bs_bs_banner

Copyright © 2015 International Center for Artificial Organs and Transplantation and Wiley Periodicals, Inc.

Thoughts and Progress

Atraumatic Pulsatile Leukocyte Circulation for Long-Term In Vitro Dynamic Culture and Adhesion Assays *†Giulia Mazza, ‡Martin Stoiber, *§1Dagmar Pfeiffer, and ‡1Heinrich Schima *Center for Biomedical Technology and †Center for Integrated Sensor Systems, Danube University Krems, Krems; ‡Center for Medical Physics and Biomedical Engineering, Ludwig Boltzmann-Cluster for Cardiovascular Research, Medical University of Vienna, Vienna; and §Institute of Cell Biology, Histology and Embryology, Medical University Graz, Graz, Austria Abstract: Low flow rate pumping of cell suspensions finds current applications in bioreactors for short-term dynamic cell culture and adhesion assays. The aim of this study was to develop an atraumatic pump and hemodynamically adapted test circuit to allow operating periods of at least several hours. A computer-controlled mini-pump (MP) was constructed based on non-occlusive local compression of an elastic tube with commercial bi-leaflet valves directing the pulsatile flow into a compliant circuit. Cell damage and activation in the system were tested with whole blood in comparison with a set with a conventional peristaltic pump (PP). Activation of circulating THP-1 monocytes was tested by measuring the expression of CD54 (ICAM1). Additionally, monocyte-endothelial interactions were monitored using a parallel-plate flow chamber with an artificial stenosis. The system required a priming volume of only 20 mL, delivering a peak pulsatile flow of up to 35 mL/ min. After 8 h, blood hemolysis was significantly lower for MP with 11 ± 3 mg/dL compared with PP with 100 ± 16 mg/ dL. CD142 (tissue factor) expression on blood monocytes was 50% lower for MP. With MP, THP-1 cells could be pumped for extended periods (17 h), with no enhanced expression of CD54 permitting the long-term co-culture of THP-1 with endothelial cells and the analysis of flow pattern effects on cell adhesion. A low-damage assay setup was developed, which allows the pulsatile flow of THP-1 cells and investigation of their interaction with other cells or surfaces for extended periods of time.

Key Words: Leukocytes—Pump—Adhesion Blood trauma—Leukocyte activation.

assay—

Flow chamber bioreactors play an active part in many endothelial cell culture experimental setups, providing the physiological mechanical stimuli missing in static culture (1). In recent years, the demand for low-volume bioreactors using cellcontaining fluids such as blood or cell fractions (2,3) has increased. At the same time, progress in the development of hybrid organs indicates that small pumping systems may become even more sought after in the future. Laboratory experiments, in fact, often require working with small priming volumes for extended periods of time, resulting in high recirculation rates and prolonged contact of the cells with the pumping apparatus (4). Such circuits, where low traumatization of the cellular component is a prerequisite, find application in long-term cell adhesion assays, perfusion of hybrid organs, and cell chromatography. Flow chamber bioreactors with defined flow properties have been developed for decades, in order to analyze how the behavior of endothelial cells and leukocyte attachment depends on biochemical and also on fluid mechanical factors such as shear stress (5–7). To address the problem of long-term investigations, such as the simultaneous stimulation of endothelial cells and leukocytes or, for example, testing of adsorbers, we developed a computercontrolled mini-pump (MP) based on non-occlusive local compression of an elastic tube. Furthermore, a compliant test circuit to optimize the pulsatile flow shape was developed and the atraumatic properties were evaluated with hemolysis, leukocyte activation tests, and visualization of the leukocyte-endothelium interaction. MATERIALS AND METHODS

doi:10.1111/aor.12474 Received August 2014; revised November 2014. Address correspondence and reprint requests to Dr. Dagmar Pfeiffer, Institute of Cell Biology, Histology and Embryology, Medical University Graz, Harrachgasse 21, 8010 Graz, Austria. E-mail: [email protected] 1 These authors contributed equally to this work.

