Stem cells and cell culturing

Characterization of the adipose-derived stem cells showed the mesenchymal origin of the cell lines used in the experiments. Only the hematopoietic surface marker CD34 showed higher expression values than suggested in the ISCT criteria for adipose-derived stem cells, which is typical when human serum is included in the culturing media. The expres-sion of this marker is shared by endothelial cells and hematopoietic stem cells (Bourin et al. 2013). Since the absolute majority (99%) of the hematopoietic cells should not adhere to cell culturing dishes (Gordon et al. 2006) and the endothelial cells in general are very dependent on shear stress and supplements, they may dedifferentiate or trigger apoptosis when cultured statically (Baer & Geiger 2012). Since the analyzed cells were at passage 1, it can be assumed that the expression of CD34 would decrease along passaging.

The 3 different cell lines showed distinct proliferative properties. Cell line 1 proliferated particularly fast - the cells started readily to curl on top of each other and the flasks were confluent only a few days after passaging. Cell line 2 had a moderate cell proliferation rate. The third cell line, used for the third set of ScCO2 processed scaffold experiments and for the PLDLA 96/4 scaffolds, showed quite fast proliferation, although not as active as the first cell line. These differences might affect the results of the cell number analysis and the efficiency of the USPIO-labeling, as the iron oxide particles become diluted (Küstermann et al. 2008).

7.3 Cell seeding methods

The study was based around 6 different cell seeding methods. Out of these, the static method was considered to be the standard control method due to its easiness and sim-plicity (Dai et al. 2009; Vitacolonna et al. 2013). In this method, no external forces are applied to the cells, which in some cases leads to non-uniform cell distribution (Ding et al. 2008). However, it also means that the cells are not subjected to forces that could possibly alter their viability or function. A uniform distribution of cells using static seed-ing requires a loose enough scaffold structure to enable cell migration and passive diffu-sion of cells into the scaffold. Another advantage of the method is that it is not dependent on the scaffold type.

The squeezing method was hypothesized to lead to better seeding results by means of suction forces created by squeezing the elastic ScCO2 processed scaffolds. Squeezing be-fore the cells were applied clearly had an effect on these scaffolds: especially the Sc-COMP50 scaffolds felt hard at the beginning despite the pre-wetting at +37ºC. After the pre-squeezing, the scaffolds were softer and more elastic. In addition to the sectional forces, it was postulated that the inner microstructure of the scaffolds might get broken during the pre-squeezing, which could result in an even better interconnectivity of the pores. The squeezing method was also applied to the knitted and non-woven PLDLA 96/4 scaffolds, but the knitted scaffolds could not be squeezed much without breaking the scaf-fold structure. Even low force squeezing resulted in deformation of the scafscaf-fold shape.

As expected, the non-woven scaffolds recovered poorly from the stress.

One concern with the squeezing method was the usage of the sterile plastic pouches. Han-dling and orienting the scaffold in a desired way was often complicated due to the slippery pouch. Touching the sides of the pouch with the top side of the scaffold (where the cells were seeded) was tried to be avoided, often unsuccessfully. This, as well as the flow of medium and cell suspension from the scaffold during the squeezing process admittedly led to losing some of the cells to the pouch during the seeding process.

One of the methods with a significantly larger cell suspension volume was centrifuga-tion. The way it was done in this work is partially dependent on the scaffold shape and mechanical properties. Soft non-woven fabrics and knitted scaffolds seemed to detach

from the bottom of the falcon tube. At the same time, pressing the scaffolds too tightly needs to be avoided or the openness of the pores might be compromised. Some minor problems were related in removing the scaffold from the tube – caution was needed to avoid contaminations and to keep track which was the top side where the cells were seeded. One question regarding this method is associated with the hypothesized fluid flow through the scaffold: of the cell suspension volume, how much actually flows through the pores and what portion goes directly to the bottom of the falcon tube from the sides? To avoid wasting cells, it is possible to re-suspend the cells from the bottom of the tube be-tween centrifugation (Roh et al. 2007), but this makes the process also more time-con-suming and complicated.

Similar problems with the fluid flow were encountered with both of the syringe methods.

Before starting the cell seeding experiments, the methods were tried out with distilled water and Sc-COMP50 scaffolds. These scaffolds fit the syringes tightly enough to pre-vent the liquid from flowing past the scaffold before applying pressure with the syringe plunger. Later it was noticed that this was not the case with the other scaffold types, or even with some of the COMP50 scaffolds. Instead, the cell suspension flowed often freely past the scaffold. Also the COMP50 scaffolds fitting the syringe tightly raised a question whether the fluid actually flows through the scaffold - some signs were noticed that at least part of it slips past the scaffold. Ideally, this should be prevented by tight-fitting the scaffold into the syringe, yet leaving the porous structure open for fluid flow.

Concerning syringe 1 cell seeding method, the looseness of the scaffold-syringe con-struct meant that most of the 1 ml cell suspension had already flown out from the syringe before the plunger was used. The effect of the pressure created by the plunger was anyhow clearly seen, as the remaining cell suspension splashed forcefully out from the syringe.

At the same time, the appearance of the more loosely structured pre-wetted knit and non-woven scaffolds turned paler and drier, due to the cell culturing media partly escaping the scaffold.

The larger cell suspension volume means that a large amount of cells are wasted with the excess cell suspension. The rationale behind it was to create fluid flow inside the pore network of the scaffolds and to avoid too high cell densities in the suspension, which might result in aggregates on top of the scaffold. In syringe 2 method, the idea was to actualize the fluid flow from both sides of the scaffold. With most of the scaffolds, the tightness of the construct was not sufficient. In practice, this meant that the seeding method turned out to be a semi-static seeding method: the scaffolds floated in cell sus-pension as the syringe constructs were handled according to the protocol.

Out of these 6 seeding methods, the injection method was probably the most dissatisfy-ing in terms of repeatability. First of all, the injection depth needed to be estimated in every injection. Pushing the needle too far resulted in injecting most of the cell suspension directly to the bottom of the well. Another concern was dividing the cell suspension

evenly into the 5 injection spots. Not only estimating such small volumes was hard, but achieving the even division in practice required extreme precision. In addition, the method damages the scaffold structure slightly. In the case of Sc-COMP50 scaffolds, dark traces of metal particles were left to the injection sites, indicating that the needle is ground by the ceramic content.