Since the 1980s, stem cells’ shape-shifting abilities have wowed scientists. With proper handling, a few growth factors, and some time, stem cells can be cooked up into specific cell types, including neurons, muscle, and skin.
However, stem cells know more than they’re given credit for. Over the past decade, researchers have discovered that, left to their own devices, stem cells will generate multiple cell types that assemble into structures resembling an organ. These organoids have been made for many body parts, including the retina, liver, intestine, kidney, and even the brain.
We knew we could make stem cells, but we thought to turn them into organs would require a lot of engineering. It turns out that a lot of these programs are built in, so we really do nothing— just stick the cells in a gel,” says Hans Clevers of the Hubrecht Institute in Utrecht, The Netherlands (Figure 1). Clevers pioneered the use of adult stem cells to generate miniature versions of intestine and now makes lung, stomach, liver, pancreas, breast, and prostate organoids. “We just need to get the stem cells on the right track, and then they sort of take over from there,” says Madeline Lancaster of the MRC Laboratory of Molecular Biology in Cambridge, United Kingdom (Figure 2). Lancaster developed brain organoids while working with Juergen Knoblich at the Institute of Molecular Biotechnology at the Austrian Academy of Sciences in Vienna.
These self-driving stem cells are taking scientists for a wild ride of possibilities. Although tiny, organoids made from human cells provide useful platforms for drug and toxicology testing, venues for studying human-specific aspects of development, and potential patient-specific tissue sources that could remedy a damaged organ. To help organoids reach their full potential, bioengineers are exploring ways to steer their self-organizing properties. This includes incorporating vessel-like networks into organoids to deliver oxygen and nutrients and finding materials that mimic the milieu in which organs grow.
Along the way, organoids promise to reveal more about the basics of development. “I think bioengineering has a very important role to play to move this field to the next level,” says Matthias Lutolf of the École Polytechnique Fédérale de Lausanne, Switzerland (Figure 3). “It can help generate more reproducible and standardized organoid systems, which could open up new applications.’’
Like a blank slate that can become anything, a stem cell divides to give rise to all of the cell types in our bodies. During development, stem cells and the new cells they spawn send and receive biochemical and physical signals to and from their surroundings, (that is, other cells or the extracellular matrix in which cells sit). This complicated back and forth gradually specifies the identity of a cell and gives rise to the complexity of our bodies: the curvature of a liver, the undulations of the intestine, and the labyrinth of 100 trillion neural connections contained within the human brain.
Renowned for their activities during embryogenesis, stem cells can also operate in full-grown adults, most notably in skin, which constantly renews itself. When an organ is damaged, dormant stem cells within the organ also awaken to divide and replace lost cells. Clevers discovered that such adult stem cells can divide indefinitely, yet are restricted to making the specific tissue they come from—unlike embryonic stem cells or stem cells that were reprogrammed to become pluripotent.
Such adult stem cells formed the basis of Clevers’ intestinal organoids, also called “miniguts,” which are now the most well-established organoid system. In 2009, a team including Clevers showed that, when a single adult stem cell from mouse intestine was put into a gelatinous culture environment, it readily divided to form a clump a few millimeters wide . Closer inspection revealed an orderly side-by-side arrangement of multiple cell types as they are found in the intestinal lining, complete with troughs called crypts and fingerlike projections called villi, which absorb nutrients from food in the gut. “They have every component that you see on the inside of the gut, all the cell types are there, the structure is the same, it’s just very small,” Clevers says. “And you can make massive amounts of them.”
The ease with which miniguts develop, including those derived from human cells, means that researchers can generate enough tissue to run tests on a large scale. For example, Clevers has devised an assay to predict whether a drug will help someone with cystic fibrosis (CF), a disease caused by mutations in a gene encoding a channel in the lungs and intestine. About half of patients have the same mutation, but the other half have less frequent mutations in other parts of the gene; this disqualifies them from available drugs, which can cost about US$250,000 a year.
The assay starts with a biopsy of colon tissue from the CF patient, isolates and grows the adult stem cells into intestinal organoids, then exposes the organoids to each drug. If the organoid swells—an indicator that the malfunctioning channel has begun to work—then the drug is likely to work for the patient. The test requires only two weeks, and insurance companies are paying for it to be performed on all CF patients in The Netherlands. For the unlucky ones for whom the drugs don’t work, their miniguts may be frozen and stored away for future tests with new drugs.
