In vitro irradiation system for radiobiological experiments

The irradiation of any biological system generates a succession of processes that can be grouped into physical, physical-chemical and biological phases, all characterized by specific events occurring at different times. The principal task of radiobiology is to analyze the effects that may occur in the weeks, months and years after radiotherapy and propose improvements to current therapies.

In recent years, researchers have focused on molecular mechanisms triggered by irradiation which could help cells to escape death. Such studies are key to obtaining information on potential molecular targets for the setting up of novel radiosensitization approaches or for the design of mathematical algorithms suitable for tailoring radiotherapy [1]. It is also crucial for scientists working in this specific research area to carry out experiments on in vitro biological models that are capable of reproducing and mimicking clinical conditions.

Although human tumor cell lines grown in monolayer are mainly used for research and provide important information on basic tumor biology and radiobiology, they cannot be considered as fully representative of clinical tumors. Multicellular tumor spheroids (MCTS) have a three-dimensional architectural organization in which the tumor cells are not uniformly exposed to nutrients and oxygen, a condition that closely mimics the organization of human tumors. Indeed, the potential of three-dimensional (3-D) cell cultures for new anti-cancer therapeutic strategies is gaining recognition and is believed to improve the pre-animal and pre-clinical selection of the most promising treatment modalities [2‐6].

A bioreactor developed at NASA [8] (Figure 1a) was used to grow MCTS starting from single cell suspensions derived from established human cancer cell lines (Figure 1b-e). The pathophysiological features of solid tumors (necrosis, apoptosis, dermal junctions and mitosis) in the MCTS sections stained with hematoxylin and eosin are shown in Figure 6 a,b. In particular, areas of necrotic debris characterized by hypereosinophilia and spherical-shaped apoptotic bodies showing a loss of contact with the adjacent cells were detected. The chromatin of these cells was degraded and condensed, giving the nucleus a heterochromatic appearance. The simultaneous presence of apoptotic bodies and necrosis was confirmed by immunohistochemical analysis using specific antibodies against cleaved caspase 3 and the protein HMGB1 (Additional file 1: Figure S1), respectively. Finally, structures that could be equated with the presence of dermal junctions, normally present in tissue between adjacent cells, were observed. These features were no longer visible when the same cells were grown as monolayers. Under such conditions, numerous mitotic figures randomly distributed alongside large cells with eosinophilic cytoplasm and polymorphic nuclei were observed (Figure 6c,d).

In addition, ultrastructural electron microscopy analysis revealed features present in human lung tissue, such as desmosomes, confirming what hematoxylin and eosin staining had highlighted, and lipid clusters [11‐13] (Figure 7a,b). Such structures were not observed in monolayer cultures of established tumor cell lines, as confirmed by the electron microscopy image of monolayer CAEP cells that showed polygonal shaped cells with no detectable desmosomes and only a few lipid clusters in the cytoplasm (Figure 7c).

In the present paper we report preliminary results after a 20 Gy irradiation of CAEP- and A549–derived MCTS pretreated or not with 20 μM cisplatin (Figure 8). Results were compared with those obtained after the same treatments in cell lines grown as monolayers. The cell growth of MCTS was analyzed over a 20-day period and different effects were observed in relation to the cell line type used. In particular, in CAEP-derived MCTS, significant cell growth inhibition was observed independently of the type of treatment used (Figure 8a). Furthermore, a significant increase in radiotoxicity was observed 20 days after the end of irradiation in the spheroids pre-treated with cisplatin compared to those treated with cisplatin or irradiation alone. Conversely, although the growth of A549-derived MCTS was initially strongly hampered by the combined radio-chemotherapy treatment, at the end of the 20-day period such inhibition was similar to that induced by cisplatin or irradiation alone (Figure 8b).

We also tested the same treatment schemes on monolayer CAEP and A549 cells (Figure 8c,d), analyzing cell growth over a 7-day period. Notably, the duration of the experiment was significantly shorter than that used for 3-D cultures because the cells reached confluence about seven days after the end of treatment, showing cell-cell contact inhibition and a consequent growth arrest. Under such conditions, the different treatments tested induced the same effect independently of the type of cell line used. In particular, while 20 Gy of radiation in a single dose did not influence cell growth, both cisplatin alone and cisplatin plus 20 Gy radiation caused 100% cell death.

