What Some Federal Money Buys

Amazing investigations into the life sciences abound under the sponsorship of the Defense Advanced Research Projects Agency (DARPA) in the U.S. Department of Defense. Under program director Alan Rudolph, more than 25 projects around the nation and overseas will receive a total through 2004 of about $84 million. Launched in 1998, the research efforts are divided into three broad categories: Controlled Biological Systems that deal directly with living organisms; Tissue-Based Biosensors that are biological-artificial hybrids; and Biomimetics that reverse-engineer the mechanics of organisms into devices. Successful innovations will be applied to help cope with various threats of chemical and biological warfare, but much of the work derives from previous research aimed at eventual biomedical uses. Following are reports on six of the projects, selected on the basis of their appeal to readers of The Scientist. Liver on a Chip The fit between medical and defense research is seamless for Massachusetts Institute of Technology associate professor of bioengineering and environmental health Linda G. Griffith and her colleagues. In recent years, they have been trying to create "a little organoid in culture." The team has constructed complex, degradable biopolymer scaffolds on which liver cells can survive, aiming at a transplantation therapy for end-stage liver disease patients.1 Their DARPA assignment is to develop a multicellular biosensor in culture that will demonstrate the feasibility of detecting toxins and pathogens, by coupling capillary bed-size tissue units with an optical spectroscopy sensing system. Eventually, the cell lines will be placed in channels on a silicon wafer inside a tiny polycarbonate bioreactor. Griffith explains that interaction between different cell types in culture is necessary for detection of biochemical weapons and other toxins. Similarly, pharmaceutical companies want toxicity tests that involve more complex physical models of human tissue than are currently available. Key cells in their DARPA project are hepatocytes, because the liver is sensitive to viral and chemical toxins. Thus, she says, "This project was meant as a way to enable the building of a transplantable liver." The liver has about a trillion cells in it, and at least a billion cells would be needed for a transplantable liver, she notes. Because artificial scaffolds could not be used for such large-scale assembly in real time, a major goal is to manipulate the cellular self-assembly process. Another challenge is the death after a few weeks of embryoid bodies that develop from cultured tissue. "The intrinsic question is, do they die off because they're limited in biological age, or because they don't have a circulatory system?" The project employs a unique, flow-through microarray that may provide answers. "Vascularity is a huge, huge issue," she asserts. "It's the giant next challenge of tissue engineering." Embryonic stem cells, which not only can differentiate but can be cultivated and increased in vitro, are also central to the project. If feasibility is demonstrated, much of what is learned should be applicable to other organ tissues. "We actually would like to build, eventually, the human body on a chip." A light micrograph of Nostoc commune showing the long filaments of photosynthetic cells and a single differentiated heterocyst (center of filaments--responsible for nitrogen fixation). Robo-Lizard A gecko robot capable of traversing walls and ceilings is the goal of a biomimetic project under way at IS Robotics, headquartered in Somerville, Mass. A sixth prototype now under construction called "Hexa-Gecko" will have six legs, like a cockroach, instead of the rotary tracks or the two legs of its predecessors. Drawing on the locomotion and foot adhesion properties of both animals, each of the robot's legs will have three degrees of freedom of movement, equivalent to the number of its joints. The multijointed conformation and physical effects of its small size will enable the robot to climb obstacles much larger than itself, unlike traditional rovers. The lab of University of California at Berkeley (UCB) integrative biology professor Robert J. Full has studied the climbing dynamics of various organisms to extract design principles for the machine. Kellar Autumn, a biology professor at Lewis and Clark College, discovered how geckos produce the forces they need to run up smooth, vertical surfaces and even upside down. A paper is in press. IS Robotics research director Polly Pook says the gecko's foot adhesion is unique. Its musculoskeletal structure and a blood pressure mechanism enable the five toes on each foot to roll in and out, rather like a New Year's Eve party blower. A peeling action of the toes avoids the large forces that could result from attachment and detachment. "We don't pull down on tape to take it off a wall. We peel it away. This is exactly what the gecko does with its special toes," Pook explains. The attachment mechanism is not well understood, but the toes are composed of millions of hairlike setae with even smaller branches at their ends. Researchers at UCB and Lewis and Clark are