, La Jolla, USA, 2von Karman Institute for Fluid Dynamics, Rhode-St-Genese, Belgium

Verifying numerical predictions with experimental data is an important aspect of modeling studies. In the case of the lung, the absence of direct in-vivo flow measurements makes such verification almost impossible. We performed computational fluid dynamics (CFD) simulations in a 3D scaled-up model of an alveolated bend representative of the alveolar region of the human lung and compared numerical predictions with experimental flow measurements made in the same model by Particle Image Velocimetry [1]. The bend had a 1450 angle, a lumen and outer diameter of 15 and 30mm, respectively. Using a polyhedral unstructured mesh (STAR-Design, CD-Adapco), simulation was performed using STAR-CCM+ (CD-Adapco) for a flow of 0.84 ml/s, which was representative of acinar flow during normal breathing (500 ml/s at the mouth). The maximum simulated velocity found along the centerline of the model was 9.1 mm/s, 4% smaller than measured experimentally [1]. Regions of high velocity upstream and downstream of the bend could be identified. The velocity inside the alveoli was -2 orders of magnitude smaller than the mean lumen velocity. Velocity profiles in the lumen were parabolic and no bulk convection existed between the lumen and the alveoli. These observations were similar to those obtained experimentally [1]. This suggests that CFD techniques we used to model acinar flow can satisfactorily predict in-vivo behavior since the model incorporated enough essential characteristics of alveolar structures. This validation will ensure a great degree of confidence in the accuracy of predictions made in more complex models of the alveolar region of the lung using similar CFD techniques. This study was supported by grant ES011177 from the NIEHS at NIH. C. van Ertbruggen is a Belgian-American Educational Foundation - Henri Benedictus Fellow of the King Baudouin Foundation, Belgium.

References [1] N. Buchmann, Experimental modeling of aerosol particles within lung bifurca-

tions, Diploma thesis. 2005; von Karman Institute for Fluid Dynamics, Belgium.

6373 We-Th, no. 11 (P64) Alveolar macrophages probe and respond to the stiffness of their substrate S. F6r6ol, R. Fodil, S. Galiacy, B. Labat, V.M. Laurent, B. Louis, E. Planus, D. Isabey. Inserm U651, Equipe Biom#canique Cellulaire et Respiratoire, Facult# de M#decine, Cr#teil, France

In order to understand how Alveolar Macrophages probe and respond to the stiffness of their substrate, we have developed a new cellular model of macrophages cultured on substrates of increasing Young's modulus: (i) a monolayer of alveolar epithelial cells representing the supple (-0.1 kPa) physiological substrate, (ii) polyacrylamide gels with two concentrations of bis-acrylamide representing low and high intermediate stiffness (respectively 40kPa and 160kPa) and, (iii) a highly rigid surface of plastic or glass (re- spectively 3 MPa and 70 MPa), the two latter being or not functionalized with type-I collagen. The macrophage response was studied through their shape characterized by 3D-reconstructions of F-actin structure and their cytoskeletal stiffness estimated by transient twisting of magnetic RGD-coated beads and thoroughly corrected for actual bead immersion. Macrophage shape dramati- cally changed from rounded to flattened as substrate stiffness increased from soft [(i) and (ii)] to rigid (iii) substrates, revealing a sensitivity of alveolar macrophages to substrate stiffness but without generation of F-actin stress fibers. Macrophage stiffness was also increased by large change in substrate stiffness increase but this increase was not due to an increase in F-actin internal tension. The mechanical sensitivity of AMs could be partly explained by a numerical model describing how low cell height enhances the substrate- stiffness-dependence of the apparent (measured) AM stiffness. The increase in AM stiffness depends also on whether substrate adhesion is specific or not. Altogether, these results suggest that macrophages are able to probe and respond to their physical environment but the mechanosensitive mechanism behind appears quite different from tissue cells.