Although the use of tumor cell lines grown in monolayer provides very important information on basic tumor biology and radiobiology, it does not fully represent clinical tumors. In particular, compared to cells cultured on a flat surface, 3-D cell culture techniques better mimic the natural environments of tissues and organs, where cells exist in a 3-D microenvironment with intricate cell-cell and cell-matrix interactions and complex transport dynamics for nutrients and cells [14, 15]. Furthermore, tumor spheroids show a three-dimensional organization in which cells are not uniformly exposed to oxygen or nutrients, which is more representative of the organization of human tumors consisting of regions of regularly dividing cells, hypoxic cells, and necrosis [6, 16]. Conversely, monolayer cultures consist of cells generally growing in a nutrient rich liquid environment or on an agar surface with an ample supply of oxygen, which form a virtually homogeneous colony.

In the present study, MCTS were grown using a bioreactor developed at NASA [8] which exploits the low force of gravity to assemble tumor cells and, importantly, offers the advantage compared to other systems (soft agar, liquid overlay, hanging drop, etc.) of producing a large amount of homogeneously-sized spheroids. In particular, the MCTS obtained using this approach exhibit growth characteristics and pathophysiological features of avascular tumor nodules and also of peripheral tumor microregions near blood capillaries [6]. Furthermore, it must be underlined that the MCTS experimental model permits several analytical investigations in whole (growth kinetics, histology, immune cell infiltration capacity and cell mediated cytotoxicity) or dissociated spheroids (molecular analysis, re-growth analysis, clonogenic assay, flow cytometry, biochemical assay) to be performed at baseline or after radiation treatment. Several studies have shown that cancer cells grown in a 3-D culture survive low doses of chemotherapeutic agents that would otherwise be cytocidal in the same cells grown in a two-dimensional (2-D) culture [6, 17]. Our preliminary results highlighted differences in terms of chemo- or radiosensitivity of the two cell lines grown as 3-D spheroidal colonies after a 20-day observation, a culture time that cannot be reached by 2-D cultures (maximum 7–8 days).

In the present work we also described how we resolved some of the problems relating to irradiation of the cells (the same system used for MCTS can be used to irradiate flasks or plates containing cell cultures grown as monolayer) under sterile conditions. For example, some authors place the cell-containing plates or flasks, wrapped in cling-film and tape, in basins filled with water to simulate the passing of radiation through tissues. These experimental conditions run the risk of water infiltration which could compromise the sterile conditions. We, on the other hand, inserted the 96-multiwell plates or the flasks containing the cells into a custom-built plexiglass phantom. Plexiglass was chosen because, from a physical point of view, in terms of interaction with radiation, it is the material most similar to water, the main component of human tissues passed through by radiation beams before reaching the target tumor mass. Furthermore, the recess within the plexiglass slabs was tailored for spheroid-containing microplates and used in our experiments to minimize the interfaces of different density between the plexiglass walls and that of the microplate itself. In this way, the plate containing the spheroids was closed inside the plexiglass chamber by its lid, thus avoiding further manipulations and the risk of water infiltration. This system also guarantees irradiation of the 3-D or 2-D cultures with complete coverage of the target utilizing the same equipment and calculation methods used for the planning of patient/treatments, without the need for dedicated or highly expensive instruments and dosimetric verification for each experiment. Furthermore, the duration of dose delivery ranges from a few seconds to a few minutes, which does not influence cell survival. We are thus able to perform experiments with both conventional fractionation and hypofractionation regimens using the same procedures and timing as routine clinical practice, resulting in a more realistic evaluation of the cellular and biomolecular effect of radiation therapy.

We believe that the novel system and biological model proposed represents an important experimental approach for researchers working in the area of radiobiology that could deepen our understanding of the basic mechanisms of tumor radiobiology and help to improve clinical treatment modalities.