The newest lab rats aren’t rats. Scientists are growing microscopic human tissues and challenging the need for animal testing.
From the moment you wake up, you encounter hundreds of chemicals that have never been tested for safety. The shampoo and shaving cream and makeup you apply every day, the plastic water bottle you sip from, the detergent you use to wash your clothes and dishes, the fabric protector on your sofa: many—even those labeled “natural” or “organic”—are teeming with compounds whose safety for human use has never been proved. And that’s just before you leave the house.
Our lives are awash in chemicals—more than 80,000 of them, from the clothes we wear to the food we eat to the air we breathe. The Toxic Substances Control Act, passed in 1976, was intended to regulate the safety of chemicals in commercial products. But the law grandfathered in roughly 60,000 chemicals already in use and, despite an update giving the EPA more authority in 2016, to date only about 700 chemicals have been tested under the law. Meanwhile, about 3,000 new ones come on the market annually.
“There’s so many chemicals that are released into the world every year, and we have studied such a small number of them. And we only seem to study them after the fact, when we have some sort of evidence that there’s a problem,” Donna McGraw Weiss ’89, JD, a Brown University trustee, says.
That’s because testing these tens of thousands of chemicals is time-consuming and expensive work. Toxicity assessments must be done on animals, often rodents, and every chemical must be tested many, many times, at multiple concentrations, before decisions can be made. Even then the data may be flawed; humans aren’t 150-pound rats. It can cost millions of dollars, many years, and hundreds of animals’ lives to complete toxicity testing on a single substance—a significant investment if the findings are correct, a potentially tragic one if they’re wrong.
Weiss, who studied biochemistry at Brown and began her career in a molecular biology lab, says there are other, better ways to test chemicals and protect ourselves from harm, and they’re being developed at her alma mater. She and her husband, Jason Weiss, are so confident in the know-how at Brown that they’ve committed millions of dollars to support the research.
“If you have a quick, efficient, inexpensive, human cell model, you can pre-test many chemicals before they ever get out there,” she says. “The impact of that is tremendous.”
Willing Convert
More than a decade ago, Kim Boekelheide, MD, PhD, a professor of pathology and laboratory medicine, was wrapping up a three-year stint on a National Academy of Sciences committee that examined chemical safety testing methods. An expert in toxicology testing, he says “animal testing had been what I had done my entire life.” But he came to see the flaws in this traditional methodology.
The committee brought together experts in the field from around the world to ask how they could do better science and better protect the public. “I got converted in the process of serving on that committee,” Boekelheide says. In 2007 they published their findings in the report Toxicity Testing in the 21st Century: A Vision and a Strategy. It transformed the way his field thinks about testing on animals, he says.
“We need to move away from animals into an in vitro testing approach to achieve the goal of high-throughput toxicity testing for safety assessment,” which would allow thousands of compounds to be tested quickly, Boekelheide says. “I became a proselytizer: ‘we’ve got to do this.’” He returned to campus determined to find a collaborator in this new venture.
Meanwhile Jason Weiss was looking for ways to support research on alternatives to animal testing. A former vice chair of the board of directors for the Humane Society of the United States, he contacted the organization’s chief scientific officer at the time, who knew Boekelheide. Weiss was thrilled an opportunity might be found at Brown.
Donna Weiss, however, was skeptical. As a researcher at Rockefeller University, she had worked on mouse models. She’d had “moral qualms” about it, she says, but nonetheless, like Boekelheide, “I had always thought that that was how research was done.” Now a private equity adviser and philanthropist, she’s maintained a lifelong interest in scientific research and education. The more she talked to Boekelheide about the challenges facing toxicity testing and the need for faster, cheaper alternatives, “the more exciting I thought it was.”
That’s how the Weisses and Boekelheide wound up on the doorstep of Jeffrey Morgan, PhD, a professor of medical science in the Department of Molecular Pharmacology, Physiology, and Biotechnology and a professor of engineering. By then, Morgan had for years been perfecting new, better ways to conduct tests in vitro—literally, “in glass,” like petri dishes. Morgan shared their concerns about the expense and unreliability of animal testing. But petri dishes are no substitute for Homo sapiens, either.