The MP system A small tubing pump was developed at the Center for Medical Physics and Biomedical Engineering at the Medical University of Vienna. A computercontrolled cantilever compressed periodically, via a plunger, a 3-cm long section of silicone tubing Artificial Organs 2015, 39(11):973–978

974

THOUGHTS AND PROGRESS

(HCR silicone, 4.8 mm diameter, RS Components GmbH, Gmuend, Austria), which was equipped with two valves (duckbill valve 20–4672, Vernay Laboratories Inc., College Park, GA, USA), one at the inflow and one at the outflow. Filling followed automatically when the tubing was released due to inherent elasticity. The pump flow was adjusted via the compression depth, motor speed, and pumping frequency using a self-programmed microcontroller (Atmel-ARM, San Jose, CA, USA, programmed in C++). The circuit (Fig. 1A) included three lengths of PVC tubing with diameters of 4.2 mm and a total length of 75 cm (Fresenius Medical Care, Bad Homburg, Germany), one 20 cm length of silicone tubing with a diameter of 3.2 mm (Tygon 3350, Saint Gobain, Courbevoie, France), and a T-connector with a stopcock for sample uptake. A 10-mL reservoir acted as hemodynamic compliance, a bubble trap, and gas exchanger via a 0.2-μm pore filter (VWR International, Darmstadt, Germany). This configuration provided hydraulic elasticity, resistance, and inertia, and was chosen to tune the flow pattern to a physiological shape (a concept following Schima et al. [5]). Three different test sections were introduced into the circuit depending on the experiment. The hydraulic resistances of the different circuit configurations were calculated by dividing the pressure drop measured at the two ends of the circuit by the flow rate imposed by a syringe pump (ranging from 1 to 20 mL/min). The first test section was a 15 cm length of long silicone tubing with a diameter of 1.5 mm (RCT-SIT60, Reichelt Chemietechnik, Heidelberg, Germany); in this configuration, the circuit had a hydraulic resistance of 0.60 mm Hg/mL/min. The second test section was the same 1.5-mm silicone tubing, but with a length of 5 cm; this corresponded to a circuit resistance of 0.45 mm Hg/mL/min. Finally, as the third test section, a parallel plate flow chamber (μ-Slide I 0.8 Luer, Ibidi, Martinsried, Germany) was chosen for endothelial cell culture, generating a circuit resistance of 0.33 mm Hg/mL/min. To investigate the influence of flow patterns, a stenosis was created in the middle of the flow chamber by silicone glue, reducing the chamber width from 5 to 2.5 mm (Fig. 2A). All setups resulted in a priming volume of 20 mL. A flow meter (HT110 with sensor H4XL, Transonic Systems Inc., Ithaca, NY, USA) was placed on the silicone tubing directly beside the test section. All components were pre-sterilized and mounted in a laminar hood under sterile conditions. Unless otherwise stated, all tests were performed in a cell culture incubator at 37°C in air/5% CO2. Artif Organs, Vol. 39, No. 11, 2015

FIG. 1. Schematic drawing (A) of the test circuit. (B) Flow patterns achieved at a mean flow rate of 1.6 mL/min (top) and 6.6 mL/min (bottom). Hemodynamic performance of the circuit: (C) flow generated at a frequency of 80 bpm and an impression depth of 80% with artificially increased afterload.

Hydrodynamic characterization of the pump Hydraulic pump data were obtained using a water–glycerin mixture with a viscosity of 3.5 mPa/s at room temperature. Performance of the pump at different impression rates, frequencies, and afterloads was investigated. Flow, pressure rise, and impression depth were measured during the experiment.

THOUGHTS AND PROGRESS

975

FIG. 2. Picture of the flow chamber (A) with the stenosis. The flow chamber measures 50 × 5 × 0.8 mm; at the stenosis, the width is 2.5 mm. Scheme of the flow chamber (B) showing the locations of the immunofluorescence pictures. Immunofluorescence images of THP-1 adhered on HUAEC layer before (C) and between (D) the stenosis. F-actin is shown in green, CD90 in red. White arrows indicate adhered THP-1 cells (scale bar = 150 μm).