trying to define and apply rules, thought to be relatively simple, that may enable microfabrication of artificial structures based on this complex anatomy. Meanwhile, inspired by the cockroach, wet and dry adhesion materials and retractable claws are being developed for Hexa-Gecko, which will have onboard technology enabling it to perform such tasks as search and rescue, reconnaissance, and mine detection. Photos: WEBSITE PICTURES Copyright ©Thomas HorbrandTwo stages of the giant sphinx moth, Manduca sexta: as an adult (above), and as a caterpillar (right). This moth, under study in a multi-institutional DARPA-sponsored program, attempts to exploit its sex pheromone-controlled flight behavior for gathering information about chemical and physical features of an environment. Pheromones and Flight John G. Hildebrand, neurobiology division director in the Arizona Research Laboratories at the University of Arizona (UA), has been studying the giant sphinx moth, Manduca sexta, for more than two decades. He now is coordinating a multidisciplinary and multi-institutional study focusing on the moth, which attempts to exploit its sex pheromone-controlled flight behavior for gathering information about chemical and physical features of an environment. UA associate research scientist Tom Christensen is examining various neurobiological mechanisms that underlie the moth's stereotypical behaviors. "We're trying to use the pheromone system as a window into the olfactory system," he says. That task requires understanding the functions and organization of a structure called the macroglomerular complex in the antennal lobe of the male sphinx moth. This neuronal machinery, which is not shared by the female, is responsible for collecting and processing information about her pheromones. The male brain can process two components of a complex pheromonal blend that are sufficient for attraction to the female.2 Some receptor neurons cannot discriminate between these components, some respond to only one chemical, and some respond to the other one. Still others integrate information from the two pathways to compose a temporal reaction to these intermittent odors. Simulations of odor plumes show that they are "like bubbles of pheromone, or any olfactory stimulus, interspersed with pockets of clean air," Christensen explains. Wind currents, environmental conditions, and possibly even the animal's own flight can interrupt the plume. "We've been studying connections among various neurons," he says, "to discover how they go about encoding this information that varies rapidly in time." A prototype telemetry device has been developed that can be carried on the moth's abdomen, transmitting the nervous signals of its flight behavior to a standard FM radio. The main technical problem currently is to find a lighter-weight battery, he notes. The chemicals of the moth's pheromones are known and can be synthesized and delivered to the relevant neurons, ultimately to influence flight behavior. "We might be able to help the military in sniffing out land mines or chemical weapons." Brains on a Chip University of Southern California (USC) biological sciences professor Michel Baudry works with in vitro systems to study the mechanisms of hippocampal cell death and survival. He says this interest coincided with research by USC biomedical engineering professor Theodore W. Berger to represent small hippocampal network functions by interfacing neuronal tissue with silicon chips that could be implanted in damaged brains. Their complementary DARPA projects involve attempts to create a hybrid neuron- silicon biosensor that could harness the computational capacity of cultured hippocampal circuits. Such a device might not only recognize spatial and temporal patterns, but also detect and assess cognitive impairment produced by exposure to toxins. It could be a "chronic recording and sensing device," says Baudry. "People have been growing neurons on silicon chips for a long time, but in the absence of real function, it's not clear that subtle effects of drugs can be predicted," he adds. Collaborating with them are information and computer science professor Richard H. Granger and psychiatry and human behavior professor Gary Lynch, both at the University of California at Irvine, whose project is to determine whether a hybrid device that includes spatial cells from the hippocampus could respond to processed visual stimuli sufficiently to identify locations. The collaborators have demonstrated that the slices are capable of recording inputs directly from multielectrode arrays, but the stability of these functions or how long the neurons could be maintained remains unclear. Also, "The difficulty has been to identify a class of problem that these circuitries would be good at solving," Baudry explains. Their work now is moving toward use of the cultured hippocampal slices to investigate the dynamics of reverse engineering. Berger has begun developing algorithms drawn from the dynamic properties of hippocampal networks to explore speech recognition. Desiccated, friable crusts of cyanobacterium Nostoc commune (left). Cells can be stored for centuries in a dry but otherwise viable state. The