© TOSHIRO SHIMADA/GETTY IMAGES
When biomedical engineer Jeff Karp has questions, he looks to animals for answers. In 2009, Karp gathered his team at the Brigham and Women’s Hospital in Boston to brainstorm novel ways to capture circulating tumor cells (CTCs) in the bloodstream. They mulled over the latest microfluidic devices. Then the conversation turned to the New England Aquarium, and to jellyfish.
Scientists have tried to grab cancer cells from blood ever since they discovered that tumors shed malignant cells that migrate throughout the vasculature—a process known as metastasis. “If you pluck out these cells, you have a direct indicator of what the cancer looks like,” says Karp. “Then you can screen drugs to get those that will have the greatest impact.” Doctors might also be able to detect such cells during the earliest stages of metastatic cancer, when it’s more readily treatable.
Karp wanted to change that. “We asked ourselves, ‘What creatures can capture things at a distance?’” he recalls. One of his graduate students suggested jellyfish, whose long, sticky tentacles grab prey and other food particles from water. Within a year, Karp and his colleagues had designed a microfluidic chip on which 800-micron-wide microchannels are lined with long, tentacle-like strands of DNA that bind a protein on the surface of leukemia cells as they pass through the channels. (See illustration below.) In 2012, Karp showed that the jellyfish-inspired device could process 10 times more blood than existing chips in the same amount of time and trap an average of 50 percent of circulating leukemia cells.1 Karp estimates that a device the size of the standard microscope slide could collect hundreds or thousands of tumor cells in minutes. Encouraged by such results, Karp’s team is now improving the platform, designing chips that can catch any CTC of interest.
The jellyfish is far from the only intriguing organism to have served as a blueprint for scientists in the field of bioinspired medicine. Researchers have taken cues from the adhesive chemistry perfected by mussels and marine worms to create tissue glues that stick in wet and turbulent conditions; from red blood cell membranes to help drug-carrying nanoparticles avoid immune attack; and from the slippery slides that help carnivorous pitcher plants catch prey to produce novel antibacterial surfaces. (See “Bioinspired Antibacterial Surfaces.”) Nature, it seems, provides a compendium of biomedical solutions.
“Nature has used the power of evolution by natural selection to develop the most efficient ways to solve all kinds of problems,” says Donald Ingber, founding director of the Wyss Institute for Biologically Inspired Engineering in Boston. “We’ve uncovered so much about how nature works, builds, controls, and manufactures from the nanoscale up. Now we’re starting to leverage those biological principles.”
Looking to nature is not a new concept, and bioinspiration is just one of several approaches bioengineers employ to devise new medical treatments and devices. But in the last few years, the approach has come to the fore with several promising new products, even if most of them remain a few years away from human trials. “Almost every research institute now has a center for biomimicry or biologically inspired engineering,” says Ingber. “It’s just reaching that tipping point where it’s going to begin to have an impact.”
SANDCASTLE WORM: PHEBE LI FOR THE SCIENTIST. DIGRAM: KIMBERLY BATTISTAMedical adhesion is one area where bioinspiration promises to make an impression. Stitches and staples are still the standard for suturing wounds and closing up surgical incisions, but these technologies can damage tissue, leave gaps for bacteria to infiltrate, and increase the risk of inflammation. For years, surgeons have been in need of new medical adhesives that can bond tissue strongly inside the body without provoking inflammation.
Heeding the call, bioengineers have again turned to the sea. Phillip Messersmith of the University of California, Berkeley, for example, is focused on the protein-filled secretions marine mussels use to fasten themselves to wave-battered rocks. The proteins in these liquid secretions are rich in an amino acid called dihydroxyphenylalanine (DOPA), which features reactive catechol chains. These catechol chains bond tightly with each other in a mussel’s own secretions but also bond with metal atoms present on the surface of rocks. Using this strategy as a blueprint, Messersmith and colleagues chemically synthesized a variant of DOPA to crosslink biocompatible polymers.
Their glue has successfully fastened transplanted insulin-producing islet cells to the outer surface of the liver and nearby tissues in mice.2 The technique could potentially provide an alternative to standard methods of islet transplantation in which islets are infused into the liver vasculature, where they trigger an inflammatory response that quickly kills off about half of the transplanted cells—and impairs the surviving cells’ ability to produce therapeutic insulin. The researchers are also testing the bioinspired adhesive’s ability to repair ruptured fetal membranes, which can lead to premature birth and other serious complications. (See “Mimicking Mussels,” The Scientist, April 2013.)