Blood tests: Hemolysis induction and CD142 expression The blood damage of the MP was compared with a commercial peristaltic pump (PP) (IPC pump, Ismatec SA, Glattbrugg, Switzerland). Eight test series were performed using circuits with the first test section. Additionally, a static sample of 8 mL was used as a negative control. Human blood samples of 40 mL, taken after informed written consent was obtained, were anticoagulated with heparin (50 IU/mL) and diluted with 9 mL 0.9% NaCl solution, resulting in a hematocrit of 30 to 35%. Each donation was used to fill both circuits and the static control. Blood was pumped at a mean flow rate of 10 mL/ min for 8 h at room temperature. Blood samples were taken at the 15 min, 1 h, 3 h, and 8 h time points. Free hemoglobin (fHb) per deciliter plasma was calculated spectrophotometrically using the method of Kahn (8). For CD142 expression measurements, blood probes obtained after 3 and 8 h were stained against CD14-FITC (clone M5E2), CD142-PE (clone HTF1), and immunoglobulin G (BD Biosciences, San Jose, CA, USA). The probes were measured with an FC500 cytometer (Beckman Coulter, Brea, CA, USA) and analyzed with FlowJo (Tree Star Inc., Ashland, OR, USA). Monocytes were gated as CD14-positive events. Expression levels were reported as mean fluorescence intensity (MFI). THP-1 activation in pulsatile flow THP-1 (TIB-202, ATCC, Manassas, VA, USA) cells were cultured in RPMI-1640 (Sigma, St. Louis, MO, USA), 10% fetal bovine serum (PAA Laboratories, Cölbe, Germany), HEPES (Sigma), and penicillin-streptomycin (Sigma) growth medium at

densities between 0.3 and 1 × 106 cells/mL and used for experiments up to passage 18. To prove the nonactivation of monocytes during moderate pulsatile shear stress, experiments were performed at two different flow rates. The circuit was mounted using the second test section. THP-1 cells were resuspended in growth medium at a density of 106 cells/mL and, in order to avoid cell sedimentation during extended test times, the density of the medium was increased by adding 20% vol/vol Percoll (Sigma). For moderate shear stress, the suspension was pumped at a mean flow rate of 1.6 mL/min (n = 6, Fig. 1B) for 17 h in a cell culture incubator at 37°C. As a reference setting of positive activation, the suspension was pumped at a flow rate of 6.6 mL/min for 4 h (n = 3) and cells were kept in static culture as a negative control. CD54 (ICAM-1) expression was measured by flow cytometry and reported as MFI. THP-1 and endothelial cells co-culture under flow As an example of a possible application of the system in adhesion assays, the circuit was mounted with the third test section. Primary human umbilical artery endothelial cells (HUAEC) were isolated (9), cultured in endothelial growth medium, and used for experiments until the third passage. HUAEC were seeded in the flow chamber and left to adhere for 3 h. Thereafter, the flow chamber was connected to the MP system. A 20-mL suspension, as used in THP-1 activation tests, was perfused over the endothelial layer at a mean flow rate of 1.6 mL/min (Fig. 1B). After 17 h, the cells in the flow chamber were fixed in 2% formaldehyde at 37°C, and permeabilized and stained with AlexaFluor488 phalloidin (Life Technologies, Carlsbad, CA, USA) and Texas-Red antiCD90 (Dianova, Hamburg, Germany). Pictures were taken with a TCS SP2 confocal microscope (Leica Artif Organs, Vol. 39, No. 11, 2015