“The classic petri dish forces cells to stick to the dish, the plastic, and spread out, and they look like a fried egg,” with the nucleus as the yolk, Morgan says. Cells in a two-dimensional layer can’t reliably predict how our three-dimensional tissues will react to a chemical exposure.
So Morgan invented the 3D Petri Dish, a patented, three dimensional cell culture mold that’s used to produce microtissues: tiny clusters of cells that naturally aggregate and form structures like capillaries and ducts. They can do this because, Morgan says, the cells are nestled in the small microwells of a molded agarose gel, “like an egg carton.” The gel is non-adhesive, so, with nothing else to stick to, the cells’ surface adhesion molecules stick to each other, as they would in a living organism.
“They’re now interacting with each other, forming a ball of cells, and they’re doing it spontaneously,” Morgan says. “And not only do they do that, they’re now free to do morphogenesis”—a developmental process in which tissues acquire their functional shapes. “We’re really just standing back and letting the biology do its thing,” he says.
And does it ever. Within weeks, brain microtissues, grown from adult human stem cells, are producing electrical signals. Heart microtissues are beating. And each is composed of just one, two, or three types of cells.
But is that enough to decide whether a chemical is safe for human use? Morgan and Boekelheide are only at proof-of-concept stage, they say, so time will tell. But they’re feeling good about their chances. “Everyone soberly acknowledges that this technology … will only get you so far,” Morgan says. “There won’t be an entire human in there. But there will be some aspect of biology that’s valuable, that’s organ specific, that will be sufficient to make a decision.”
“And that’s all you need,” Boekelheide says.
New Beginnings
Well, you also need a lot of data. And analyzing all that data is the rate-limiting step, Donna Weiss says.
That’s where she and her husband’s interest in the project really got the ball rolling. Boekelheide and Morgan had recruited other faculty to work with them to refine the technology and demonstrate its utility. “It’s been quite a journey. We started out on a shoestring,” Boekelheide says. Then the Weisses came forward with several major gifts.
“That definitely catalyzed it,” Morgan adds, by allowing the group to buy a high-throughput confocal microscope, the Opera Phenix, with which they can rapidly acquire high-quality, 3-D images of microtissues. He held up one of the bright green molds that researchers use to cast a 3D Petri Dish. Smaller than a fingertip, the plate forms 96 wells, each of which has four microwells. The cells, once added, naturally aggregate and form a spherical microtissue in each microwell. They’re then treated with different chemicals at various concentrations.
“We go to the microscope,” Morgan says, “we open the door, we put [the 3D Petri Dish]in there, close it, go to lunch. We come back, and now we have 17,000 images,” serial sections of the 384 microtissues, dyed with different fluorescent colors. With the Opera Phenix’s software, “we can assess the health or disease of those tissues based on their structure and function,” he says.
In 2017, the group became an official campus entity: the Center to Advance Predictive Biology (CAPB). Morgan and Boekelheide partner effortlessly as director and associate director, respectively, joking about each other’s hair (or lack thereof), bantering about biochemical engineering like it’s last night’s playoff game, and finishing each other’s sentences.
The center’s equally congenial members, most of them biologists, engineers, or doctors, meet regularly to share expertise and troubleshoot, and annually at a retreat for progress updates. The five teams are using microtissues to assess the effects of chemical exposures on brain, heart, lung, ovary, and prostate cells, and the Opera Phenix to gather the data. The next step is automating the process, so they can gather more data even faster. It’s only with data—massive quantities of data—that they can prove the technology’s viability as an alternative to animals in toxicity testing.
The microtissues are “being exposed across a concentration range to the 80,000 environmental chemicals … to find out, what is the concentration of exposure that produces an adverse effect?” Boekelheide says. Though each microtissue may contain only a couple of human cell types—cardiomyocytes and cardiac fibroblasts in the heart models; pulmonary macrophages, epithelial cells, and fibroblasts in the lung—it may be enough to be predictive. “The goal is to develop a panel of cells … to have confidence that you’re covering the biologic response landscape in sufficient degree to make a call on safety,” he says.