$595

6414 We-Th, no. 12 (P64) Hyperoxia combined with high tidal volume mechanical ventilation reduces alveolar type II cell adhesion: reduced phosphorylation of focal adhesion proteins and effect of KGF

C.M. Waters 1 , L.P. Desai 1 , S.E. Sinclair 2, A. Hassid 1 , K.E. Chapman 1 . 1 Department of Physiology, University of Tennessee Health Science Center, Memphis, TN, USA, 2Department of Medicine, University of Tennessee Health Science Center, Memphis, TN, USA

The combination of high stretch and hyperoxia during mechanical ventilation may induce increased lung injury in patients. We hypothesized that the com- bination of high stretch and hyperoxia impairs epithelial repair mechanisms necessary for restoration of barrier function. We ventilated rats for 2 hr with high tidal volume ventilation (25ml/kg) and hyperoxia (50% O2) and investigated the signaling mechanisms important in regulating alveolar type II (AT2) cell attachment and repair. AT2 cells isolated from rats exposed to hyperoxia and high tidal volume mechanical ventilation (MVHO) exhibited significant down- regulation of phosphorylated focal adhesion proteins focal adhesion kinase (FAK), p130Cas, and paxillin, as compared to control rats (NO), rats exposed to hyperoxia without ventilation (HO), or rats ventilated with normoxia (MVNO). The combination of hyperoxia and high tidal volume ventilation also increased activation of RhoA leading to decreased attachment of AT2 cells. Treatment of MVHO cells with keratinocyte growth factor (KGF) for lhr upon isolation reduced RhoA activity and restored attachment to control levels. The effect of KGF on MVHO cell adhesion and migration was similar to the effect of over- expression of a dominant negative form of RhoA. Moreover, overexpression of a FAK mutant lacking the non catalytic carboxyl-terminal protein binding domain (FRNK), significantly delayed attachment and inhibited migration of control cells. Expression of a substrate domain-deleted p130Cas decreased adhesion and migration, whereas overexpression of wild-type p130Cas blocked the inhibitory effect on adhesion and migration. We conclude that decreased phosphorylation of focal adhesion proteins and increased RhoA activity results in decreased adhesion and cell migration in AT2 cells limiting repair mecha- nisms. KGF accelerates repair mechanisms of AT2 cells under conditions of hyperoxia and high tidal volume ventilation by preventing the upregulation of RhoA activity.

6599 We-Th, no. 13 (P64) Experimental modelling of airflow in airway bifurcations K.B. Heraty, N.J. Quinlan. NCBES, National University of Ireland, Galway, Ireland

Pulmonary drug delivery utilises the airflow mechanism within the airways to transport the drug from the mouth to the required pulmonary region. For successful drug delivery, understanding of the flow of air and particles in the lung is essential. Most of the experimental work done in this area to date has been applied to idealised bifurcation geometries based on the morphological work of Weibel (1963). The computational study of Nowak et al. (2003) raised questions over the validity of using idealised geometries to define the airway. The aim of this work is to compare the airflow structures in an anatomically realistic pulmonary bifurcation to an idealised Weibel based model. The non- intrusive optical flow measurement technique Stereoscopic Particle Image Velocimetry (SPIV) is used to characterize the flow field. To improve the spatial and temporal resolution of the experiment, the model geometries are scaled up by a factor of 6. Dynamic similarity between the flow in the model and the true physiological flow is preserved by ensuring that the dimensionless Reynolds and Womersley numbers are maintained in both cases. Optically clear experimental bifurcation models were made by a lost core process, described by Hopkins et al. (2000). The models are tested under steady state and physiologically realistic breathing patterns. The flow is generated by a custom built piston in-line pump. Results indicate that the flow structures observed in the realistic geometry are not captured by the idealised geometry. These differences in flow structure are due to the non-cylindrical nature of the walls in the parent and daughter branches of the realistic model.

References Hopkins L.M., et al. (2000). Particle image velocimetry measurements in complex

geometries. Experiments in Fluids 29: 91-95. Nowak N.K., et al. (2003). Computational fluid dynamics simulation of airflow and

aerosol deposition in human lungs. Ann. Biomed. Eng. 2003; 31: 374-390. Weibel E.R.(1963). Morphometry of the Human Lung. New York: Academic,

pp. 136-143.