976

THOUGHTS AND PROGRESS

Microsystems, Wetzlar, Germany). Concentrations of tumor necrosis factor (TNF)-α, interleukin (IL)-6, IL-8, and IL-10 were measured at the end of the experiment using the Luminex multiplex assay (BioRad, Hercules, CA, USA). Statistics Samples were compared using the paired Wilcoxon signed-rank test with the statistics software SPSS (SPSS Inc., Chicago, IL, USA). The null hypothesis was rejected at P values ≤0.05. RESULTS Hydrodynamic characterization of the MP The developed setup provided a physiologically shaped pump flow, as seen in peripheral arteries (2) (Fig. 1B). The flow rate was adjusted by changing the impression depth. At pump frequencies between 40 and 90 bpm, the mean flow rate showed a rather linear relation to the frequency. The pump showed low afterload sensitivity (Fig. 1C) and a maximum pump performance of 30 mL/min against a pressure of 190 mm Hg. Maximum frequencies up to 140 bpm were achieved. Blood tests: Hemolysis and CD142 expression After 8 h, free hemoglobin induced by the MP was 10 times lower than that by the PP (11 ± 3 vs. 100 ± 16 mg/dL, n = 8, P ≤ 0.05, Fig. 3A). CD142 was not expressed on monocytes kept under static conditions, but it was slightly up-regulated after 3 h of pumping in both circulating samples: MFI was 984 ± 101 for MP, 1041 ± 124 for PP, and 712 ± 74 for the static control (n = 5, Fig. 3B). The maximum expression was achieved by the monocytes after 8 h in the peristaltic circuit (MFI of 3877 ± 1577 for PP vs. 2208 ± 595 for MP, n = 5, P ≤ 0.05). THP-1 activation in pulsatile flow The flow of 1.6 mL/min caused no up-regulation of CD54 on THP-1 (MFI of 1561 ± 457 for MP vs. 1257 ± 405 for static, P > 0.05). The positive control showed CD54 expression of MFI = 4181 ± 749 (Fig. 3C). THP-1 and endothelial cells co-culture in pulsatile flow THP-1 accumulation in small stagnant regions caused cell activation and the production of TNF-α (106 ± 4 pg/mL), IL-6 (5926 ± 79 pg/mL), IL-8 (7834 ± 396 pg/mL), and IL-10 (5.5 ± 0.1 pg/mL). Numerous THP-1 cells adhered in the stagnation zones before and after the stenosis (schematic gray Artif Organs, Vol. 39, No. 11, 2015

FIG. 3. (A) Time course of plasma free hemoglobin for MP, PP, and static (n = 8, *P ≤ 0.05). (B) CD142 (tissue factor) expression on monocytes after 3 h and 8 h pumping (n = 5, *P ≤ 0.05). (C) Boxplot of the measurements of CD54 expression on THP-1 after 17-h circulation (n = 3 for the positive control, n = 6 for the other samples; NS = P > 0.05). NS, not significant.

THOUGHTS AND PROGRESS zones Fig. 2B). Some THP-1 cells even adhered to those areas with high shear stress in the middle of the stenosis (Fig. 2D), with HUAEC f-actin fibers oriented in the flow direction. DISCUSSION With increased performance and advanced applications of the bioreactors, the use of cellular fluids and the mimicking of natural flow patterns with low flow volumes have become increasingly desirable. The MP system described here represents a novel approach allowing circular flow with low priming volume and low traumatization and activation. The two-valve design prevents backflow and exhibits low afterload sensitivity. This setup allows the use of cellular suspensions for many hours without fluid exchange: a desirable feature for cell and hybrid organ perfusion bioreactors. Hemolysis results clearly show the superiority of the MP to conventional occlusive pumps, both in the amount of fHb generated as well as in the rate of fHb accumulation. In our tests, the fHb increase was limited to the first 3 h, while it was constant over time in the PP. This probably indicates that in the initial MP system, cell damage affected mostly an older and therefore weaker subpopulation of erythrocytes (10). Furthermore, the MP induced a less pro-coagulatory phenotype of the monocytes than the PP, as indicated by CD142 expression on monocytes. CD54 was chosen as a marker to further evaluate the biocompatibility of the pumping system on leukocytes. CD54 is an activation marker that is known to be up-regulated in THP-1 by chemical stimuli-like drugs (11) or lipopolysaccharides (12), and is considered one indicator of monocyte function (13). THP-1 cells were an extremely useful model of human monocytes during the optimization process of the MP system, as they solved the problem of donor variability, thereby guaranteeing good reproducibility of the results. We observed that CD54 up-regulation on THP-1 was in good correlation with the circuit resistance and the flow rate values. THP-1 activation could be reduced and then eliminated in pilot experiments by the careful selection of tubings, reservoir, air volume, air filter, and mean flow. During the pump developing process, the main issues that had to be addressed to prevent mechanical damage of the cells were the choice of valves with low forward pressure and the prevention of vibrations induced by the plunger. The circuit was later adapted and optimized to avoid high shear stress regions by the careful selection of tubing