Miniguts may also eventually be used as regenerative medicine. Another of Clevers’s teams has successfully grafted miniguts onto the insides of damaged intestines in mice , but this approach is not yet ready for humans. Most organoids are grown in Matrigel, a gelatinous mixture of proteins produced by mouse sarcomas, which makes transplantation into a human a no go. Also, as cells divide to form organoids over and over again, mutations may accrue during DNA replication, although so far their genomes seem clean. “We really need to work out how far you can expand organoids to feel safe, how you should check the cells that you are going to give back to the patients, and how to deliver them,” Clevers says. “I think it is a promise for the future, but the time is not immediately upon us yet.”
Brain in a Dish
The inaccessibility of the human brain has not kept it out of the organoid surge. In 2013, Lancaster and her colleagues introduced minibrains grown from human skin cells that had been reprogrammed to a primordial stem cell state . From there, they were nudged toward a neural fate, then suspended in Matrigel. “What was remarkable was that it just kept on growing and reached the size of a pea,” Lancaster says. The resulting light tan ball of cells contains resemblances to different regions of the human brain, including the retina, the layers of cortex, the memory center (hippocampus), and even the spinal cord (Figures 4 and 5).
Lancaster is using these minibrains to understand what makes human brains unique, particularly their expanded size compared to other species. Specifically, she is trying to understand which steps in development go awry to lead to conditions such as microcephaly. “Most of what we know about human brain development doesn’t actually come from humans, but rodents mainly,” she says. Others are seizing upon brain organoids to understand the effects of genetic glitches linked to autism or schizophrenia, for which hundreds of genes have been implicated. In addition, eventually spinal cord cells derived from paralyzed patients may help heal their severed spinal cords.
Brain organoids can also help gauge the toxicity of compounds, according to a recent study by researchers with the University of Wisconsin, Madison . After exposing organoids to known toxic and nontoxic compounds, the team collected RNA from the minibrains to get a readout showing which genes switched off and on in response to the compound. They then trained a computer model to recognize toxic and nontoxic signatures of gene expression. “Now, if we have gene expression profiles of our tissues in response to something unknown, we can feed it into this computational model and out comes the answer: toxic or nontoxic,” says Bill Murphy, a bioengineer who devised the study (Figure 6). It worked well: when organoids were exposed to a new set of compounds, the model correctly identified nine out of ten compounds as toxic or nontoxic. The one it didn’t get correct might have been a shortcoming of their experiment because drugs don’t normally bathe the brain but enter through blood vessels. “So now we are trying to figure out how to get plumbing into the system,” Murphy explains.
Plumbing and Scaffolding
The plumbing problem looms large for organoid researchers, who need ways to efficiently deliver oxygen and nutrients to keep their miniature organs healthy and allow them to grow beyond their current millimeter scale. Now, this is accomplished by stirring and shaking, which only goes so far. Even in the case of the pea-sized brain organoid, the center contains a wasteland of dead cells. “If you really want to start to capture organ-level functionality, you probably also need larger sizes,” says Lutolf, who is investigating ways to grow tubes of intestine to study how bacteria colonize the gut and how drugs are absorbed within it. “It cannot be a tiny balloon of cells.”
One approach is to coax endothelial cells, which line blood vessels, to form networks within organoids. Murphy’s brain organoids contain such networks as a result of introducing endothelial cells during organoid assembly. However, the resulting three-dimensional (3-D) vasculature was left unused. His group is now working on how to perfuse the vessels, which could result in larger structures.
Another approach builds artificial networks of channels within the gelatinous medium in which organoids are grown. For example, a 2012 study used a 3-D printer to form vessels from sugar. Cells were grown around this vessel cast, and, once they were established, the sugar was dissolved, leaving behind a network of channels that could be perfused with blood . Endothelial cells could also be seeded into these artificial channels, where they quickly lined the walls. Similarly, microfiber networks spun from cotton-candy machines can create microcapillary, bedlike channels in gel mediums .
Bioengineers are also coming up with synthetic alternatives to Matrigel, composed of polymerized proteins organized in lattices. These hydrogels should avoid the batch-to-batch variability of the mouse-derived Matrigel, which contains a diverse mix of 1,800 different proteins. A well-defined synthetic hydrogel could help standardize the quality of organoids grown within it and make them suitable for human transplantation.
