By Jeffrey M. Perkel
Perhaps no aspect of life in a modern molecular biology lab is more ubiquitous – or more important -- than cell culture. Cell culture is at the heart of almost every experiment -- the source of every protein, RNA, and genomic DNA sample, of every microscope slide and transfection dataset.
Most biologists grow cells the way their advisors did it, and their advisors before them: As monolayers on two-dimensional polystyrene surfaces. There's just one problem: The resulting data may very well not accurately represent what happens in vivo.
"A number of publications have documented gene expression or protein markers that are generally lost in 2D culture that are not lost when cells are grown in 3D," says Stephen Navran, Chief Scientific Officer at Synthecon.
After all, cells in the body generally don't grow as monolayers but in complex three-dimensional structures, structures that provide physical scaffolding, cell-cell communication, migratory paths, and other cues. When it comes to cell morphology and behavior, as well as gene expression, physiology, and drug responses, those factors can make a big difference.
Cells grown in so-called three-dimensional substrates can form tissue-like structures called spheroids, which, for instance, more accurately model the drug, nutrient, and gas diffusion properties of tissues in vivo than do cell monolayers. Primary endothelial cells can form microvascular networks in BD Biosciences' BD Matrigel 3D matrix that they don't make in standard 2D culture, says Marshall Kosovsky, technical support manager at BD Biosciences. And human pluripotent stem cells seeded onto AMSBIO's Alvetex® matrix with Basement Membrane Extract and induced to differentiate can form gland-like structures, says managing director, Alex Sim.
"If cells are growing in a three-dimensional environment, then the connections between them are more native-like, so the proteins being expressed and the cell-cell communication is more realistic. That allows your downstream experiments to provide more practical output," says David Welch, senior manager for market development for primary and stem cell systems at Life Technologies.
Several companies now support 3D cell culture applications. Some provide natural or artificial scaffolds that cells can migrate into, whereas others use hardware to recapitulate a 3D environment. Whether the application is stem cell maintenance or drug development and ADME/Tox, the goal of these systems is to emulate tissues in vivo, and thus to create a more realistic – and more reliable – model of cellular behavior.
3D matrices
Perhaps the most common approach to 3D cell culture is the porous matrix. Cells generally are plated atop these matrices (though it sometimes is possible to mix the cells and scaffold prior to plating). The cells can then either grow on top of the 3D substrate, or burrow inside the matrix to form more complex structures.
Matrices run the gamut from natural to synthetic, and some companies offer a range of options. For instance, BD Biosciences offers its BD Matrigel, an extracellular matrix preparation derived from EHS mouse tumors, complete with growth factors, but also BD Laminin/Entactin Complex, a partially purified Matrigel formulation highly enriched in laminin but lacking collagen IV and growth factors. BD PuraMatrix Peptide Hydrogel is a synthetic, animal-free peptide-based hydrogel suited for a variety of applications such as stem and primary cells.
Similarly, Life Technologies offers Geltrex™, which like Matrigel is an ECM preparation derived from a mouse tumor; CELLstart™ CTS™, an xenobiotic-free preparation specifically intended for stem cells; and Algimatrix, a growth factor-free, seaweed-derived alginate.
The key to selecting a matrix, says Wendy Bray, global product manager for cell culture research at Life Technologies, is to consider the cells and their needs. "A hepatocyte has different matrix requirements than a keratinocyte or a neuron," she says. "So what is its normal environment?" Hepatocytes, for instance, might prefer to form spheroids, and thus would thrive in a plain scaffold such as Algimatrix, whereas mammary cells might do better in Geltrex, as it more closely mimics those cells' native environment, Bray says.
Other 3D matrices are wholly synthetic. That's especially important when doing drug testing, says Carlos Caicedo-Carvajal, research scientist at 3D Biotek, because it eliminates potential variability caused by animal-based products, which may act as "black boxes" during drug testing. "If you are assessing cellular response in the presence of animal products, there is uncertainty about the response of the cell as a result of the linked effect of animal-derived factors and drug treatment."
3D Biotek's 3D Insert™-PS scaffolds (available in 6- to 96-well formats) are made of polystyrene, for instance, the same material widely used for tissue culture. So is AMSBIO's alvetex, a 200-micron-thick polystyrene "sponge" growth surface launched in April at the American Association for Cancer Research annual meeting. According to Sim, polystyrene offers advantages in both physical properties – it doesn't autofluoresce and is compatible with standard molecular and cellular assay techniques, he says -- and researcher comfort. "You are creating a 3D extendable culture material without moving away from substrates you are already familiar with," he says.
Also using a synthetic material is the Scivax 3D Cell Culture System, available in the US from InfiniteBio. The bottom of the Scivax NanoCulture® Plates is a thin film containing a nanoscale raised square or honeycomb grid on which the cells sit. According to Akiko Futamura, president and CEO of InfiniteBio, while NCPs do not present a true 3D surface in the same way that Matrigel does, for instance, their grid prevents cells from adhering to the surface as tightly as they would on a plain polystyrene dish. The surface's uniformity, she adds, promotes spheroid formation. "Compared to other coatings, [the NCP] has the most reproducible results in terms of the size of the spheroids you get," she says.
3D surfaces on demand
Grant Cameron of TAP Biosystems (formerly The Automation Partnership), says the problem with most biological scaffolds is that the resulting hydrogels are too weak – that is, that the concentration of materials in the scaffold is lower than in vivo.
