Sample pieces of the sintered titanium surface found in the HeartMateII left ventricular assist device were obtained from Thoratec Corporation, Pleasanton, CA, and used in the experiments described herein.
In preparation for imaging with our scanning electron microscope, Philips XL30 ESEM TMP (FEI Company, Hillsboro, OR), samples were dehydrated in an ethanol series, dried in a Pelco CPD2 critical point dryer (Ted Pella Inc., Redding, CA), and coated with Au/Pd (60/40%) in a Hummer 6.2 sputter coater (Anatech Ltd, Springfield, VA). Au/Pd was sputtered from a single alloy source to a film thickness of 6-7 nm. Of note, the resulting film is enrichment in Pd due to the preferential sputtering of Pd over Au due to differences in surface binding energies and recoil densities in the Au/Pd alloy (Betz, 1980 ).
A series of experiments was devised (seeTable 1 ) to find the optimal method of regenerating the titanium surface after Au/Pd coating. The success of the various cleaning protocols was assessed by evaluating the surface chemistry of the titanium samples before and after each experiment with X-ray photoelectron spectroscopy (XPS). XPS was performed on a Kratos Axis Ultra instrument (Chestnut Ridge, NY). Samples were attached to the sample holder via copper tape. They were then inserted into the sample transfer chamber and allowed to reach a pressure of 5.0 × 10-7 torr. Following this, the samples were transferred to the sample analysis chamber where the measurements were performed. Spectra were obtained at ∼2.0 × 10-8 torr. A monochromated Al X-ray source at a power of 15 kV and an emission of 10 mA was used to obtain scans. The lens mode used was hybrid, i.e. magnetic and electrostatic. The slot aperture was used for these experiments and a resolution of 160 eV was used for survey scans. Scans were performed with the charge neutralizer on to counteract any sample charging effects and were taken over the range of 5 - 1200 eV with a step eV of 1 and a dwell time of 200 msec. Each scan constitutes an average of 2 individual scans.
The atomic percentages were obtained by determining the area under the primary peaks for the element under consideration. The areas were determined for the C1s at 284.5 eV, O1s at 529.7 eV, N1s at 400.9 eV, Au4f at 84.1 eV, Pd3d at 335.1 eV, Ti2p at 453.8 and the sensitivity factors were C=0.278, O=0.78, N=0.477, Au=6.25, Pd=5.356, Ti=2.001. A linear background subtraction was performed. Analysis was achieved using CasaXPS software.
As a control experiment, a sample piece of bare sintered titanium surface (Thoratec Corporation, Pleasanton, CA) was analyzed by XPS. This XPS spectrum served as a baseline (Table 1 , control, Figure 1a ). The titanium surface was then coated with Au/ Pd as described above and imaged with SEM using standard high vacuum imaging at 25kV (Figure 1b ). Following, five sintered titanium surfaces (Thoratec Corporation, Pleasanton, CA) were precoated with bovine fibronectin (10μg/ml) at room temperature for a duration of 12 hours and seeded with 1 × 106 porcine mesenchymal stem cells by incubation in 6 well plates for 24 hours at 37°C (Figure 2a ). All samples were fixed in 3.7% formaldehyde (Ricca Chemical Company, Arlington, TX) overnight before sputter coating as described above.
Porcine mesenchymal stem cells were obtained from bone marrow of Yorkshire swine by isolating mononuclear cells by density gradient centrifugation according to our established protocol (Werner and others, 2003 (link)). Cells were then washed with Dulbecco's Phosphate Buffered Saline (D-PBS), (Gibco, Invitrogen Corporation, Carlsbad, CA) suspended in 10% dimethylsulfoxide (DMSO), (Sigma-Aldrich, St. Louis, MO), 90% Fetal Bovine Serum (FBS), (HyClone, Thermo Fisher Scientific, Logan, UT) and plated onto fibronectin coated chamber slides in Endothelial Cell Basal Media-2 (EBM-2), (Lonza, Clonetics, Walkersville, MD) with EGM-2 Single Quots (Lonza, Clonetics, Walkersville, MD) and 10% FBS (HyClone, Thermo Fisher Scientific, Logan, UT). Cells were incubated at 37°C in a water-jacketed incubator at 5% CO2 for 7 days when nonadherent cells were removed. Cultures were then incubated an additional day before cells were harvested with 0.25% Trypsin/EDTA (Lonza, Clonetics, Walkersville, MD) and replated into T25 flasks in media. Passage 3 and 4 cells were used for our experiments.
The first of the samples coated with biologic material and Au/Pd was submerged in aqua regia (1:3 solution of concentrated nitric and hydrochloric acid) for 5 min at ambient temperature without stirring. No attempts were made to control the temperature; however, the temperature did reach ∼ 30°C upon mixing the acids together and the Ti pieces were immediately immersed into that solution. This was followed by sonication (Fisher Scientific Solid State/Ultrasonic FS-14, Pittsburgh, PA) in deionized water (DI H20) for 5 min and the Ti pieces were then dried with (N2) nitrogen gas. XPS spectra were obtained before and after cleaning.
The second coated sample was also submerged in aqua regia for 5 min, followed by sonication in DI H20 for 5 min. It was then treated by ozonolysis (UV-ozone cleaner, UVO-Cleaner Model No. 42, Jelight Company Inc., Irvine, CA) for 45 min.