The physical scaffolding surrounding an organoid may be just as important as the soluble growth factors and hormones that cell biologists have traditionally supplied to stem cells to get them to differentiate. Cells sense the stiffness, stickiness, and shapes of their physical environment, and their fates change accordingly. Thus, bioengineers are finding ways of controlling these properties, including sculpting topography into hydrogels, such as nanopits, nanopillars, and microgrooves . Murphy’s group is systematically altering these properties to identify the four to five variables that encourage a certain organoid to develop. “You might say we’re looking for the Goldilocks conditions— that sweet spot optimized for a particular tissue type,” he says.
Growth-guiding soluble molecules can also be attached to these physical scaffolds. Lutolf et al. have developed a way to robotically spot combinations of bioactive molecules into hydrogels , and microfluidic technology may soon allow more naturalistic gradients to be created.
Dynamic aspects of a developing organ’s environment may also be mimicked. For example, Lutolf has experimented with how light-sensitive, cross-linking molecules may be embedded into the hydrogel; when exposed to light, additional cross-links are formed, leading to a localized stiffening of the gel. Likewise, photoactivation can attach or release biochemical signals. Of course, orchestrating the precise spatiotemporal dynamics of these signals presupposes a strong working knowledge of development itself. However, building organoids is a good test of this knowledge. As Lutolf puts it, “If you can really control development, then you also begin to understand it.”
- T. Sato, R. G. Vries, H. J. Snippert, M. van de Wetering, N. Barker, D. E. Stange, J. H. van Es, A. Abo, P. Kujala, P. J. Peters, and H. Clevers, “Single Lgr5 stem cells build crypt-villus structures in vitro without a mesenchymal niche,” Nature, vol. 459, pp. 262–265, May 2009.
- S. Yui, T. Nakamura, T. Sato, Y. Nemoto, T. Mizutani, X. Zheng, S. Ichinose, T. Nagaishi, R. Okamoto, K. Tsuchiya, H. Clevers, and M. Watanabe, “Functional engraftment of colon epithelium expanded in vitro from a single adult Lgr5+ stem cell,” Nat. Med., vol. 18, Mar. 2012.
- M. A. Lancaster, M. Renner, C.-A. Martin, D. Wenzel, L. S. Bicknell, M. E. Hurles, T. Homfray, J. M. Penninger, A. P. Jackson, and J. A. Knoblich, “Cerebral organoids model human brain development and microcephaly,” Nature, vol. 501, pp. 373–379, Sept. 2013.
- M. P. Schwartz, Z. Hou, N. E. Propson, J. Zhang, C. J. Engstrom, V. Santos Costa, P. Jiang, B. K. Nguyen, J. M. Bolin, W. Daly, Y. Wang, R. Stewart, C. D. Page, W. L. Murphy, and J. A. Thomson, “Human pluripotent stem cell-derived neural constructs for predicting neural toxicity,” Proc. Nat. Acad. Sci. USA, vol. 112, pp. 1,2516–1,2521, Oct. 2015.
- J. S. Miller, K. R. Stevens, M. T. Yang, B. M. Baker, D. H. Nguyen, D. M. Cohen, E. Toro, A. A. Chen, P. A. Galie, X. Yu, R. Chaturvedi, S. N. Bhatia, and C. S. Chen, “Rapid casting of patterned vascular networks for perfusable engineered three-dimensional tissues,” Nat. Mater., vol. 11, pp. 768–774, Sept. 2012.
- J. B. Lee, X. Wang, S. Faley, B. Baer, D. A. Balikov, H.-J. Sung, and L. M. Bellan, “Development of 3D microvascular networks within gelatin hydrogels using thermoresponsive sacrificial microfibers,” Adv. Healthcare Mater., vol. 5, no. 7, pp. 781–785, Feb. 2016.
- W. L. Murphy, T. C. McDevitt, and A. J. Engler, “Materials as stem cell regulators,” Nat. Mater., vol. 13, pp. 547–557, June 2014.
- A. Ranga, S. Gobaa, Y. Okawa, K. Mosiewicz, A. Negro, and M. P. Lutolf, “3D niche microarrays for systems-level analyses of cell fate,” Nat. Commun., vol. 5, no. 4324, July 2014.