2.1 Materials
Material | Information |
AlexaFluor488 secondary antibody | Sigma-Aldrich, Darmstadt, Germany, SAB4600044, 1:1000 |
BSA | AppliChem, Darmstadt, Germany, A1391,0025, 1% |
Calcein-AM and EthD-III | Biotrend, Cologne, Germany, 30002-T |
CD68-primary antibody | Abcam, Cambridge, UK, ab955, 1:100 |
DAPI | Sigma-Aldrich, Darmstadt, Germany D8417, 1:1000 |
DMEM/F12 | Thermo Fisher, Waltham, USA, 11330057 |
FCS | Sigma Aldrich, Darmstadt, Germany, S0615, 10% |
Filter: Alizarin/Xylenolorange; Cy5&AF647; Tetracyclin; FITC/Cy5 H Dualband Filter | AHF, Tuebingen, Germany |
Methanol | Sigma-Aldrich, Darmstadt, Germany, 32213-1L, -20°C |
Olympus BX51 fluorescent microscope, equipped with a 10x objective | Olympus, Tokyo, Japan |
PBS | Sigma-Aldrich, Darmstadt, Germany,08662 |
Thermanox™ Coverslip membrane | Thermo Fisher, Waltham, USA, 174985 |
UV-Lightsource X-Cite Series 120 Q | Excelitas Technologies, Waltham, USA |
2.2 β-TCP
The Robert Mathys Foundation (RMS) produced the β-TCP ceramics according to our specifications. A mixture of eighty grams α tricalcium phosphate (α- TCP; Ca3(PO4)2), 20 g tricalcium phosphate (Art. No. 102143, Merck, Switzerland), 60.0 ± 0.2 g solution of 0.2 M Na2HPO4 and 1% polyacrylic acid (Art. No. 81132, Fluka, Switzerland; Mw = 5.1 kDa) was produced. After intensively stirring for 2.5 minutes the paste was poured into a plastic syringe of 70 mm length and 23 mm diameter and left for 45 minutes. The paste was then covered with 10mL of phosphate-buffered saline (PBS) (Item No. P5368, Sigma, USA), pH 7.4 solution, and incubated for 3 days at a temperature of 60°C. Afterwards the green bodies were dried at 60°C and sintered for 4 h at 1250°C using a heating and cooling rate of 1°C/min. The ceramic cylindrical-shaped bodies were now cut to 7 mm diameter and 25 mm length. To remove wear particles and organic residues by combustion the ceramics were washed in an ethanol bath and calcined at 900°C (26).
β-TCP ceramics produced with this method were analyzed in other works using environmental scanning electron microscopy (ESEM) and were found to have a mean pore diameter of 4.8 ± 1.2 µm
(27) or an average pore radius of 2.1 ± 0.3 µm (28).
β-TCP dowels were sawn into 6 mm or 2 mm thick slices and subsequently placed into an ultrasonic bath starting with 10 minutes in 70% ethanol followed by 10 minutes in distilled water. Afterwards the slices were autoclaved and ready to be loaded into a flow chamber.
2.3 Sampling
Buffy coats (BC) and EDTA whole blood were provided by the Institute for Transfusion Medicine and Gene Therapy, University Freiburg. BC was used without further preparation and 60 ml EDTA whole blood were pooled to further process into either PRP (platelet-rich plasma) or PPP (platelet-poor plasma). Blood counts were performed by the Institute for Clinical Chemistry and Laboratory Medicine, University Freiburg. Analyzes were performed on a Sysmex XN analyzer. Platelets were measured using the impedance method (resistance measurement). Monocyte counts as part of a differential blood count were determined by flow cytometry as a 6-part differential. BC, PPP, PRP were analyzed in a 1:10 dilution and pooled EDTA whole blood without dilution.
PRP was created using centrifugation according to Cattaneo et al. 2013 (29). 60 ml pooled EDTA whole blood was centrifuged at 200 g for 10 minutes (no breaks). The supernatant of the erythrocyte fraction (plasma and the leukocyte + platelet rich middle layer) was then pipetted into a new tube resulting in PRP with an increased monocyte count. The target for range for platelet counts was 600–800,000/µl to contrast higher platelet counts in BC.