It’s the simplicity and practicality of this approach that attracted Paul Carmichael, PhD, to the project. Carmichael, science leader and senior toxicologist at the Safety and Environmental Assurance Centre at Unilever in the UK, sits on the CAPB external advisory committee. His company, which makes and sells hundreds of consumer products around the world, has funded some of the center’s research.
“You could create as an academic exercise … the perfect ex vivo lung,” Carmichael says. “But for toxicity testing, it might be that we don’t need that. … A chemical agent can have a toxicity because it perturbs a common aspect of the biology in that cell system or that organ tissue system.” He continues, “We think we can be successful … by having a number of key sentinel tissues with the higher biology provided by these sophisticated culture systems.”
Of Mice and Men
Agnes Kane, MD, PhD, a professor of pathology and laboratory medicine, studies the health effects of asbestos fibers and engineered nanomaterials on the lung. She abandoned mouse research about a dozen years ago. “We were trying to cure mesothelioma, and we did cure mesothelioma in the mice, but we also killed them,” Kane says. “I said, that’s enough of this. I can’t do this anymore. By then, I’d gone through thousands of mice, and there’s got to be a smarter way to do this.”
But when Boekelheide suggested she switch to microtissues, “I’d say, no, Kim, we can’t do this, it’s not going to reproduce any chronic pathology,” Kane says. It took him six months to wear her down. “I finally said, I’ve got to give this a try.” She laughs with delight, her eyes sparkling at the memory. “It started working, and I was just amazed.”
Her lab’s human lung microtissues are composed of macrophages, epithelial cells, and fibroblasts, which she chose to reproduce fibrosis, or collagen scar tissue. A lung has about 80 types of cells, and most pulmonary diseases arise from complex interactions between them. But disease produced by inhaling engineered nanomaterials, she says, “is due mostly to the interaction between epithelial cells lining the airways; macrophages, which are supposed to clean up any particles that get inhaled into the lungs, and then the fibroblasts,” which produce collagen.
It took three years of tinkering to get the microtissues to behave as they would in a human, to identify biomarkers that indicate fibrosis in response to asbestos or carbon nanotube exposure. Kane hopes that, if the technology works, her model might be used to prevent disease from new and existing nanomaterials. “Limit airborne exposure. The workers, make sure they’re protected,” she says. “There is a way to be more proactive about it.” She says with improvements the lung microtissues could also be used to test new therapies.
Nonetheless, “it is a model,” Kane says. “You have to be very, very careful in choosing the cells and choosing what you’re trying to assess.” Such “fit-for-purpose” models can be made, though they’ll be technically challenging. Her group also wants to compare effects on microtissues made from female and male cells, as there are hormonally driven differences in lung diseases. And they haven’t yet exposed the lung microtissues to chemicals.
The to-do list is long, but Kane finds it all exciting and worthwhile. The microtissue model “probably will be better than testing it in a mouse, and be more closely reflective of how humans would respond,” she says. “We have to do this. Because we can’t keep making more chemicals and keep using all of these chemicals.” Furthermore, she adds, “sometimes we’re exposed to several of these all at once. Now what’s going to happen? There’s no way you can predict this. You’d be using mice till there’s no tomorrow.”
Some Assembly Required
Jess Sevetson PhD’21 had misgivings about working with animals before she joined the neuroscience graduate program. She wanted to study neural system regeneration, but to do that she’d “have to injure an animal and then watch it recover,” she says. But then she learned about the lab of Diane Hoffman-Kim PhD’93, an associate professor of medical science and of engineering, who was using brain microtissues, which Sevetson now uses to study the effects of traumatic brain injury.
