A recent study can be seen as a major step forward in learning to mimic the mechanisms of the human body — or as a gigantic “oops” for years of research into neural development.
Growing in the body, cells stretch out into three dimensions, but for years researchers have cultured cells on the flat, two-dimensional bottom of a Petri dish. Since the ultimate applications of much research relies on knowing how cells act in the human body, Diane Hoffman-Kim, an assistant professor of medical science and engineering at Brown University, examined how two-dimensional and three-dimensional culture methods affect gene expression and cell growth. In analyzing nerve cells, she and her lab discovered that more than 1,700 genes were expressed differently when grown under three-dimensional conditions as compared to being grown on a standard Petri dish. These genes included ones associated with the cytoskeleton or the extracellular matrix and with neurite outgrowth. This three-dimensional method of culturing cells might bring researchers a step closer to mimicking the human body.
“We want to do experiments that are relevant to disease states in your body,” Hoffman-Kim says.
She grew human neuroblastoma cells on Petri dishes just coated with collagen or in a gel of collagen I. After a day of growth, during which neuroblastoma cells usually attach to their substrate and extend their neurites, the cells were removed for analysis.
Then Hoffman-Kim looked at cellular morphologies and global gene expression. Visualization of the cells through scanning and transmission electron microscopy showed that cells grown on the two dimensional media had flatter, more spread-out cell bodies with short neuronal extensions, while the three-dimensionally grown cells were rounder with longer neuronal extensions. She and her team followed up those observations with microarray analysis, surveying the expression patterns of each condition. The finding: neuroblastomas grown in the three-dimensional collagen I matrix had 1,766 genes that were either up- or down-regulated differently than those grown in only two dimensions.
“It was interesting to see all those changes. The gold standard after that is to confirm with RT-PCR,” says Hoffman-Kim. And so she did, focusing on cell spreading and neurofilament genes. The PCR step showed that the genes did indeed differ in expression levels — some registered a nearly two-fold change in expression.
What this means for all those experiments performed in two dimensions is not yet known — but it certainly does raise questions and create complications. Experiments that are easily done in two dimensions, such as axon guidance studies, suddenly seem more daunting in three dimensions. Furthermore, the chemical and mechanical properties of ideal three- dimensional media, to get it that much closer to the human body, still need to be worked out.