LFH is one of the most crucial factors in degenerative LSS, which not only affects quality of daily life due to pain and discomfort but also leads to disability, especially in elderly patients. However, due to the lack of basic understanding of this disease, invasive surgery has remained the major avenue of treatment. Therefore, one notable goal of this study was to provide a proof-of-concept for a spheroid cluster model for studying in vivo human LFH using an in vitro culture system with improved similarities. Primary LF cells are usually investigated in 2D cell culture on petri dishes, and thus the characteristics of the proposed 3D spheroid culture were compared with those of conventional 2D culture.
Among various assessment tools, morphology is the most direct method to study the form and structure of living cells. Cytoskeletal actin filaments are particularly abundant polymers under the plasma membrane and respond to environmental cues by establishing mechanical support, determining cell shape, and facilitating cell-substrate movement.[21] Therefore, staining the cytoskeletal structure may help provide a better understanding of the peculiar shapes observed in different cell culture conditions in an intuitive manner (Fig. 1). The results indicated that 2D planar culture on a hard plastic surface may be much less representative than spheroid culture, which provides improved cell-cell and cell-matrix interaction and 3D geometric arrangement.
The choice of the spheroid culture platform was based on the high reproducibility and accessibility of conventional ULA 96-well plates, which are simple to harvest for further analysis.[22] One crucial feature of ULA 96-well plates is the convenience of direct staining and observation; consequently, spheroid formation can be easily and continuously observed over time. Although the sphericity of a spheroid depends on the intrinsic property of cell type, LF cells seem to favor spheroid formation immediately after 1 day (Fig. 1a). This rapid spheroid formation may be due to the fibroblast-like nature of LF cells; a few fibroblast-based studies have highlighted the tendency of fibroblasts to form spherical spheroids [23–25]. Another interesting aspect revealed from the time-lapse investigation was the different growth trends depending on the initial LF seeding concentration. These results may be explained from a homeostatic perspective, considering the secretion of metabolic waste, nutrient diffusion and oxygen consumption in media. The 100-cell spheroids, due to the lower diffusion barrier, may encourage cell proliferation or ECM secretion to expand in volume over time; by contrast, 1000-cell spheroids appeared to reach a plateau throughout the period of investigation, whereas 5000-cell spheroids primarily underwent a process of compaction, as also reported elsewhere [26, 27].
Given the paucity of information on 3D culture of LF cells, it is imperative to ensure the viability of primary cell growth in the proposed platform. Ethidium homodimer-1 binds to nucleic acids in cells with damaged membranes. Only a minor amount of cells in spheroids showed bright red fluorescence, and the majority of cells were green, suggesting that the LF cells had good viability over the 28-day culture period. Long-term maintenance of the LF spheroids demonstrated the repeatability and reliability of the culture conditions for the growth of primary LF cells. Because the growth trends differed under distinct conditions, cell cycle analysis was used to evaluate whether the different culture conditions had an effect on the cell cycle (Figure S2). Cell cycle analysis by flow cytometry was performed after 4 days of 2D culture and 3D culture (Figure S2a-c). Most of the LF cells were in the G0/G1 phase, which means that most of the cells were quiescent in both the monolayers and spheroids. However, a smaller proportion of cells were in the S and G2/M phases, indicating that the cells grew and proliferated more slowly. In 2D culture, the cell cycle distribution was not significantly different between the 1000-cell and 5000-cell cultures (Figure S2d). These results indicated that cell number does not significantly affect the cell cycle of LF cells in 3D culture compared with monolayer culture. In fact, a previous study revealed that long-term culture and cloning of primary human bronchial basal cells can maintain pluripotent differentiation capacity and cystic fibrosis transmembrane conductance regulator channel function. [33] Therefore, long-term maintenance of in vitro primary culture provides future possibilities for functional research. We cultured LF cells for 28 days to ensure that the cells could be maintained in long-term culture.
The expression of a matrix rich in collagen type I and III and fibronectin has been found in LFH tissue, [28] and the cultured LF cells acquired an LF phenotype with uniform expression of fibronectin, elastin and collagen type I and III in both 2D and 3D conditions as validated by immunofluorescence cell staining. The initial cell seeding concentration did not appear to change the acquisition of the LF phenotype in terms of the analyzed ECM expression profile. Western blot analysis further showed that fibronectin primarily differed between the 1000-cell spheroids and 2D culture, whereas collagen I and III and elastin were expressed at similar levels in both culture conditions. Because elastin degradation is an important characteristic of LFH, the minimal discrepancy in elastin expression may be attributed to the degenerative nature of the source hypertrophic tissue, underscoring the need to investigate ECM expression levels at different cell seeding concentrations and times. Further investigations are needed to clarify the details.
Histological examination using H&E staining is a standard assessment in cytopathology for studying morphological changes, such as the arrangement of the LF structure and the severity of elastin degradation, i.e., loss, fragmentation and disorganization [20]. Compared with non-hypertrophic LF tissue with a defined ECM orientation, the hypertrophied LF cells not only showed fragmented and disorganized fibers but also cell proliferation and cluster formation (Fig. 6). LF cells from the LSS group displayed high elastin degradation, consistent with previous studies. [22–24] Notably, the morphological features of the LF cells in the 3D spheroid model were more similar to those of LFH tissue, including random arrangement and cell-cell clustering, than 2D-culture cells. These observations suggest that this platform may provide a pan-pathophysiological in vitro model that more closely mimics part of the LFH compared with conventional 2D culture.
In this study, a proof-of-concept for the generation of an in vitro 3D LF model was demonstrated using conventional ULA 96-well plates. This work is the first to attempt to develop a conducive in vitro model using a 3D culture system and to report cell clustering resembling that found in primary tissue from LFH patients. The preliminary findings suggest a high dependence of the growth of LF spheroids on the cell seeding concentration, whereas cell viability is less of a concern when culturing in ULA 96-well plates. In addition, the current platform conforms with the standard well-plate formats used by most laboratories with high ease of control and observation. The pure cellular-based model facilitates the study of ECM expression. Although the current pioneering approach has demonstrated considerable robustness and potential utility, there are some limitations that should still be taken into account. First, 100-cell spheroids can be difficult to integrate with subsequent cell cycle and Western blot analysis because of the insufficiency of the cellular contents retrieved from each ULA 96-well plate. Acquiring sufficient cellular material would significantly increase the cost and involve laborious procedures. Second, the source of the primary sample greatly affects the quality of spheroid formation. We observed that clusters could not be formed from a few primary LF cells, but the exact cause remains elusive. Last but not least, the process of spheroid culture is profoundly dependent on development over time, as the 100-cell and 5000-cell spheroids did not appear to reach a plateau or valley of growth over the 28-day investigative period. However, other proposed platforms may have similar issues.