Special Delivery
From the exploration of other planets to the meanderings of single cells through our bloodstream and into our tissues, Caltech researchers are thinking about transportation in unexpected ways. They’re using transformative delivery methods to land on Mars, collect data in hard-to-reach locales, and shepherd drugs to the brain, and they’re doing so in order to better be able to ask big questions about the origins of life, to monitor the earth’s emissions and overall health, and even to treat some of the most devastating diseases we encounter.
Dispensing Drugs
Chemical engineer Mark Davis didn’t start his scientific journey looking for new ways to treat cancer. But it’s where his research has taken him in the years since his wife, Mary, was diagnosed with the disease in 1995. She beat it, but only after a long stay in intensive care as a result of the side effects of her chemotherapy. So she turned to her husband and, in essence, said “make this better.”
At first he was taken aback— after all, Davis is an expert in designing catalytic materials that speed up chemical reactions, not drugs that can attack some of the most deadly and difficult-to-treat diseases. But then he did what any self-respecting scientist would do: research. Davis realized that a major part of the solution for fewer side effects from chemotherapy lies in the ability to set the drugs on a particular path to the tumor. “If you can keep cancer drugs away from healthy tissue, the awful side effects that are typically manifested would go away,” he says. The problem is, most cancer drugs are like pharmaceutical carpet bombs. Because the molecules comprising these drugs are so tiny, they can squeeze their way through the wall of a blood vessel and wend their way to healthy tissues where they penetrate and destroy cells.
“They go into bone marrow and kill cells that make up your immune system, into hair follicles and make your hair fall out, and into other organs, causing failure,” says Davis.
Building on his experience with nanomaterials as a result of his catalysis work, he came up with the idea of building a nanoparticle delivery vehicle that would encapsulate the therapeutics and carry them to where they were supposed to go. While small in size, the nanoparticles are bigger than the chemotherapy molecules they are ferrying and also too big to slip out of the bloodstream into healthy tissue.
In theory, Davis designed nanoparticles that should stay in the blood until they reach a tumor and then release their payload in the tumor—thus allowing the drugs to destroy solid tumors, like those of lung and breast cancer, while sparing healthy tissue.
Over time, he and his colleagues settled on nanoparticles made of cyclodextrins, which are a form of sugar. “Cyclodextrins are very biocompatible molecules, with a low toxicity,” Davis says. “So, in humans, they sneak past the immune system.” They soon recognized that 50 nanometers was the sweet spot for size, since the nanoparticles were then big enough to carry many cancer drug molecules yet small enough to travel in the blood and penetrate a tumor.
Davis’s first success was with mice and used a cyclodextrin nanoparticle carrying a well-known chemotherapy drug, camptothecin. Davis and his collaborators tested the drug on various cancers—pancreatic, lung, breast, and more—and found that their delivery vehicle did indeed deliver. Because tumors are always growing new blood vessels, they give the nanoparticle access to the tumor through blood. Once the nanoparticles are inside, chemical sensors within the nanoparticles control the release rate of the delivered payload. By design, the remnants of the disassembled nanoparticle are then flushed harmlessly out of the body in the urine.
Davis’s nanotherapy—developed and tested primarily by Cerulean Pharma Inc., a company that Davis is a consultant for and holds stock in—has now been used in over 10 clinical trials, many of which have been Phase II trials that test for both safety and indications of efficacy. Data from ongoing trials will be used later this year to assess whether the nanoparticle will enter Phase III trials that can be used to enable FDA approval.
“So far the side effects in all these trials have been very low,” Davis says.
In March 2016, Davis and his coworkers reported in the Proceedings of the National Academy of Sciences on a nanoparticle clinical trial, where the nanoparticles were given intravenously to patients with stomach cancer. Biopsies showed evidence that the nanoparticle delivered its drug only to the tumors in the nine patients treated, and not in their adjacent healthy tissue.
“Right now, if a doctor wants to use multiple drugs to treat a cancer, they often can’t do it because the cumulative toxic effects of the drugs would not be tolerated by the patient,” Davis says. “With targeted nanoparticles, you have far fewer side effects, so it is anticipated that a drug combination can be selected based on biology and medicine rather than the limitations of the drugs.”
