In the present study we demonstrate that quality of recellularization with endothelial cell line EA.hy 926 in acellular kidney scaffold depends on the insensitivity of the beforehand used decellularization protocol. Specifically, relative cell count after five days of dynamic culture more than doubled after gentle (i.e. 0.66% SDS) compared to aggressive (i.e. 3% SDS) decellularization. Thus, our present results demonstrate that gentle decellularization offers more suitable culture conditions for the human endothelial cell line EA.hy 926 which benefits attachment and proliferative behavior in acellular renal scaffolds. For culture of human osteoblasts, our group surprisingly discovered improved niche conditions in acellular kidney scaffold after decellularization with SDS at a concentration of 3% compared to 0.66%. These results could not be verified for the human endothelial cell line EA.hy 926. Regarding observations by Burgkart, Tron (20) and Schmitt, Csiki (24) who could already prove suitable growth conditions for human osteoblasts in renal scaffold after decellularization at 0.66% SDS this decellularization protocol should also be preferably used for future coculture experiments.
SDS is known for effective removal of cellular components  and is more potent to do so even in cell dense tissues compared to other detergents such as triton x-100 [26, 27]. For the purpose of whole organ decellularization it is typically used at concentrations between 0,1%-1% depending on organ size  ranging up to 4% in combination with triton x-100 at a concentration of 3% performed in rat kidneys [29, 30]. At the same time SDS also has adverse effects on the remaining extracellular matrix such as reducing amount of growth factors and sulfated glycosaminoglycans [31, 32] as well as disrupting its ultrastructure . Besides concentration of the used agents, their exposure time is the second key variable defining how significant these effects are. Compared to decellularization protocols of rat kidney described in the literature [29, 30, 34–36], the method by Burgkart, Tron (20) used in this study ist not only time saving but also comparatively gentle due to short exposure time. Efficacy was furthermore confirmed by absence of cellular remnants, intact ultrastructure and positive immunohistochemical staining for laminin, fibronectin, and collagen IV. However, aggressive decellularization with SDS at 3% led to insufficient recellularization with only singular cells adherent to the vascular basement membrane. As compared to noticeably improved results after gentle decellularization at 0.66% SDS this might most probably be caused by an impairment of endothelial cell niche conditions in the acellular kidney scaffold due to SDS-induced alterations of the ECM. Observations of our group although indicated improved results for long-term cultivation of human osteoblasts seeded via the ureter after denaturing decellularization with SDS at 3% compared to 0.66%. It is known that perfusion pressure and shear stress differ in between the tubules and the vasculature system leading to uneven physical effects on cellular binding for both compartments. In which compartment cells are reseeded might therefore have enormous implications for recellularization success due to their physical properties.
For recellularization of acellular kidney scaffolds, 20-40x106 immortalized human endothelial cells EA.hy 926 were injected into the renal artery followed by dynamic culture at perfusion pressure up to 100mmHg matching physiologic conditions in vivo . As demonstrated in previous studies by Burgkart, Tron (20) reseeding with HUVECs could be successfully performed after decellularization at 0.66% SDS following this procedure. While both mentioned cell types share similar phenotypes , expression of integrin β1,- α2 und α5-subunits is significantly lower in endothelial cells EA.hy 926 than in HUVECs impairing their ability to establish a firm adhesion to fibronectin and collagen I in static culture . As perfusion pressure and concomitant shear stress is known to negatively affect reendothelialization  weak binding to the ECM can furthermore lead to cell detachment and removal under dynamic conditions. Despite impaired adhesive properties of EA.hy 926 endothelial cells, the present work still demonstrates their ability to adhere and proliferate in acellular kidney scaffold after gentle decellularization at 0.66% SDS. Predominantly in the cortex area, they reached an acceptable density partly forming a monolayer on the vascular basement membrane. With these results endothelial cells EA.hy 926 can be considered as an viable alternative for primary endothelial cells such as HUVECs which were so far frequently used for dynamic culture studies in acellular kidney scaffolds. By applying EA.hy 926 endothelial cells instead of HUVECs, some of their natural limitations can even be overcome. This particularly includes their short life spawn after which they senesce, transform to giant cells, and ultimately undergo apoptosis as well as donor-specific properties leading to less comparable results depending on their origin .
A major and yet unsolved problem of reendothelialization is to obtain an even distribution of endothelium throughout the vascular system. The dispersion of cells within the scaffold is strongly affected by how they are brought into it during the seeding process giving this step enormous importance . As performed in the present study, the primary method to insert endothelial cells into acellular kidney scaffolds is seeding through the renal artery. On this way cells will distribute within arterial vessels and glomeruli but only barely reach the peritubular capillaries . To overcome this issue, successive injecting anterograde though the renal artery and retrograde through the renal vein is described to be beneficial . Furthermore applying negative pressure during seeding  or preincubation with CD-31 antibodies  were reported giving significant improvements. According to existing literature our experiments also resulted in insufficient distribution with a relatively low cell count in the medulla area even after decellularization with SDS at 0.66%. As discussed above, optimal seeding techniques for the endothelial cell type EA.hy 926 need to be further investigated.
However, the present work has some limitations. First, no recellularization of acellular kidney scaffold was performed after decellularization with SDS at concentrations in between 0.66% and 3%. For recellularization purpose only one cell type (i.e. EA.hy 926 endothelial cells) was utilized giving less opportunity for comparative evaluation. To our knowledge comparative analysis do not exist in the literature and must be further investigated. Only a total of 6 experiments with n = 3 for each group (i.e., SDS at 0.66% and 3%) were obtained and analyzed while each experiment was stopped after 5 days of dynamic culture. Therefore, no long-term studies e.g., 14 or 21 days were applied. Lastly, no coating or preincubation of the vascular system within the acellular renal scaffold for potential optimization of recellularization was tested.