The University of Utah’s Russell Stewart is drawing inspiration from another marine creature: the sandcastle worm (Phragmatopoma californica), which builds reef-like shelters by gluing together grains of sand. Like mussels, the worm produces secretions full of proteins that crosslink to form strong bonds. But unlike mussels, “sandcastle worms are putting adhesive directly into the water,” says Stewart. That makes them an ideal muse for the development of tissue glues that can stick things together in moist environments of the body.
PORCUPINE, SPINY-HEADED WORM: PHEBE LI FOR THE SCIENTIST. DIAGRAMS: KIMBERLY BATTISTAStewart and his colleagues discovered that the sandcastle worm’s underwater glue is produced in two separate secretions: one that contains negatively charged polyphosphate proteins and another that harbors positively charged polyamine proteins. When mixed together under the right conditions, these oppositely charged proteins form electrostatic bonds and associate into a separate fluid that does not dissolve in water. This condensed fluid partially sets in 30 seconds and, over a few hours, solidifies to the consistency of shoe leather as crosslinks form between the proteins. “[Sandcastle worms] came up with this chemistry of using the electrostatic interactions between oppositely charged polyelectrolytes to ensure it doesn’t mix with the water, and then forms a strong and permanent bond,” Stewart says. “That’s a mechanism that hasn’t previously been used in medical adhesives.”
Stewart’s team made synthetic polyelectrolytes with the same characteristics and in the same ratios as those produced by the sandcastle worm. The result is an injectable fluid that can be applied underwater and that does not shrink or swell once in place. In 2012, Stewart demonstrated that the glue could patch fetal amniotic membrane ruptures in an in vitro model.3 Now his team is working with researchers at the University of Texas Medical School to test it in a live pig model of preterm fetal surgery. Karp’s lab is also taking inspiration from sandcastle worms and slugs to design a polymer-based, light-activated glue that can bind cardiac tissue in beating pig hearts.4 (See “Next Generation: Strong Surgical Glue on Demand,” The Scientist, January 8, 2014.)
Meanwhile, Karp is developing another type of surgical suture inspired by a very different animal. The North American porcupine protects itself with a coat of roughly 30,000 quills, each of which slips into flesh easily but is hard to pull out—as many an unsuspecting dog can testify. Intrigued, Karp and his colleagues bought dozens of porcupine quills on eBay to take a closer look and saw that the tips were covered in microscopic backward-facing barbs. Then the researchers performed a few mechanical tests.
Barbed quills required roughly half as much force to penetrate pigskin as quills on which the barbs had been sanded off, but took four times as much force to remove. That’s because the barbs are flexible, meaning they flatten as they slip into flesh and flare out and snag on tissue fibers when pulled in reverse. “It’s a beautifully engineered system,” says Karp. Duly inspired, he and postdoc Woo Kyung Cho fashioned plastic quills modeled on the porcupine’s barbs. Sure enough, a patch of the synthetic quills slipped easily into muscle tissue or pig skin, but required 30 times more force to pull out than a patch covered with barbless quills.5 Barbs could be made of biodegradable materials that would release the patch after sufficient time for healing. The team is currently developing a scalable prototype that Karp hopes will eventually enter human trials for suturing wounds or internal tissues.
Porcupines aren’t the only spiky species prodding Karp’s imagination. “When we thought about it, we figured that parasites would have all sorts of interesting ways of latching on to their hosts,” says Karp. Searching Google images, the group came across the spiny-headed worm (Pomphorhynchus laevis), a parasite with a spine-covered proboscis that swells immediately after entry to fasten itself to the intestinal wall of its fish host. In 2013, postdoc Seung Yun Yang and Karp mimicked this design by creating conical microneedles: a polystyrene core surrounded by a layer of polystyrene and polyacrylic acid, a superabsorbent chemical used in diapers. This outer layer immediately absorbs water when it enters flesh, and the swollen tips mechanically lock the needles into place. (See “Sticking Power,” The Scientist, July 2013.)
Karp’s microneedle patches could be used to hold skin grafts in place, he says. Unlike staples, they don’t tear tissue or create easy entry points for bacteria. And because they lock tissue together across the whole graft and have pores for fluid to escape, they prevent fluid from pooling in the middle, which can prevent tissue fusion. Indeed, Yang and Karp’s 100-needle patches held pigskin onto chicken muscle tissue with roughly four times the strength of metal staples.6 His team is now working on a biodegradable version that could be used to fasten internal tissues.
Cleaning up blood
In 2011, when the Defense Advanced Research Projects Agency (DARPA) asked the Wyss Institute’s Ingber to develop a blood-cleansing device to treat sepsis, he focused on the spleen. Sepsis is an often fatal condition in which inflammation goes into overdrive as a result of rampant systemic infection. It kills around 6 million people worldwide every year and is a major cause of death among injured soldiers in the field. It’s hard to treat because doctors often can’t identify the cause of the infection, and thus the right antibiotic, before it’s too late. “[For] about 70 percent of people who get sepsis, [we] don’t know the cause and just treat blindly,” says Ingber.