Next he hopes to surmount one of the biggest challenges in creating new therapeutics—penetrating the blood-brain barrier (BBB) for delivery of drugs to the brain. The BBB is a cellular barrier that controls the entrance and exit of molecules into the brain, e.g., it lets nutrients in but efficiently keeps foreign substances out, including most therapeutic drugs. “This is a huge goal,” says Davis, “not only for treating brain cancers and other diseases of the brain like Parkinson’s and Alzheimer’s diseases, but also because many cancers that start in the liver or breast or elsewhere in the body can metastasize to the brain and become the cause of death.”
Chemical engineer Mark Davis has been able to send experimental nanoparticles through the blood (green layer), transporting them across the blood-brain barrier (white) and into the brain (orange layer) using a mechanism called transcytosis. Next, he will try to pair the nanoparticles with a chemotherapeutic. If successful, it would represent an important breakthrough for treating brain cancers.
Last year, Davis and his coworkers reported in the Proceedings of the National Academy of Sciences a big step toward this goal by borrowing from biology to successfully send 80-nanometer particles across the BBB in a mouse study using a mechanism called transcytosis. Transcytosis is the process by which various macromolecules, including proteins, are transported from the blood, across the cells that make up the BBB, and into the brain. In Davis’s study, nanoparticles containing an ironbinding protein (called transferrin), which naturally crosses the BBB to bring iron into the brain, hijacked the transcytosis process to get past the BBB and into the brain.
The next step, Davis says, is to pack a nanoparticle with both the transferrin protein and a therapeutic; if that, too, can be shown to deliver as expected, he would want to eventually move the technique on to human clinical trials.
While those are Grand Canyonesque goals, Davis thinks the use of nanotherapeutics will someday become commonplace. They could well become a primary delivery system for personalized medicine, he says. “In the ultimate manifestation of the concept, one could envision even prophylactic treatments. For example, you have a family history of a certain disease or you have an X or Y gene that makes you susceptible to something bad. Your doctor will take a finger prick of blood, and you will be given a personalized nanoparticle containing the right drug that will circulate throughout your body, preventing the disorder from ever gaining a foothold. It would have few if any side effects, bypass healthy tissue, and you wouldn’t even think about it. It will be like taking an aspirin.”
Collecting Information
Engineer Mory Gharib is focused on finding a better way to get a very different kind of job done. Instead of transporting drugs throughout the body, his focus is on figuring out the best ways to take the already nearly ubiquitous autonomous drones and turn them from playthings into a workforce that can do the boring, expensive, and/or dangerous jobs that humans shouldn’t be doing, like 24/7 monitoring of pipelines and rapidly delivering alerts of potential leaks; traveling to the way-distant shores of space to collect information; and entering collapsed buildings or damaged nuclear power plants and reporting back on conditions inside.
“The majority of drones and robots out there now are still toys,” says Gharib (PhD ’83), the Hans W. Liepmann Professor of Aeronautics and Bioinspired Engineering and vice provost. “There’s nothing wrong with that, but we are now at the cusp where autonomous drones can have a meaningful impact on humans. The opportunities for drones to enhance science and our lives are numerous.”
Gharib anticipates drones that will do the heavy lifting on construction sites, for example, or monitor farmland to detect a threat from insects and then do localized spraying with the minimum amount of pesticide needed.
He believes they will also improve the delivery of everyday goods while reducing the carbon footprint. He notes, however, that there are fundamental challenges that must be overcome to allow this technology to reach its full potential. That’s why this spring Caltech established—with the help of a generous gift from investor and philanthropist Foster Stanback and his wife, Coco—the Center for Autonomous Systems and Technologies (CAST). Directed by Gharib, the center is poised to attack the multifaceted challenges of autonomous systems by taking advantage of the expertise that cuts across Caltech’s divisions and JPL. Among the baker’s dozen of scientists affiliated with CAST, Beverley McKeon, professor of aeronautics and assistant director of the Graduate Aerospace Laboratories, builds small-scale models of aircraft to study turbulent flow. This is not just to help passengers avoid nausea when their airplane hits an unruly pocket of air but also to design more streamlined drones and thus avoid them being knocked off course when strong winds are blowing.
Richard Murray, the Thomas E. and Doris Everhart Professor of Control and Dynamical Systems and Bioengineering, recognizes that drones and robots have the potential to exceed the capabilities of humans, but—like humans—they will have to be able to adjust to their environment in real time and also be able to self-correct. So Murray and his research group are exploring decision-making, resource allocation, and fault handling in unmanned, autonomous vehicles and mission systems.