For the CAPB, Sevetson was part of Hoffman-Kim’s team evaluating how the microscopic spheres of brain cells respond when exposed to domoic acid, a potent neurotoxin. Preliminary results were consistent with the real-life effect, though the team noted challenges like a lack of a blood-brain barrier as well as the sheer complexity of measuring neurological response. That’s where Sevetson comes in, using calcium imaging to observe neuron activity.
“I’ve gotten good results in microtissues that are composed of about 8,000 cells,” she says. “All it takes is a couple of weeks for the neurons to put out extensions and make synapses. … They’re doing all the microscale processes that brains are doing.” And that was using microtissues made with only a few types of cortical cells, which assemble somewhat randomly; despite that, “it works about the same way every time,” Sevetson says.
As she’s gotten more involved with microtissue research, Sevetson has been rethinking her postgraduate career. “It’s pushing the boundaries. It’s work with broader implications in a lot of different directions,” she says. “I’d love to cure a disease—we all would—but I’d really love to make it easier and more efficient for other researchers to cure a lot of diseases.”
Sevetson follows developments of other organ models out there, like organ-on-a-chip technology (microchips that use human cells to model organ function) and more complex organoids to study cancer or developmental diseases like autism. “I’m thinking about that for a postdoc,” she says. “The whole field is advancing just so fast.”
Sending trainees into the world with not only knowledge of microtissues but a new perspective on animal research is one of Donna Weiss’s goals for the CAPB. “They’ll go out to other institutions and bring that with them,” she says. “It’s really seeding a whole field. … It’s really a way of us leading.”
About 25 graduate and undergraduate students and postdocs work in the labs affiliated with the center, Boekelheide says. Molly Boutin PhD’16, who worked on brain microtissues in Hoffman-Kim’s lab, is now on the 3-D Tissue Bioprinting Team at the NIH’s National Center for Advancing Translational Sciences. “The impact is the people the center trains and where they go and what they carry with them,” Morgan says.
Hearts and Minds
Kareen Coulombe, PhD, an assistant professor of engineering and of molecular pharmacology, physiology, and biotechnology, is working on cardiac microtissues for the CAPB with Ulrike Mende, MD, professor of medicine, and Bum-Rak Choi, PhD, associate professor of medicine. Using tissues derived from female human induced pluripotent stem cells, which spontaneously beat (“We call them beaters,” Coulombe says, describing how one pulsed so vigorously it popped out of the mold), the group has shown that bisphenol A (BPA) blocks ion channels, which can cause arrhythmia. It’s unlikely they would have found that with animal tests. “Rodents are notoriously terrible models for predicting cardiac arrhythmias,” she says.
But Coulombe does use rats in other research. She engineers myocardium from stem cells with the ultimate goal of transplanting it to people who’ve had a heart attack, to regenerate the injured heart. For now, though, she is using rats, sewing the macrotissues onto an organ about the size of your thumbnail.
“As an engineer, you always come up with a model that is as simple as possible to answer your question,” she says. Furthermore, “every animal protocol requires you to justify the use of animals.” Creating tissue that would re-engineer the contractility of the heart simply can’t be done on a microscopic ball of cells.
No one in the CAPB is anti-animal testing. Most have done it, some still do; it’s required to get a new drug approved, to get certain grants. “That is the status quo,” Morgan says. “What we’re trying to do is change the status quo. Do better science with alternatives.” There are endless examples of animal tests failing to predict human response, from thalidomide 60 years ago to a potential Alzheimer’s treatment that was scrapped last year, in the phase 2 clinical trial, due to liver toxicity. “They’d spent a bazillion dollars,” Morgan says, “they had a stack of data, animal data … to go to humans, and it still failed.”
The European Commission in 2013 banned testing cosmetics and their ingredients on animals, and selling such products anywhere in the EU. For that reason Paul Carmichael says Unilever’s interest in non-animal testing technology “wasn’t just all altruism. It’s perfect business sense,” he says. The market is a driver, too. While the public is comfortable with testing drugs on animals, “the opposite is often true of the cosmetics world and consumer products. People really are choosing those things that say ‘not tested on animals,’” he says. “We have to go with that, while always assuring safety.”