SPLEEN: PHEBE LI FOR THE SCIENTIST. DIAGRAM: KIMBERLY BATTISTA, REDRAWN FROM NAT MED, 20:1211-16, 2014. So the challenge was to make something that quickly removes all microbial pathogens and toxins—but none of the native components of human blood—without the need to first identify the infectious agent. This is essentially the job of the human spleen.
To recreate the spleen’s intricate microarchitecture, a team led by Joo Kang, a postdoc in Ingber’s lab, designed a microfluidic device with a series of horizontal slits connecting two vertical channels: a high-flow channel through which contaminated blood is perfused and a low-flow channel perfused with saline. Before entering the device, the septic blood is mixed with magnetic nanobeads coated with a genetically engineered version of the natural human blood protein mannose-binding lectin (MBL). These proteins latch on to sugar molecules found on the cell surface of most microorganisms and bind toxins released by dead microbes, allowing them to be cleared in the spleen. As blood flows through the spleen-mimicking device, pathogens and toxins bound to the MBL-coated nanobeads are pulled into the saline-filled channel and then into a collection vial by a magnet, while the cleansed blood is returned to the patient. (See “Next Generation: Blood Cleansing Device,” The Scientist, September 14, 2014.)
“We’re not replicating the entire spleen here; we took a natural capture protein from the blood and a little bit of the structure and flow dynamics of the spleen,” says Ingber. “Now we have a relatively simple device that clears out many different bugs and the toxins that trigger the inflammatory cascade in sepsis, so we don’t need to figure out what the cause of the infection is before we treat.”
When the researchers tested their device on rats infected with E. coli or Staphylococcus aureus, the device removed more than 90 percent of the bacteria from the blood. After five hours of treatment, 89 percent of the rats whose blood had been filtered were alive, whereas only 14 percent of controls—rats untreated or merely attached to the “biospleen” without MBL beads—survived.7 In addition, the researchers showed that the device could cleanse blood at a rate of up to 1.25 liters per hour, meaning it could handle the volume of blood in the average adult human in about five hours—and multiple devices can be linked in parallel.
Ingber thinks the device could also help to treat viral diseases such as influenza, HIV, and Ebola because MBL is known to bind proteins on the surface of viruses, though he hasn’t tested it yet. His team is now trying a simplified version of the device in living pigs and is close to establishing a company to take the technology from the lab to the clinic. “We’re very excited about this,” says Ingber.
Drugs in disguise
JEREMIAH’S CPS/WIKIMEDIA COMMONS; SUMMERDROUGHT/WIKIMEDIA COMMONS; PASCAL DEYNAT/ODONTOBASE/WIKIMEDIA COMMONSBiology on the nanoscale also holds clues that could inspire new ways to fight infection and disease. Take drug delivery, for example. Researchers have tried for years to create nanoparticles that ferry drugs to particular tissues or cells. The challenge has been to design particles that can evade the immune system and survive in the bloodstream long enough to reach their target sites. Coating nanoparticles in polyethylene glycol (PEG) delays but does not prevent the immune system from tagging them with proteins that tell macrophages to attack. Indeed, fewer than half of such PEG-coated nanoparticles remain in circulation after just 16 hours. It’s also possible to coat nanoparticles with chemically synthesized versions of “self” peptides found on the body’s own cells that signal macrophages not to attack. But it’s devilishly difficult to replicate such complex arrays of proteins and lipids, and the immune system readily sniffs out fakes.
Liangfang Zhang, a biomedical nanoengineer at the University of California, San Diego, decided to use the real thing—bits of red blood cell (RBC) membranes. “Red blood cells circulate in the bloodstream for long periods”—up to 120 days—“and the immune system does not attack them because the proteins on their surface say, ‘Don’t eat me,’?” explains Zhang. He and his team separated RBCs from proteins and other cells in blood, then soaked them in a solution that made the cells burst. Next the researchers chopped the deflated RBC membranes into fragments, which spontaneously coated nanoparticles when mixed with them. In the end, the nanoparticles were wolves in red blood cells’ clothing. And, sure enough, they fooled the immune system into letting them pass. In 2011, Zhang’s team demonstrated in mice that half of the camouflaged nanoparticles remained in the bloodstream 40 hours after injection.8 “This should give the nanoparticles enough time to reach their targets,” says Zhang, who is now working with biotech companies to develop large-scale fabrication techniques and to push the product into clinical trials.