PPP was meant as a negative control with low monocyte and platelet numbers and created according to Jo et al. 2013 (30): We centrifuged 60 ml pooled EDTA whole blood at 900 g for 5 minutes (no breaks). Transferred the supernatant of the erythrocyte fraction (plasma and the leukocyte and platelet rich layer) into a new vessel and centrifugated at 1500 g for 15 minutes (no breaks).
Afterwards the upper 2/3 of the liquid supernatant were pipetted into a new vessel to obtain platelet-poor plasma with a low monocyte count (PPP). To receive even lower platelet counts the last centrifugation step (1500 g x 15 minutes) was repeated in one of the samples.
2.4 Loading via flow chambers
TCP was loaded into stainless-steel flow chambers described as by Seidenstuecker et al. 2015 (31).
Five 2 mm slices were placed into a silicone seal and put into a flow chamber. The slices are later referred to in accordance with the flow direction of the sample material. Slice 1 is the first slice resembling 0 mm depth, slice 2 with 2 mm depth, slice 3 with 4 mm depth, slice 4 with 6 mm depth and slice 5 as the last slice with 8 mm depth (Fig. 1). Up to four flow chambers were loaded per sample.
Integrity of slices 1 and 5 was visually controlled during the whole loading process to prevent a sample flow through cracks.
After attaching transparent vacuum hoses to both sides of the chamber, a mild vacuum of 350 mbar gets applied. When reaching the target pressure, 1ml sample material of either BC, PRP or PPP was pipetted into the vacuum hose connecting to the front side of the flow chamber. Normally, loading did not lead to the suspension being completely drawn through the chamber and was terminated approximately 15–20 seconds after the sample came to a standstill. The chamber was now reopened and the silicone seal carrying the 2 mm slices is stored in DMEM/F12 + 10% FCS until further treatment on either the same day (day 0) or the next day (day 1). If a sample tube had to be stored overnight, it was placed inside an incubator at 37°C.
2.5 CD68-Immunofluorescence
Slices were retrieved from the silicone tube and allowed to dry for 15 minutes. Every slice was separately washed in PBS, fixated in -20°C Methanol for 10 minutes, blocked in 1% BSA for 60 minutes and then incubated with CD68-primary antibody over night. On the next day, slices were stained with AlexaFluor488 secondary antibody and counterstained with DAPI. Slices were repeatedly washed with PBS between each step.
Monocytes were identified by their size of 15–22 µm (32) combined with CD68-positivity (33) and their origin from peripheral blood as by nature of our samples. Additionally, DAPI-counterstaining had to be in direct relation to the observed CD68 fluorescence for us to assume that the monocytes were intact.
2.6 Live/dead assay
Slices loaded with BC, PRP or PPP were stained using Calcein-AM and EthD-III on either day 0 or day 1. Living cells were identified by green fluorescence and dead cells by red fluorescence. Cells with a size ranging from 18–22 µm were seen as monocytes in the live/dead assay. Cells with a range of 15–17 would technically also qualify as monocytes but would not be distinguishable from eosinophil granulocytes with a size range of up to 17 µm. Cells smaller than 15 µm were valued as granulocytes. (32)
A Thermanox™ Coverslip membrane incubated with BC was used as a positive control.
2.7 Microscopy
We performed fluorescence microscopy to examinate the front side of each slice immediately after either live/dead or CD68-immunofluorescence staining using an Olympus BX51 fluorescent microscope equipped with a 10x objective. A UV-Lightsource X-Cite Series 120 Q was used for excitation and pictures were taken using filters Alizarin/Xylenolorange and Tetracyclin for Life/dead pictures and filters Cy5&AF647 and FITC/Cy5 H Dualband Filter for CD68-immunofluoresence pictures.
2.8 6 mm slice
To verify that used samples go through the microporous structure of the ceramics and positive results in the 2 mm slice model are no product of samples flowing around the edges of each slice we also performed loading, immunofluorescence staining and live/dead staining of 6 mm slices for BC, PRP and PPP. Chambers were loaded with a singular 6 mm slice inside a silicone tube. The slice was cracked open vertically by using pliers before staining. We then analyzed the breaking edge resembling the inside of the slice using fluorescence microscopy in accordance to the 2 mm slices.