Pietro Perona, the Allen E. Puckett Professor of Electrical Engineering, teaches machines how to “see” like humans do. A drone that can not only capture and deliver images but also understand what those images show would be useful to scientists and to itself. Perona is mainly interested in the study of visual categorization of scenes, objects, and behavior a vehicle navigating in complex environments may encounter. For example, an unmanned rescue vehicle in an earthquake-stricken city has to be able to recognize and classify other vehicles, people, signals, equipment, etc. It also has to be able to judge the actions and intentions of humans and animals. Gharib’s own interest in the field is in applying his expertise in smallscale fluid flow and bioinspired fluid dynamics to help design more effective and versatile drones. What has held back the final advance of the necessary software, he says, is a common barrier: money and commercial interest. But that is starting to change, as the work of Lance E. Christensen shows.
A 2003 Caltech doctoral graduate in physical chemistry, Christensen is now a senior atmospheric scientist at JPL. He doesn’t build drones; instead, he uses them to carry his instrumentation to go out and, as he says, “sniff stuff.” He invents tunable laser spectrometers that indeed basically sniff the atmosphere for trace measurements of gases. Such spectrometers measure the abundance of atmospheric gases such as methane, water vapor, and carbon dioxide.
Christensen was part of the team that developed the Tunable Laser Spectrometer (TLS) for the Sample Analysis at Mars (SAM) suite of instruments on the Mars Science Laboratory (MSL) Curiosity rover. TLS investigates the composition of the planet’s atmosphere and compounds extracted from the surface of Mars. These days, Christensen’s work has expanded to include collaboration with private industry; he is partnering with the Pipeline Research Council International (PRCI) and Pacific Gas & Electric (PG&E), a gas and electric utility in Northern California.
“There is a natural relationship between industry and science,” says Christensen. “For example, the amount of methane in our atmosphere has been growing now for the last decade after a five-year pause. Why don’t we know where it’s coming from? How can we not know our Earth’s systems? Is it all these leaky pipes? Is it fracking? These questions cross over with industry.” As the inventor of the Open Path Laser Spectrometer (OPLS), which can measure small natural gas leaks (< 1 standard cubic foot per hour) hundreds of meters downwind, Christensen adapted the instrument for industry to act as safety equipment and, when placed aboard drones, to look for leaks along thousands of miles of natural gas pipelines.
When Christensen was first starting out, he placed his datacollecting instruments on highaltitude balloons; later, he moved on to NASA aircraft for science campaigns such as the Mid-latitude Airborne Cirrus Properties Experiment. Today, he places his instruments on a quadcopter drone that he can hold in his hand; in February, along with colleagues from UC Merced and PRCI, he flight tested the OPLS in order to see how far downwind and how high he could detect methane leaking out of the ground.
“The quadcopter is stable, it doesn’t crash, and it doesn’t get tired,” he says, adding that he’d like to integrate an electrical landing pad into the system. “When a drone runs low on power, it would land on the pad to recharge while another drone lifts off to take its place.”
It’s an ideal solution for industry, particularly those companies that are responsible for working with or maintaining the nation’s aging energy infrastructure. In PG&E’s case, drones would be a much easier and costeffective way to monitor pipelines like those in the Bay Area’s hilly regions.
One last hurdle to widespread use of drones, says Christensen, is the government’s concerns around privacy issues and the safety of other aircraft. For those reasons, the Federal Aviation Administration is moving cautiously on developing regulations for their use.
Christensen understands the agency’s slow pace. “People have mixed feelings about drones,” he says. “And with the rise of miniaturization and the growing capability of this technology, it gives a lot of us pause for thought, sometimes keeping us up at night. “But if people can get rid get rid of their preconceived notions of drones, their utility could be endless,” says Christensen. “Think about having tiny drones floating just above the tree line, monitoring leaks from transmission lines. That’s something humans will never be able to do. The public might go for that.”
Transporting Tools
Geochemist Ken Farley doesn’t have to worry about public approval. He’s the project scientist for Mars 2020, the new rover mission, which—if the popularity of the Curiosity rover is any indication—will draw the excited and curious eyes of citizens around the globe.