Elsewhere, Omid Farokhzad, a nanotechnologist at Harvard Medical School and the Brigham and Women’s Hospital in Boston, is looking to human physiology—specifically, infant physiology—to improve drug delivery. “It is widely understood that delivering biologics is extremely difficult to do orally,” he says. Large, complex molecules are broken down by enzymes in the gastrointestinal tract or don’t penetrate the intestinal membrane, so they must be delivered by infusion or injection. “But when babies breastfeed, they absorb maternal antibodies to become passively immunized against diseases,” Farokhzad says. “If it’s so difficult, then how come babies do it so well?”
Scanning the scientific literature, he and his colleagues learned that the neonatal Fc receptor (FcRn), found on the surfaces of cells lining the walls of the intestine in babies and adults alike, grabs the stalks of the Y-shaped maternal antibodies—known as the fragment crystallizable (Fc) region—and shuttles the antibodies into the bloodstream. Using this as a model system, the researchers worked with experts in FcRn biology to create insulin-filled nanoparticles whose surfaces are decorated with the maternal antibodies’ Fc fragments. When administered orally to mice, these Fc-coated nanoparticles crossed the intestinal epithelium and released insulin into blood 11 times more efficiently than control nanoparticles.9 Enough insulin was delivered to produce a 30 to 45 percent reduction in the mice’s blood sugar levels.
“This was the first and is still the only example of an actively targeted nanoparticle for oral delivery by transcytosis across the gastrointestinal lining,” says Farokhzad. In the future, he adds, the approach could be used to develop new protein-based treatments for diabetes and could potentially be applied to other diseases, including cancer.
Farokhzad has also borrowed design ideas from viruses to develop a nanoparticle-based vaccine against nicotine addiction, meant to help smokers quit and to prevent relapse. The product, known as SEL-068, consists of nanoparticles engineered with an array of proteins and peptides meant to mimic the surfaces of viruses, and coated with nicotine. As the nanoparticles flow into the lymph nodes, they trick the immune system into creating antibodies against nicotine. This keeps nicotine in the bloodstream from reaching the brain, removing the pleasurable sensations it causes. Farokhzad cofounded Selecta Biosciences to develop the product, which entered Phase 1 clinical trials in 2011. It is one of several virus-inspired nanoparticle products to have progressed to human studies.
Most bioinspired medical innovations are still a few years away from that milestone, however. When it comes to progressing from lab bench to hospital bedside, bioinspired products are no different from those generated through other approaches: translation is extremely challenging. “You can solve a problem in the lab and get great results, but if you can’t manufacture at large scale and at low cost it will never be a solution,” says Karp, who has established two start-up companies to shepherd his innovations into clinical trials. Paris-based Gecko Biomedical, which he cofounded with MIT engineer Robert Langer, plans to push its sandcastle worm–inspired tissue glue into human trials later this year.
As researchers reveal nature’s tricks in ever-finer detail, bioinspiration should continue to yield promising innovations. “From a clinical perspective there is a long way to go, as there always is,” says Langer. “But I do feel it is gaining momentum. . . . If we can dissect the unique ways nature has solved all these problems, we still have a lot to learn.”
Daniel Cossins is a freelance science writer living in London, U.K.
- W. Zhao et al., “Bioinspired multivalent DNA network for capture and release of cells,” PNAS, 109:19626-31, 2012.
- C.E. Brubaker et al., “Biological performance of mussel-inspired adhesive in extrahepatic islet transplantation,” Biomaterials, 31:420-27, 2010.
- L.K. Mann et al., “Fetal membrane patch and biomimetic adhesive coacervates as a sealant for fetoscopic defects.” Acta Biomater, 8:2160-65, 2012.
- N. Lang et al., “A blood-resistant surgical glue for minimally invasive repair of vessels and heart defects,” Sci Transl Med, 6:218ra6, 2014.
- W.K. Cho, “Microstructured barbs on the North American porcupine quill enable easy tissue penetration and difficult removal,” PNAS, 109:21289-94, 2012.
- S.Y. Yang et al., “A bio-inspired swellable microneedle adhesive for mechanical interlocking with tissue,” Nature Commun, 4:1702, 2013.
- J.H. Kang et al., “An extracorporeal blood-cleaning device for sepsis therapy,” Nat Med, 20:1211-16, 2014.
- C.M.J. Hu et al., “Erythrocyte membrane-camouflaged polymeric nanoparticles as a biomimetic delivery platform,” PNAS, 108:10980-85, 2011.
- E.M. Pridgen et al., “Transepithelial transport of Fc-targeted nanoparticles by the neonatal Fc receptor for oral delivery,” Sci Transl Med, 5:213ra167, 2013.