The third sample was subjected to aqua regia, sonication and ozonolysis and then further sonicated in water saturated with a soap solution (Contrex labware detergent, Decon Labs Inc., King of Prussia, PA). Finally, the sample was sonicated in DI H20 for 1 hour and dried with nitrogen gas.
The fourth sample was processed like the third but without the ozonolysis step. The fifth sample processed like the third sample but without submersion in aqua regia. For all of the samples, XPS spectra were obtained before and after cleaning.
In order to determine whether aqua regia exposure would significantly alter the surface structure of titanium microspheres by etching the titanium, we used atomic force microscopy (AFM) to image titanium samples before and after aqua regia treatment. AFM images were obtained in contact mode using a Digital Instruments Dimension 3100 AFM (Veeco Instruments Inc., Plainview, NY) and a Veeco Probes DNP cantilever. Images were acquired at a scan rate of 2 Hz using 512 samples/line and 512 lines. Each image was second order flattened. The surface area evaluated was 930 μm2 on average per titanium sample.
In preparation for imaging with our scanning electron microscope, Philips XL30 ESEM TMP (FEI Company, Hillsboro, OR), samples were dehydrated in an ethanol series, dried in a Pelco CPD2 critical point dryer (Ted Pella Inc., Redding, CA), and coated with Au/Pd (60/40%) in a Hummer 6.2 sputter coater (Anatech Ltd, Springfield, VA). Au/Pd was sputtered from a single alloy source to a film thickness of 6-7 nm. Of note, the resulting film is enrichment in Pd due to the preferential sputtering of Pd over Au due to differences in surface binding energies and recoil densities in the Au/Pd alloy (Betz, 1980 ).
A series of experiments was devised (see
The atomic percentages were obtained by determining the area under the primary peaks for the element under consideration. The areas were determined for the C1s at 284.5 eV, O1s at 529.7 eV, N1s at 400.9 eV, Au4f at 84.1 eV, Pd3d at 335.1 eV, Ti2p at 453.8 and the sensitivity factors were C=0.278, O=0.78, N=0.477, Au=6.25, Pd=5.356, Ti=2.001. A linear background subtraction was performed. Analysis was achieved using CasaXPS software.
As a control experiment, a sample piece of bare sintered titanium surface (Thoratec Corporation, Pleasanton, CA) was analyzed by XPS. This XPS spectrum served as a baseline (
Porcine mesenchymal stem cells were obtained from bone marrow of Yorkshire swine by isolating mononuclear cells by density gradient centrifugation according to our established protocol (Werner and others, 2003 (link)). Cells were then washed with Dulbecco's Phosphate Buffered Saline (D-PBS), (Gibco, Invitrogen Corporation, Carlsbad, CA) suspended in 10% dimethylsulfoxide (DMSO), (Sigma-Aldrich, St. Louis, MO), 90% Fetal Bovine Serum (FBS), (HyClone, Thermo Fisher Scientific, Logan, UT) and plated onto fibronectin coated chamber slides in Endothelial Cell Basal Media-2 (EBM-2), (Lonza, Clonetics, Walkersville, MD) with EGM-2 Single Quots (Lonza, Clonetics, Walkersville, MD) and 10% FBS (HyClone, Thermo Fisher Scientific, Logan, UT). Cells were incubated at 37°C in a water-jacketed incubator at 5% CO2 for 7 days when nonadherent cells were removed. Cultures were then incubated an additional day before cells were harvested with 0.25% Trypsin/EDTA (Lonza, Clonetics, Walkersville, MD) and replated into T25 flasks in media. Passage 3 and 4 cells were used for our experiments.
The first of the samples coated with biologic material and Au/Pd was submerged in aqua regia (1:3 solution of concentrated nitric and hydrochloric acid) for 5 min at ambient temperature without stirring. No attempts were made to control the temperature; however, the temperature did reach ∼ 30°C upon mixing the acids together and the Ti pieces were immediately immersed into that solution. This was followed by sonication (Fisher Scientific Solid State/Ultrasonic FS-14, Pittsburgh, PA) in deionized water (DI H20) for 5 min and the Ti pieces were then dried with (N2) nitrogen gas. XPS spectra were obtained before and after cleaning.
The second coated sample was also submerged in aqua regia for 5 min, followed by sonication in DI H20 for 5 min. It was then treated by ozonolysis (UV-ozone cleaner, UVO-Cleaner Model No. 42, Jelight Company Inc., Irvine, CA) for 45 min.
The third sample was subjected to aqua regia, sonication and ozonolysis and then further sonicated in water saturated with a soap solution (Contrex labware detergent, Decon Labs Inc., King of Prussia, PA). Finally, the sample was sonicated in DI H20 for 1 hour and dried with nitrogen gas.
The fourth sample was processed like the third but without the ozonolysis step. The fifth sample processed like the third sample but without submersion in aqua regia. For all of the samples, XPS spectra were obtained before and after cleaning.
In order to determine whether aqua regia exposure would significantly alter the surface structure of titanium microspheres by etching the titanium, we used atomic force microscopy (AFM) to image titanium samples before and after aqua regia treatment. AFM images were obtained in contact mode using a Digital Instruments Dimension 3100 AFM (Veeco Instruments Inc., Plainview, NY) and a Veeco Probes DNP cantilever. Images were acquired at a scan rate of 2 Hz using 512 samples/line and 512 lines. Each image was second order flattened. The surface area evaluated was 930 μm2 on average per titanium sample.