But what Farley does have to worry about is the mission’s overall scientific success. He also has the obligation to meet the “very hard” launch date of 2020—when Mars and Earth are closest in orbit to each other. And he’s the guy who must help define the science goals for the mission and determine how to pack an assembly of all-new scientific instruments onto an existing rover.
This illustration shows a prototype for hardware to cache samples of cores (left) drilled from martian rocks for possible future return to Earth. A major objective for NASA’s Mars 2020 rover, as described by Caltech’s Ken Farley, would be to collect and package a carefully selected set of up to 35 samples in a cache that could be returned to Earth by a later mission.
That rover will be a souped-up version of Curiosity, which still putters along to this day. That’s good, says Farley, in that the scientists know they have a proven and reliable platform. “The fundamental design is in place,” he says “but our challenge is that there’s a whole assembly of new instruments we have to cram onboard.”
The overarching mission of the 2020 rover, per NASA, is to “seek the signs of life.” Specifically, it has four main objectives, one of which is to prepare the way for human exploration of Mars. To that end, the rover will include a weather station to help scientists better understand the martian atmosphere and an in situ resource utilization (ISRU) instrument, which will be tested for its ability to convert atmospheric carbon dioxide into oxygen, both for future human consumption and for future propellant.
Each Mars mission builds on previous successes and, like the MSL, which launched in 2011, the 2020 rover will perform an extensive exploration of its landing site to understand the geological processes that helped form the surface of the planet. While the Curiosity rover is seeking (and finding!) evidence of habitable conditions, the 2020 mission will seek actual biosignatures— physical structures or molecules that show evidence of past or present life— in the rocks on Mars.
The subject of whether or not there was once life on Mars is certainly a fascinating one, but Farley believes that important questions will arise even if that search fails. “If we bring all the tools to bear on such an environment and don’t find signs of life, what does that mean?” he says. “If that’s the case, what, then, was the ‘spark’ that jumpstarted life on our planet?” He notes that some researchers now believe life as we know it on Earth actually originated on Mars.
The biggest challenge of the mission, however, is likely to be the collection and preparation of returnable geologic samples for possible delivery back to Earth by a future mission. The word “returnable,” Farley says, has a technical definition—the cache has to meet a series of criteria, and one is that it has to have enough scientific merit to be worth the expense of bringing it back.
“There is some number of samples, probably between 20 and 35, that would make that worth doing,” says Farley. “If it’s less than that, it may not be worth bringing back. So, in some sense, we have a gun to our heads to collect a large number of samples.”
That said, Farley is prepared to be patient. “We’ve learned from Curiosity that everything takes a long time,” he explains. “Driving and drilling takes a long time. That’s motivated a lot of the discussion of landing sites. You’ve got to have targets you wish to drill that are close together, and they can’t be a long drive from where you land. But there also has to be diversity because you don’t want 15 copies of the same sample.” Plus, the Mars 2020 rover has a “warranty” of roughly only a couple of years, Farley says, so it’s critical for the rover to be able to collect its samples in that time. He yearns for the speedy rover used by Matt Damon in the movie The Martian. “I wish,” he laughs. “We tell the rover to go here, it moves 50 meters, very slowly. Autonomy, and autonomous driving, is a challenge.” In 2017, Farley and the other mission scientists will decide exactly where the rover will land. He notes that the proposed sites break roughly into two environments: crater lakes with deltas and hydrothermal sites. “They are the most likely to have ancient life in them and to have preserved the evidence of it,” he notes. “But even if we don’t find signs of life on Mars, we are likely to bring back rock samples that will have the prebiotic soup,” says Farley. “From those we’ll be able to ask: What were the chemical building blocks? It’s the question of the origin of life, and I find that very exciting.”
Mark Davis is the Warren and Katharine Schlinger Professor of Chemical Engineering. His work on nanoparticles is funded by the National Cancer Institute.
Mory Gharib is the Hans W. Liepmann Professor of Aeronautics and Bioinspired Engineering, director of Caltech’s Graduate Aerospace Laboratories, and vice provost.
Lance E. Christensen is a senior atmospheric scientist at JPL. His work is supported by the Pipeline Research Council International.
Ken Farley is the W. M. Keck Foundation Professor of Geochemistry. The Mars 2020 mission is funded by NASA.