"Collagen is typically between 2 and 5 mg/ml off the shelf," Cameron says. But the normal physiologic concentration of collagen is 50 to 100 times higher (100-200 mg/ml).
TAP Biosystems' in-beta RAFT (Real Architecture for 3D Tissue) system is an instrument designed to produce more physiologically accurate 3D scaffolds, says Cameron. The system starts with a mixture of cells and collagen at the standard concentration of 2-5 mg/ml. Once a hydrogel forms, the system uses an absorbant plug to slowly wick away the water, compressing the cell-infused gel into a thin wafer while raising the collagen concentration.
"We believe we are moving from these '3D-but-not-physiological' matrices to more physiological matrix strengths," Cameron says. As proof, Cameron notes that the company is working with researchers in London to develop corneal transplants. The implants are created by first creating a RAFT fibroblast stromal layer, on top of which limbal epithelial stem cells are seeded. "It's essentially a transplant of the outer two layers of the cornea," he says.
In the lab, RAFT can produce 3D surfaces in 96-, 24-, and 12-well plate formats, as well as in permeable membrane well-inserts that can be used for co-culture and signaling experiments. The system requires a piece of hardware whose price has not yet been fixed. Cameron estimates it will cost between $27,000 and $30,000, and will be out of beta "towards the end of this year."
Cells in space
Another strategy for 3D cell culture comes straight from NASA. The space agency wanted to better understand the impact of weightlessness on astronaut physiology. So, researchers there began developing a culture system that would mimic that process. Very quickly, says Synthecon's Navran, they noticed that when cells were submerged in liquid and rotated, they would form 3D aggregates reminiscent of tissues.
"That, in a nutshell, was the impetus for forming our company," says Navran.
Today Synthecon, which was founded in 1990, develops rotary cell culture systems for 3D culture. Though several designs are available, says Navran, "The essence of the technology is a cylindrical fluid-filled reaction vessel that rotates along its long axis." Anchorage-dependent cells grown under these conditions "go into orbit," he says, because they are submerged in a completely fluid-filled container. But they also have better access to nutrients and oxygen than in static cultures. As a result, he explains, they are weightless and well-fed, but experience no shear (as opposed to systems that keep cells suspended through stirring or turbulence).
"With the exception of endothelium in blood vessels, most cells don't experience shear in vivo," Navran explains. Doing so in culture could force the cells to alter their expression patterns to adapt.
According to Navran, the advantage of 3D culture isn't so much quantitative as qualitative. For instance, he says, researchers have shown that neural progenitor cells grown in this manner can form spheroid structures and differentiate into something akin to brain cortex. "It's amazing how different it is from a standard 2D culture," he says.
Synthecon offers several systems based on this rotary culture premise, the simplest of which uses disposable reaction vessels and costs about $2,600. On the high end are systems with external perfusion and sensors that cost about $16,000. Culture vessels range in size from 1 mL to 3L.
Spheroids on demand
Sometimes researchers are more interested in coaxing cells into specific 3D forms on top of matrices, rather than within them. Those researchers might consider using the 3D Petri Dish™ from Microtissues, which can be used to generate controlled and reproducible embryoid bodies, spheroids, mammospheres, and more.
In a 3D Petri Dish, explains company president Jeffrey Morgan, the cells cannot interact with the matrix, as agarose, the material from which they are made, is not adhesive. Instead, the cells interact with each other. “It maximizes cell-to-cell interactions, as opposed to cell-scaffold or cell-gel interactions,” Morgan says.
3D Petri Dishes are agarose culture plate inserts (for 12- and 24-well plates) containing microscale recesses (like pits) on which cells may aggregate and grow. Like Jell-O molds, those inserts are cast by the user out of molten agarose using autoclavable plastic molds, which Morgan calls “precision micromolds,” and are available in a variety of configurations, including spheroids, rods, toroids, and honeycombs. The autoclavable and reusable molds come in packages of six identical units (which cost $500 USD), but users can also purchase a “Tech Evaluation” package containing eight different molds.
According to Morgan, the 3D Petri Dish can be used to study both homogeneous and heterogeneous cell-cell interactions. By mixing stromal and cancerous cells, for instance, researchers can model a cancer’s interaction with its surrounding tissue. Harvesting the resulting spheroids, he adds, is simple: because they do not interact with their growth substrate, the cells can be collected simply by inverting the agarose in media.
Should I go 3D?
Though all agree cells are more physiological when grown in 3D cultures, that doesn't mean researchers should abandon 2D completely.
Common 3D applications, Kosovsky says, include researchers growing primary cells or stem cells, or who are interested in co-culture experiments. "There may be a benefit in seeing greater functionality and demonstration of cell behavior that is more representative of the situation in vivo," he says. But for general cell maintenance and expansion, 2D will often suffice, he says.
If you are planning to migrate your 2D cultures to 3D, expect at least some transition time, Welch says, though like everything else culture-related, each cell type is different. Experimental protocols also will have to change, Kosovsky says, as variables like plating density, doubling time, and harvest procedures must be adapted to the new growth conditions as well.
Bray suggests looking for a system that most closely mimics the cells' in vivo environment, just as when selecting growth media. In some cases, products can be mixed – Geltrex-coated polystyrene scaffolds for instance – but no one system will work for everything, she adds. "There are many different kinds of products and you have to think about what's the best choice for your cells and application."
The image at the top of this article is Alvetex matrix from AMSBIO.