dark coloration reflects the presence of UV radiation-absorbing pigments. Desiccation Row At Virginia Polytechnic Institute and State University, biochemistry professor Malcolm Potts and others are engaged in a program to engineer biopolymers that can withstand extremes of temperature, radiation, and other environmental influences. Their "role models" are extremophiles, in particular the terrestrial cyanobacterium called commune. It can withstand desiccation or complete loss of moisture for many hundreds of years, Potts says, recovering its functions after rehydration. By studying the mechanics of the extremophile at its structural, physiological, and molecular levels, the researchers are discovering how it withstands desiccation and are applying those principles to achieve long-term storage of such labile biological components as platelets and mammalian cell lines. Potential military and medical applications are many. Eventually, microenvironments could be reverse-engineered that would resemble artificial cells, their membranes decorated with compounds to protect them from environmental extremes. They also could be used, and reused, as biosensors. Through expression cloning, the researchers already have discovered a Nostoc commune gene that encodes for sucrose-phosphate synthase, an enzyme responsible for sucrose synthesis by a range of organisms, which helps to withstand desiccation. Expressing the gene in sensitive cell lines, they have achieved marked stabilization in desiccation. A paper is being prepared. Meanwhile, Potts is collaborating with biotech company Life Technologies of Rockville, Md., on transfer technology enabling mammalian cell lines to integrate the engineered gene. "We proved that we can, in fact, stabilize mammalian cell lines," he says. "What we have to do is better understand how various human cells function together." Despite increasing complexity as an organism grows, he believes it may be possible some day to stabilize an entire organism with such technology. Photos: WEBSITE PICTURES Copyright ©Thomas Horbrand Pupa of the Manduca sexta Mind-Operated Machine Monkeys are being trained to respond to visual and tactile cues in preparation for an exciting experiment. Microchips to be implanted in their brains will sample neuronal signals from motor, tactile, and visual regions involved in reacting to the cues they have learned. The monkeys will be taught an arm movement that corresponds to moving a robotic arm placed 100 feet away. This neuronal information will be integrated by the microchips, then telemetered to a data processing unit that will mathematically transform it and send the signals required to move the robotic arm in real time. With cables instead of telemetry, the technology was proven effective in rats, which eventually stopped pressing a lever to control the robotic arm, achieving the same result and earning their reward merely by expressing the brain wave pattern associated with pressing the lever.3 Not only could the technology be employed by soldiers to move robots in potentially dangerous areas, or to monitor animal "scouts," but it could enable paralysis patients to manipulate machines by mind power. "Hopefully, one day, this will not be invasive at all," says Duke University Medical Center associate professor of neurology Miguel A. Nicolelis. "You could put it on the scalp." He adds that the research may reveal how the brain resolves tactile and visual cues that conflict (for example, if a joystick vibration instructs "go left," and a visual cue says "go right"). This will help to explain the different ways sensory input is recorded and output produced. "Can we look at the brain of the animal and predict what solution it will have to this confusion? That's a key question," he notes. At Allegheny University of the Health Sciences, anatomy control professor John K. Chapin and colleagues are developing a new electrode that may be able to collect signals from several hundred neurons, compared to a previous best of about 100. An eventual goal is to reach perhaps 1,000 neurons for a resolution that could enable a variety of directed tasks. Nicolelis notes that technological progress has led to a paradigm shift from traditional studies of single neurons to such network studies. "The only thing comparable to the number of neurons in the brain is the number of stars in the sky. And you can't understand the universe by examining only one star." Steve Bunk (sbunk@uswest.net) is a contributing editor for The Scientist. References 1. S.S. Kim et al., "Survival and function of hepatocytes on a novel three-dimensional synthetic biodegradable polymer scaffold with an intrinsic network of channels," Annals of Surgery, 228:8-13, 1998. 2. T. Heinbockel et al., "Temporal tuning of odor responses in pheromone-responsive projection neurons in the brain of the sphinx moth Manduca sexta," Journal of Comparative Neurology, 409:1-12, June 21, 1999. 3. J.K. Chapin et al., "Real-time control of a robot arm using simultaneously recorded neurons in the motor cortex," Nature Neuroscience, 2:664-70, July 1999.

What Some Federal Money Buys

Liver on a Chip

Interested in reading more?

Become a Member of