977

diameter, connectors, and mean flow. Together, these modifications resulted in the reduction and further elimination of THP-1 activation; we conclude that the final parameters selected here provide “safe” conditions for THP-1 against mechanical activation. The system permitted the examination of longterm interactions between endothelial cells and leukocytes, which is useful for investigating the impact, for example, of specific vessel geometries. The challenge of this approach was the long experimental time required to permit the accumulation and subsequent contact activation of THP-1 in the stagnant regions. Live imaging of the cells during dynamic culture should also permit the analysis of flow effects on leukocyte adhesion. CONCLUSION The presented circuit with the newly developed non-occlusive pulsatile pump allows the atraumatic pumping of monocytes over extended time periods. It enhances modeling of the intravasal environment, reproducing physiological mechanical stress and allowing the co-culture and co-stimulation of different cell types. Acknowledgments: The authors want to thank Thomas Posnicek and Karl-Heinz Kellner for technical support and Dr. Carla Tripisciano for correction of the manuscript. This study was funded by the Government of Lower Austria and Give2Asia. REFERENCES 1. Morigi M, Zoja C, Figliuzzi M, et al. Fluid shear stress modulates surface expression of adhesion molecules by endothelial cells. Blood 1995;85:1696–703. 2. Lawson C, Rose M, Wolf S. Leucocyte adhesion under haemodynamic flow conditions. Methods Mol Biol 2010;616: 31–47. 3. Bianchi E, Molteni R, Pardi R, Dubini G. Microfluidics for in vitro biomimetic shear stress-dependent leukocyte adhesion assays. J Biomech 2013;46:276–83. 4. Macey MG, Wolf SI, Wheeler-Jones CP, Lawson C. Expression of blood coagulation factors on monocytes after exposure to TNF-treated endothelium in a novel whole blood model of arterial flow. J Immunol Methods 2009;350:133–41. 5. Schima H, Tsangaris S, Zilla P, Kadletz M, Wolner E. Mechanical simulation of shear stress on the walls of peripheral arteries. J Biomech 1990;23:845–51. 6. Viggers RF, Wechezak AR, Sauvage LR. An apparatus to study the response of cultured endothelium to shear stress. J Biomech Eng 1986;108:332–7. 7. Jones DA, Smith CW, McIntire LV. Leucocyte adhesion under flow conditions: principles important in tissue engineering. Biomaterials 1996;17:337–47. 8. Bednar R, Bayer PM. Free hemoglobin in plasma— comparison of two spectrophotometric methods. Interference due to bilirubin. LaboratoriumsMedizin 1994;18:196–9. Artif Organs, Vol. 39, No. 11, 2015

978

THOUGHTS AND PROGRESS

9. Martin de Llano JJ, Fuertes G, Garcia-Vicent C, Torro I, Fayos JL, Lurbe E. Procedure to consistently obtain endothelial and smooth muscle cell cultures from umbilical cord vessels. Transl Res 2007;149:1–9. 10. Yeleswarapu KK, Antaki JF, Kameneva MV, Rajagopal KR. A mathematical model for shear-induced hemolysis. Artif Organs 1995;19:576–82. 11. Endo S, Toyoda Y, Fukami T, Nakajima M, Yokoi T. Stimulation of human monocytic THP-1 cells by metabolic activation of hepatotoxic drugs. Drug Metab Pharmacokinet 2012; 27:621–30.

Artif Organs, Vol. 39, No. 11, 2015

12. Michee S, Brignole-Baudouin F, Riancho L, Rostene W, Baudouin C, Labbe A. Effects of benzalkonium chloride on THP-1 differentiated macrophages in vitro. PLoS ONE 2013;8:e72459. 13. Hiesmayr MJ, Spittler A, Lassnigg A, et al. Alterations in the number of circulating leucocytes, phenotype of monocyte and cytokine production in patients undergoing cardiothoracic surgery. Clin Exp Immunol 1999;115:315–23.

Atraumatic Pulsatile Leukocyte Circulation for Long-Term In Vitro Dynamic Culture and Adhesion Assays.

Low flow rate pumping of cell suspensions finds current applications in bioreactors for short-term dynamic cell culture and adhesion assays. The aim o...
398KB Sizes 0 Downloads 8 Views