In September, the EPA committed to a phaseout of animal testing, simultaneously announcing elimination of its support for mammal studies by 2035 and funding for the development of alternative technologies. Compared to Europe, where several labs and private companies are working on new models, there are just a few in the US, such as the Wyss Institute at Harvard and Johns Hopkins’ Center for Alternatives to Animal Testing. “We do really have the opportunity to be one of the premier institutions in this work,” Donna Weiss says.
In her former life as a biomedical researcher, Weiss worked on the Human Genome Project, a joint endeavor among dozens of universities and labs around the world. “In order to develop this field of alternatives to animal models and using human cells for toxicity testing,” she says, “you’re going to need to have multiple facilities and centers that are really focused on this. … I’m about to see Brown being one of those core centers.”
“It takes a village to do this kind of thing,” agrees Les Reinlib, PhD, of the National Institute of Environmental Health Sciences. He’s familiar with similar projects other NIH institutes are funding. “Even though different institutes are sponsoring different things, we should be able to bring it all back together and say, here’s the entire cardiovascular system, plus the entire respiratory system, plus the entire lymphatic system,” Reinlib says. “We’ll be able to use what we want to in a much more in-depth way, much, much more similar to nature than in current tissue models.”
And Then There Were Two
As an administrator and program director in the NIEHS’s Exposure, Response, and Technology Branch, Reinlib helps oversee the $2.6 million, five-year Bioengineering Research Partnerships (BRP) grant that is funding much of the center’s current work. He also sits on the CAPB Scientific Steering Committee and, in that role, will help decide next summer which two of the organ microtissue models will move forward. There’s simply not enough money in the BRP for all five projects to generate the amount of data needed to present microtissues as a viable toxicity testing alternative to private companies as well as federal regulators.
“It’s a shame, in a way, to have invested a lot of upfront efforts—and, by the way, your money—into a project like this, and then it’s going to have to scoot around and try to find additional funding,” Reinlib says, though he adds, “I’m confident in the long run they’ll all be OK.”
The researchers say they’ll find new grants to continue their work in the center if their projects aren’t among the chosen ones. By the end of the second year of the BRP, this summer, all had showed significant progress over year one and had papers published or in the works. “We’re very collegial,” Agnes Kane says. “I think we’ll be able to make the decision. It’s a little too early to tell yet. … Things can change pretty rapidly now because we’ve got most of the technology advanced to a critical point.”
In the meantime Morgan and Boekelheide (who head up the ovary and prostate projects, respectively) have other plans for the center as it becomes a more established campus entity. They’re hiring a new faculty director; they’re working on a new, catchier name for the center; they’re planning a campus seminar series.
And they’ll keep pushing the technology, beyond the scope of the BRP grant: the Boekelheide and Morgan labs filed a provisional patent on a new coculture model in which an organ microtissue is surrounded by liver cells that would, theoretically, metabolize chemicals or drugs in a more lifelike way. They see center members eventually building more complex organoids or even whole organs—“the next frontier,” Morgan calls it.
But first things first. “The onus is on these systems to be replacements,” he says, “to demonstrate they work. So that’s a very high bar.”
“Although, on the animal side, there’s never been an onus,” Boekelheide adds, “because there wasn’t an alternative. … They’ve gotten a pass, historically—”
“That we’re not getting,” Morgan puts in. “And we don’t mind.”
Boekelheide is fond of saying that when he wrapped up his service on the National Academy of Sciences panel, he predicted it would take 50 years to get where the field is now, just over a decade later. “There’s been a huge, huge revolution in the way people are thinking,” he says. And at Brown, “we have made such tremendous progress in the last three years. I mean, there was no center three years ago.”
Donna Weiss hearkens back to the Human Genome Project. “It was years and tens of millions of dollars to sequence the first genome. And now, it’s less than a day,” she says. “That’s been over the course of my working lifetime.
“So that’s where I would love to see this go,” she continues. “And if it does, it will have so many positive impacts on science and the world.”
